Why Government Fails So Often: And How It Can Do Better (63 page)

Better information can be especially valuable to low-income people who are more likely to be isolated from the institutions and informal grapevines that serve others so well. For example, economists Caroline Hoxby and Sarah Turner show that many high-achieving but low-income students who would not otherwise apply to high-ranked colleges and universities that grant substantial financial aid and have high graduation rates can be induced to apply and often gain acceptance
via a simple program that mails information packets and follow-up materials to them
at a cost of only six dollars per student
.
⁶⁸

Other information-based techniques, increasingly common in the private sector, might be used by government agencies to improve their efficiency. An all-too-rare example of this is a novel use of crowdsourcing by the Defense Advanced Research Projects Agency to improve the design of military equipment.
⁶⁹

In 1990, Congress enacted the Federal Credit Reform Act, which requires agencies to use a method for accounting for the costs of federal loan programs that systematically misrepresents those costs, invariably by
underestimating
them. This method misinforms the public and officials about the true cost of government loans by calculating their net future cash flows to the government from the loans and then discounting that amount to present value for the year in which the loan is disbursed. The discount rate assumes, however, that the risk of default by private borrowers is the same as the vanishingly small risk of default by the U.S. Treasury—an absurdity that the Congressional Budget Office (CBO), an arm of Congress, urges replacement with a “fair-value accounting” method that would reflect the true private default risks.
⁷⁰
This absurdity enabled senator Elizabeth Warren and some colleagues to erroneously claim, with straight faces, that the student loan program is highly profitable—despite the huge default losses described in
chapter 8
.
⁷¹
More generally, it suggests, incorrectly, that government can extend credit more cheaply than private lenders can. Congress should implement the CBO’s recommendation for all federal loan programs.

RIGIDITY

As discussed in
chapters 3
and
4
, our policy-making system and political culture create many veto points, cementing the status quo bias described in
chapter 5
and elsewhere in this book. Indeed, these veto points make it even harder to change a policy once it is established than to institute it in the first place because the inertial political forces that normally resist change are now augmented by the interests that
have come to benefit from the current policy and do not want it changed, regardless of the need for reform.

Nevertheless, this bias, while powerful, is neither wholly implacable nor irremediable. Writing in 1997, I called attention to “how many complex, controversial issues Congress has managed to address and resolve (usually for better, sometimes for worse) in the last two decades even as special interest groups—much to the consternation of most commentators—proliferated and became more politically engaged. The period since the early 1970s has been one of truly remarkable policy innovation at all levels of government,” in contrast to the dire predictions of Mancur Olson and others about what Jonathan Rauch called
Demosclerosis
in a 1994 book by that title. I offered many examples of sound public policy changes in the past that, with a few exceptions (see just below), remain largely in place: “the substantial, if not total, economic deregulation of basic industries such as telecommunications, financial services, energy, trucking, railroads, buses, and airlines; renovation of the welfare and social security programs entrenched since the New Deal; reform of the federal budgeting and inter-governmental systems; a dramatic reduction in the military’s share of the national economy and budget; a vast expansion of the national park and wilderness systems; and an overhaul of the major tax, immigration, housing, agricultural, transportation, campaign finance, environmental, intellectual property, and civil rights laws.”
⁷²
The main exceptions are agricultural subsidy reform (repealed), budgeting reform (mostly ignored in practice), and federal mandate reform (also ignored). Some, like tax reform, need much updating.

In that 1997 article, I proposed some partial remedies for the policy system’s status quo bias; I repeat them here in abbreviated fashion.

Congress and the executive branch should require proponents of subsidies and other expenditures to propose offsetting reductions in other areas, and should insist that lobbyists do so as well. This helps to place special interests in a zero-sum game where any costs that they add must come at the expense of other special interests, and it
contrasts with the logrolling incentives (discussed in
chapter 5
) that so often lead to inefficient and inequitable policies. Such a rule, called “pay as you go” or “paygo,” was enacted in 1990 with bipartisan support, but it expired in 2002; it was widely considered a success until a budget surplus reduced its bite. Paygo was reinstated in 2002 by congressional rule, not statute; it was mostly ignored, and no sequester (the enforcement mechanism) was ever invoked. In 2010, a bitterly divided Congress agreed to a statutory paygo system,
⁷³
but the enforcement sequester is still routinely waived—partly because budget legislation in recent years has been enacted in an omnibus form as part of a “grand bargain” rather than in more discrete pieces. Until this changes, the paygo principle can perhaps best be vindicated through a combination of more pointed CBO disclosure of paygo violations and efforts to mobilize public awareness of the violations.

If paygo rules increasingly constrain special interests, however, the pressures generated by these groups do not disappear. Instead, they are simply channeled into regulations, private-sector mandates, special trust funds, borrowing authorities, and other off-budget areas of governmental activity where these constraints do not now apply, or are weaker. By extending the paygo approach to these off-budget areas, policy makers can further reduce special-interest abuses while still exploiting the informational resources that those groups bring to the policy process.

Paygo only affects
new
subsidies, but the status quo bias is of course most strongly reflected in those already on the books. Policy makers should review and rationalize the current programs that subsidize the activities of private business interests in the name of such attractive goals as regional development, job creation, export promotion, price stabilization, and national defense. Many of these “corporate welfare” programs, which take multiple forms, are unfair to their competitors, costly to taxpayers, and ineffective. In 2011, for example, the Export-Import Bank gave almost half its loan guarantees and other subsidies to those unfortunates Boeing and General Electric.
⁷⁴
Rigorous CBA would likely expose many of these as policy failures.

Government can seek to increase the accountability of interest groups both to their members and to the public by requiring the groups to disclose information about their transactions that the public has a right to know. Even where disclosure is already required, as with federal election campaign contributions, much improvement is possible, particularly through the Internet and other “open government” mechanisms discussed above. Disclosure is often (not always) a better way to protect public values than imposing substantive restrictions on the groups’ activities.

Sunset provisions for federal programs cannot assure that policy makers will bring a fresh perspective to old programs and ask the right questions about their continuing effectiveness, but they can help. One item on the legislative checklist proposed above should be a sunset date. The prospect that a program will face a sunset review can be a spur to better performance and needed reforms, but it does entail several disadvantages. First, it is common for Congress to simply extend or reauthorize programs without conducting a genuine substantive review. At the time of this writing (September 2013), for example, the No Child Left Behind Act of 2001 remains in this posture. Second, sunset reconsideration of existing policies may magnify the problem of government credibility, to which I now turn.

CREDIBILITY

I noted in
chapter 6
that the democratic accountability of government necessitates that it change its policies in response to elections, other strong expressions of public opinion, and perceptions of policy need—but that this very uncertainty deters those whose cooperation is essential for policy success. Private actors and lower levels of government will enter an opaque, labile policy environment only reluctantly and at a higher price (in some form) reflecting the risk that Washington will change signals in ways that harm their interests. This problem impedes many otherwise mutually desirable arrangements. Earlier chapters presented a number of examples, most vividly in the
Oakland Project (see
chapter 8
) and most currently in the refusal of many states to expand their Medicaid programs under the Affordable Care Act for fear that once the three-year guarantee of full federal funding ends, the states will be locked into much higher costs. (Wags liken this prospect to “the gift of a baby elephant.”) This problem is both constitutional and tactical in nature: no Congress or president may bind a subsequent one, which is always free, within its constitutional authority, to change the rules, and even if the president could somehow bind himself, he cannot as a political matter bind Congress—and vice versa.

As explained in
chapter 6
, there are no good solutions to this endemic problem.
⁷⁵
The government can enter into contracts, issue performance bonds, and take on other legally enforceable obligations, but this will not affect the important regulatory and tax policies that Congress often changes, thus engendering great uncertainty costs. Regulatory agencies, bound by Congress’s policies, are even more constrained in changing direction abruptly, even when the courts uphold them.
⁷⁶

That said, Washington can encourage private actors and other levels of government to rely on and cooperate with it by reducing the risks of doing business with it even though it might later change its mind. Law-and-economics scholars Michael Abramowicz and Ian Ayres have explored how such credibility enhancements might be designed, most notably with “commitment bonds.”
⁷⁷

IMPLEMENTATION

As explained in
chapter 8
, many obstacles impede federal policy implementation. In earlier work, I focused on the special, additional difficulties that federal judges face in attempting to implement their complex institutional reform decrees (sometimes called “structural injunctions”) in prisons, schools, housing projects, police departments, and other agencies. I proposed that they should not issue such orders before conducting a “decree implementation analysis” through public
hearings and other techniques to help them predict what different requirements might actually entail in the real bureaucratic world that they hope to reform, and to fashion their decree in light of that better understanding.
⁷⁸
A similar practice for legislative and executive policy makers before they make difficult policy decisions would be salutary; any delay produced by such consultation and analysis would likely be compensated for by better-designed, more effective policies. Policy implementation planners should recruit skeptics to perform the challenging function that opposing counsel provides in structural injunction proceedings. Manufacturers deciding whether to invest in a costly new product line would surely do no less, yet the obstacles to successfully implementing complex public policies are just as predictable and opaque as those in markets; in fact, they are probably more so.

As
chapter 7
explained, many federal laws are enforced weakly, or not at all, and improper payments and fraud (which are not the same thing) are rife. One straightforward remedy is to hire more inspectors, auditors, and other enforcement personnel in situations—tax collection and child support enforcement, for example—where the delinquencies are enormous and additional enforcement personnel pay for themselves many times over. (The Treasury’s Financial Management Service estimates that a dollar spent in federal debt collection in 2011 produced $52.42 in recoveries.
⁷⁹
) Politics, not cost-effectiveness reasons, account for Congress’s penny-wise, pound-foolish refusal to expand enforcement personnel in such areas.
⁸⁰
Even more basic, the long-vacant inspector general positions in many agencies (see
chapter 10
) should be filled and generously funded.

Whistle-blowers privy to illegal conduct may be a valuable adjunct to agency enforcement, especially for regulators that supervise large organizations whose compliance levels are opaque to their regulators. Law professor David Engstrom has systematically analyzed a variety of whistle-blower approaches:
qui tam
lawsuits under the False Claims Act, a recently revised tax bounty program, a bounty scheme under the Dodd-Frank financial regulation overhaul, and the cash-for-information programs of the Securities and Exchange Commission
and the Drug Enforcement Administration, among others. Today, such efforts are largely limited to remedying procurement, tax, and securities fraud, but Engstrom notes the relevant trade-offs in possibly extending them to other policy areas.
⁸¹