INTRODUCTION
In Showdown at Gucci Gulch (Birnbaum and Murray, 1987), chronicling
the passage of the Tax Reform Act of 1986, the issues of state and local
tax preferences are front and center. The lobbying effort to remove from
the Administration's proposal the repeal of state and local tax
deductions was "one of the most persistent and pervasive lobbying
campaigns of the tax reform story" (p. 113). That effort succeeded
in preserving, in this watershed tax reform, the full deduction in the
House, and all but the sales tax deduction in the Senate. The final
proposal eliminated the sales tax deduction, but in 2004 the deduction
was restored as an option to the income tax deduction.
In addition, tax exemption for state and local bonds "was one
of the most intensely lobbied parts of the tax bill" (p. 137). The
debate, however, was not about the general exemption, but the alarming
growth in private activity bonds, which, despite previously enacted
restrictions, had surged to two-thirds of the bond market. And even in
this case, numerous carve-outs for many types of activities, as well as
capped small issue bonds, remained. Although private activity bonds were
included in the minimum tax, the general tax exempt preference was not.
As the tax reform drama played out, we saw the federal government as
protective of the states' tax benefits and slow to react even when
these benefits were being abused--an indulgent approach. At the same
time, as we show in the following discussion, in many cases federal tax
revisions have taken place with consequences for the states, but with
little attention to states' concerns.
In order to review both the ways in which federal tax policy
affecting the states is made and how important those effects are, it is
useful to sort the interaction between federal tax policy and state and
local government activity into four broad categories. The first is the
explicit preferences granted state and local governments in the federal
tax law. The second is indirect effects on the costs of financing state
and local activities from apparently unrelated tax provisions. The third
is how tax changes interact with the states' desires to conform
their tax systems to the federal system to simplify administration and
compliance of their own tax systems. Finally, there are federal
restrictions on state tax policy; recent provisions are tracked by the
Congressional Budget Office under the Unfunded Mandates Reform Act of
1995. Following the discussion of these interactions, the concluding
section discusses implications for the attitudes of federal policymakers
towards state and local governments in formulating tax policy and the
outlook for future changes.
EXPLICIT PREFERENCES IN TAX LAW
The deduction for state and local taxes and exemption of interest
on tax exempt bonds have been in the income tax since its beginning and
together constitute the main tax expenditures for state and local
activity in the federal income tax. As shown in Table 1, out of the $80
billion of official tax expenditures associated with state and local
governments, slightly over half (55 percent) are itemized deductions for
state and local taxes, and virtually all of the remainder are with tax
exempt bond interest. (Note that it is not strictly correct to add the
preferences, particularly for itemized deductions, because of
interactions.) The tax expenditure list allocates these expenditures by
functional category and does not classify the property tax or the
exemptions for interest on private activity bonds as supporting state
and local government activity.
Two additional items, not in the official tax expenditure list,
could be considered subsidies. One is the deduction from the federal
income tax base for state and local corporate income taxes, estimated at
approximately $20 billion. If corporate income tax deductions were
considered a preference, and a case can be made for doing so, the cost
would rise by $20 billion, to $100 billion, with tax deductions
accounting for almost two-thirds of the total.
The second item that could be considered an income tax expenditure
is the exemption from income tax of the profits of state and local
business enterprise. These activities are also activities undertaken by
the for-profit sector. The Congressional Budget Office (CBO) has
estimated, for FY 2002, sales of $287 billion associated with these
enterprises (CBO, 2005). The largest share of these sales is for
hospitals (23 percent), higher education (21 percent), electric power
(19 percent), and water (12 percent). Other activities include gas
utilities, lotteries, air transport, solid waste management, parks and
recreation, toll highways, liquor stores, special assessments, water
transport, and parking. The benefit of excluding this income from the
federal corporate tax is uncertain; CBO (2007) has estimated a revenue
gain of $0.8 billion from taxing electric utilities; if this same amount
relative to revenue occurred for other business enterprises, the total
benefit would be $4.2 billion. This number may be too large given the
capital intensity of electric power production, but may be understated
because of direct funding for activities such as higher education. Given
the lack of precedent in taxing these activities, the small size of the
effect, and the conferring of similar benefits on the tax-exempt sector,
this area seems one unlikely to be subject to legislation. Nevertheless,
CBO (2007) included a provision for taxing state electric utilities,
where no non-profit counterparts exist, in their budget options study.
Prior estate tax law allowed a credit against the federal estate
tax, up to a limit, for state death taxes. If the estate tax repeal
enacted in 2001 is allowed to expire, the state death tax credit, which
was repealed by the end of 2004, would be restored. In 2001, this death
tax credit was $6.3 billion. The credit was actually a free ride for the
states, since each dollar of the death credit up to the limit in the tax
law (and states generally constructed their taxes to conform to those
limits) reduced the federal tax dollar for dollar. The current smaller
estate tax now has a deduction, which is less valuable because it
reduces the tax by taxes paid times the estate tax rate, and many states
no longer have a death tax credit.
The repeal of the estate tax and the elimination of the state death
tax credit are examples entirely different from the treatment of
preferences in the 1986 reform act. In this case, virtually no attention
was paid to the implications for the states. Indeed, the state death tax
credit was phased out more rapidly than the estate tax itself. On the
other hand, a dollar-for-dollar state death tax credit is an
extraordinarily generous federal tax provision that involves a direct
transfer of federal revenues to the states, limited only by the size of
the federal estate tax.
For a variety of reasons, many believe that the complete repeal of
the estate tax may not be feasible. In 2006, a number of proposals were
made to substitute a much smaller permanent estate tax for the
elimination of the estate tax. Interestingly, elimination of the
deduction of state death taxes in computing the federal estate tax was
also proposed to limit the revenue impact, a case where tax policy
makers were pursuing their own interests without retaining any benefit
for the states.
The remainder of this subsection discusses the two major tax
preferences--deductibility of state and local taxes and tax exempt
interest--in more detail.
Deductibility of State and Local Taxes
Tax subsidies for state and local itemized deductions amounted to
$44 billion in FY 2007, or about six percent of state and local tax
receipts, based on receipts reported in the Economic Report of the
President, 2006. The subsidies vary substantially both by tax and by
state (and locality). The largest relative subsidy is for individual
income taxes, where the tax expenditure is 11 percent of state and local
individual income tax collections. For property taxes, the subsidy is
much less, four percent, because property taxes also apply to businesses
and owners of rental property (where the tax would be deducted as a
business expense in any case) as well as owner-occupied homes. The
additional benefit due to the sales tax deduction is less than one
percent of sales and gross receipt taxes. This small effect occurs
because the current sales tax deduction is allowed as a substitute for
income tax deductions and the benefit is measured as the additional
deduction allowed because of this option, which tends to be small.
(Because disaggregated state and local tax data are only available for
2004, and aggregate data, only for 2006, these estimates assume each
revenue source grew at the same rate from 2004 to 2006, and by aggregate
GDP growth from 2006 to 2007.)
A curious issue is why deductions for state and local income taxes
by corporations are not considered a tax subsidy. The general rule for
determining whether a tax deduction constitutes an income tax
expenditure is whether it reduces the taxable base for purposes of the
federal income tax, or more broadly, whether it results in a
mismeasurement of income. The exceptions to this rule are provisions
such as personal exemptions and standard deductions that constitute part
of the progressive structure of the federal income tax and items whose
inclusion is not practical (such as imputed net rent on owner-occupied
housing). Deducting state and local corporate profit taxes reduces
corporate income in the same way that state and local individual income
tax deductions reduce individual income. The notion that state and local
corporate taxes are part of the firm's cost structure is not
persuasive. There seems no more reason that state and local taxes on the
profits of corporations should be considered a necessary cost of earning
income than is the case of unincorporated businesses or, for that
matter, wage earners or investors. Indeed, many of the firms that have
become unincorporated in recent years, due to the rapid growth of
Subchapter S firms, now have their tax deduction treated as a preference
when the same view is not applied to incorporated business. Mere legal
form should not alter the view of a subsidy. And the revenue loss from
state and local corporate tax deductions is considerable. According to
Census data, state and local corporate taxes were $57.5 billion in 2006.
At a 35 percent tax rate (and one presumes that firms with state and
local corporate profit taxes were taxable for federal purposes), the
value of the subsidy is $20 billion.
This view of the corporate state and local income tax deduction as
a subsidy, although compelling on economic grounds, has not been
included in proposals, such as those by the administration in 1986 or
the recent President's Advisory Panel (2005) proposals.
Note that the this argument for counting tax deductions as tax
expenditures cannot be as made for property taxes or sales taxes. In
measuring income, sales taxes are appropriately deducted as a cost by
firms, whether corporate or unincorporated, in determining profit; the
subsidy occurs at the individual level when the sales tax is deducted a
second time as an itemized deduction by consumers. Property tax
deductions by firms and landlords are similarly an appropriate deduction
because they are a cost that arises before income is determined, and
these deductions are not considered subsidies. The property tax
deduction for owner-occupied housing is a subsidy, not because of the
property tax per se, but because imputed rent is not included in income.
Were imputed gross rent to be included in income, the costs would be
appropriately deductible. Whether property tax deductions on
owner-occupied homes are treated as a subsidy to state and local
activities or a subsidy to home ownership is, however, a legitimate
question.
Many economists and tax reformers are critical of deductions for
state and local taxes. State taxes provide untaxed benefits, and an
income tax that excludes these payments from the base fails to tax
comprehensive income. There is an alternative view, however: that they
are mandatory payments and should be deducted on those grounds and that
costs and benefits do not match closely. The deduction also distorts the
level of state and local services and the mix of taxes and fees. Thus,
there are two very different types of issues surrounding the
deductibility of state and local taxes: those of fairness in the
distribution, and those of incentives.
These tax deductions reduce both revenue and the progressivity of
the income tax, but, assuming the rate structure could be adjusted to
offset these effects as was the case in 1986, to many the important
equity issue is the disparate treatment of taxpayers in different
states. Although itemized deductions are taken by 35 percent of
taxpayers overall, the share itemizing varies substantially across the
states (Maguire, 2006). In 2004, the shares itemizing varied from almost
50 percent in Maryland to 17 percent in West Virginia. Only Maryland had
a share above 45 percent, but eight states and the District of Columbia
had shares of 40 percent to 45 percent. Of the nine states without a
broad income tax, only three (Nevada, New Hampshire, and Washington)
were among the states with a share above the average, and all of those
fell into the 35 percent to 40 percent itemizing. Reuben (2005),
studying the issue using data from 2002 before the sales tax option was
added, concludes that itemization occurs at higher rates with states
with high income and wealth and progressive income taxes. These
characteristics persist with the sales tax deduction. According to data
in Maguire (2006), only a third of states without an income tax had an
itemization rate above the average, while over one-half of those with an
income tax did. Thus, even with the optional sales tax deduction in
place, having an income tax made itemization more likely.
The addition of the sales tax deduction, which was extended but
remains temporary, primarily benefited taxpayers in the states without
an income tax (Alaska, Florida, Nevada, New Hampshire, South Dakota,
Tennessee, Texas, Washington, and Wyoming). These states only accounted
for half of the sales tax deductions claimed, but many are small states,
and in the case of other states with an income tax, the net benefit (in
excess of the income tax option) is likely small.
One policy concern is that the deduction increases the size of the
state and local sector by reducing the state and local tax burden for
itemizers. At the same time, some applaud the preferences for income
taxes and property taxes, which tend to be more progressive than sales
taxes, and even with the sales tax option, the deduction of income taxes
dominates the deduction for sales taxes ($202 billion in income taxes
compared to $19 billion in sales taxes in 2004). Arguments are also made
that federal deductibility mutes distortions in interstate location
decisions as a result of differential taxes. While there are concerns
that the deduction influences the magnitude of the state and local
sector and the mix of taxes, most studies find mixed effects. The two
most recent studies, Stotsky (1990) and Courant and Gramlich (1990),
like most of the previous ones (which they reviewed), tend to find
modest effects. Reuben (2005) points out that no state lowered its sales
tax rates after the 1986 revision eliminating deductibility and 15
states had a higher sales tax rate in 1989 than in 1985. No state
without an income tax introduced one to substitute for sales tax
revenues. However, an argument has been made that the lowering of
marginal tax rates was a more serious reduction in the value of income
tax deductions than of state and local tax deductions because income
taxes were more important for the high-income taxpayers with the largest
rate reductions (Stark, 2004). Chirinko and Wilson (2005) review the
evidence on interstate substitution, which suggests limited effects of
state and local tax differentials on location decisions. These results
may suggest to federal tax policy makers that the deduction has no
significant distorting effects on states' mix or level of taxes,
and should not be an important issue in the consideration of these
policies. Concern may instead be directed at equity issues and, if so,
the likelihood of retaining the sales tax deduction appears high.
Equity issues may also factor in a number of legislative proposals.
Arguments have been made that the elimination of state and local tax
deductibility is a good offset for eliminating the Alternative Minimum
Tax (AMT), where the major preference is the state and local tax
deduction, and the reach of the AMT will increase dramatically without
legislative change that is costly in revenue loss. Such proposals were
explored by Reuben (2005) and in recent simulations by Burman, Gale,
Leiserson, and Rohaly (2007). The President's Advisory Panel's
(2005) tax reform proposals proposed to eliminate the deduction along
with the AMT. A different approach to addressing inequities across the
states was taken by Senator Ron Wyden and Congressman Rahm Emmanuel, who
proposed replacing the itemized deduction with a credit for state and
local taxes available to all taxpayers in their comprehensive tax reform
proposal, introduced in the Senate as S. 1111.
Tax Exempt Bonds
State and local issuance of debt has been going on virtually
throughout the existence of the republic. (The history discussed in this
section is based on Zimmerman (1991).) Throughout the early to mid
1800s, state public debt began increasing. States turned more and more
to aiding private developers. With the industrial revolution, railroad
development was actively sought by state and local governments, to the
point of serious fiscal problems by 1837. Indeed, the sharp increases in
debt issued to aid private developers along with the depression sparked
a multitude of restrictions, generally restrictions against lending to
private individuals and corporations, and debt limits on general
obligation bonds that continue to this day. However, with the
limitations came numerous methods for avoidance such as special
districts, special funds, and public authorities.
These concerns all seemed to stem from fears over state fiscal
collapse, not from the overlap of state and federal government tax
authorities. In fact, prior to the civil war there was no federal income
tax, making the question of tax-exempt bonds a moot point. Although many
advocates of exemption status may note, and use, a first instance of the
concept of intergovernmental tax immunity with Justice Marshall's
1819 decision and statement "the power to tax involves the power to
destroy," this famous claim actually referred to the states ability
to tax an instrument of the federal government.
While there were repeated attempts after ratification of the 16th
amendment in 1913 to remove tax-exempt bonds in their entirety, it was
not until after World War II that there was ever a discussion of
limiting the particular use of tax exempt bonds. Nineteen hundred and
sixty-eight marked the first attempt to restrict state uses of tax
exempt bonds. The 1968 Congress began by limiting industrial development
bonds (IDBs) and arbitrage bonds in response to a growth of bonds issued
for private purposes. In fact, the limits on IDBs began the first
qualified private activity bonds. Congress set the future trend not by
defining public purpose but rather by defining what was not. The 1968
restrictions outlined the still-current two-part use and security test
of a private activity bond, although at the time the use and security
tests required 25 percent or more of the use or securities to be
private. This test is a test for "bad" bonds, that is, if the
share is greater than 25 percent the bond passes the test for a taxable
bond. This and later restrictions were met with an onslaught from the
states to exempt specific activities. For example, transition rules were
added, in part to acquiesce to congressional pressures on behalf of
their states, which included "exceptions to tax-exempt-bond-limits
for a new stadium for the Miami Dolphins football team, a convention
center in Miami Beach, a midtown Miami redevelopment project, and two
new heating and cooling systems for the Florida Region" (Birnbaum
and Murray, 1987, p. 147).
The states' pleas that were met as private activities
qualifying for tax exempt status continued to include residential
property; sports facilities; facilities for convention or trade show;
airports; docks, wharves, parking facilities, and mass commuting
facilities; and other waste, pollution, and utility facilities were
allowed to qualify for tax exempt status.
The Tax Reform Act of 1986 made substantial changes to the exempted
facilities and narrowed the use and security tests to ten percent (i.e.,
ten percent or more of the use or the securities were private). Gold
(1990) points out the dramatic fall in state and local bond financing in
response to changes in TRA 1986. Municipal financing fell from $204.3
billion in 1985 and $151.0 billion in 1986 to $105.5 billion in 1987 and
$117.0 billion in 1989.
Despite the federal government's historical relaxation and
subsequent constriction of tax exempt municipal bonds, states and local
governments have a wide variety of activities that can be financed with
tax exempt bonds, particularly government obligation bonds that have few
limits. For example, according to the Government Accountability Office
(2006), 15 percent of golf courses were owned by state and local
governments. Of these, four percent were associated with resort or real
estate facilities. In addition, 39 hotels associated with convention
centers, airports or golf courses were identified by finance experts as
being financed with tax--exempt bonds. The report also mentions over 300
government-owned convention centers and cites instances of state and
local government support of gaming facilities.
Table 2 shows the distribution of bond issuance across broad
categories of activities that can be considered private in nature. Of
states' bond issues, at least half are for these activities. (Total
bond issuances exclude general purpose since there is no reliable way to
allocate those bonds to private or public purpose. Bonds issued for
general purpose account for roughly a quarter of total bond issuances.)
This table also includes tax-exempt, taxable, and qualified private
activity bond issues, which account, respectively, for 85.1 percent, 7.4
percent, and 7.5 percent of total (including general purpose) issuances
over the six-year period.
According to Table 1, above, explicitly private activity bonds
account for 23 percent of the tax expenditure for bonds. This number is
larger because of the existing stock. As with the property tax
deduction, the tax expenditure list classifies these bonds in functional
categories outside state and local activities. Nevertheless, the states
and localities play an important role in allocating these resources.
While some of the uses in Table 2 could be argued to be traditional
functions of state and local governments, convention centers, stadiums
and arenas, and other private-entertainment-related activities are
questionable.
ECONOMIC EFFECTS OF OTHER FEDERAL TAX POLICIES
State and local governments are also affected indirectly by other
tax policies of the federal government. The most obvious are the
marginal tax rates themselves. As important, in terms of incentives, as
proposals to disallow state and local tax deductions was the fall in the
top marginal tax rate from 70 percent in 1980 to 28 percent in 1986.
Although part of this rate reduction was reversed in 1993, with rates
then reduced again in 2001, there has been little attention to this
effect. Of course, states were hardly in a position to protest the lower
rates because these rates benefited their taxpayers. Moreover, there may
have been little in incentive effects for state and local taxes. These
changes affect the benefits of interest exclusion for state and local
bonds because the value of the exclusion is higher for higher marginal
tax rates.
There are several other important tax policy decisions affecting
tax burdens on capital income that affect tax exempt borrowing,
including the tax rates on corporate source income (the corporate tax
rate, capital gains tax rates, and tax rates on dividends), business
subsidies, and tax preferred savings provisions. These effects depend on
which assets are closer substitutes for municipal bonds.
Consider the corporate tax rate. If interest-bearing assets are
much closer substitutes for state and local bonds than equities, the
corporate income tax, which might normally be expected to burden
competing business investment, could make state and local bond finance
more costly by driving up the interest rate. Interest is deductible at
the corporate level, leading corporations to favor debt finance, and
their increased d