Why Government Fails So Often: And How It Can Do Better (61 page)

Another moral hazard reform, here directed at the Federal Deposit Insurance Corporation, might use the stress testing process mandated by the Dodd-Frank law to provide a more reliable, refined basis for differentiating premiums according to risk. Indeed, targeted publication of that risk information could enable private insurers to bid for the deposit insurance business.

Reforming the system of mortgage guarantees dominated by Fannie Mae and Freddie Mac to reduce moral hazard is imperative, as Congress recognized in putting it into conservatorship. This system, discussed in
chapters 5
and
8
, encouraged a housing bubble and overleveraging by homeowners and financial institutions that produced a financial catastrophe in 2008 whose immense effects will continue to be felt for years to come. Experts disagree about precisely how to allocate responsibility for this calamity and therefore about the best remedies. No one doubts, however, that a primary cause was the agencies’ promotion of subprime lending and overleveraging in the secondary mortgage market and by homeowners, and that these behaviors were spurred and abetted by the implicit government guarantee, private securities rating agencies, and supervising congressional committees.
²¹
At a minimum, we should drop the pretense that these agencies are private entities and should include their operations on the government’s balance sheet.

Plausible proposals have been advanced to eliminate the government’s exposure to losses in this important market. As the study by Dwight Jaffee and John Quigley discussed in
chapter 5
shows, the policy rationales for Fannie and Freddie’s housing programs are not convincing, and the western European experience, especially in Denmark, confirms this.
²²
If the government must continue to play a role, these economists argue, the appropriate model is not the government-sponsored enterprise but the Federal Housing Administration (FHA), which for almost eighty years underwrote loans to low-income home buyers on an actuarial basis without government subsidy.
²³
Unfortunately, as
chapter 8
explained, the FHA recently abandoned its successful traditional model in the wake of the Fannie/Freddie meltdowns, with costly results for taxpayers.

The Housing Commission of the Bipartisan Policy Center has offered a different proposal. Cofounded by former Senate majority leaders of both parties with members representing a broad range of interests and ideologies, the commission also represents the interests of Fannie Mae and Freddie Mac and of groups that received substantial
funding from them.
²⁴
It claims that its plan would sharply reduce the government’s exposure to home mortgage losses by (1) phasing out Fannie and Freddie; (2) relying almost entirely on private capital to finance home purchases; (3) using the government only as a “public guarantor” exposed only after three layers of private capital subject to stricter capital requirements have paid; (4) requiring safer, more transparent servicing and securitization processes; and (5) returning the FHA to its traditional role: assisting first-time home buyers with down payment resources.
²⁵
Critics of this proposal worry that the “public guarantor” will still have poor incentives, remain vulnerable to fraud, and improperly price risks.
²⁶
After all, many of these same regulators, funders, and agencies utterly failed to foresee, much less prevent, the last crisis, despite some experts’ dire warnings.
²⁷
Other regulatory proposals are being developed, but all involve significant risk to taxpayers
²⁸
—which makes Jaffee and Quigley’s privatization approach for this market a promising way to avoid another huge bailout. Policy makers should resist the siren song of those who plump for new subsidies by assuming that they will be more successful than existing ones are.
²⁹

Reducing moral hazard on the part of government agencies and program beneficiaries is only one way to improve incentives. Another is to strengthen the long-standing efforts at “management by objective” (MBO), sometimes called “performance management.” This technique has long been promoted by reform-minded officials and academics.
³⁰
Reaching back at least to president Lyndon Johnson’s Planning-Programming-Budgeting System, it requires each program to specify and quantify its goals in advance insofar as their nature permits.
³¹
Properly implemented, it encourages officials to “buy in” to the pursuit of certain outcomes whose attainment can be measured, and then holds them accountable for the results. It also tends to discipline their predictions by introducing sanctions against tactical over-promising. (Hopefully, incentives not to
under
promise already operate.) Some goals cannot be quantified, of course, but good managers can often devise imperfect but serviceable proxies for them. Once
policy makers are confident that MBO is being properly implemented at the program level, they can fashion financial and other incentives to reward success. (Congressional authorization for such changes to the bureaucratic compensation system is probably required, as in the 1978 civil service reform law.) President Bill Clinton’s much-touted National Performance Review, led by vice president Al Gore, included many interesting ideas and some significant improvements.
³²

In reality, however, such reforms have been long on promise and short on achievement. According to Jack Donahue, who advocates them, these efforts have had little effect: “The transformation of America’s public sector to date is both limited and, perhaps more importantly, distorted. Some eminently sensible changes remain stubbornly stalled. Some second-order, silly, or questionable reforms have outpaced the fundamentals.”
³³

Almost all government programs designed to provide services to the public do so by funding providers in the expectation that they will render the desired service to the program beneficiaries. Because providers tend to be far better organized than the consumer/beneficiaries, however, this arrangement disproportionately serves the former’s interests. Once they have the funds, their incentive is to use them with only minimal oversight by the agency and with little or no accountability to the consumers whom they are supposed to be serving—other than whatever indirect effect consumer feedback may have on the politicians who authorize and fund these programs. Congress should instead direct that federally funded services be provided, insofar as possible, in the form of consumer vouchers subject to appropriate conditions to protect policy goals. Congress has done this with Section 8 housing and with food stamps, and many states have done so for private schooling. This form of program delivery, which Milton Friedman proposed more than fifty years ago,
³⁴
is certainly no panacea for the problems facing the poor: for example, many landlords will not accept Section 8 vouchers, quality control of charter schools by parents is difficult, and food stamps may be used to buy unhealthy meals. But it does allow consumers to make their own trade-offs, increases providers’ incentives to perform effectively and responsively,
and informs policy makers about how well public funds are being used.
³⁵
The logic of this approach extends to all publicly funded services, not just those for the poor. Many of the Affordable Care Act’s problems might have been avoided by providing coverage in the form of income-and wealth-tested vouchers rather than through the act’s highly complex, jerry-built system. Many private employers already use choice vouchers to fund health care for their employees, and will likely do so even more under the act.
³⁶
Indeed, Congress should make vouchers, regulated to prevent fraud or consumer abuse, the default form of federally provided access to noncash benefits.

Policy makers should attend more to the architecture of individual incentives and choices—particularly their status quo and other cognitive biases—and then take this architecture into account in designing specific programs. I discuss below the difficulties of this approach, associated with scholars Richard Thaler and Cass Sunstein,
³⁷
(under “Limits of Law”) in connection with simplification strategies and “nudges.”

Competitive incentives can improve government performance by rewarding effectiveness and punishing failure. In
chapter 4
, I discussed the Tiebout effect, under which communities vie with one another in providing different combinations of services and taxes in order to attract residents who “vote with their feet” and hence a greater tax base. States and localities, however, also compete for financial capital, talent, infrastructure, and successful programs, as in the public school systems. Although this competition already exists, federal policy makers might heighten and exploit it. An encouraging (but allegedly politicized) example is the Obama administration’s Race to the Top program in public education; a counterexample is its stifling of voucher-based school choice in its own backyard testing ground, the District of Columbia. Another approach, of ancient lineage, is for the government to incentivize not only scientific discoveries (which is common) but also new, demonstrably effective techniques of policy design, assessment, performance, and enforcement. A recent instance is the Federal Trade Commission’s prize for the best software to block illegal robocalls.
³⁸

A more radical move toward competition would be for Congress or an executive order to require each agency—before opting to provide a new service itself—to analyze and publish its findings as to whether that service can be provided as well or better and at the same or lower cost (all things/costs considered) by privatizing it, much as agencies have been required to reduce paperwork, conduct cost-benefit analyses (CBA; see
chapter 2
), and consider effects on small businesses. The agency should remain free to provide the service directly, but only after explaining publicly why it is doing so in light of this analysis. This implies, however, that the agency’s capability to manage contracts with private providers must be strengthened so that the federal responsibility is preserved, not abdicated.

Congress, even more than the agencies, should also systematically reconsider whether and to what extent regulated activities should instead be subjected, under certain conditions, to competitive bidding (Medicare is especially ripe for this),
³⁹
deregulated (e.g., sugar and other crop subsidies), or privatized (e.g., the US Postal Service and Amtrak). Whether Congress would heed this recommendation, of course, is an entirely separate question—and doubtful, judging from its shackling of the Postal Service. Most tariffs and other trade restrictions tax consumers for the benefit of inefficient industries; they too should be repealed. But again, designing each of these reforms requires detailed study beyond the scope of this book. And as with performance management, a reality check is warranted: after a strong outsourcing push by the administration of president George W. Bush, the movement of federal nondefense jobs to the private sector was relatively trivial, perhaps one in a thousand.
⁴⁰

Short of deregulation, privatization, and outsourcing, policy experts have advanced a host of recommendations for legal changes that would likely improve governmental effectiveness across a broad range of policy domains. For example, almost all economists favor using taxes to reduce a broad range of harmful activities rather than relying on other forms of centralized regulation. Such Pigouvian taxes would make almost everyone better off and could be designed to be
revenue-neutral or revenue-raising, depending on concerns about their distributive impacts.
⁴¹

Government-sponsored innovation is shadowed by the Solyndra fiasco and other bad bets on emerging technologies (discussed in
chapter 8
), but numerous important counterexamples exist. Mariana Mazzucato’s 2013 book
The Entrepreneurial State
heralds a new appreciation for the role that the government has played and might (if carefully implemented) play in the future in spurring socially transformative technologies.
⁴²
A recent Kauffman Foundation study proposes wider dissemination of scientific research; growth-oriented tax reforms; improved financial regulation and reporting; enhanced efficiency and fairness of public and private law rules and processes; and legal reforms to protect identity and privacy and refine intellectual property rights.
⁴³
Another volume presents novel approaches to health and safety regulation;
⁴⁴
yet another study analyzes various forms of “collaborative governance” between the government and private providers (both regulated and unregulated, both for-profit and nonprofit) in a variety of policy areas (not all federal) that can often exploit private advantages in information, legitimacy, productivity, and access to financial resources.
⁴⁵
Careful studies of successful, marketable innovations are sure to yeild many other lessons for government policy.

IRRATIONALITY

Earlier chapters, especially
chapter 5
, identified many powerful forces that combine to produce policies that, while rational for the often concentrated, well-organized interests that stand to gain from them, tend to be irrational from the perspective of the diffuse public that must usually bear the costs of policy failure.