[Economic Report of the President (1996)]
[Administration of William J. Clinton]
[Online through the Government Publishing Office, www.gpo.gov]

[DOCID: f:erp_c4._]
Economic Report of the President - - - - - - - - - - - - H. Doc. 104-161
[From the online service of the U.S. Government Printing Office]
[wais.access.gpo.gov]


CHAPTER 4

Devolution

THE APPROPRIATE ROLE of the Federal Government in the U.S. economy
has been a fundamental issue in this past year's debate over the budget.
At issue are both the role for government in general and the division of
responsibility between Federal and State governments. Chapter 1 of this
Report addressed the question of the broader role of government. This
chapter addresses how responsibilities might best be apportioned among
the levels of government.
This Administration has dedicated itself from the outset to making
government work better. Improving the efficiency of government requires
a rational division of responsibility among Federal, State, and local
entities. Today many support the notion that, in several policy areas,
authority ought to be devolved from Federal agencies to States,
localities, and individuals, to foster more creative and responsive
solutions to the problems of diverse communities.
This Administration has been at the forefront of efforts to empower
State and local governments by removing impediments to innovation and
experimentation in public health, welfare, public housing, and
environmental protection, and by reducing the proliferation of Federal
unfunded mandates. But devolution of responsibilities must be done
carefully, to ensure that national objectives are still met. In
particular, a profound national interest lies in maintaining a social
safety net, to guarantee at least a minimum standard of living for
today's vulnerable families, and in promoting investment in education,
research, and infrastructure, to ensure high living standards for
tomorrow's families. The Federal Government also has a clear interest in
ensuring that all of its expenditures, including those over which States
and localities have some degree of control, are spent in the manner
intended. Devolution that merely inserts an extra level of bureaucracy
makes little sense: in many cases it is far better to empower
individuals directly than to dispense funds to State and local
governments on their behalf.

FACTS ON FEDERALISM

Despite major changes in our economy and in government programs over
the past 25 years, the roles of the States and the Federal Government in
the economy have remained relatively stable.

TRENDS OVER TIME

Total government expenditures--Federal, State, and local--have rose
slowly as a share of gross domestic product (GDP) over the past three
decades, from roughly 28 percent in the early 1960s to over 34 percent
today (Chart 4-1).



The Federal Government accounts for the largest share of this
spending. In 1993, if expenditures on State and local grants are
included, the Federal Government accounted for 69 percent of total
government spending. As Chart 4-2 shows, this share has not changed
dramatically over the past 25 years: the Federal share of expenditures
rose from 67 percent to 72 percent between 1970 and 1984, but has shrunk
back to 69 percent since then.

COMPOSITION OF SPENDING

The Federal Government's major responsibilities include national
defense, Social Security, and Medicare. States and localities have



primary responsibility for public education, police and fire protection,
and sewerage and sanitation. Highways are generally maintained by States
and localities, but funds for new construction are largely provided by
the Federal Government. Medicaid and some welfare programs are jointly
financed by the Federal and State governments but administered by the
States. Table 4-1 documents the current division of responsibility
between the Federal Government and State and local government.
This division of responsibility has evolved gradually. Public roads
and support for the needy, for example, are two areas where
responsibility has traditionally rested with States and localities, but
in both areas the Federal Government has assumed an increasingly
important role. The Highway Revenue Act of 1956 created the Highway
Trust Fund and dedicated the revenue received from taxes on diesel fuel
and gasoline to this fund. These funds were used to build the interstate
highway system, which has changed the face of America.
The growth of the Federal role in welfare arose in part out of the
widely shared view that all children, no matter where they were born or
who their parents were, should be entitled to certain basic standards of
nutrition, housing, and health--common decency in a country as rich as
the United States demanded no less. Although the acceptance of this
national obligation was fundamentally a

Table 4-1.--Composition of Government Spending, 1993
------------------------------------------------------------------------
Percent of
Percent of  non-    expenditures
Spending by function              interest        financed with
expenditures     Federal grants
------------------------------------------------------------------------
Federal Government:

National defense..................              26.6  ................
Social security...................              23.4  ................
Medicare..........................              13.2  ................
Veterans benefits and services,
welfare and social services, and
housing subsidies................               9.0  ................
Civilian and military retirement..               4.9  ................
Other.............................              22.9  ................

State and local government:

Education.........................              37.5               4.7
Medicaid..........................              15.9              57.9
Welfare and social services.......               8.0              58.0
Highways..........................               7.5              26.1
Police and fire protection........               6.2                .8
Corrections.......................               3.7                .7
Water, sewerage, and sanitation...               1.5              15.5
Other.............................              19.6              19.7
------------------------------------------------------------------------
Note.--Data are on a national income and product accounts (NIPA) basis,
and are as published in the Survey of Curent Business, September 1994.
No later data are available.
In this table, Federal grants-in-aid to State and local governments are
not included in Federal Government expenditures.

Source: Department of Commerce.

moral decision, it was supported by self-interest, in the recognition
that the cost to society of not providing these minimal standards--in
terms of lost wages, higher crime rates, and the like--could be very
high.

THE RATIONALE FOR A FEDERAL ROLE

The reasons for the creation and expansion of these Federal programs
provide considerable insight into the forces that drive the expanded
Federal presence in American society. Yet a sensible allocation of
responsibilities for governments in the future must be based on more
than historical precedent.
Some might make the case that the Federal Government should do
nothing other than national defense. After all, States and localities
are better able to tailor their programs to meet the different needs and
preferences of their residents, and competition among the States may
enhance efficiency and innovation, just as it does in the private
sector. But this view ignores the benefits of Federal action in a number
of areas. The enumeration of powers given to the Federal Government
under the Constitution suggests that our forefathers, even in the early
infancy of the Republic, recognized the advantages of Federal
involvement across a broad range of endeavors. The economic strength of
the United States rests in part on our vast national market, fostered
not only by the free flow of commerce without artificial trade barriers,
but also by national standards and a national transportation and
communications system.
Economists have sought to identify some general principles that
would elucidate a ``rational'' division of responsibilities between
levels of government. At least four categories of arguments justifying
Federal action can be identified.

THE NEED FOR UNIFORMITY

Although diversity among State government programs is often
desirable, in some cases the benefits of uniform government action
across the States tip the scale toward Federal involvement. Uniformity
of standards and regulations may improve efficiency. For example,
uniform rules for interstate commerce preserve one of America's
strengths: our large national market. Conflicting State regulation could
fragment this market and impede producers' ability to take advantage of
economies of scale. Likewise, uniformity in minimum safety net benefits
would guarantee that all needy Americans, regardless of where they
lived, enjoyed at least a certain level of well-being, and would avoid
distorting and inefficient movements of households in response to
differences in benefits.

DIRECT SPILLOVERS

Actions taken or not taken by States sometimes affect residents of
other States. Residents of a State might be willing to tolerate
pollution of their ground water, but the contaminated water could seep
across State boundaries and harm residents of other States. States may
also engage in activities that unintentionally benefit the residents of
other States. For example, one State's successful innovation in its
schools can lead the way for other States to reform their education
systems, and States' efforts to prevent communicable diseases can
benefit the health of nonresidents. Similarly, when States invest more
in education, and incomes rise as a consequence, they confer a positive
benefit on all taxpayers: the Federal Government reaps some of the
rewards of the higher incomes in the form of higher Federal tax
revenues. When the policies of one State affect the residents of others,
for good or for ill, States may lack the right incentives to provide an
appropriate level of public services, because the effects of policies on
nonresidents may not factor strongly in their decisionmaking.

THE EFFECTS OF POLICY-INDUCED MOBILITY

The freedom of people and firms to move at will from State to State
promotes competition among State governments. Although this competition
can enhance the efficiency of government, it can also make it difficult
for States to pursue certain worthwhile policies. For example, the fear
of welfare-induced migration may cause States to reduce welfare benefits
to a level below what they would otherwise provide. Similarly, State
competition for jobs may limit the generosity of unemployment insurance
programs.

INEQUALITY OF RESOURCES

States that are poorer than the average, or that are experiencing
temporary downturns, are able to raise less revenue, yet have to spend
more than other States to provide services for the needy. Clearly, only
the Federal Government can transfer resources among the States. Not only
does such redistribution help poorer States, but financial assistance
from the Federal Government that increases during economic downturns can
also help to stabilize regional economies. This assistance can be given
through a number of channels: direct transfers of cash or in-kind
benefits to lower income individuals, grants to lower income States or
localities, matching grants to State or local programs for the needy, or
direct provision of public services in poor communities. The role of the
Federal Government in transferring resources to States and localities is
more complicated, both in theory and in practice, than is often
recognized, and will be discussed at greater length below.
These rationales for a Federal role are not mutually exclusive, and
sometimes it is their interaction that makes a strong case for a Federal
role in policy. For example, national safety standards, when desirable,
might evolve on their own, were it not for spillovers. States could
simply agree to a set of voluntary standards, and each State would weigh
the benefits and costs of complying. In doing so, however, it would
ignore the costs it might impose on others. A State might adopt more lax
safety regulation for its cars, but then when its cars cross over into
another State, the other State bears part of the costs. Federal action
is therefore needed to ensure uniform national standards that avoid
these spillover effects.

DEVOLUTION OF POLICYMAKING RESPONSIBILITY

Determining which level of government should be responsible for a
particular program or activity is a delicate balancing act. It requires
weighing the benefits of innovation, greater responsiveness, and
competition that State and local control offers against the rationales
for Federal involvement just outlined. Sometimes the answer is simple
and obvious: either purely Federal control and financing or purely State
control and financing. But many cases call for a sharing of
responsibilities.
All government activities have three basic elements: policymaking,
financing, and administration. These activities can be apportioned
between the Federal Government and State and local governments in
various ways. The current debate centers largely on how the policymaking
role for programs that receive financing from the Federal government
should be shared. At one extreme, the Federal Government could provide
funds to States with no strings attached--States would not even be told
on which programs to spend the money. Such an arrangement, used in other
countries and in the past in the United States (where it was called
``general revenue sharing''), is not currently under consideration.
Instead the debate has focused on whether to convert existing programs
into block grants (grants that can be used to fund programs in broad
policy areas) and on how much discretion to allow States in determining
how those grants should be used.
This Administration strongly supports enhancing the role of States
and localities in policymaking. In many areas--job training, community
development, and welfare, for example--enhanced flexibility for States
and local communities is likely to yield better results. But it is
important that this enhanced flexibility be provided in a way that
protects the national interest. For all the reasons cited earlier, some
Federal role in policy may need to be maintained. Furthermore, the
Federal Government has a significant role in financing programs, it also
should have some role in policy in order to ensure accountability.

ENSURING GOVERNMENT ACCOUNTABILITY

The Administration is committed to ensuring that government funds
are spent wisely, whether the Federal Government or States and
localities are doing the spending. A problem with revenue-sharing
arrangements or pure block grants is that the level of government making
the policy decisions is no longer the one responsible for financing the
program. This separation of functions may increase the likelihood that
taxpayer money is not well spent. Indeed, some evidence suggests that
States spend money they receive from the Federal Government differently
from funds they raise themselves--and restrictions on spending imposed
by the Federal Government do not account for all of the difference.
Thus, the availability of Federal highway money results in more spending
on highways than States would otherwise undertake, even though, at the
margin, most States pay 100 percent of each additional dollar of highway
spending (Box 4-1). Evidently, State taxpayers are content to give
government officials more discretion over funds coming from Washington
than over funds contributed by their own State tax dollars.
This is a two-edged sword. On the one hand, it means that the
Federal Government can influence the pattern of State spending more
easily than it might otherwise: Federal money may not just

Box 4-1.--Federal Grants and the ``Flypaper Effect''

The Federal Government provides substantial grants to States and
localities--over $228 billion in 1995. Most of this grant money can be
used for projects that these governments might otherwise fully fund
themselves, and most do not require that the State or locality
contribute any matching funds. Because these grants can simply serve to
free up State and local government funds for other uses, they can be
viewed as equivalent to pure transfers of cash from the Federal
Government. From an economic perspective, then, one would expect States
and localities to spend these grants in the same manner as they would
any other increase in income. For example, States might allocate 5 to 10
cents of each grant dollar to increases in their spending, and the rest
would simply be used to reduce State taxes.
Researchers have consistently found, however, that Federal grants
have much larger effects on State and local government spending than
this logic would suggest. Recent studies find that the actual increase
is on the order of 40 to 65 cents on the dollar. This result has been
dubbed the ``flypaper effect'': the money sticks where it hits.
Moreover, not only does State and local spending increase when Federal
grants increase, but the money tends to remain in the program area for
which the grant was intended: grants for education tend to increase
education spending, grants for infrastructure tend to increase
infrastructure spending, and so on. Some of the grant money is used to
finance other areas of government and to finance tax cuts, but such
``leakages'' are much smaller than economic theory would predict.
substitute for State money, as many critics of block granting have
feared. (And, as is discussed later in the chapter, it is precisely the
Federal Government's desire to influence patterns of State spending that
justifies a Federal role at all.) On the other hand, if the substitution
of Federal for local funding leads to less diligent monitoring by
taxpayers, the money may not be well spent.
Federal actions can also impose costs on the States. And just as
States may spend Federal money more readily than money raised through
State taxes, so the Federal Government may spend State money more
readily than funds raised through Federal taxes. Federal legislation
that raises States' costs--so-called unfunded mandates--has recently
received considerable attention. Legislation passed in 1995 attempted to
address some of the most important problems posed by unfunded mandates
(Box 4-2).

Box 4-2.--The Unfunded Mandates Reform Act

The Unfunded Mandates Reform Act, enacted in early 1995, will
restrict the ability of the Congress to impose costly mandates on
States, localities, and tribal governments. This legislation requires
the Congressional Budget Office (CBO) to analyze the costs of any
proposed mandates on State and local governments. Mandates certified by
the CBO as costing States and localities $50 million or more in any of
the first 5 years after becoming effective are not permitted. However,
majority votes in both the House and the Senate can waive this
prohibition. The CBO also is required to estimate the cost of any
mandate on private companies which exceeds $100 million in any year over
this same 5-year period.
The unfunded mandates legislation was enacted to restore
equilibrium to the relationships between Federal, State, and local
governments. For too long, Washington has placed overly burdensome
mandates on States and localities. The new law rectifies this imbalance
but also permits mandates that are in the national interest. For
example, some unfunded mandates may be designed to control cross-
jurisdictional externalities. A State that dumps garbage in a river,
polluting the shores of a neighboring State, causes an externality every
bit as important as that generated by a private firm. The law provides a
flexible way of addressing unfunded mandates: it requires the disclosure
of relevant information, without being overly prescriptive. With a
majority vote, the Federal Government could, for instance, still
proscribe States from dumping garbage in ways that adversely affect
neighboring States. To do so imposes costs on States, but these are
costs that they should rightly bear.
The legislation also requires Federal agencies to assess the
qualitative and quantitative costs and benefits of any proposed
regulatory actions that would result in annual expenditures of $100
million or more by State, local, and tribal governments or the private
sector before promulgating such actions. Agencies must ``. . . [1]
identify and consider a reasonable number of regulatory alternatives and
[2] from those alternatives select the least costly, most cost-effective
or least burdensome alternative that achieves the objectives of the
proposed rule,'' or explain their decisions if a different action is
adopted. Finally, the legislation requires the Advisory Commission on
Intergovernmental Relations, an independent agency, to make
recommendations on paring existing mandates.

DEVOLUTION AND THE PROVISION OF PUBLIC SERVICES

In many cases, government action can correct inefficiencies in the
private market--so-called market failures--and so improve the overall
allocation of resources. As discussed in Chapter 1, to correct market
failures, government may need to provide certain goods directly (so-
called public goods), adopt or enforce standards that apply to other
goods (such as safety standards), or encourage, through subsidies or
regulation, private firms to provide goods that would otherwise be
underprovided (i.e., those with positive externalities). All of these
activities can be viewed as providing public services.
State and local governments have many advantages in providing these
public services. They can more easily address the differing needs and
preferences of particular communities. For example, building codes
should reflect local weather and geological conditions, and communities
should be able to choose their own level of community services. Having a
number of communities with different mixes of services (and of taxes to
pay for them) improves overall efficiency, if people can choose to live
in the jurisdiction that best meets their needs and desires.
Competition among localities can enhance this efficiency by making
it easier for people to hold their local government accountable for the
decisions it makes. For example, if a city, by operating efficiently, is
able to maintain a high level of public services with relatively low
taxes, residents of nearby cities may demand equally efficient
government from their policymakers--and use the threat of relocation to
the efficient city to make their demands resonate.
But the problems described above require some Federal role in the
provision of many public services. Uniformity of regulations or of
standards, such as safety standards, can improve the effectiveness and
efficiency of certain policies. A uniform set of minimum water and air
quality regulations ensures that all Americans, regardless of where they
live, have clean air to breathe and water to drink. Cross-jurisdictional
spillovers also can be important. Some types of public services, like
national defense and subsidies to scientific research, need to be
provided by the Federal Government because the spillovers from
government action are so large. Public services and goods, like national
defense, that can only be provided effectively at the national level are
called national public goods. Those whose benefits accrue exclusively to
residents of a particular locale are called local public goods.
Between purely local and purely national public goods are many
intermediate cases: many public services create some spillovers, but
still much of the benefit accrues within the community. Highways are a
prominent example. Many highways are used primarily by residents of the
State where the highway is located. But these highways also provide
significant benefits to out-of-State residents who travel on them and
who purchase goods that have been transported on them. If State
residents had to pay for all the costs of building highways in their
State, their choices regarding highway construction might take into
account only the benefits they expect for themselves. Thus they would
construct fewer roads with smaller capacity than would be socially
desirable.

MATCHING GRANTS

One method used to solve this problem is the categorical matching
grant, in which the Federal Government pays a fraction of the overall
costs of the program. For example, the Federal Government could match
additional State spending on a 1-to-1 basis, or on a 2-to-1 or 4-to-1
basis. Ideally, the match rate would be set so that the fraction of the
total costs paid by the States equals the fraction of the total benefits
that accrue to their residents. Under such a financial arrangement, the
decision on the level of expenditures can be delegated to the States.
Because the spillover effects are taken into account in the ``price''
States have to pay, they will set the level of expenditures at an
efficient level.
In practice, however, a large share of Federal grants for public
infrastructure, education, and social services is not in the form of
matching grants, but rather in the form of categorical unmatched grants
(grants that provide funds for particular purposes, such as education of
the disabled or tuberculosis control, but do not require States or
localities to put up any of their own money). Furthermore, while there
are grant programs that do require States to spend their own funds in
order to receive Federal money, many are in the form of capped matching
grants, which place a ceiling on the total amount that the Federal
Government will pay. From an economist's perspective, capped matching
grants are much like categorical grants. Once the cap on Federal grants
is reached, State and local governments bear the full cost of additional
projects. And since, for many capped matching grant programs, States
likely do spend more than the amount required to receive the maximum
allowable Federal grant, the grants probably do little to change the
incremental costs of projects, but simply allow States and localities to
shift resources to other projects. Capped matching grants may thus
insufficiently address the problem of underspending arising from cross-
jurisdictional spillovers. Surprisingly, however, evidence indicates
that categorical grants and capped matching grants do stimulate a
significant amount of additional investment in the targeted activities
(Box 4-1), although they also serve to free up State funds for other
purposes. Open-ended matching grants, which would change the marginal
cost to States, could have significantly greater effects on State
spending decisions, because they would affect the prices faced by the
States at the margin.

PUBLIC SERVICES AND DIFFERENCES IN LOCAL RESOURCES

One of the rationales cited above for a Federal role in provision of
public goods is that some jurisdictions lack the resources to finance
public services at a level deemed adequate by the Nation as a whole. But
a lack of sufficient State resources to provide services does not
necessarily imply that the Federal Government should provide those
services instead. In principle, the Federal Government could, instead of
providing grants for public services to poor States, provide income
transfers to poor individuals. Just as individuals, not the government,
should decide how their income is spent, so too individuals should
decide for themselves about the level of consumption of local public
goods.
If taxpayers closely monitored their policymakers, the level of
public services would not depend on whether resources were transferred
to State and local governments or directly to taxpayers, and the
transfer of resources to the States would simply substitute for State
governments levying taxes. But the evidence cited earlier suggests that
the way money is distributed does matter. Direct transfers to
individuals force State and local policymakers to justify their choices
of public services.
This general principle has two exceptions. First, Americans believe
that society has a special responsibility to children, regardless of the
economic condition of their parents. Providing services that go directly
to children, rather than providing cash to their parents, may be a more
effective way of making sure that it is children who ultimately benefit.
More generally, society may not care so much about inequality of income
as about inequality in the consumption of certain goods, and so may
prefer to provide these goods instead of cash. To the extent that these
goods are public services--like health care, clean water, decent
schools, good job opportunities, and safe places for children to play--
Federal funding of such services for poor neighborhoods is warranted.
A second reason why it may be better for the Federal Government to
provide direct financing for public services is to save on
administrative costs. Indirect financing, through Federal transfers to
citizens residing within the jurisdiction, involves two steps:
disbursing funds to individuals and collecting the money once again at
the State or community level. Because each step has its costs, direct
transfers to State and local governments may save on overall transaction
costs.

BETTER GOVERNMENT THROUGH COMMUNITY AND INDIVIDUAL EMPOWERMENT

Over time, a large number of Federal programs have evolved primarily
to meet certain perceived needs that were not being adequately addressed
at the State and local level. Although these programs direct attention
and resources to real problems, in some cases they leave too little
discretion to States and localities in allocating the funds, and Federal
paperwork requirements lead States and localities to devote too large a
share of their resources toward administrative costs. Furthermore, in
some cases these funds could be more effective if they were used to
empower individuals, by providing them the wherewithal and the
information to make appropriate choices, rather than having government--
Federal, State, or local--in the driver's seat.
This Administration has put forward a new approach to Federal
grants:
 The Federal Government would provide States and local
governments with greatly enhanced flexibility: funds from
numerous programs would be consolidated, and regulations would
be pared back.
 Accountability would be ensured not by restrictions on the
use of funds but by performance measures. Programs that live up
to their stated goals could receive more funding.
 Individuals benefiting from government programs would also
be given as much discretion as possible to choose how those
funds should be spent, reducing the possibility that they would
be spent unwisely.
One example of this new approach is the Administration's proposed
G.I. Bill for America's Workers (Box 4-3). Under the current Job
Training Partnership Act, States are provided the funds to obtain
training for dislocated workers. Under the Administration's proposal,
funds would instead be dispensed directly to individuals, in the form of
``skill grants'' which they could use for tuition at private or public
institutions. States and localities would create one-stop career
development centers, which would provide individuals with the
information necessary to make good choices about how to use their skill
grants, would track participant earnings and job retention, and would
work with businesses to help match newly trained workers with jobs.
Allowing individuals to make informed choices about what skills to
obtain and where to obtain them will ensure that only those institutions
that provide high-quality, relevant training will survive.
The Administration has also encouraged legislative efforts, such as
the proposed Local Empowerment and Flexibility Act, that would waive
programmatic regulations for local communities that have a federally
approved ``Local Flexibility Plan'' from certain Federal laws and
regulations that impede their efforts to meet their plan. Similarly, as
part of its overhaul of environmental regulation, the Administration has
initiated Project XL, which gives responsible companies and other
regulated parties the flexibility to replace the requirements of the
current regulatory system with their own alternative strategies to
achieve better bottom-line environmental results.
These efforts are similar to the project currently under way to
revitalize distressed communities: The Empowerment Zone and Enterprise
Community (EZ/EC) initiative provides block grants, tax subsidies, and
regulatory flexibility to a number of designated communities that have
formulated innovative strategic development plans. A major element of
these plans, and of the EZ/EC initiative, is the inclusion of
performance benchmarks, so that policymakers can learn what works and
what does not.
In cases where local control has not done the job, a reconsideration
of the intergovernmental partnership is in order. Public housing is one
example of a program that needs major change (Box 4-4). In its plan to
reorganize the Department of Housing and Urban Development (HUD), the
Administration proposed providing greatly increased flexibility to well-
performing housing agencies and overhauling those public housing
agencies that are chronically troubled. In some cases, residents of
severely distressed units will be provided with rental vouchers, which
could be used to obtain private housing. After all, individuals have the
best incentive to ensure that the dollars they receive for housing are
well spent.

DEVOLUTION AND THE SAFETY NET

This country has reached a general consensus that providing a
minimum level of subsistence for our most vulnerable citizens,
regardless of where they live, is an essential government role. But
because differences exist across States--in job opportunities, in family
characteristics, and even in views on the appropriate level of support
for the poor--States also have a role in providing and administering the
safety net.
At the same time, safety net programs--programs that provide
assistance to those meeting certain income or asset tests, such as Aid
to Families with Dependent Children (AFDC) or Medicaid--present several
problems that require some Federal role. One problem stems from the
mobility of the population. For example, whenever one State chooses to
expand its welfare program--by raising benefits or relaxing eligibility
criteria--it may encourage poor people from other States to move in. As
they do, the welfare program becomes more expensive, forcing the State
either to reduce benefits or eligibility, raise taxes, or both. If the
State raises taxes to pay

Box 4-3.--Rethinking Devolution: The Job Training Partnership Act

The history of the Job Training Partnership Act (JTPA) shows that
simply shifting accountability and policy discretion to the States does
not always improve performance. When enacted in 1982, JTPA was designed
as a block grant to the States. JTPA reduced the role of the Federal
Government, enhanced the role of the States, and retained a strong role
for policymaking and initiative at the local level. However, the program
became the subject of a growing number of reports. The General
Accounting Office concluded that Federal dollars were being misused,
while the Department of Labor's Office of the Inspector General found a
serious lack of uniform control and guidance. JTPA's problems led the
previous Administration and a coalition in Congress to reassert Federal
accountability through a set of new rules and regulations enacted in
1992.
The 1992 legislation was an understandable response, but it made
JTPA less flexible. The dilemma facing JTPA is one of the reasons why
the present Administration has proposed a G.I. Bill for America's
Workers. The new bill is based on a different model, one that replaces
bureaucratic accountability with market-driven accountability based on
individual empowerment, informed customer choice, and competition among
providers. It establishes appropriate and complementary roles for all
three levels of government--Federal, State, and local--in the design,
implementation, and oversight of effective workforce development
systems. It also provides for the close participation of businesses,
labor organizations, and local elected officials to facilitate effective
training and placement.
for the more expensive welfare program, residents with higher incomes
may migrate to other States with lower taxes, again making it harder for
the State to finance its established level of benefits. Accordingly,
States and localities are discouraged from providing safety net
benefits. This phenomenon--sometimes labeled the ``race to the
bottom''--limits the ability of States to offer their residents welfare
benefits that are as generous as they would like in the absence of
migration.
The severity of this problem depends on how prone people are to move
in response to differences in the generosity of welfare benefits across
States. The evidence is inconclusive. Some researchers have found that
low-income households are indeed more likely to move from low-benefit to
high-benefit States, whereas others have found no evidence of welfare-
induced migration. Even when welfare bene-

Box 4-4. Rethinking Devolution: The Case of Public Housing

Since 1937 the Federal Government has invested some $90 billion in
Federal housing. The legacy of that investment is mixed. Public housing
does provide affordable shelter for approximately 1.3 million
households. But many public housing projects are in abject disrepair.
One problem with the current system is the lack of accountability.
The discipline of the real estate market seldom extends to public
housing. Instead, local public housing agencies administer the public
housing stock, subject to the rules and regulations of a distant Federal
bureaucracy. Under the reorganization plan for the Department of Housing
and Urban Development, well-performing public housing agencies will be
given greater flexibility to improve their housing stock, through
modernization or demolition, and to attract and retain a broader range
of families by setting their own rules for admission to public housing.
But public housing agencies that exhibit persistent management
deficiencies will be overhauled. And some projects, such as Chicago's
infamous Cabrini-Green, will be demolished. In many cases, residents of
demolished units will be given rental vouchers to live in better housing
in the private market. Vouchers permit tenants to demand quality
housing, and also make it easier for them to seek out gainful employment
and jobs that maximize their income, regardless of where they are
located. In other communities, a new form of public housing is being
tested. Instead of mammoth apartment buildings, small-scale, townhouse-
style housing is being constructed that would provide housing to
residents with a wide range of incomes. Instead of purely public
ownership and management, this housing will be owned and managed by
partnerships between public entities and for-profit and non-profit
developers.
fits are found to affect migration, the effects are generally small and
slow to happen. But even if the effects are small on average, they could
be substantial for neighboring States with population centers in close
proximity. Furthermore, the studies examining the effects of differences
in AFDC benefit levels on migration were all done within the context of
the current AFDC program, which does impose some limits on the
differences across States. For instance, although average benefit levels
and eligibility requirements vary widely, States are required to provide
coverage for all families meeting the State income and asset
requirements. Interstate competition might be more of a problem if some
States imposed rigid time limits on welfare recipiency or denied
benefits to certain families while others did not.
Some State legislatures have taken the position that welfare-induced
migration occurs and should be discouraged. As a result, under waivers
granted by the previous Administration, California and Wisconsin were
permitted to create ``two-tier'' welfare systems, in which new residents
on AFDC could receive a different level of benefits than longer term
residents of those States. However, some have questioned the legality of
the two-tier system: California's waiver was voided by the Court of
Appeals, and Wisconsin's is the subject of pending litigation.
Disparities in State resources--particularly in relation to the
demands put upon them--provide another rationale for a Federal role.
Poorer States feel that they cannot afford the same level of safety net
protection that wealthier States can. As in the case of public services,
this rationale does not necessarily imply that the Federal Government
should finance the safety net programs. Just as the government can help
provide public services in two ways, it also has two ways of helping
individuals: directly, and indirectly by first giving it to States and
communities. The direct method can save on transaction costs, and the
resulting empowerment of individuals may enhance the efficiency of the
funds. On the other hand, in cases where benefit recipients also require
other government help--for example, in finding child care or getting job
training--transfers to States or communities to fund such services may
prove more effective.
Some States have historically been poorer than others, and these
differences are not likely to change any time soon. But in addition to
these persistent disparities, shorter term disparities arise from
fluctuations in the business cycle. In the past, Federal funding has
acted in part as a form of insurance against these shocks, with those
States experiencing increases in their poverty population receiving
greater Federal funding. To some extent States can insure themselves
against temporary economic shocks if they maintain ``rainy day'' funds
or if they permit themselves to borrow during hard times. However,
political constraints that States face, such as balanced budget
requirements, may reduce their ability to insure their safety net
programs against adverse economic shocks.

THE FEDERAL ROLE IN THE SAFETY NET

All these considerations argue for a strong Federal role in safety
net programs. And in fact, most safety net programs are either federally
run or run jointly by the Federal and State governments. The Federal
Government finances and makes policy decisions for Food Stamps and
Supplemental Security Income (SSI, the cash assistance program for the
low-income aged, blind, or disabled); States do have an administrative
role in both, however, and many States supplement SSI benefits with
their own funds. Medicaid and AFDC--which along with Food Stamps are the
largest programs for the nonelderly poor--are administered by the
States, but States and the Federal Government share responsibility for
funding and for policymaking. Other safety net programs, like housing
subsidies and energy assistance, are provided by both the Federal
Government and the States.
The Federal share of spending on safety net and social service
programs increased with the introduction of Medicaid, SSI, and Food
Stamps: from roughly 44 percent of the nationwide total in 1960 to over
70 percent in 1976, and has remained relatively stable since then.
Under current law the Federal Government provides open-ended
matching grants to States for Medicaid and AFDC, with the Federal share
of expenditures in 1995 varying, from 50 percent to 79 percent,
according to State income. This open-ended matching reduces the States'
marginal price of providing benefits, giving States an incentive to
provide higher levels of benefits than they otherwise would. Federal
matching also helps offset the problems of States offering lower
benefits for fear of becoming welfare magnets or because of insufficient
resources. Yet despite their significantly higher Federal matching
rates, poorer States still tend to pay lower AFDC benefits (Table 4-2).
Although the theoretical arguments supporting a Federal role in
welfare are strong, almost all observers, including welfare program
participants themselves, agree that the welfare system is not working
well. For too long, it has undermined the values of work and personal
responsibility, not strengthened them.
Welfare policy presents a dilemma with which the Nation has been
struggling for 60 years: providing adequate support to low-income
families who fall upon hard times, and especially to their children,
without generating dependency. Despite a broad consensus that the goal
of welfare reform should be to move people from welfare to work, how
best to accomplish this goal is still unclear.
In such uncertain circumstances, the potential value of innovation
and experimentation is high, and States have shown increasing interest
in trying new approaches. This Administration has used waivers
effectively to allow States to engage in valuable experimentation. The
Administration has made clear that it is open to States' proposals for
alternative approaches to providing welfare support. Since January 1993
the Administration has approved welfare demonstration projects in 37
states. In an average month these demonstrations will cover more than 10
million people, or approximately 73 percent of all AFDC recipients.

Table 4-2.--Typical Maximum AFDC Payments for a Family of Three
[Dollars per month]
----------------------------------------------------------------------------------------------------------------
Three-
person                                         Three-person
State                        family                 State                  family  typical
typical                                            maximum
maximum
----------------------------------------------------------------------------------------------------------------
Alabama........................................  164      Montana...........................                375
Alaska.........................................  923      Nebraska..........................                364
Arizona........................................  347      Nevada............................                348
Arkansas.......................................  204      New Hampshire.....................                550
California.....................................  607      New Jersey........................                424
Colorado.......................................  356      New Mexico........................                381
Connecticut....................................  581      New York..........................                577
Delaware.......................................  338      North Carolina....................                272
District of Columbia...........................  420      North Dakota......................                431
Florida........................................  303      Ohio..............................                341
Georgia........................................  280      Oklahoma..........................                307
Hawaii.........................................  712      Oregon............................                460
Idaho..........................................  317      Pennsylvania......................                403
Illinois.......................................  377      Rhode Island......................                554
Indiana........................................  288      South Carolina....................                200
Iowa...........................................  426      South Dakota......................                430
Kansas.........................................  403      Tennessee.........................                185
Kentucky.......................................  262      Texas.............................                188
Louisiana......................................  190      Utah..............................                426
Maine..........................................  418      Vermont...........................                656
Maryland.......................................  373      Virgin Islands....................                240
Massachusetts..................................  579      Virginia..........................                291
Michigan.......................................  459      Washington........................                546
Minnesota......................................  532      West Virginia.....................                253
Mississippi....................................  120      Wisconsin.........................                517
Missouri.......................................  292      Wyoming...........................                360
----------------------------------------------------------------------------------------------------------------
Note.--``Typical maximum'' is amount paid for basic needs to a family (including one adult) with no income or
special needs in State's highest caseload area.

Source: Department of Health and Human Services.

In their reform efforts, many States have sought to reduce welfare
dependency by beginning to experiment with time limits on families'
welfare benefits. Others have sought to facilitate the movement from
welfare to work by setting strict job search or work requirements, or by
providing subsidies to private employers who hire welfare recipients.
Many States require recipients to sign ``personal employability plans''
or ``self-sufficiency agreements,'' with specific goals and deadlines.
Failure to meet the deadlines can result in reduction or denial of
benefits.
The Administration has reinforced these state welfare reform efforts
with other policies that reward work over welfare. In 1993 the
President's economic plan cut the taxes of 15 million working families
through the earned income tax credit. The Administration has also
proposed raising the minimum wage to ensure that, in combination with
the Earned Income Tax Credit, a single parent with two children working
full-time would escape poverty. The Administration has also strengthened
collection of child support, enabling more single parents to support
themselves through a combination of child support and work, instead of
welfare.

MOVING FORWARD: WELFARE REFORM

The Administration has called for comprehensive, bipartisan welfare
reform legislation to impose time limits, work requirements, and tough
child support enforcement nationwide. Many in the Congress, believing
that the waiver process is still too burdensome and uncertain, have
proposed converting AFDC into a block grant program, providing States
the flexibility to design their own approaches to welfare reform without
the need for waivers from Washington, and putting an end to the open-
ended entitlement funding structure.
Converting AFDC to a pure block grant could have a number of
effects. First, under pure block grants, States would no longer receive
additional funding for increases in benefits arising from economic
downturns or population growth, making it more difficult to provide
needed benefits. Second, under a block grant program, States would
receive a fixed amount of money from the Federal Government, independent
of the level of State expenditures. The elimination of the Federal
matching program would mean that States would no longer receive extra
Federal money when they raised benefits, nor lose Federal support when
they cut benefits. This change in incentives (which would be greater for
low-income States since they now have the most generous Federal match
rates) might induce some States to cut their welfare spending.
On the other hand, converting AFDC to a block grant program would
also mean that States that managed to get people off welfare and into
jobs would realize all the resulting welfare savings. Under the current
program most State job training expenditures are not matched, even
though the Federal Government receives a large fraction of the resulting
welfare savings.
In any reform of the welfare system, the Administration has
consistently argued for crucial safeguards to promote work and
responsibility and to protect children. It has insisted on a strong
maintenance-of-effort requirement so that States keep their welfare
spending at adequate levels, and sufficient resources to pay for child
care so that recipients can leave welfare and go to work. Finally, the
Administration has required that additional resources be made available
to States during economic downturns. Under the current system, this
occurs automatically through the Federal match, but an adequately
financed contingency fund with an effective trigger mechanism could also
accomplish this goal.
Because the current system frustrates taxpayers and recipients
alike, the Administration plans to work with the Congress to enact a
bipartisan welfare reform bill. As part of its 7-year balanced budget
proposal, the Administration has proposed repealing the AFDC program and
replacing it with a new Federal program with strict time limits on
welfare benefits. The new program would require parents to go to work
after 2 years or lose their benefits; after 5 years benefits would end
unconditionally. States would enjoy new flexibility to tailor their
welfare systems to local conditions. At the same time, the plan would
provide vouchers to protect children whose parents reach the time limit.
Because the Federal government would continue to match State welfare
spending, States would be protected in the event of economic downturns
or caseload growth.

MOVING FORWARD: MEDICAID

This Administration has insisted upon maintaining the Federal
entitlement to Medicaid, for two main reasons. First, this
Administration believes that all Americans should be guaranteed access
to quality medical care, regardless of income or State of residence.
Second, the Medicaid program is not performing badly: it needs reform,
not repeal. Although overall Medicaid expenditures have been increasing
at a rapid rate, part of this increase is attributable to legislated
increases in the eligible population.
This Administration's insistence upon maintaining the guarantee of
health care coverage for poor families in no way contradicts its
commitment to flexibility, innovation, and experimentation. The
President's plan expands State flexibility in administering Medicaid
programs, but maintains protection for beneficiaries and for States
facing population growth or economic downturns. To this end, the
Administration is committed to working in partnership with the States to
test new approaches to Medicaid through the waiver process. The
Administration shares States' interest in developing innovative delivery
systems, improving quality of care, and expanding coverage to uninsured
Americans. To date, the Administration has approved 12 comprehensive
health care reform demonstrations. These waivers have allowed States to
greatly increase their use of managed care, to subsidize health
insurance for employed but uninsured workers, and to expand Medicaid
eligibility by eliminating asset tests and increasing income limits.
Furthermore, the Administration has granted 14 States Medicaid waivers
as part of larger welfare reform projects. These waivers enable States
to continue providing essential health care services while encouraging
independence from welfare. The Administration's 7-year budget plan would
give States further flexibility to modify their programs. In particular,
it would no longer require States to obtain a waiver in order to expand
coverage to any person whose income is at or below 150 percent of the
poverty line, to use managed care plans to provide health insurance to
their Medicaid population, or to move people from nursing homes to home-
and community-based settings. The plan also repeals the Boren Amendment,
thus allowing States greater flexibility in establishing their provider
payment rates.

THE CHALLENGES OF DEVOLUTION

This Administration is committed to making government more efficient
and effective. Designing government programs so that activities are
performed at the appropriate government level--Federal, State, or
local--is one of the most difficult challenges associated with this
task. Although in many areas the answers are clear--national defense is
a Federal responsibility, whereas sewage treatment and water supply are
local responsibilities--in many other areas the advantages of Federal
responsibility must be balanced against the advantages of State and
local responsibility. Federal grants to fund certain public services can
reduce the problems of spillovers, but if the sense of accountability
for Federal funds is different from that for funds raised through State
or local taxes, Federal grants may be spent unwisely. Restrictions on
the use of Federal funds may reduce this problem, but they may also
impose significant administrative burdens and severely limit State
innovation.
One approach to solving this problem is to ensure accountability
through results-oriented measures, rather than through conventional
rules and regulations. A results-oriented approach allows States much
more flexibility without severely hampering efficiency. The
Administration has proposed using this approach in housing, job
training, the environment, welfare, and numerous other policy areas.
Subjecting government expenditures to this discipline is likely to be
the best way to improve government efficiency. Furthermore, when
possible, government should use the private market directly. For
example, individuals can be provided housing vouchers that permit them
to live wherever they choose, and those in need of job training can
receive funds to pay for training at the institution of their choice. In
this way, individuals are provided the wherewithal to choose what is
best for them, and only those providers that bring desirable services to
market at the lowest cost--whether it be rental housing or job
training--will survive.
States must also be provided with greater flexibility where no
consensus has emerged on how to accomplish the goal. In these cases,
experimentation and innovation by the States could prove invaluable. But
this enhanced flexibility must be provided in a way that protects the
national interest and advances the objectives of the programs. What is
appropriate in one program may not be appropriate in another. In some
cases the solution may entail Federal regulation as a ``default
option,'' with wide latitude for waivers to allow for State and local
adaptation. In other cases, block grants with little Federal policy
involvement may be called for.
Devising policies that ensure accountability and that protect the
national interest, while also allowing for flexibility, adaptability,
and innovation at the State, local, and individual levels is a great
challenge. What worked in the past may no longer work today. Carefully
balancing the advantages and disadvantages to find the right mix of
policies is vital if government is to work at its best.