Fiscal Federalism: Theory and Practice

June 3, 2020 R46382

Fiscal Federalism: Theory and Practice
June 3, 2020
The field of fiscal federalism studies how to divide responsibilities (including finances)
among federal, state, and local governments to improve economic efficiency and
Grant A. Driessen
achieve various public policy objectives. Determining the optimal division of
Analyst in Public Finance
responsibilities is difficult because of varying subjective views about what the role of

government should be. As a result, fiscal federalism research general y renders no
Joseph S. Hughes
judgment on the proper level of total government intervention or what types of services
Research Assistant
governments should provide. The research focuses instead on how responsibilities are

assigned across multiple layers of government once policymakers have decided to
implement a given policy, and what trade-offs may be involved in administering it.

For example, a more prominent federal government role may improve efficiency when taxpayers can easily move
among localities and states to minimize taxes, or when there are substantial spil over effects from providing goods
and services. A more active state and local role may be beneficial in other cases, such as when there is a high level
of variation among constituents in the desired amount of government-provided services, or when obtaining
enough information to effectively administer a program is difficult (e.g., public education or local housing
initiatives). Theories of fiscal federalism can be useful to policymakers when analyzing policies that could involve
several layers of government.
General interest in fiscal federalism has increased following the economic downturn accompanying the
Coronavirus Disease 2019 (COVID-19) crisis. Early evidence suggests there has been a significant shift in how
fiscal responsibilities are divvied up among the federal government, and state and local governments. This shift
has included $150 bil ion in direct federal assistance to state and local governments and the Federal Reserve’s
support of up to $500 bil ion in state and local debt issuances.
This report introduces a basic model of fiscal federalism. After developing the basic model and discussing its
implications for governments’ roles, the analysis explores the effects of various extensions to incorporate more
realistic assumptions. A real-world example is presented with each modification to the basic model’s assumptions
to draw a connection between theory and practice.
Variation in preferences across individuals and firms incentivizes more activity from state governments, which
can respond to any differences in demand for government from their residents. Al owing mobility across
jurisdictions, meanwhile, can exacerbate state-level variation but can also introduce market inefficiencies that
benefit from federal intervention.
Differences in the effectiveness of tax and spending programs across physical space al ow for public benefits from
state government activity so long as the variation is confined to the state. Introducing externalities, or program
effects on anyone other than the government and program target, may warrant increased federal activity if
externalities are present across multiple states. Al owing for differences in administrative costs and budget
flexibility can also influence fiscal federalism choices.
Practical examples with implications for fiscal federalism include relative federal, state, and local government
activity during economic downturns, spending in policy areas like education and transportation, and the type and
total level of taxation present across areas.

Congressional Research Service

link to page 18 link to page 21 Fiscal Federalism: Theory and Practice

Introduction
The field of fiscal federalism studies how to divide responsibilities (including finances) among
federal, state, and local governments to improve economic efficiency and achieve various public
policy objectives.1 This report develops a basic model (referred to henceforth as the basic model)
of fiscal federalism to explore the economic rationales behind assigning responsibility for certain
public policies to the various levels of government.2
General interest in fiscal federalism has increased following the economic downturn
accompanying the Coronavirus Disease 2019 (COVID-19) crisis. Early evidence is suggestive of
a large shift in how fiscal responsibilities are divided among the federal government and state and
local governments. This shift has included $150 bil ion in direct federal assistance to state and
local governments and the Federal Reserve’s support of up to $500 bil ion in state and local debt
issuances.3
Under the basic model’s assumptions, the conclusion is rather straightforward: the federal and
state governments only need to ensure that the total level of government services maximizes the
collective wel -being of al residents (often referred to as utility), and do not need to worry about
which level of government provides those services. Subsequent sections of this report modify the
basic model’s assumptions to be more realistic, and explore what those changes suggest about
relative government activity levels. A number of these modifications have clear implications for
the type of government that would produce the most benefit. Other modifications suggest that
how governments divide responsibilities is highly dependent on context. A real-world example is
presented with each modification to the basic model’s assumptions to draw a connection between
theory and practice.
Determining the optimal division of responsibilities is difficult because of varying subjective
views about what the role of government should be. As a result, fiscal federalism research
general y renders no judgment on the proper level of total government intervention or what types
of services governments should provide. The research focuses instead on how responsibilities are
assigned across multiple layers of government once policymakers have decided to implement a
given policy, and what trade-offs may be involved in administering it.
For example, a more prominent federal government role may improve efficiency—and increase
collective wel -being—when taxpayers can easily move across localities and states to minimize
taxes, or when there are substantial spil over effects from the federal provision of goods and
services. A more active state and local role may be beneficial in other cases, like when there is a
high level of variation in the desired amount of government-provided services, or when obtaining
enough information to effectively administer a program is difficult (e.g., public education or local
housing initiatives). Theories of fiscal federalism can be useful to policymakers when analyzing
policies that could involve several layers of government.

1 For a basic introduction to fiscal federalism, see Richard A. Musgrave, The Theory of Public Finance (New York:
McGraw Hill, 1959).
2 Simplified formulas corresponding to the t heoretical discussion are provided in Appendix A, along with relevant
outside research where applicable. A summary of recent trends and levels in U.S. fiscal federalism can be found in
Appe ndix B.
3 Both provisions were created by the CARES Act (P.L. 116-136). For more information on the CARES Act, see CRS
Report R46299, Coronavirus Aid, Relief, and Econom ic Security (CARES) Act: CRS Experts, by William L. Painter
and Diane P. Horn.
Congressional Research Service

1

Fiscal Federalism: Theory and Practice

Although this report focuses on the economic rationales for assigning certain responsibilities to
the various levels of government, policymakers also consider many noneconomic factors when
weighing the merits of potential changes to government activity levels. For example, this report
does not examine the role of federalism as a political system or as a protection against tyranny,
nor does it evaluate the legal issues inherent in a federalist system relative to other political
structures. For more on those subjects, see CRS Report R45323, Federalism-Based Limitations
on Congressional Power: An Overview.
The Basic Model
This section presents a basic model to examine the economic incentives of assigning tax and
spending activities to various levels of government. As with al models, it relies on a set of
assumptions that may not always accurately reflect certain aspects of the real world.
Acknowledging this, the basic model is extended in the sections that follow to account for a
number of real-world observations and scenarios.
In the basic model there are three different economic actors: governments, individuals, and firms.
The assumptions about these parties are as follows:
 Assumptions about government behavior:
 Governments are equally effective in collecting taxes and carrying out
spending programs. Potential cost differentials, including administrative
costs and program targeting, are ignored.
 Government budgets are balanced. Government spending is exactly equal to
receipts.
 Governments work to maximize the collective well-being of all residents.4
The federal government chooses its tax and spending levels with the goal of
maximizing the collective wel -being of al national residents, while state and
local governments (which for simplicity wil be described as state
governments) independently make their own tax and spending choices, with
the goal of maximizing the wel -being of al the residents in their
jurisdiction.5
 Governments have full information about the actions of other participants.
The federal government makes its decisions with an understanding of how
state governments wil respond, and state governments similarly know how
the federal government wil operate.
 Assumptions about individuals and firms:
 All individuals and firms are identical. There is no variation in preferences
across individuals for government activity, private goods, or leisure. (In
economic jargon, al individuals have the same utility function.) Individuals
supply capital and labor to firms. In exchange, each individual receives an
identical wage and return on capital that are fixed shares of firm revenues.
Individuals and firms have preferences for total government spending while
being indifferent to what level of government provides those services. Firms

4 Extensions of this basic model can prioritize the welfare maximization of certain segments of the population (e.g.,
low-income households or households with children, elderly, and/or disabled persons).
5 Fiscal federalism models can also be constructed to allow for ordered choices (e.g., either the federal government or
state governments choose first). A simultaneous structure is presented here for sake of simplicity.
Congressional Research Service

2

Fiscal Federalism: Theory and Practice

are perfectly competitive and produce an aggregate good with identical
endowments of labor and capital (nonlabor) resources.
 Individuals and firms cannot move across states. Individuals can only
respond to public policy changes through changes in consumption, and firms
can respond only through changes in production.
To complete the basic model, assumptions about public goods and services (collectively cal ed
services for simplicity) and taxes are as follows:
 Assumptions about public services and taxes:
 Services are purely public and taxes are nondistorting. Pure public services
are nonexcludable (anyone in the system has complete access to them) and
nonrival (one individual’s program use does not affect its availability to
others). A nondistorting tax does not influence the payer’s market choices.
 Public services and taxes do not have spillover effects. Spil over effects, also
cal ed externalities, occur when someone other than the individual or firm
that paid a tax to fund a public service receives value (positive or negative)
from it.
 Tax and benefit levels do not depend on individual or firm location within a
jurisdiction. A state government produces an equal benefit for al individuals
in a state, while a federal benefit does the same for al individuals
nationwide.
Under the basic model’s assumptions, the conclusion is that al governments only need to ensure
that the optimal level of total government services—the level that maximizes total wel -being—is
provided, and do not need to worry about who provides those services. This result stems
specifical y from the assumption that individual preferences for public services are identical and
not dependent on which government is providing them. This means that the split between federal,
state, and local activity is indeterminate. Everyone would be just as happy with a system where
the federal government spent $10 tril ion and states (collectively) spent $5 tril ion as they would
in a situation where the federal government spent $5 tril ion and the states spent $10 tril ion.
Extending the Basic Model
Extending the basic model to accommodate more realistic assumptions reveals information about
how various economic conditions influence the preferred distribution of government activity and
how those choices translate to overal wel -being. The following presents a number of extensions
of the basic model in an attempt to gain some insight into how government responsibilities are
divided among federal, state, and local governments. In many cases, there are a number of
conflating factors, economic and noneconomic, that help explain observed outcomes. When
possible, the discussions reference research that may be relevant.
Varying Preferences
Perhaps the simplest deviation from the basic model is to al ow for differences in preferences
among individuals and firms. Individuals who choose to live in New York City, rural Wyoming,
or suburban Florida are likely to have variation in their demands for different types of goods,
services, and leisure, which wil translate into differences in what (and how much) government
intervention they would like. This also generates differences among firms, which wil vary their
production of goods and service in response to individuals’ varying demands. The modification
Congressional Research Service

3

link to page 7 link to page 7 Fiscal Federalism: Theory and Practice

affects the decisions of state governments, which now face differences in the quantity of services
their residents demand. In turn, the federal government wil respond with changes in the amount
of services it provides (and taxes it levies).
Varying preferences wil tend to reduce the maximum level of federal government services that
wil be provided in equilibrium (i.e., when collective wel -being is maximized). In the basic
model, each state demanded the same amount of government activity, and the federal government
could provide al of those services just as wel as if state governments provided them. Because
preferences across states now differ, in order to maximize general welfare, the federal
government cannot intervene more than the minimum level of total government demanded by any
state. Otherwise, total government provisions wil be overprovided relative to demand in at least
one area. For example, assume two identical y sized states with total government demand of $10
bil ion in the first state and $30 bil ion in the second state. The model wil require the second state
government to provide at least $20 bil ion in services, with the remainder provided by the federal
government. State governments therefore need to tailor the level of total government involvement
to meet whatever excess demand exists in their areas after accounting for federal government
behavior. The greater the differences are in government preferences across states, the larger the
state government role must be to maximize wel -being.
Practical Example: Effective Tax Rates
To some degree, differences in individual taxation levels across states may indicate variation in
the size of government preferred in different areas.6 Table 1 shows the five states with the highest
effective tax rates and the five states with the lowest effective tax rates in 2017 (the most recent
data year), along with the U.S. average effective tax rate. Effective tax rates are calculated as the
amount of state and local tax revenue collected divided by the amount of state personal income.
In 2017, state and local tax collections nationwide amounted to 9.8% of al personal income.
Four of the five states with the highest effective tax rates shown in Table 1 have a top marginal
tax rate on personal income that is higher than the national average. Likewise, the five states with
the lowest effective tax rates al have top marginal income tax rates below the national average.
Three of those states—Florida, Tennessee, and Alaska—have no state tax on personal earned
income whatsoever. (Alabama and Oklahoma both have a top marginal tax rate of 5%.)
Similarities between effective and marginal tax rates indicate that receipt levels are, at least to
some degree, an intentional function of the state tax system.
Table 1. Highest and Lowest Effective Tax States, 2017
State and Local Taxes
Personal Income
State
($ billions)
($ billions)
Effective Tax Rate
New York
177.8
1,281.1
13.9%
District of Columbia
7.5
55.5
13.4%
North Dakota
5.0
39.5
12.7%
Hawai
9.5
75.4
12.6%
Vermont
3.8
32.6
11.7%
United States Total
1,652.8
16,820.3
9.8%
Alabama
16.4
198.9
8.3%

6 T his does not mean that all residents in a state will agree on govern ment’s role.
Congressional Research Service

4

Fiscal Federalism: Theory and Practice

Oklahoma
13.9
174.4
8.0%
Florida
77.7
1,000.6
7.8%
Tennessee
22.9
305.7
7.5%
Alaska
3.0
42.3
7.2%
Sources: Census Bureau, 2017 State & Local Government Finance Historical Datasets, and Bureau of Economic
Analysis, Regional Economic Accounts.
State variation in the types of taxes used may weaken the relationship between effective income
tax rates and the amount of government preferred in an area. For example, Alaska collects a
majority of its revenues from oil and natural gas production, including royalties, petroleum
property taxes, and corporate income taxes. In some years, revenues from these resources exceed
90% of the state’s discretionary funds.7 These revenue streams are not available to most state
governments. North Dakota’s high income tax collections, meanwhile, are fueled by a recent
surge in oil and natural gas production that may not necessarily reflect residents’ preferences
regarding government.8
Mobility
Relaxing the basic model’s assumption that individuals and firms cannot move across states
introduces another incentive with relevance to fiscal federalism. In practice, location choices are
products of many different factors. For instance, an individual may move to take advantage of a
job market that is better suited to her skil s. A firm could similarly move its headquarters to take
advantage of another state’s lower tax rates. The basic model did not give individuals and firms
direct influence over how responsibilities are divided among federal and state governments. With
this adjustment, it is stil assumed that governments maximize their residents’ overal wel -being,
but now individuals are able to maximize their own wel -being and firms are able to maximize
their profits along another dimension by choosing where to locate.
How much interstate movement occurs when the mobility restriction is lifted depends on how
costly that movement is for individuals and firms. A key finding of early fiscal federalism
research is that if it is costless to relocate (i.e., full mobility) and if there is preference variation,
individuals with similar government preferences tend to cluster together in states, an outcome
described as “Tiebout sorting” or “voting with your feet.”9 Making movement completely free
increases the total level of desired mobility, whereas high movement costs wil hamper state
switches that would otherwise improve wel -being. (If preferences are stil identical, individuals
and firms wil have no incentive to move even if it is al owed.) Put another way, al owing
individuals and firms to move across jurisdictions exacerbates the differences in government
preferences among individuals across states. This gives state governments a bigger role as the
difference in preferences between states grows. The costlier it is to relocate, the less motivation
individuals and firms wil have to move across states, and the smal er the increase wil be in state
government activity.

7 “Understanding Alaska’s Revenue,” Understanding Alaska’s Budget (website): http://www.alaskabudget.com/
revenue.
8 Patrick Springer, “ North Dakota tax collection grows since Great Recession tops all states,” Bismarck Tribune, April
28, 2019, at https://bismarcktribune.com/news/state-and-regional/north-dakota-tax-collection-grows-since-great -
recession-tops-all/article_c09ae687-fe54-5976-bcc2-2ddfc75d83cb.html.
9 Charles M. T iebout, “An Economic T heory of Decentralization,” Public Finances: Needs, Sources, and Utilization,
National Bureau of Economic Research, 1961, p. 79.
Congressional Research Service

5

link to page 9 Fiscal Federalism: Theory and Practice

Research has also found, however, that al owing freedom of movement can introduce interstate
inefficiencies that can be addressed with increased federal government activity.10 Governments
may be interested in keeping people and firms in their jurisdictions for a number of reasons,
including to minimize negative spil over effects (discussed later) or to maximize benefits from
increasing returns to scale (meaning that additional inputs are increasingly productive) from
firms. This can lead to equilibrium conditions characterized by a state’s “race to the bottom.” In a
race to the bottom, states tax at levels lower than what is actual y preferred to reduce individual
and firm emigration that would otherwise reduce total wel -being. Often in these tax-competition
models, the federal government is the only entity that can compensate residents of jurisdictions
with suboptimal state preference levels by adopting a more active approach.11 Additional y,
al owing individuals and firms to move means that state governments are no longer assured of a
given level of demand for services or of revenue.
Practical Example: Tax Structure
There is some evidence that governments consider the mobility of residents and firms when
devising their tax structures. Changing tax jurisdictions may be easier in some cases than in
others. For example, a large national corporation may find it relatively inexpensive to change the
state where it locates its corporate headquarters, whereas a household may experience significant
expense (e.g., moving costs and the costs of changing employment) to move to a jurisdiction with
lower taxes.12
Table 2 shows the composition of tax revenue across levels of government. Taxes other than
income taxes, including property and sales taxes, comprise the majority of revenues at the state
(58%) and local (94%) levels. States and localities may tax income less in part because income
taxes are relatively easy to avoid, whereas property taxes and sales taxes often have points of sale
that are easier to identify (e.g., where property and retail centers are located).
Table 2. Percentage Composition of Tax Revenue by Government Level, 2017
Type
Federal
State
Local
Individual income
51.8%
37.2%
4.7%
Corporate income
6.9%
4.7%
1.1%
Other taxes
41.2%
58.1%
94.2%
Sources: Census Bureau, 2017 State & Local Government Finance Historical Datasets, and Office of Management
and Budget, Historical Tables, Table 2.1.
Notes: Federal “other taxes” includes social insurance and retirement receipts, excise taxes, and estate and gift
taxes. State and local “other taxes” is calculated as total taxes less income taxes.
Mobility may also play a role in explaining differences in tax structure within state and local
government levels. Whereas most local governments in the United States have no local corporate
income tax, New York City levies a corporate income tax of 8.85%, higher than most state
governments. This may in part reflect New York City’s position as a center of national and

10 Michael J. Keen and Christos Kostogiannis, “T ax competition in federations and the welfare consequences of
decentralization,” Journal of Urban Economics, vol. 56, no. 3 (November 2004), pp. 397-407.
11 For an overview of these models, see John Douglas Wilson, “T heories of T ax Competition,” National Tax Journal,
vol. 52, no. 2 (June 1999), pp. 269-304.
12 For a general discussion of behaviors motivated by tax minimization, see Nadine Riedel, “Quantifying International
T ax Avoidance: A Review of the Academic Literature,” Review of Economics, vol. 69, no. 2 (June 2018), pp. 169-181.
Congressional Research Service

6

link to page 11 link to page 11 Fiscal Federalism: Theory and Practice

international commerce, home to the stock exchanges with the two largest market capitalizations
in the world.
Spatial Effects
The basic model did not al ow for what are known as spatial effects (i.e., how the geographic
location of a household or firm can affect how much it is able to benefit from certain programs).
Spatial effects may be a function of proximity; for instance, public parks tend to be utilized more
intensely by those who can access them easily. In other cases, service effectiveness can depend on
physical characteristics. A fire station that is five miles away from a household, for example, wil
be more effective if the household can be reached by a highway than if it is accessible only along
a winding one-lane road.
Al owing for spatial effects in the model means that the wel -being of residents depends not only
on the programs themselves, but also on their geographic location. Specifical y, a public
program’s impact on an individual’s wel -being depends on the individual’s location relative to
the program’s center. The precise nature of how an individual’s location affects the individual’s
benefits from certain public services—including the way benefits taper off as an individual’s
distance increases from the center, as wel as the size of the total service area—is also important,
because it wil determine how a given proximity to a program’s center translates into individual
benefits.
The role that various levels of governments should have in providing certain public programs
depends on the distribution of those benefits within and across states. In most cases, programs
that produce benefits that are confined entirely within a state wil maximize wel -being when
delivered by the state government. Because the federal government has limited ability to tailor
program benefits or taxes to particular areas in the model, it is incapable of adjusting programs to
accommodate spatial variation in the same manner as state governments.
The federal government general y has a larger role to play, however, in cases where benefits may
spread across state lines, as state governments would not factor the benefits of out-of-state
residents in their program choices. Federal and state governments might choose different
locations for benefit centers, such as hospitals near state borders that would serve both in-state
and out-of-state patients. The practical implications of spatial effects can be quite complex, even
when they are only introduced for a single service.13
Practical Example: Education
The structure of some spending across levels of government may reflect the proximity of service
locations to the benefits they provide. Evidence of this effect can be found in spending on
education, shown in Table 3. Elementary and secondary education primarily provides a highly
localized benefit to the children and families in a particular school district. Although it is in a
nation’s interest to have a wel -educated citizenry (i.e., there may be positive externalities,
discussed in the next section with public education), the direct benefit of elementary and
secondary school spending is the knowledge bestowed on the individuals in those schools and
their families. In contrast, higher education benefits are much more spatial y diffuse. Although a
locality may benefit from proximity to a large research university or teaching hospital, the
primary service recipients are the students, who come from across the state and country. Table 3

13 More detail on how equilibrium outcomes vary with spatial characteristics can be found in Charles M. T iebout, “An
Economic T heory of Decentralization,” Public Finances: Needs, Sources, and Utilization, National Bureau of
Economic Research, 1961.
Congressional Research Service

7

link to page 15 Fiscal Federalism: Theory and Practice

highlights this dynamic, as spending on elementary and secondary education overwhelmingly
takes place at the local level. Higher education spending, however, mostly takes place at the state
level with assistance from the federal government.
Table 3. Education Spending by Government Level, 2017
(Bil ions of Dol ars)

State and Local
State
Local
Federal
Elementary & secondary
660.4
7.0
653.5
40.6
Higher education
296.5
253.1
43.4
71.8
Sources: Census Bureau, 2017 State & Local Government Finance Historical Datasets, and Office of Management
and Budget, Historical Tables, Table 3.2.
Notes: Federal elementary and secondary education includes vocational education.
Externalities
Another modification to the basic model is to al ow public programs to generate externalities.
Externalities occur when the production or consumption of a service (or good) imposes benefits
or costs on a third party. Externalities (also known as spil overs) may be positive or negative. For
example, a local y financed bridge generates benefits for out-of-town drivers who may also utilize
the bridge (a positive externality). In contrast, second-hand smoke imposes health costs on
individuals other than the smoker (a negative externality). In practice, externalities can be
difficult to measure: what is the value to out-of-town drivers of a more direct route? How much is
bystanders’ health impacted by al owing smoking in restaurants?
In al owing for externalities, the model must reflect that the choices made by one actor can affect
another’s wel -being. In the basic model, state governments maximized the happiness of their
residents, whose wel -being depends on their own private consumption, their own state’s actions,
and the actions of the federal government. With the presence of externalities, however, the wel -
being of individuals in one state may also depend on the actions of other state governments. This
modification may also change the federal government’s behavior, both because of externality
effects on the demand for total government and due to changes in state government behavior.14
How the presence of externalities affects the relative structure of governments’ roles depends on
the way spil overs affect individuals and firms across state boundaries. Certain cases, like a local
access road, may have spil over effects that do not extend beyond the state, and thus can be
addressed by any level of government. In cases where externalities may spil over state lines,
however, state governments may overprovide negative-externality goods and underprovide
positive-externality goods. In these cases, a given state government does not consider the wel -
being of individuals outside its jurisdiction. When this occurs, the federal government may be
better suited to administer the program and increase total wel -being because it considers the
wel -being of al residents in its decisions. In practice, there may also be a threshold of interstate
spil over effects where state governments can be the primary administrators with relatively little
consequence.

14 T his ignores the presence of international economies, as discussed in the “ Conclusion” section.
Congressional Research Service

8

link to page 12 link to page 12 Fiscal Federalism: Theory and Practice

Practical Example: Excise Taxes
Excise taxes are one of the main ways governments can address goods with negative externalities.
By imposing taxes on such goods, the government may both discourage their use (because raising
the price of the good wil decrease its consumption) and raise revenue that can be used to address
any fiscal ramifications of their use on outside parties.15 Some of the most prominent excise taxes
in the United States are imposed on motor fuel, products containing tobacco and alcohol, aviation
fuel, and firearms, though not al of these are intended to address externalities. Externalities
represent only one motivation for excise taxes, of course, as the benefits of increased revenues are
potential y attractive to governments regardless of their effect on demand for certain activities.
Table 4 shows the distribution of excise tax receipts collected across levels of government in
2017. Some negative externalities, such as damage from motor fuel consumption and drunk-
driving costs, may impose externalities that are confined to the immediate area around the good’s
consumption, which could explain some of the state and local government use. In other cases, as
with general health care system stress caused by tobacco and alcohol consumption or motor fuel’s
contributions to pollution and global warming, the effects are more widespread, which could
serve as justification for a federal excise tax levy. In total, Table 4 shows that 70% of al 2017
excise tax receipts flowed to state and local governments with the remaining 30% going to the
federal government.
Table 4. Excise Tax Collections by Government Level, 2017
(Bil ions of Dol ars)

State and Local
State
Local
Federal
Excise taxes
191.3
157.1
34.2
83.8
Sources: Census Bureau, 2017 State & Local Government Finance Historical Datasets, and Office of Management
and Budget, Historical Tables, Table 2.1.
Notes: State and local excise tax col ections include amounts reported as selective sales taxes.
Government Effectiveness
The basic model assumed that federal and state governments are equal y effective in
administering tax and spending programs. There are reasons this may not be true in al cases, in
ways that lend advantages to either federal or state interventions. Governments may not always
have perfect information: for spending programs, identifying the intended recipients can be
difficult, whereas with certain taxes, the tax base may have incentives to avoid or evade levies.
Without perfect information, governments may not be able to comprehensively target their entire
jurisdictions, although the situation may be aided if some of the population that is difficult to
identify participates in other activities coordinated by the government.
Administrative costs may also differ across governments for certain programs, even when
considering a model that otherwise conforms to the basic model. Certain programs that would
lead to duplicative administrative costs if organized at the state level may lend themselves to
more centralized coordination. Two cities with recycling that can be collected in one day per
week, for instance, would save money on vehicle costs by pooling their resources and sharing a
vehicle fleet—or simply having a higher level of government coordinate the service.

15 For more on excise taxes, see CRS Report R43189, Federal Excise Taxes: An Introduction and General Analysis, by
Sean Lowry.
Congressional Research Service

9

link to page 13 Fiscal Federalism: Theory and Practice

The model can incorporate differences in effectiveness across governments by treating activities
as distinct goods, services, or taxes when administered by different governments. With this
adjustment, services provided by federal and state governments can differ both in how they affect
wel -being and in their cost. In other words, the model acknowledges that federal and state
governments do not provide identical goods and services, or provide them at identical costs, when
tasked with a particular activity.
The effect that these changes have on governments’ equilibrium behavior is general y intuitive. If
changes incorporate information and administrative costs that are higher for the federal
government, then the equilibrium condition wil shift to reduce the federal role and increase state
activity. If, instead, information and administration costs are higher for state governments, then
federal activity in equilibrium wil increase and state activity wil decrease.
Practical Example: Transportation
Table 5 examines three different types of transportation spending across levels of government.
Differences in the federal share of spending on these services may be a function of variations in
efficiency across governments. The use of highway and ground transportation services mostly
takes place at the local and regional level (e.g., school and work commutes and access to market
for local firms), so local and state governments are better able to structure these services to fit
users’ needs.
Water and air transportation, however, often involve transportation across state or international
lines. The use of these services is significantly more diffuse than highway and ground
transportation. Given the difficulties in administering interstate or international service provision,
the share of funds for these types of transportation is weighted more toward the federal
government.
Table 5. Transportation Spending by Government Level, 2017
(Bil ions of Dol ars)

State and Local
State
Local
Federal
Highways
182.0
109.8
72.2
45.8
Air
26.0
2.3
23.6
16.8
Water
6.4
2.1
4.3
4.4
Source: Census Bureau, 2017 State & Local Government Finance Historical Datasets, and Congressional Budget
Office Public Spending on Transportation and Water Infrastructure, 1956 to 2017.
Budget Imbalances
The final extension of the basic model considered in this report introduces budget imbalances,
al owing tax and spending levels to differ. Governments may wish to generate short-term budget
imbalances for any number of reasons. One reason is to smooth the effects of business cycles,
which describe the pattern of expansion and contraction of the economy. With strict balanced-
budget requirements, adverse economic conditions that lead to a drop in individual wages and
firm revenues may also require a cut in public spending to fal in line with decreased tax revenues
collected from individuals and firms. Such austerity measures are general y not supported by the
Congressional Research Service

10

link to page 15 link to page 21 Fiscal Federalism: Theory and Practice

economic literature, and reduce long-term wel -being under most fiscal federalism model
specifications.16
Another potential cause for short-term budget imbalances is mismatches in the timing of public
spending and revenues for certain programs. For example, a government may be interested in
building a bridge paid for with a toll levied on future (automobile) users. That project is likely to
generate most of its public costs earlier, as the bridge is constructed, and most of its revenues
later, when the bridge is in use. Modifying the project’s tax and spending structure so that budgets
are always balanced may sacrifice some of the public good principles discussed earlier, such as
spil over minimization.
One adjustment that can accommodate this change is to introduce multiple time periods to the
model. Incorporating more than one time period would provide for multiple sets of decisions for
each actor. It would also al ow for short-term budget deficits and surpluses (if the requirement is
also stretched over the course of multiple periods), which would al ow governments to incur
deficits (where spending exceeds revenues) in some periods, as long as they are balanced by
surpluses (revenues exceeding spending) in others. Economic business cycle effects can then be
included as determinants of some combination of firm productivity, wages, and government
revenue. This modification would also al ow for the inclusion of public projects with mismatched
spending and revenue patterns.
Another possible modification is to al ow budget deficits up to a certain level. For example,
raising permissible government spending from exactly the level of revenues to that level plus a
particular percentage effectively sets an al owable deficit limitation similar to those that exist in
the European Union and elsewhere.17 Regardless of the budget constraint’s duration or exact
value, the model needs some form of a budget restriction in order to produce useful outcomes.
Without a budget restriction in place, governments have no incentive to place any limit on the
amounts they can spend, and thus the relative levels of federal and state spending cannot be
identified.
In general, the more overlap there is in economic conditions across states, or the greater the
simultaneous spread of recession (or expansion) at once to al areas of the country, the greater the
benefit there is to having the federal government take on borrowing responsibilities. The
preference for relaxed federal budgeting is particularly strong in models that have free mobility of
individuals and firms. The reasoning for this is similar to the incentives described earlier: states
may “race to the bottom” in adjusting budgets to economic shocks to avoid individual and firm
emigration. Coordination costs not accounted for in the model lend additional support to federal
budget deficits.
Practical Example: Government Spending in Recessions
Figure 1 shows the share of total government spending attributable to the federal and combined
state and local levels from FY2005 through FY2012. In the years prior to the Great Recession, the
distribution of spending was roughly in line with historical trends (as shown in Figure B-1). The
state and local share then took a precipitous drop during the Great Recession, averaging 38%
between 2009 and 2012. The need for many state and local governments to balance budgets

16 For further discussion, see Alan S. Blinder, “Keynes, Lucas, and Scientific Progress,” American Economic Review,
vol. 77, no. 2 (May 1987), pp. 130-136; and Alan J. Auerbach and Yuriy Gorodnichenko, “ Effects of Fiscal Shocks in a
Globalized World,” IMF Econom ic Review, vol. 64, no. 1 (2016), pp. 177-215.
17 Details on the requirements in the European Union can be found in European Central Bank, “Five things you need to
know about the Maastricht T reaty,” February 15, 2017, available at https://www.ecb.europa.eu/explainers/tell-me-
more/html/25_years_maastricht.en.html.
Congressional Research Service

11

link to page 16

Fiscal Federalism: Theory and Practice

during this period contributed to the decline, stemming largely from reduced spending to offset a
loss in tax revenues resulting from decreases in property values, lower consumer spending, and
high unemployment. The federal government, meanwhile, enacted a number of countercyclical
measures to increase spending during this time, including direct aid to state and local
governments through the creation of a State Fiscal Stabilization Fund. Federal deficits averaged
9.0% of GDP from FY2009 to FY2011, the largest recorded values since the end of World War II.
Figure 1. Federal and Combined State and Local Expenditures FY2005-FY2012
(as a percentage of combined expenditures)

Source: Federal Reserve Bank of St. Louis.
Note: Shaded area denotes period of the Great Recession, as identified by the National Bureau of Economic
Research.
Early evidence suggests that the federal government wil record a similar uptick in the percentage
of total expenditures in the period surrounding the COVID-19 crisis. That increase is partial y
attributable to several pieces of enacted legislation that increased short-term federal spending.
Relevant provisions in those laws included the creation of the Coronavirus Relief Fund,18 which
provided $150 bil ion in direct federal assistance to state and local governments, and the
establishment of the Municipal Lending Facility,19 which provided the Federal Reserve with the
capacity to support up to $500 bil ion in outstanding state and local government debt.
Conclusion
This report developed a basic model of fiscal federalism to explore the economic rationale behind
assigning certain types of public activities to various levels of government, as summarized in
Table 6. In certain cases, practical aspects of economic behavior (e.g., preference variation and
budget imbalances) have clear implications for the type of government that would produce the
most benefit: in other situations (e.g., mobility of businesses and individuals, externalities), the

18 For more information on the Coronavirus Relief Fund and a brief discussion of the State Fiscal Stabilization Fund,
see CRS Report R46298, The Coronavirus Relief Fund (CARES Act, Title V): Background and State and Local
Allocations, by Grant A. Driessen.
19 For more information on the Municipal Lending Facility and policy issues related to state and local debt, see CRS In
Focus IF11502, State and Local Governm ent Debt and COVID-19, by Grant A. Driessen.
Congressional Research Service

12

Fiscal Federalism: Theory and Practice

economic incentives for fiscal federalism are highly dependent on context. (In addition, the
overal implications of fiscal federalism may depend on examining how some of the
aforementioned model extensions interact with one another, which is beyond the scope of this
report.) As shown in the practical examples, fiscal federalism choices may have ramifications
across the policy spectrum.
Table 6. Summary of Relative State or Federal Government Responsibilities from
Extended Assumptions to the Basic Model of Fiscal Federalism
Implication of Relative Responsibility of Level of
Factor
Government
Varying Preferences
State role increases (and federal role decreases) with
level of variation across states.
Mobility
Depends on context. High mobility would exacerbate
variation in preferences and result in a larger state role,
or it may result in a “race to the bottom” implying a
larger federal role.
Spatial Effects
Depends on context. Highly concentrated spatial effects
tend to increase state role, while broadly dispersed
effects tend to increase federal role.
Externalities
Depends on the level of these effects across state lines.
The federal government may more efficiently address
large spil over effects that spread far into other
jurisdictions, whereas other cases are ambiguous.
Government Effectiveness
Increase in the role of the government with relatively
smal er information and administration costs.
Budget Imbalances
Federal role increases as there is more variation in
economic conditions across states. Federal role may also
increase with relatively greater flexibility in budget
outcomes (as is true in the United States).
Source: CRS analysis.
Several practical adjustments are not considered as part of the exercise extending the basic model,
some of which are at least partly addressed in existing research. Although most fiscal federalism
research assumes that federal, state, and local government operations are distinct, multilevel
coordination can and often does occur for many programs.20 Such cooperation can include
intergovernmental transfers, collaborations where each government type makes some sort of
contribution or collection, or intergovernmental loans. Historical y, fiscal cooperation efforts
included the general revenue sharing (GRS) program that was active from 1972 through 1986.21
Under GRS, state and local taxing authority was redirected to the federal government, which
would then al ocate tax revenues to state and local governments as unconditional grants using a
formula based on population, tax effort, and total personal income.
The model also omits the presence of foreign governments, which may affect domestic economic
conditions and present competition to federal and state governments for resources. In a sense,

20 For more on how cooperation between the federal government and state and local governments operates in practice,
see CRS Report R40638, Federal Grants to State and Local Governm ents: A Historical Perspective on Contem porary
Issues, by Robert Jay Dilger and Michael H. Cecire.
21 T he GRS program was established by the Local Fiscal Assistance Act (P.L. 92-512). For more on the GRS program,
see CRS Report RL31936, General Revenue Sharing: Background and Analysis, by Steven Maguire.
Congressional Research Service

13

Fiscal Federalism: Theory and Practice

international actors may be thought of as potential competitors for the federal government, and
thus may reduce the potential benefit of federal involvement in al eviating externalities and other
inefficiencies. Final y, this basic framework adopts an approach that forgoes more sophisticated
models of business activity, which can be found in the academic literature and which add
complexity to equilibrium findings.22 Despite these limitations, fiscal federalism research
provides a basis for evaluating the effectiveness of activity across levels of government that can
be useful in designing and implementing public policy.

22 For example, see Rainald Borck and Michael Pfluger, “Agglomeration and tax competition,” European Economic
Review, vol. 50, no. 3 (September 2006), pp. 647 -668.
Congressional Research Service

14

Fiscal Federalism: Theory and Practice

Appendix A. Theoretical Framework
Basic Model
A theoretical representation of the initial model is shown in Equations 1 through 4. Because
individuals are identical across jurisdictions, these choices converge to the maximization
constraint provided in Equation 1. Individuals derive utility (U) from their own private
consumption and from the combination of any goods they receive from federal, state, and local
governments, al purchased with their after-tax income.
Sufficient federal and state taxes are levied to meet the balanced budget requirement (Equations
2 through 4). The federal government chooses a level of spending (C) and taxes (T) that applies
equal y to al jurisdictions, while the spending and tax levels (c, t) of state governments—which,
for the moment, combine to form one representative actor—apply exclusively to their residents.
(1) Max U = f(G)
(2) G = C + c = T+t
(3) C = T
(4) c = t
As shown in Equations 5 and 6, equilibrium (denoted with asterisks) can be achieved with any
level of federal and state activity that leads to total government activity at the optimum (G*),
which is then consumed by the individual to maximize utility.
(5) U* = f(G*)
(6) C + c = G*
Varying Preferences
Accommodations for varied preferences in the model are provided in Equations 7 through 9.
The decisions facing federal and state governments are now distinct. State governments
(Equation 7), identified with j, are each tasked with choosing a level of spending (cj) that
maximizes the utility of their unique set of individuals, subject to their budget constraint
(Equation 8). The federal government chooses a level of spending applicable to all jurisdictions
that maximizes total utility (Equation 9), while meeting its budget constraint (Equation 10).
(7) Max Uj(cj) = fj(Gj)
(8) cj = tj for al j states
(9) Max ∑j Uj(C) = fj(Gj), where ∑j is the sum across al j states
(10) C=T
Al owing preferences to vary generates an equilibrium that places an upper limit on the level of
federal government intervention that wil maximize public wel -being. Since individuals stil only
have preferences for total government services and not which government is acting, the federal
government can contribute to utility optimization so long as it restricts its total activity to the
minimum level of intervention desired when looking across al states (Equations 11 and 12).
(11) C*< Minj(Gj)
(12) Gj = C + cj
Congressional Research Service

15

Fiscal Federalism: Theory and Practice

Mobility
Al owing for public mobility across states introduces individual choice to the model, accounted
for in Equation 13. Each individual (i) selects the state of jurisdiction with the combination of
spending and taxes that maximizes individual utility (u). Al else equal, those preferences tend to
produce clustering of similarly dispositioned individuals, which exacerbates differences in
preferences across states.
(13) Max ui(j) = f(cj, tj) for al individuals
Though the decision set facing state governments is unchanged, the lack of certainty over the
number of residents in their jurisdiction can lead to modifications in government behavior.
Research has found that, under most specifications, mobility can cause underprovision of
government services (c below optimal levels) across competitors, in this case state governments.23
In a model with a federal actor, such outcomes may be offset by increases in federal activity (C).
Spatial Effects
Cost differentials that are a function of spatial considerations (including proximity to a service
center) require a different adjustment. In these cases, rather than the level of services provided or
revenues collected depending on the level of government application, the nature of the services or
revenues themselves changes with where the provisions are applied. In other words, the model
considers the provision of one good by the federal government to be different from the provision
of that same good by a state government. Formulaic representation of that shift is seen in
Equations 14 and 15, with the utility function now separating federal and state goods and
services to account for this distinction. Theoretical research has found the effects of spatial
variation to be highly dependent on model specifications and the level and type of variation
present.24
(14) Max Uj(cj) = fj(cj, C)
(15) Max ∑j Uj(C) = fj(cj, C)
Externalities
Accounting for externalities also involves a modification to the utility function that governments
are seeking to maximize. The model now must account for the fact that the wel -being of
residents in one state is affected by the decisions made by other state governments. This shift is
represented in Equations 16 and 17, which incorporate the addition of other state government
outcomes (coth) on state and federal actors, respectively.
(16) Max Uj(cj) = fj(cj, C, coth)
(17) Max ∑j Uj(C) = fj(cj, C, coth)
Government Effectiveness
Administrative cost differentials are incorporated into the model by modifying the balanced
budget conditions, as shown in Equations 18 and 19. Government spending is now the product of
revenues collected and a multiplier (z for state governments, Z for the federal government) that

23 John Douglas Wilson, “T heories of T ax Competition,” National Tax Journal, vol. 52, no. 2 (June 1999) pp. 269-304.
24 Charles M. T iebout, “An Economic T heory of Decentralization,” Public Finances: Needs, Sources, and Utilization,
National Bureau of Economic Research, 1961, p. 79.
Congressional Research Service

16

Fiscal Federalism: Theory and Practice

accounts for frictional costs in the application of spending or collection of revenues. Situations
where the state government has an efficiency advantage would be characterized by z<Z, while
instances with a federal efficiency advantage would produce the opposite relationship, z>Z.
(18) cj = ztj
(19) C = ZT
Budget Imbalances
Equations 18 and 19 can also accommodate the introduction of government imbalances to the
model. Cases where z, Z >1 may be taken to represent deficits, as government spending (c, C) is
otherwise greater than taxes (t, T), while z, Z<1 would describe fiscal surpluses.
Alternatively, short-term budget imbalances but a long-term binding constraint are accounted for
through the introduction of multiple time periods, which provide governments with the
opportunity to incur deficit spending in one period that is offset by matching surpluses in other
periods. Equations 20 through 24 introduce two time periods to the initial model.
(20) Max U1 + U2 = f(G1) + f(G2)
(21) G1 = C1 + c1
(22) G2 = C2 + c2
(23) C1 + C2 = T1 + T2
(24) c1 + c2 = t1 + t2
Congressional Research Service

17

link to page 21 link to page 22 link to page 23 link to page 22 link to page 23

Fiscal Federalism: Theory and Practice

Appendix B. Recent Trends and Levels in Federal
and State and Local Activity
Figure B-1 shows the balance of spending across federal, state, and local levels since 1970. State
and local spending has general y comprised 35%-45% of al government spending in the United
States. The state and local share was at its lowest point during the 1970s and 1980s. After rising
through the 1980s and 1990s, the state and local government share declined through much of the
2000s. The state and local share then experienced a more significant drop after the Great
Recession, averaging 38% between 2009 and 2012. This drop reflects the federal government’s
stimulus spending during and immediately following the recession, as wel as the loss in tax
revenues to state and local governments resulting from decreases in property values, lower
consumer spending, and high unemployment.
Figure B-1. Federal and State and Local Expenditures, 1970-2019
(as a percentage of combined annual expenditures)

Sources: U.S. Bureau of Economic Analysis, State and Local Government Current Expenditures and Federal
Government Current Expenditures retrieved from FRED, Federal Reserve Bank of St. Louis. Percentage
calculation by Congressional Research Service.
Note: Shaded areas denote years with an economic recession, as identified by the National Bureau of Economic
Research.
Table B-1 and Table B-2 offer more detail on the composition of revenues and spending across
major categories for the federal government and the governments (state and local combined) in
each state. As shown in Table B-1, state and local governments derive their revenues from a
range of general sources, including federal transfers, whereas the federal government col ected
just over half of its receipts from taxes on individual and corporate income in 2017. Table B-2,
meanwhile, shows a much greater spending focus on education among state and local
governments, and on other expenditures (of which health expenditures are the most prominent) at
the federal level.
Congressional Research Service

18

Fiscal Federalism: Theory and Practice

Table B-1. Federal and State and Local Revenues by Source, 2017
(as percentages of total revenue)
Property
General
Income
Federal
Other
State
Tax
Sales Tax
Taxes
Transfers
Revenue
Federal
0.0%
0.0%
52.6%
n/a
47.4%
All State and Local
13.4%
9.9%
11.2%
18.1%
47.4%
Alabama
5.7%
10.2%
8.6%
22.0%
53.6%
Alaska
11.0%
1.6%
0.6%
24.6%
62.1%
Arizona
11.5%
15.4%
6.7%
23.6%
42.8%
Arkansas
7.1%
14.5%
10.1%
26.1%
42.2%
California
10.3%
8.0%
15.3%
16.4%
50.1%
Colorado
14.0%
12.1%
11.8%
15.9%
46.2%
Connecticut
21.8%
8.6%
17.9%
16.4%
35.4%
Delaware
7.4%
0.0%
11.9%
20.5%
60.2%
District of Columbia
16.2%
9.1%
16.7%
26.7%
31.3%
Florida
14.5%
14.4%
1.2%
15.6%
54.2%
Georgia
13.3%
10.2%
13.1%
17.9%
45.5%
Hawai
8.7%
17.0%
11.2%
15.2%
47.9%
Idaho
11.3%
10.7%
12.1%
17.9%
48.1%
Il inois
19.3%
9.7%
10.9%
15.6%
44.6%
Indiana
11.0%
12.0%
11.3%
24.0%
41.6%
Iowa
12.4%
8.6%
10.2%
14.9%
53.9%
Kansas
14.0%
13.4%
8.5%
13.8%
50.3%
Kentucky
8.2%
7.7%
14.4%
26.9%
42.8%
Louisiana
8.1%
16.5%
6.2%
24.9%
44.3%
Maine
19.8%
10.0%
11.8%
21.2%
37.2%
Maryland
13.3%
6.3%
21.0%
19.2%
40.2%
Massachusetts
17.5%
6.5%
17.7%
18.0%
40.2%
Michigan
13.2%
8.6%
10.5%
21.4%
46.4%
Minnesota
12.1%
8.3%
16.6%
16.0%
46.9%
Mississippi
9.3%
10.6%
6.7%
26.0%
47.3%
Missouri
10.4%
10.5%
11.4%
19.6%
48.2%
Montana
14.1%
0.0%
11.0%
27.7%
47.2%
Nebraska
15.3%
9.1%
10.1%
14.1%
51.4%
Nevada
9.5%
17.4%
0.0%
17.7%
55.3%
New Hampshire
31.6%
0.0%
4.5%
17.2%
46.6%
New Jersey
25.0%
8.2%
13.8%
15.6%
37.4%
New Mexico
6.4%
12.9%
5.5%
28.4%
46.8%
New York
15.0%
8.0%
17.6%
17.6%
41.8%
Congressional Research Service

19

Fiscal Federalism: Theory and Practice

Property
General
Income
Federal
Other
State
Tax
Sales Tax
Taxes
Transfers
Revenue
North Carolina
9.7%
10.2%
12.4%
19.8%
48.0%
North Dakota
11.7%
10.4%
3.6%
18.2%
56.2%
Ohio
11.0%
10.8%
10.3%
18.7%
49.2%
Oklahoma
7.7%
12.4%
8.8%
20.4%
50.7%
Oregon
10.6%
0.0%
15.7%
18.5%
55.2%
Pennsylvania
13.1%
7.7%
13.3%
21.6%
44.3%
Rhode Island
19.3%
7.6%
10.4%
21.9%
40.8%
South Carolina
11.4%
7.2%
8.5%
18.7%
54.1%
South Dakota
15.6%
15.8%
0.3%
18.3%
50.0%
Tennessee
9.4%
14.9%
3.1%
19.0%
53.6%
Texas
19.2%
14.6%
0.0%
16.6%
49.6%
Utah
10.0%
10.3%
12.3%
15.7%
51.7%
Vermont
20.3%
4.8%
10.4%
24.9%
39.7%
Virginia
15.2%
5.9%
15.1%
12.9%
50.9%
Washington
11.1%
18.3%
0.0%
15.6%
55.0%
West Virginia
8.7%
6.9%
9.7%
25.9%
48.8%
Wisconsin
14.9%
8.7%
13.6%
19.5%
43.2%
Wyoming
12.5%
6.7%
0.0%
31.8%
49.0%
Sources: Census Bureau, 2017 State & Local Government Finance Historical Datasets and Tables, and Bureau of
Economic Analysis, National Income and Product Accounts.
Notes: CRS calculations. State and local government figures draw from Census data, while federal results draw
from Bureau of Economic Analysis data.
Table B-2. Federal and State and Local Expenditures by Function, 2017
(as percentages of total expenditures)
Defense and
Environment
Other
State
Education
Social Services
Transportation
Public safety
and Housing
Expenditures
Federal
3.1%
24.2%
2.4%
15.0%
2.3%
53.0%
All State and
27.9%
26.6%
5.9%
7.0%
5.8%
26.8%
Local
Alabama
29.1%
32.3%
5.7%
5.4%
4.3%
23.2%
Alaska
22.2%
22.0%
13.0%
6.5%
6.9%
29.4%
Arizona
26.1%
27.5%
5.2%
9.5%
5.0%
26.7%
Arkansas
30.4%
32.7%
7.9%
5.7%
5.0%
18.4%
California
23.4%
29.1%
3.9%
8.2%
6.1%
29.3%
Colorado
29.1%
19.8%
6.8%
7.2%
7.1%
29.9%
Connecticut
33.3%
15.5%
5.7%
6.2%
5.3%
34.0%
Congressional Research Service

20

Fiscal Federalism: Theory and Practice

Defense and
Environment
Other
State
Education
Social Services
Transportation
Public safety
and Housing
Expenditures
Delaware
33.6%
25.8%
7.6%
6.8%
5.1%
21.1%
District of
Columbia
16.2%
25.0%
2.6%
6.4%
9.6%
40.3%
Florida
24.5%
24.7%
8.4%
10.1%
9.0%
23.4%
Georgia
33.2%
22.6%
6.4%
7.3%
5.9%
24.6%
Hawai
20.5%
25.6%
7.2%
6.1%
8.6%
31.9%
Idaho
27.5%
26.3%
7.2%
8.8%
8.2%
22.0%
Il inois
26.3%
19.1%
7.4%
7.0%
5.6%
34.6%
Indiana
31.0%
32.9%
4.6%
6.0%
4.9%
20.5%
Iowa
31.9%
28.6%
8.2%
4.6%
5.4%
21.3%
Kansas
32.5%
27.6%
6.7%
6.1%
5.0%
22.2%
Kentucky
28.3%
32.6%
5.7%
5.0%
4.7%
23.7%
Louisiana
25.6%
30.0%
6.3%
7.2%
7.2%
23.7%
Maine
28.0%
29.0%
8.2%
6.1%
6.7%
21.9%
Maryland
30.1%
22.1%
5.7%
8.7%
8.1%
25.2%
Massachusetts
25.3%
27.7%
4.9%
5.5%
5.7%
30.9%
Michigan
31.5%
27.0%
5.0%
6.5%
6.0%
23.9%
Minnesota
29.3%
26.7%
7.9%
5.5%
6.8%
23.8%
Mississippi
27.4%
36.0%
6.2%
5.5%
4.3%
20.5%
Missouri
28.1%
28.8%
4.9%
6.7%
5.8%
25.6%
Montana
27.4%
25.9%
8.6%
6.9%
7.0%
24.1%
Nebraska
32.0%
18.2%
7.3%
5.6%
5.0%
31.9%
Nevada
26.0%
20.9%
9.1%
10.6%
6.4%
27.0%
New Hampshire
33.3%
20.9%
6.4%
7.6%
5.4%
26.4%
New Jersey
33.6%
21.1%
4.6%
6.5%
4.8%
29.3%
New Mexico
27.9%
32.0%
4.1%
7.4%
5.2%
23.5%
New York
24.8%
27.2%
4.3%
6.3%
4.9%
32.6%
North Carolina
29.6%
31.4%
5.9%
7.1%
5.5%
20.5%
North Dakota
28.2%
16.9%
17.0%
5.8%
8.3%
23.8%
Ohio
28.1%
28.9%
5.1%
6.3%
5.2%
26.4%
Oklahoma
30.6%
26.4%
8.0%
7.1%
5.1%
22.7%
Oregon
26.0%
27.9%
5.0%
7.2%
5.8%
28.2%
Pennsylvania
28.8%
29.4%
7.1%
5.9%
4.9%
24.0%
Rhode Island
27.2%
26.8%
4.3%
8.0%
5.5%
28.2%
South Carolina
30.2%
32.1%
5.7%
5.5%
4.8%
21.7%
Congressional Research Service

21

Fiscal Federalism: Theory and Practice

South Dakota
30.6%
18.9%
13.0%
6.2%
8.1%
23.3%
Tennessee
24.7%
27.9%
4.6%
6.7%
5.3%
30.8%
Texas
34.2%
24.0%
7.0%
7.1%
4.4%
23.3%
Utah
33.1%
22.4%
7.4%
5.6%
5.4%
26.0%
Vermont
35.3%
27.7%
8.3%
5.9%
5.5%
17.3%
Virginia
32.5%
23.4%
7.5%
8.1%
5.8%
22.7%
Washington
27.3%
25.2%
6.1%
6.5%
7.3%
27.7%
West Virginia
29.5%
31.5%
6.9%
5.1%
5.3%
21.7%
Wisconsin
29.3%
20.4%
8.8%
3.0%
1.1%
37.5%
Wyoming
30.1%
8.5%
6.7%
2.2%
1.9%
50.6%
Sources: Census Bureau, 2017 State & Local Government Finance Historical Datasets and Tables, and Office of
Management and Budget, Historical Tables, Table 3.2.
Notes: CRS calculations. State and local government figures draw from Census data, while federal results draw
from Office of Management and Budget data. State and local figures taken from calendar year 2017, while federal
figures are from fiscal year 2017.

Author Information

Grant A. Driessen
Joseph S. Hughes
Analyst in Public Finance
Research Assistant

Disclaimer
This document was prepared by the Congressional Research Service (CRS). CRS serves as nonpartisan
shared staff to congressional committees and Members of Congress. It operates solely at the behest of and
under the direction of Congress. Information in a CRS Report should n ot be relied upon for purposes other
than public understanding of information that has been provided by CRS to Members of Congress in
connection with CRS’s institutional role. CRS Reports, as a work of the United States Government, are not
subject to copyright protection in the United States. Any CRS Report may be reproduced and distributed in
its entirety without permission from CRS. However, as a CRS Report may include copyrighted images or
material from a third party, you may need to obtain the permission of the copyright holder if you wish to
copy or otherwise use copyrighted material.

Congressional Research Service
R46382 · VERSION 1 · NEW
22

Download PDF

Download EPUB

Revision History

Jun. 3, 2020

HTML · PDF

Metadata

Topic areas

Economic Policy

Report Type: CRS Report

Source: CRSReports.Congress.gov

Raw Metadata: JSON