Government Performance and Accountability: Tax Expenditures	 
Represent a Substantial Federal Commitment and Need to Be	 
Reexamined (23-SEP-05, GAO-05-690).				 
                                                                 
Numerous federal programs, policies, and activities are supported
through the tax code. As described in statute, tax expenditures  
are reductions in tax liabilities that result from preferential  
provisions, such as tax exclusions, credits, and deductions. They
result in revenue forgone. This report, done under the		 
Comptroller General's authority, is part of an effort to assist  
Congress in reexamining and transforming the government to meet  
the many challenges and opportunities that we face in the 21st	 
century. This report describes (1) how tax expenditures have	 
changed over the past three decades in number, size, and in	 
comparison to federal revenue, spending, and the economy, and (2)
the amount of progress made since our 1994 recommendations to	 
improve scrutiny of tax expenditures.				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-690 					        
    ACCNO:   A37958						        
  TITLE:     Government Performance and Accountability: Tax	      
Expenditures Represent a Substantial Federal Commitment and Need 
to Be Reexamined						 
     DATE:   09/23/2005 
  SUBJECT:   Comparative analysis				 
	     Productivity in government 			 
	     Tax administration 				 
	     Tax expenditures					 
	     Performance appraisal				 
	     Accountability					 
	     Losses						 
	     Transparency					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-05-690

                 United States Government Accountability Office

                         GAO Report to Agency Officials

September 2005

GOVERNMENT PERFORMANCE AND ACCOUNTABILITY

Tax Expenditures Represent a Substantial Federal Commitment and Need to Be
                                   Reexamined

                                       a

GAO-05-690

[IMG]

September 2005

GOVERNMENT PERFORMANCE AND ACCOUNTABILITY

Tax Expenditures Represent a Substantial Federal Commitment and Need to Be
Reexamined

                                 What GAO Found

Whether gauged in numbers, revenues forgone, or compared to federal
spending or the size of the economy, tax expenditures have represented a
substantial federal commitment over the past three decades. Since 1974,
the number of tax expenditures more than doubled and the sum of tax
expenditure revenue loss estimates tripled in real terms to nearly $730
billion in 2004. The 14 largest tax expenditures, headed by the individual
income tax exclusion for employer-provided health care, accounted for 75
percent of the aggregate revenue loss in fiscal year 2004. On an
outlayequivalent basis, the sum of tax expenditure estimates exceeded
discretionary spending for most years in the last decade. For some budget
functions, the sum of tax expenditure estimates was of the same magnitude
as or larger than federal spending. As a share of the economy, the sum of
tax expenditure outlay-equivalent estimates has been about 7.5 percent of
gross domestic product since the last major tax reform legislation in
1986.

All federal spending and tax policy tools, including tax expenditures,
should be reexamined to ensure that they are achieving their intended
purposes and designed in the most efficient and effective manner. The
nation's current and projected fiscal imbalance serves to reinforce the
importance of engaging in such a review and reassessment. Although data
and methodological challenges exist, periodic reviews of tax expenditures
could establish whether they are relevant to today's needs; if so, how
well they have worked to achieve their objectives; and whether the
benefits from specific tax expenditures are greater than their costs. Over
the past decade, however, the Executive Branch made little progress in
integrating tax expenditures into the budget presentation, in developing a
structure for evaluating tax expenditure outcomes or in incorporating them
under review processes that apply to spending programs, as we recommended
in 1994. More recently, the Administration has not used its Program
Assessment Rating Tool process to systematically review tax expenditures
or promote joint reviews of tax and spending programs sharing common
goals.

Sum of Tax Expenditure Estimates Compared with Total Federal Outlays,
1981-2004 Dollars in billions (in constant 2004 dollars) 1,400

1,200

1,000

800

600

400

200

0

1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Fiscal year

Mandatory spending Discretionary spending Tax expenditure
outlay-equivalent estimates Source: GAO analysis of OMB budget reports on
tax expenditures, fiscal years 1983-2006.

Note: Summing tax expenditure estimates does not take into account
interactions between individual provisions.

United States Government Accountability Office

Contents

  Letter

Results in Brief
Background
Tax Expenditures Have Represented a Substantial Federal

Commitment over Time

Systematic Review of Tax Expenditures Is Integral to Reexamining the
Federal Base, but Little Progress Has Been Made Since 1994 to Increase
Scrutiny

Conclusions
Recommendations for Executive Action
Agency Comments and Our Evaluation

1 4 7

19

43 71 73 74

Appendixes                                                             
                Appendix I:             Scope and Methodology              79 
               Appendix II:  Comments from the Office of Management and    82 
                                               Budget                     
              Appendix III:     How Tax Expenditures Are Measured and      92 
                                              Reported                    
                             Compilation of Tax Expenditures Reported by  
              Appendix IV:                    Treasury                    
                                           (1974 to 2004)                 110 
                Appendix V:                   Glossary                    119 
              Appendix VI:         GAO Contact and Acknowledgments        124 

Selected Bibliography

  Related GAO Products

Tables	Table 1: Table 2:

Table 3: Table 4:

Examples of Types of Tax Expenditures Available to
Taxpayer Groups 10
Revenue Loss Estimates for the Largest Tax Expenditures
Reported for Fiscal Year 2004, with Taxpayer Group and
Budget Function 34
Comparison of Tax Expenditure Reporting by JCT and
Treasury 96
List of Tax Expenditures Reported by the U.S. Department
of the Treasury and the Joint Committee on Taxation for
Fiscal Year 2004 100

                                    Contents

Table 5:	Sum of Revenue Loss Estimates for Tax Expenditures by Budget
Function Reported by the U.S. Department of the Treasury and Joint
Committee on Taxation, Fiscal Year 2004 109

Table 6:	Tax Expenditures Reported by the U.S. Department of the Treasury
(1974 to 2004) 110

Figures	Figure 1: Figure 2: Figure 3: Figure 4: Figure 5: Figure 6:

Figure 7: Figure 8:

Figure 9:

Examples of How Each Type of Tax Expenditure Relates
to the U.S. Individual Income Tax Return (Form 1040)
Number of Tax Expenditures Reported by Treasury,
1974-2004
Duration of Tax Expenditures Reported by Treasury,
1974-2004
Sum of Tax Expenditure Revenue Loss Estimates,
1974-2004
Sum of Tax Expenditure Revenue Loss Estimates with
Outlays for Refundable Tax Credits, 1974-2004
Tax Legislation Enacted From 1974-2004 That May Have
Influenced the Sum of Revenue Loss Estimates for Tax
Expenditures
Sum of Revenue Loss Estimates by Taxpayer Group,
1974-2004
Sum of Tax Expenditure Outlay-equivalent Estimates
Compared with Total Mandatory and Total Discretionary
Outlays, 1981-2004
Sum of Tax Expenditure Outlay-Equivalent Estimates
Compared to Total Mandatory and Discretionary Outlays
and Receipts as a Percentage of GDP, 1981-2004

12 22 24 26 27

30 32

36

38 39

41 45

46 94 98

Figure 10: Size of Tax Expenditure Outlay-Equivalent Estimates by Budget
Function, 1981-2004

Figure 11: Tax Expenditure Outlay-equivalent Estimates Compared with
Federal Outlays by Budget Function, Fiscal Year 2004

Figure 12: Composition of Spending as a Share of GDP under Baseline
Extended

Figure 13: Composition of Spending as a Share of GDP Assuming
Discretionary Spending Grows with GDP after 2005 and All Expiring Tax
Provisions Are Extended

Figure 14: Treasury's Supplemental Reporting for Comprehensive Income and
Consumption Tax Baselines

Figure 15: Supplemental Estimates Developed by Treasury and JCT

Contents

Figure 16: Sum of Revenue Loss Estimates Reported by the Joint Committee
on Taxation and the U.S. Department of the Treasury (1987-2004) 99

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

A

United States Government Accountability Office Washington, D.C. 20548

September 23, 2005

The Honorable Joshua Bolten
Director
Office of Management and Budget

The Honorable John W. Snow
The Secretary of the Treasury

Over the last decade, a culture of performance has taken root in the
federal
government. This culture has placed a greater premium on the importance
of accountability for the effective use of public resources and
authorities.
Periodic reexamination of existing federal policies and operations offers
the prospect of weeding out ineffective or outdated programs, policies,
functions, or activities while strengthening and updating those that are
retained. While inefficient and ineffective programs and policies are
never
appropriate, the nation's long-term fiscal imbalance provides an
additional
impetus for reexamining all major spending programs and tax provisions.1
This includes tax incentives and subsidies intended to promote various
social and economic objectives. Tax preferences-which are legally
known as tax expenditures-result in forgone revenue for the federal
government due to preferential provisions in the tax code, such as
exemptions and exclusions from taxation, deductions, credits, deferral of
tax liability, and preferential tax rates. These tax expenditures are
often
aimed at policy goals similar to those of federal spending programs;
existing tax expenditures, for example, are intended to encourage
economic development in disadvantaged areas, finance postsecondary
education, and stimulate research and development. Tax expenditures
have a significant effect on overall tax rates-in that, for any given
level of
revenue, overall tax rates must be higher to offset the revenue forgone
through tax expenditures-as well as the budget and fiscal flexibility.
They
also contribute to the growing complexity of the federal tax system.
Regardless of the level of revenue deemed appropriate, tax expenditures-
like other federal programs or activities-should be reviewed to determine
their effectiveness and continued relevance. Yet, tax expenditures and
their relative contributions toward achieving federal missions and goals
are
often less visible than spending programs which are subject to more
systematic review.

1For more information, see GAO, 21st Century Challenges: Reexamining the
Base of the Federal Government, GAO-05-325SP (Washington, D.C.: February
2005).

A little over a decade ago, GAO examined the growth in tax expenditures
and opportunities to focus policymakers' attention on them.2 Our review
found that tax expenditures were not integrated into annual federal budget
review processes, and most were not subject to reauthorization, even
though revenues forgone through tax expenditures were substantial. In
turn, policymakers had few opportunities to make explicit comparisons or
trade-offs between tax expenditures and federal spending programs.
Therefore, we made several recommendations to the Office of Management and
Budget (OMB) intended to encourage more informed policy debate about tax
expenditures and to stimulate joint reviews of related tax and spending
programs. OMB agreed that our recommendations were generally reasonable
and reported that the Executive Branch had initiated efforts to integrate
tax expenditures in the budget review process and develop performance
measures for some tax expenditures.

We have prepared this report under the Comptroller General's authority as
part of a continuing effort to assist Congress in reexamining the base of
federal programs, policies, and activities critical to achieving fiscal
discipline in the budget as a whole. This report updates our 1994 work and
specifically describes (1) how tax expenditures have changed over the past
three decades in number and size and in comparison to federal revenue,
spending, and the economy, and (2) the amount of progress that has been
made since 1994 in how the Executive Branch scrutinizes tax expenditures.

This update draws on our previously issued work on tax expenditure trends
and individual tax provisions as well as results-oriented management and
performance reporting. We also reviewed relevant literature related to tax
expenditure measurement and reporting, and trends in the use of tax
expenditures over time. We interviewed relevant agency officials and tax
policy experts to obtain a greater understanding of information gained
through our literature review. To gauge trends in the numbers and size of
tax expenditures for fiscal years 1974 to 2004, we relied on the
Department of the Treasury's (Treasury) annual list of tax expenditures
and estimates for each tax expenditure of the associated revenue loss-the
amount of revenue that the government forgoes-and the outlay-equivalent
value-the amount of outlays required to deliver the

2GAO, Tax Policy: Tax Expenditures Deserve More Scrutiny,
GAO/GGD/AIMD-94-122 (Washington, D.C.: June 3, 1994).

same after-tax income as provided through the tax expenditure.3 We added
the tax expenditure revenue loss estimates for each fiscal year to
approximate the total revenue forgone through tax expenditure provisions.
While sufficiently reliable as a gauge of general magnitude, the sum of
the individual revenue loss estimates has important limitations in that
any interactions between tax expenditures will not be reflected in the
sum. In addition, tax expenditure revenue loss estimates for specific
provisions do not take into account potential behavioral responses to
changes in these provisions on the part of taxpayers, and, in turn, no
potential behavioral response would be reflected in the sum of the
estimates. Thus, the revenue loss from all or several tax expenditures
together might be greater or less than the sum of the estimated revenue
losses from the individual tax expenditures, and no measure of the size or
the magnitude of these potential interactions or behavioral responses to
all or several tax expenditures is available. Growth in the sum of tax
expenditure estimates across the three-decade period is presented in
inflation-adjusted 2004 dollars and measured relative to the economy as a
share of the gross domestic product (GDP). We compared the sum of tax
expenditure outlayequivalent estimates to total federal mandatory and
discretionary spending and spending by budget function. Finally, we used
Treasury's estimates to examine trends in the size of tax expenditures by
taxpayer group and identify the largest tax expenditures in 2004.

To examine progress over the last decade in how the federal government
scrutinizes tax expenditures, we examined actions taken by the Executive
Branch to implement our 1994 recommendations for (1) presenting tax
expenditures in the annual budget, (2) developing a structure for
conducting reviews of tax expenditures' performance, (3) conducting case
studies to assess performance review structure, and (4) incorporating tax
expenditures into the annual budgetary review process. We reviewed efforts
to include tax expenditures under the Government Performance and Results
Act's (GPRA) statutory framework for strategic planning,

3We used revised estimates, which incorporate the most recent changes in
tax policy and economic activity, developed by the U.S. Department of the
Treasury and reported in each year's budget in either the Special
Analyses, Appendixes, or Analytical Perspectives. We chose the tax
expenditure estimates reported in the budget for our analysis because
Treasury develops (1) revised estimates based on changes in tax policy and
economic activity for the year prior to the reported fiscal budget year
and (2) outlay-equivalent estimates that facilitate comparison to federal
spending. Although they are the last available estimates reported,
Treasury's estimates are still projected estimates and may not reflect
additional policy changes.

performance measurement, and program reporting and evaluation.4 We also
considered any related activities to include tax expenditures under OMB's
Program Assessment Rating Tool (PART)-its current framework for assessing
the performance of federal programs. (See app. I for details on our scope
and methodology.)

Our work was conducted from August 2003 through July 2005, in accordance
with generally accepted government auditing standards. In July 2005, we
requested comments on a draft of this report from the Director of the
Office of Management and Budget, the Secretary of the Department of the
Treasury, and the Commissioner of the Internal Revenue Service (IRS). We
received comments from OMB's Associate Director for Economic Policy in a
letter dated September 2, 2005 (see app. II). Treasury did not provide
separate comments, instead deferring to OMB. IRS staff provided a
technical correction that we incorporated.

Results in Brief	Whether gauged in absolute numbers and revenues forgone
or in comparison to federal spending or the size of the economy, tax
expenditures have represented a substantial commitment of federal support
over the past three decades. Between 1974 and 2004, tax expenditures
doubled in number from 67 to 146, and while some were repealed or allowed
to expire, considerably more were added to Treasury's list. Based on our
analysis of Treasury's estimates, the sum of revenue loss estimates
associated with tax expenditures, adjusted for inflation, tripled from
approximately $240 billion to nearly $730 billion over the period.5 The 14
largest tax expenditures, headed by the largest single tax expenditure-
the income tax exclusion for employer-provided health care-accounted for
75 percent of the aggregate revenue loss in fiscal year 2004. The sum of
the revenue loss estimates for tax expenditures that are used by
individual taxpayers increased in real terms from approximately $190
billion to nearly $650 billion. Since 1981 when outlay-equivalent
estimates were first available, the sum of the outlay-equivalent estimates
for tax expenditures has been similar in magnitude to discretionary
spending, and this sum

4Pub. L. No. 103-62, Aug. 3, 1993.

5Aggregate tax expenditure estimates must be interpreted carefully because
of inherent limitations in the meaning of the summed estimates. The sum of
the individual tax expenditure estimates is useful for gauging the general
magnitude of revenue forgone through provisions of the tax code but does
not take into account interactions between individual provisions.

exceeded total discretionary spending for most years during the last
decade. As a share of the U.S. economy, the sum of tax expenditure
outlayequivalent estimates remained relatively stable at about 7.5 percent
of GDP since the last major tax reform legislation in 1986. Across budget
functions, the size of tax expenditures varied, and for some budget
functional areas, such as housing and education, tax expenditures were the
same magnitude as, or larger than, federal spending.

Although tax expenditures are substantial in size, little progress has
been made in the Executive Branch to increase the transparency of and
accountability for tax expenditures. The entire set of tools the federal
government can use to address national objectives-including discretionary
and mandatory spending, tax provisions, and loans and loan
guarantees-should be subject to periodic reviews and reexamination to
ensure they are achieving their intended purposes and designed in the most
efficient and effective manner. The nation's current and projected fiscal
imbalance serves to reinforce the importance of engaging in such a review
and reassessment. Tax expenditures may not always be efficient, effective,
or equitable and, consequently, information on these attributes can help
policymakers make more informed decisions as they adapt current policies
in light of our fiscal challenges and other overarching trends. In
addition, some tax expenditures, at least as currently designed, may serve
to exacerbate other key private sector and public policy challenges, such
as controlling health care costs. Although data and methodological
challenges may impede studies of some tax expenditures, periodic reviews
of tax expenditures could help establish whether these programs are
relevant to today's needs; if so, how well tax expenditures have worked to
achieve their objectives; and whether the benefits from particular tax
expenditures are greater than their costs. Over the past decade, however,
the Executive Branch has made little progress in integrating tax
expenditures into the budget presentation, in developing a structure for
evaluating the performance of tax expenditures, or in incorporating tax
expenditures under review processes that apply to spending programs, as we
recommended in 1994. Also, more recently, OMB has not used its PART
process to systematically review tax expenditures and promote joint and
integrated reviews of tax and spending programs sharing common,
crosscutting goals. One of the key impediments to moving forward in
evaluating tax expenditures' performance is the continuing lack of clarity
about the roles of OMB, Treasury, IRS, and departments or agencies with
outlay program responsibilities.

We are recommending that the Director of OMB, in consultation with the
Secretary of the Treasury, take several steps to ensure greater
transparency of and accountability for tax expenditures by reporting
better information on tax expenditures' performance and more fully
incorporating tax expenditures into federal performance management and
budget review processes.

In providing comments on a draft of this report, OMB's Associate Director
for Economic Policy disagreed with our recommendations, raised concerns
about our use of tax expenditure estimates developed by Treasury and
reported in the annual federal budget, and implied that increasing the
attention paid to tax expenditures due to the severity of the nation's
longterm fiscal imbalance would lead to tax increases. Pursuant to the
Congressional Budget Act of 1974, the term tax expenditure, as our draft
stated, has been used in the federal budget for three decades, and the tax
expenditure concept-while not precisely defined-is a valid representation
of one tool that the federal government uses to allocate resources. In
addition, OMB's implication that focusing more attention on tax
expenditures would automatically lead to increased taxes is unfounded. Our
report states that the revenues forgone through tax expenditures reduce
revenues available to fund other federal activities or they require higher
tax rates to obtain a given amount of revenue. Thus, if the evaluations of
tax expenditures we call for lead to reducing or eliminating some tax
expenditures, the net change after rate adjustments could, depending on
overall congressional priorities and preferences, result in tax reductions
for many taxpayers. Furthermore, although specific tax expenditures, such
as the earned income tax credit (EITC) and Liberty Zone tax benefits, have
received varying degrees of scrutiny, efforts to date have not provided
the Congress and others with an integrated perspective on the extent to
which programs and tools-including tax expenditures- contribute to
national goals and position the government to successfully meet 21st
century demands. The lack of a requirement to disclose tax expenditures in
agencies' annual performance and accountability reports may result in
important performance and cost related data not being fully considered
with other federal resources allocated to achieve similar objectives.
Although challenges must be overcome to provide systematic reviews of tax
expenditures, these challenges cannot be addressed absent effective
leadership within the Executive Branch. For these reasons, we believe our
recommendations, if fully implemented, will ensure that policymakers and
the public have the necessary information to make informed decisions and
to improve the progress toward greater scrutiny of tax expenditures.

Background	To understand the trends in the size of tax expenditures, it is
helpful to understand how tax expenditures are defined and how the
different types affect taxpayer liability. For this report, we also
provide an overview of the broad purposes of tax expenditures-one method
the federal government can use to achieve national objectives-and a
discussion of how tax expenditures interact with the federal budget.

Tax Expenditures Defined	Tax expenditures are revenue losses-the amount of
revenue that the government forgoes-resulting from federal tax provisions
that grant special tax relief for certain kinds of behavior by taxpayers
or for taxpayers in special circumstances. These provisions may, in
effect, be viewed as spending programs channeled through the tax system
and are classified in the U.S. budget by the same functional categories as
other spending, such as energy and health. Tax expenditures are provisions
that are exceptions to the "normal structure" of the individual and
corporate income tax necessary to collect government revenues.6 Deciding
whether an individual provision should be characterized as a tax
expenditure is a matter of judgment, and disagreements about
classification stem from different views about what should be included in
the income tax base.7 As a practical matter, the term tax expenditure has
been used in the federal budget for three decades, and the tax expenditure
concept-while not precisely defined-is a valid representation of one tool
that the federal government uses to allocate resources.

Both the congressional Joint Committee on Taxation (JCT) and Treasury's
Office of Tax Analysis annually compile a list of tax expenditures and
estimates of their cost.8 (App. III provides additional information on how

6The concept of tax expenditures extends beyond the income tax. Tax
expenditures also exist for other types of taxes such as excise and
payroll taxes; however, this report considers only tax expenditures for
the federal income tax system.

7Some object that the distinction between those tax provisions labeled tax
expenditures and those that are not is arbitrary and, that the very notion
of labeling these provisions expenditures implies that all income could be
taxed and thus, all income inherently belongs to the government. See, for
example, Joint Economic Committee, Tax Expenditures: A Review and Analysis
(Washington, D.C.: August 1999).

8Office of Management and Budget, Analytical Perspectives, Budget of the
United States Government, Fiscal Year 2006 (Washington, D.C.: 2005); and
Joint Committee on Taxation, Estimates of Federal Tax Expenditures for
Fiscal Years 2005-2009, JCS-1-05, (Washington, D.C.: Jan. 12, 2005).

tax expenditures are measured and reported and perspective on differences
among the lists of tax expenditures reported by JCT and Treasury.)
Treasury's tax expenditure estimates are included as an informational
supplement to the annual federal budget by the OMB. The revenue loss is
estimated for each tax expenditure separately by comparing the revenue
raised under current law with the revenue that would have been raised if
the single provision did not exist, assuming all other parts of the tax
code remain constant and taxpayer behavior is unchanged. Revenue loss
estimates are intended to provide information about the value of tax
expenditures. However, tax expenditure estimates do not incorporate any
behavioral responses and thus do not necessarily represent the exact
amount of revenue that would be gained if a specific tax expenditure were
repealed.9 For example, when the consumer interest deduction was phased
out gradually beginning in 1987, some taxpayers shifted to
interest-deductible home equity loans to finance consumption, thereby
affecting the revenue gain from eliminating the consumer interest
deduction.

In addition to estimating revenue loss, Treasury also measures tax
expenditures on an outlay-equivalent basis. Outlay-equivalent estimates
represent the amount of budget outlays that would be required if the
government were to provide taxpayers with the same after-tax income they
receive through the tax expenditure. Outlay-equivalent estimates are often
higher than revenue loss estimates to reflect that a comparable outlay
program could result in additional taxable income to recipients.10
Outlayequivalent estimates are useful to compare tax expenditures and
other parts of the federal budget. For example, the outlay-equivalent
estimate for the tax exclusion for housing and meal allowances for
military personnel reflects the additional pretax income that military
personnel would have to be paid to raise their income after federal taxes
by the amount of the tax expenditure. The outlay-equivalent estimate can
be used to compare this tax expenditure with other outlays for defense
compensation on a more consistent basis.

9Changes in taxpayer behavior are taken into account when JCT and Treasury
prepare revenue estimates for proposed legislation.

10Although outlay-equivalent estimates are often higher than revenue loss
estimates, the net estimated revenue loss for the government is unchanged.
For more information on how outlay-equivalent estimates are measured, see
app. III.

Types of Tax Expenditures	The Congressional Budget and Impoundment Control
Act of 197411 lists six types of tax expenditures: exclusions, exemptions,
deductions, credits, preferential tax rates, and deferral of tax
liability. Some tax expenditures apply only to individual taxpayers, such
as deductions and exclusions for employer-provided contributions for
medical insurance, and some only to corporate taxpayers, such as a tax
credit for corporations doing business in U.S. possessions. Other tax
expenditures, such as accelerated depreciation, apply both to corporations
and to individual taxpayers with income from businesses such as sole
proprietorships or partnerships.12 Table 1 shows examples of each type of
tax expenditure and the taxpayer group that may claim a particular type.

11Pub. L. No. 93-344, Sec. 3, 88 Stat. 299 (July 12, 1974) (codified at 2
U.S.C. sec. 622(3)).

12Tax expenditures have been described by some analysts as being directed
toward social or business interests. A social tax expenditure provides
transfer payments related to old age, education, occupational and health
benefits. A business tax expenditure has as its central goal the provision
of financial support to corporations, partnerships, or individuals in any
industry. See Eric Toder, The Changing Composition of Tax Incentives:
1980-99, (Washington, D.C.: The Urban Institute, March 1999).

  Table 1: Examples of Types of Tax Expenditures Available to Taxpayer Groups
                                Taxpayer groups

Both corporate and Types of tax expendituresa Individual taxpayersb
Corporate taxpayers individual taxpayers

Exclusion from taxable income	Exclusion of employer Extraterritorial
income exclusion Exclusion of interest on contributions to medical
insurance public purpose state and premiums and medical care local bond

Exemption from taxable income	Parent personal exemption for Exemption of
credit union income Special employer stock student age 19 and over
ownership plan rules

Deduction from  Deductibility of       Special Blue       Deductibility of 
taxable income  mortgage interest      Cross/Blue Shield        charitable 
                     on owner-occupied        deduction       contributions   
                           homes                             
       Credit                                                Credit for       
subtracted from    Child tax credit    Employer-provided  low-income       
        taxes                                 child care     
     ordinarily                                 credit           housing      
      computed                                                 investments    

Preferential tax rate for all or Averaging previous period taxable
Graduated corporation income tax N/A part of taxable income income for
farmers rate

Deferral of tax liability	Carryover basis of capital gains on Deferral of
income from controlled Accelerated depreciation of gifts foreign
corporations machinery and equipment

Source: GAO.

aTypes of tax expenditures that are identified in the U.S. Congressional
Budget and Impoundment Act of 1974.

b Individual tax expenditures include those available to non-corporate
forms of business such as sole proprietorships.

Figure 1 illustrates how tax expenditures appear on the U.S. Individual
Income Tax Return (Form 1040). Exclusions are those items of income that
would otherwise constitute a part of the taxpayer's gross income, but are
excluded under a specific provision of the tax code. Exclusions generally
do not appear on the Form 1040, and excluded income is not reflected in
total reported income. For example, the income tax exclusion of employer
contributions to medical insurance premiums and medical care is not
reported in a taxpayer's wages or salaries. An exemption, such as the
parent personal exemption for students over age 19 but under age 24, is a
reduction in taxable income offered to taxpayers because of their status
or circumstances. Deductions are adjustments from adjusted gross income.13
Deductions claimed before the adjusted gross income line on the Form 1040,
such as the tuition and fees deduction (this appears on line 27 in fig.
1), are sometimes called "above-the-line" deductions. Taxpayers may

13Adjusted gross income is equal to gross income less qualifying
adjustments to income, including some deductions.

also claim "below-the-line" deductions after the adjusted gross income
line; to do so, taxpayers must itemize their deductions.14

14Taxpayers list their itemized deductions on the U.S. Individual Income
Tax Return Form 1040 Schedule A-Itemized Deductions. In 2001, we reported
that less than one-third of individual taxpayers itemize their deductions;
in lieu of itemizing their deductions, most taxpayers take the standard
deduction, which is considered part of the individual income tax
structure. See GAO, Tax Deductions: Estimates of Taxpayers Who May Have
Overpaid Federal Taxes by Not Itemizing, GAO-01-529 (Washington, D.C.:
Apr. 12, 2001).

Figure 1: Examples of How Each Type of Tax Expenditure Relates to the U.S.
Individual Income Tax Return (Form 1040)

Exemption

An example of an exemption is the

parent personal exemption for

students over age 19 but under age

24 (appears on line 6c).

Exclusion

Excluded income generally does not

appear on the tax form, but can still

cause the income line to be lower

than it otherwise would be. For

example, the exclusion of employer

contributions to medical insurance

premiums and medical care causes

the reported amount of wages and

salaries to be less than it would be

otherwise (see line 7).

Deferral

An example of a deferral is the

accelerated depreciation of

machinery and equipment for a

sole proprietorship (appears on

line 12 via Schedule C-Profit or

Loss from Business and Form

4562-Depreciation and

Amortization.

Preferential tax rate

An example of this tax expenditure

type is the preferential tax rate on

capital gains of certain income

(appears on line 13 via the

Schedule D-Capital Gains and

Losses form).

Deduction

An example of a deduction above the

adjusted gross income line is the tuition

and fees deduction (appears on

line 27). An example of an itemized

deduction below-the-line is the deduction

for mortgage interest on owner-occupied

homes (appears on line 39 as an

itemized deduction via the Schedule

A-Itemized Deductions form).

Note: Adjusted gross income is equal to gross income less qualifying
adjustments to income, including some deductions.

Credit

An example of a credit is the tax credit for the elderly or disabled
(appears on line 48). An example of a refundable credit is the child tax
credit (appears on line 51 with the refundable portion on line 67 via Form
8812-Additional Child Tax Credit).

                                  Source: GAO.

Each type of tax expenditure creates tax savings in different ways and,
consequently, reduces federal revenues in different ways. The amount of
tax relief per dollar that a taxpayer receives using an exclusion,
exemption, or deduction depends on the taxpayer's marginal tax rate.15
Generally, the higher the taxpayer's marginal tax rate, the greater the
tax savings from these tax expenditure types.16 Tax credits reduce tax
liability dollar-fordollar, so the value of a credit is the same
regardless of the taxpayer's marginal tax rate. A nonrefundable tax credit
can be used to reduce current year tax liability to zero, and a refundable
credit in excess of tax liability results in a cash refund. For
preferential tax rates which reduce the tax rate on some forms of income
such as capital gains, the tax savings depend on the difference between
the preferential rate and a taxpayer's marginal tax rate. By allowing
taxpayers to reduce current tax liability by delaying recognition of some
income or accelerating some deductions otherwise attributable to future
years, a tax deferral shifts the timing of tax payments and, in effect,
provides an interest-free "loan" to the taxpayer. The benefit from a
deferral is even greater if the taxpayer expects to face a lower tax
bracket in the future. A lower-income taxpayer-with no net annual income
or with no current tax liability after claiming the standard deduction and
any personal exemptions-would not directly benefit from most tax
expenditures other than refundable credits.

Some techniques have been used to limit the benefits that taxpayers may
receive from individual tax expenditures or groups of them. Congress has
controlled the amount of revenue forgone for some tax expenditures by
adopting provisions to restrict taxpayers' eligibility for benefits. For
example, the mortgage interest deduction is limited to interest on debt up
to $1 million to buy, build, or improve first and second homes and up to
$100,000 in home equity debt. Aggregate itemized deductions are reduced by
3 percent of the amount of a taxpayer's adjusted gross income that exceeds
a certain threshold, eliminating 3 cents of itemized deductions for each
dollar of income above the threshold for higher-income taxpayers.17 Some
tax expenditures, such as tax-exempt private-activity bonds issued by each
state, are subject to volume caps limiting the aggregate amount of

15A marginal tax rate is the tax rate that applies to an additional dollar
of income.

16Marginal individual tax rates for 2004 taxes were 10 percent, 15
percent, 25 percent, 28 percent, 33 percent, and 35 percent.

17The reduction in itemized deductions for higher-income taxpayers may
also be viewed as a hidden tax rate.

benefits available. The alternative minimum tax (AMT) also affects tax
expenditures and the amount of the revenue loss for the federal
government.18 The AMT is intended to ensure that taxpayers with income
over certain thresholds pay some income tax, no matter how much they claim
in certain deductions and credits. Under the AMT, taxpayers may have to
add back some tax expenditures that they could otherwise claim under the
regular tax system, such as deductions for state and local taxes and home
equity loan interest, and they may have to include as income certain
tax-exempt bond interest that is excluded under the regular tax system.

    Objectives of Tax Expenditures

In addition to raising revenue, the federal income tax has long been used
as a tool for accomplishing social and economic objectives. The general
objectives of tax expenditures are to encourage particular types of
activities (such as saving for retirement, promoting home ownership,
investing in certain sectors, or funding research and development) and
provide economic relief to selected groups of taxpayers (such as the
elderly, the blind, and those with children). Another objective of tax
expenditures may also be to adjust for differences in individuals' ability
to pay taxes. For example, if two taxpayers have the same income, but one
has a catastrophic illness and costly medical bills (or large casualty and
theft losses), the other taxpayer is judged better able to pay income
taxes. Some tax expenditures may be enacted to compensate for other
provisions of the tax code. For example, advocates of reduced tax rates on
capital gains often explain the special treatment of capital gains income
as offsetting, in part, the assessment of taxes on the nominal, rather
than the real, value of capital gains. The rationale and reasons for a
particular tax expenditure may change over time. For example, according to
the Congressional Research Service (CRS), the income tax code instituted
in 1913 contained a deduction for all interest paid. No distinction was
made between business and personal interest expenses, although most
interest payments at that time represented business expenses. The
legislative history does not indicate that the deductibility of mortgage
interest was originally intended to encourage home ownership or subsidize
the housing industry. However, over time, encouraging home ownership,
stimulating

18The AMT is a separate tax system that applies to both individual and
corporate taxpayers. It parallels the regular individual income tax system
but with different rules for determining taxable income, different tax
rates for computing tax liability, and different rules for allowing the
use of tax credits.

residential construction and maintenance, and encouraging families to save
and invest in a major financial asset have all been offered as
justifications for the mortgage interest deduction.19

The tax expenditure tool may substitute for a federal spending program in
that the federal government "spends" some of its revenue on subsidies by
forgoing taxation on some income.20 Certain activities may be cheaper and
simpler to subsidize through the tax code than by setting up a separate
program using a different tool. For example, the incremental
administrative and compliance costs to deliver the tax credit for child
and dependent care expenses may be relatively low compared to the costs of
setting up a separate system for processing child care applications and
sending vouchers to those eligible. The administrative infrastructure
already exists for the government to collect and remit money to over 131
million individual tax filers and 6 million corporations via the tax
system administered by the IRS. In concept, the costs to implement an
incomebased benefit program through the existing tax system could be lower
than to set up separate spending programs to deliver these benefits.

In some circumstances, tax expenditures may not be the best policy choice
to deliver timely benefits or reach intended populations. For programs
that seek to provide benefits within a given year, the annual income
measure relevant for income tax purposes may not be the best way to target
benefits. Relative to spending programs, tax expenditures are limited in
their ability to directly provide benefits to nontaxpayers. For example,
tax credits must be refundable to reach low-income individuals who do not
pay taxes and otherwise would not be required to file tax returns. Tax
expenditures generally do not deliver federal resources directly to state
and local governments and tax-exempt nonprofit organizations. The
charitable contribution deduction provides an incentive for individual and
corporate taxpayers to donate to charitable, religious, educational, and

19For more information on the history of the mortgage interest deduction,
see Pamela J. Jackson, Fundamental Tax Reform: Options for the Mortgage
Interest Deduction, Library of Congress, Congressional Research Service
Report RL33025 (Washington, D.C.: Aug. 8, 2005).

20For more discussion of administering programs using tax expenditures,
see OMB, Analytical Perspectives, Budget of the United States Government,
Fiscal Year 2006

(Washington, D.C.: 2005); Eric J. Toder, "Tax Cuts or Spending-Does It
Make a Difference?" National Tax Journal, Vol. LIII, No. 3, September
2000); and Joel Slemrod and Jon Bakija, Taxing Ourselves: A Citizen's
Guide to the Debate Over Taxes, 3rd Edition (Cambridge, Mass.: The MIT
Press, 2004).

health nonprofit organizations. The deduction, in effect, is a federal
grant to the donor that reduces the out-of-pocket cost of giving. The
itemized deduction for state and local taxes directly increases an
individual taxpayer's after-tax income and thus reduces the after-tax
price of state and local taxes. State and local governments receive some
of the benefit to the extent that taxpayers may be more willing to pay
state and local taxes.

Tax expenditures are not necessarily an either/or alternative to federal
spending and may be used in combination with federal spending and
strategies to achieve national objectives. For example, the HOPE and
Lifetime Learning tax credits are used with federal education assistance,
such as student loans, all of which help individuals fund higher
education. Many tax expenditures are comparable to entitlement programs
for which spending is determined by rules for eligibility, benefit
formulas, and other parameters rather than by Congress appropriating
specific dollar amounts each year.21 With some exceptions, tax
expenditures typically make funds (through reduced taxes) available to all
qualified claimants, regardless of how many taxpayers claim the tax
expenditures, how much they claim collectively, or how much federal
revenue is reduced by these claims. Some tax expenditures resemble other
policy tools, such as grants or direct loans. A few tax expenditures are
administered like grant programs, allowing for some administrative
discretion over who receives funds. For the New Markets Tax Credit (NMTC),
those seeking the credit must apply to the Community Development Financial
Institutions (CDFI) Fund within Treasury and be chosen by a group of
evaluators to receive the tax credit. Like a grant program, the NMTC has a
maximum amount that can be allocated by CDFI.22 Tax expenditures in the
form of deferrals resemble loans, because they allow taxpayers to postpone
the time when income is recognized for tax purposes or to accelerate the
deduction of expenses, both of which effectively lower the amount of
income currently subject to tax. Deferrals can result in higher taxes in
later years when taxpayers recognize deferred income in later tax years or
have fewer deductions to claim than they otherwise would have had; the
amount of the deferral is, in effect, analogous to a government loan.

21Entitlement statutes provide the authority to make payments to any
person or government if, under the provisions of the law containing that
authority, the United States is obligated to make such payments to persons
or governments who meet the requirements established by that law.

22The NMTC legislation limits the allocation of equity eligible for tax
credits from 2001 through 2007 from $1 billion to $3.5 billion per year,
totaling $15 billion over the 7 years.

    Tax Expenditures and the Federal Budget

Tax expenditures, by definition, reduce federal revenue and thus have
implications for income tax rates, federal spending, and the federal
budget. To obtain a given amount of revenue, tax expenditures require
overall statutory tax rates to be higher. Otherwise, revenues forgone
through tax expenditures reduce the revenue base available for funding
federal spending programs. From a budgetary perspective, most tax
expenditures are comparable to mandatory spending for entitlement
programs, in that no further action is required to provide resources for
tax expenditures. Tax expenditures do not compete overtly in the annual
budget process and, in effect, receive a higher funding priority than
discretionary spending subject to the annual appropriations process.
Revenues forgone through tax expenditures-unless offset by increased taxes
or lower spending- increase the unified budget deficit and federal
borrowing from the public (or reduce the unified budget surplus available
to reduce debt held by the public).

As noted previously, both the executive and legislative branches-by
Treasury and JCT, respectively-publish annual lists of tax expenditures
and the associated revenue loss, but budgetary decisions generally are not
based on these lists. Like any spending program, newly proposed tax
expenditures and those subject to expiration, to some extent, are subject
to scrutiny, but most tax expenditures are not subject to reauthorization.
Tax expenditures may be indirectly controlled to the extent that the
Congress aims to achieve any revenue target. The tax committees consider
tradeoffs between tax expenditures, tax rates, and other parts of the tax
code.

In concept, eliminating or limiting an existing tax expenditure-like an
existing spending program-would free up resources to reduce tax rates,
increase federal spending or other tax expenditures, reduce the deficit,
or produce some combination thereof. Conversely, adding a new tax
expenditure, expanding an existing tax expenditure, or extending an
expiring tax expenditure reduces the resources available to reduce tax
rates, fund federal spending and tax expenditures, or reduce the deficit.
The overall effect on the unified budget position would depend on the
extent to which any change in tax expenditures is offset by adjustments to
the tax code or other spending programs.

  Tax Expenditures Have Represented a Substantial Federal Commitment over Time

Whether gauged in absolute numbers, by revenues forgone, or in comparison
to federal spending or the size of the economy, tax expenditures have been
substantial over the last three decades. Between fiscal years 1974 and
2004, tax expenditures doubled in number, and the sum of estimated revenue
losses associated with tax expenditures tripled, most of which was
accounted for by tax expenditures that were used by individual taxpayers.
Since 1981 when outlay-equivalent estimates were first available, the sum
of the outlay-equivalent estimates for tax expenditures has been similar
in magnitude to discretionary spending, and this sum exceeded total
discretionary spending for most years during the last decade. As a share
of the U.S. economy, the sum of tax expenditure outlay-equivalent
estimates remained relatively stable at about 7.5 percent of GDP since the
last major tax reform legislation.

    Sums of Tax Expenditure Estimates Are Useful for Gauging Magnitude of Tax
    Spending but Need to Be Interpreted Carefully

Summing the individual tax expenditure estimates is useful for gauging the
general magnitude of the federal revenue involved, but it does not take
into account possible interactions between the individual tax code
provisions. Because of this limitation, sums of tax expenditure estimates
must be interpreted carefully. The JCT and Treasury estimate the revenue
loss from each tax expenditure separately, assuming that the rest of the
tax code remains unchanged. Neither JCT nor Treasury adds tax expenditure
estimates, because summing them does not take into account possible
interaction effects among the provisions. If two or more tax expenditures
were estimated simultaneously, the total change in federal revenue could
be smaller or larger than the sum of the amounts shown for each item
separately as a result of interactions among the tax expenditure
provisions. For example, the repeal of an itemized deduction tax
expenditure might cause more taxpayers to take the standard deduction
instead of itemizing. However, the revenue loss estimate for any single
tax expenditure among the itemized deductions does not reflect this
potentially sizeable interaction with the standard deduction. Eliminating
several itemized deductions at the same time could cause significant
numbers of taxpayers to take the standard deduction, and thus, the
decrease in revenue could be less than the sum of the estimated revenue
loss estimates for each itemized deduction. To demonstrate the magnitude
of possible interactions and the potential implications for summing tax
expenditures, Treasury's Office of

Tax Analysis illustrated for us the repeal of five itemized deductions.23
Based on tax year 2002 data, the sum of the five separate tax expenditure
estimates, each calculated assuming the rest of the tax code was
unchanged, was over $175 billion.24 Assuming the simultaneous repeal of
all five provisions, Treasury estimated the revenue loss after interaction
totaled $131 billion-about 25 percent less than the sum of the separate
estimates. According to Treasury, this example cannot be generalized given
that some groups of tax provisions have substantial interactions and
others do not. For all tax expenditures, the magnitude of the difference
between the sum of the estimates and an estimate for all tax expenditures
simultaneously is not known.

Additionally, tax expenditure estimates developed by Treasury and JCT do
not take into account possible behavioral responses by taxpayers if a tax
expenditure were repealed. For example, if the HOPE scholarship tax
credit-a tax credit for the first 2 years of post-secondary education-were
eliminated, taxpayers who would have used that tax credit may instead opt
for the Lifetime Learning tax credit or other tax subsidies aimed at
higher education. In contrast, certain kinds of behavioral responses, such
as changes in the timing of transactions, income recognition, or shifts
between sectors of the economy, are taken into account when JCT and
Treasury prepare revenue estimates for proposed legislation. Potential
macroeconomic effects, such as changes to GDP, are not reflected in tax
expenditure revenue loss estimates or in revenue estimates for proposed
legislation.

To some extent, the same kinds of challenges in interpreting tax
expenditure estimates also exist in projecting the costs of spending
programs. Budget line items generally do not reflect the actual budget
savings to be gained by abolishing specific programs or groups of
programs. For instance, eliminating all veterans' benefits would reduce
the federal budget by less than the amount currently spent on those
programs because spending likely would increase in food stamps, Medicaid,
and other entitlement programs. Although interaction effects also occur
for

23The 2002 estimates for the five itemized deductions were: (1) charitable
contributions, $33.8 billion; (2) home mortgage interest expenses, $69.2
billion; (3) state and local income taxes, $43.0 billion; (4) state and
local property taxes, $24.2 billion; and (5) medical expenses, $5.5
billion.

24The sum of Treasury's 2004 revenue loss estimates for the five itemized
deductions was about $168 billion; see app. III.

spending programs, Treasury officials responsible for developing tax
expenditure estimates told us that the bias in summing tax expenditure
revenue loss estimates likely is greater than the bias for outlay
projections. Whereas historical data are reported for federal budget
receipts and outlays, the last available values for tax expenditures
remain estimates. Treasury's last reported re-estimates for past fiscal
years reflect legislation enacted, prevailing economic conditions, and the
latest taxpayer data available at the time of estimation.25 Projections of
the future costs of tax expenditures are more uncertain than projections
for future tax receipts or outlays because it is not known with certainty,
even after the fact, how much was spent for any given tax expenditure.

Despite the limitations in summing separate tax expenditure revenue loss
and outlay-equivalent estimates, these are the best available data to
measure the value of tax expenditures and make comparisons to other
spending programs. Summing the estimates provides perspective on the use
of tax expenditures as a policy tool and represents a useful gauge of the
general magnitude of government subsidies carried out through the tax
code. The estimates also can be used to compare tax expenditures to
federal spending overall and by budget function. Other researchers also
have summed tax expenditure estimates to help gain perspective on the use
of this policy tool and examine trends in the aggregate growth of tax
expenditure estimates over time.26

Tax Expenditures Have More Between 1974 and 2004, tax expenditures
reported by Treasury more than

Than Doubled in Number and doubled in overall number from 67 to 146, and
while some were dropped,

Tripled in Size	considerably more were added. For 1974, Treasury listed 67
separate exclusions, exemptions, deductions, credits, preferential tax
rates, and

25Treasury's fiscal year 2004 re-estimates were published in February
2005, still months before 2004 tax returns would be filed. JCT projects
the revenue loss for future years and does not re-estimate tax
expenditures for past fiscal years.

26See for example: Eric Toder, The Changing Composition of Tax Incentives:
1980-1999. (Washington, D.C.: The Urban Institute, 1999); Dhammika
Dharmapala, Tax Expenditures Versus Direct Subsidies: A Review of the
Issues. (Austin, Tex.: National Tax Association, 91st Annual Conference,
1998); Christopher Howard, "Tax Expenditures," The Tools of Government: A
Guide to the New Governance, edited by Lester M. Salamon (New York, N.Y.:
Oxford University Press, 2002), pp. 410-444; Congressional Research
Service, Overview of the Federal Tax System (Washington, D.C.: 2005).

deferral of tax liability as tax expenditures.27 In 1986, Treasury
reported 115 tax expenditures, and by 2004 Treasury's list grew to 146 tax
expenditures. Figure 2 shows the rise of the overall number of tax
expenditures over the last three decades. (App. IV contains a compilation
of all tax expenditures reported by Treasury between 1974 and 2004.)

Figure 2: Number of Tax Expenditures Reported by Treasury, 1974-2004

Number 160

140

120

100

80

60

40

20

0 19741975

                  19761977197819791980198119821983198419851986
    198719881989199019911992199319941995199619971998199920002001200220032004

Fiscal year

Source: GAO analysis of OMB budget reports on tax expenditures, Fiscal
Years 1976-2006.

Note: The number of tax expenditures reflects all provisions reported by
Treasury, including those enacted but effective for future fiscal years.
For example, Treasury's last available list included eight new tax
expenditures enacted in 2004 that will be effective in fiscal year 2005
and later. In addition, fluctuations in the trend lines from year-to-year
may reflect changes in OMB's methodology. For example, the exclusion of
scholarship and fellowship income and several other tax expenditures were
excluded for fiscal year 1982, but included in prior and post years,
because of changes in Treasury's income tax baseline that defines a tax
expenditure.

Of the 146 tax expenditures listed by Treasury in the President's fiscal
year 2006 budget, 32 percent were on the first list in 1974, 23 percent
were added between 1975 and 1986, and 45 percent were added since 1986.
Figure 3 shows the duration of tax expenditures listed by Treasury. Of the
67 tax expenditures listed in 1974, 21 had been dropped over the period,
leaving

27The revenue loss estimates for fiscal year 1974 were reported in the
President's Budget for fiscal year 1976, published in 1975. The list has
been required by the Congressional Budget Act of 1974 in the budget
thereafter.

46 remaining on the list in 2004. Since 1974, 143 tax expenditures were
added to Treasury's list, although 43 of them have since dropped from the
list over the period. Of the 100 added since 1974 and still reported in
fiscal year 2004, 66 were first reported for 1986 or later.

Figure 3: Duration of Tax Expenditures Reported by Treasury, 1974-2004

Duration of tax expenditures

                                       1

                                      1 5

                                       7

                                       2

                                       9

                                      2 1

                                      1 6

11 3

                                       1

                                       3

5 1 1 9

                197419751976197719781979198019811982198319841985
  1986198719881989199019911992199319941995199619971998199920002001200220032004

Fiscal year

Source: GAO.

Note: The number of tax expenditures reflects all provisions reported by
Treasury, including those enacted but effective for future fiscal years.
For example, Treasury's last available list included eight new tax
expenditures enacted in 2004 that will be effective in fiscal year 2005
and later. Fluctuations in

the trend lines from year-to-year may reflect changes in OMB's
methodology. For example, the exclusion of scholarship and fellowship
income and several other tax expenditures were excluded for fiscal year
1982, but included in prior and post years, because of changes in
Treasury's income tax baseline that defines a tax expenditure.

The number of tax expenditures reported by Treasury has changed over time
for several reasons. Some provisions expired or were repealed; others were
merged with another tax expenditure. For example, until expiration on
December 31, 1984, state and local governments were allowed to issue
tax-exempt obligations to finance the purchase of mass-commuting vehicles
for lease to government transit agencies; the Tax Reform Act of 198628
repealed the investment tax credit; and the tax expenditure that provided
5-year amortization for pollution control was merged into the investment
tax credit by the Tax Reform Act of 1976.29 Legislation also added new tax
expenditures over time, such as the child tax credit created by the
Taxpayer Relief Act of 1997.30 Some tax expenditures split into additional
listings to reflect legislation expanding existing tax expenditures. For
example, Treasury began listing the net exclusion of pension contributions
and earnings with separate estimates for employersponsored defined-benefit
and 401(k) pension plans following 2001 legislation increasing the
contribution limits for 401(k) accounts. Finally, changes in the baseline
used by Treasury to identify tax expenditures may have caused some tax
expenditures to drop off its list, while adding new tax expenditure
listings.31 For example, Treasury briefly dropped the exclusion of
scholarship and fellowship income from its fiscal year 1982 list because
it was not considered a tax expenditure under the baseline that Treasury
used that year.

As the overall number reported by Treasury doubled, the sum of the
estimated revenue loss due to tax expenditures, adjusted for inflation,
tripled from approximately $243 billion for 1974 to $728 billion for
2004.32 Figure 4 shows the sum of Treasury's revenue loss estimates over
the past

28Pub. L. No. 99-514, October 2, 1986.

29Pub. L. No. 94-455, October 4, 1976.

30Pub. L. No. 105-34, August 5, 1997.

31To determine the tax code provisions that satisfy the definition of a
tax expenditure, the existing tax law must be compared or measured against
an alternative set of tax rules that represent a baseline. App. III
discusses in more detail the baselines used by Treasury as well as how
Treasury measures and reports tax expenditures.

32The tax expenditure revenue loss estimates take into account the AMT
liability.

three decades. From 1974 to 1986, revenue losses increased by nearly two
and one-half times from approximately $243 billion for 1974 to $598
billion for 1986 (in 2004 dollars). Over the next 2 years, the sum of the
revenue losses decreased by about 28 percent to approximately $433 billion
for 1988. From 1989 through 1997, however, revenue losses increased by
approximately 16 percent to approximately $547 billion. From 1998 to 2002,
the sum of the estimated revenue loss increased by an average of about $41
billion per year, peaking at about $783 billion for 2002. The sum of the
revenue loss estimates declined to approximately $728 billion in 2004.

Figure 4: Sum of Tax Expenditure Revenue Loss Estimates, 1974-2004

Dollars in billions (in constant 2004 dollars) 900

800

700

600

500

400

300

200

100

0

              1974197519761977197819791980198119821983198419851986
    198719881989199019911992199319941995199619971998199920002001200220032004

Fiscal year

Source: GAO Analysis of OMB budget reports on tax expenditures, Fiscal
Years 1976-2006.

Note: Summing the revenue loss estimates does not take into account
possible interaction effects among the tax expenditures that we mentioned
earlier in the report. Changes in economic conditions and estimation
techniques can affect revenue loss estimates for tax expenditures, making
them differ from year to year. Changes to the number of tax expenditures
reported by Treasury would also affect the amount of revenue loss reported
if some tax expenditures were eliminated or added. Finally, revenue loss
estimates include the effect of certain tax credits on receipts only and
not the effect of the credits on outlays.

The revenue loss estimates do not reflect the outlays for the refundable
portion for certain tax credits.33 Summing these outlays along with the
sum of the revenue loss estimates provides a more complete picture of the
aggregate cost of tax expenditures throughout the period, as shown in
figure 5. The sum of the estimated revenue losses and outlays associated
with tax expenditures totaled about $770 billion for fiscal year 2004.

    Figure 5: Sum of Tax Expenditure Revenue Loss Estimates with Outlays for
                       Refundable Tax Credits, 1974-2004

                 Dollars in billions (in constant 2004 dollars)
1974197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004

Fiscal year

Tax expenditure revenue loss estimates

Tax expenditure revenue loss estimates plus outlays for refundable tax
credits

Source: GAO analysis of OMB budget reports on tax expenditures, fiscal
years 1976-2006.

Note: Reflects refundable amounts for the EITC from 1976 to 2004, the
child tax credit for 1997 to 2004, the child insurance medical premium
credit for 1992 to1993, and the tax credit for health insurance purchased
by certain displaced and retired individuals for 2004. Summing the revenue
loss estimates does not take into account possible interaction effects
among the tax expenditures that we mentioned earlier in the report.
Changes in economic conditions and estimation techniques can affect
revenue loss estimates for tax expenditures, making them differ from year
to year. Changes to the number of tax expenditures reported by Treasury
would also affect the amount of revenue loss reported if some tax
expenditures were eliminated or added. Finally, revenue loss estimates
include the effect of certain tax credits on receipts only and not the
effect of the credits on outlays.

33In fiscal year 2004, the associated outlays were $33.1 billion for the
EITC, $8.9 billion for the child tax credit, and $0.7 billion for the
health insurance tax credit for certain displaced and retired individuals.

Trends in the sum of tax expenditures are due, at least in part, to
legislation affecting the number or scope of tax expenditures or modifying
tax rates or other basic structural features of the tax code. During this
period, tax legislation directly influenced the sum of tax expenditure
estimates by repealing or limiting some tax expenditures, enacting new
ones, and extending the life of expiring tax expenditures. Even without
changes to tax expenditures, legislation affecting tax rates or the tax
structure affects the sum of the tax expenditure estimates. When a
taxpayer uses a tax expenditure, his or her effective tax rate34 is
reduced, because some part of his or her income remains untaxed or is
taxed at a lower rate. When statutory rates increase, a taxpayer's ability
to avoid tax on a portion of income is worth more; consequently, tax
expenditures are worth more. Likewise, when rates decrease, tax
expenditures are worth relatively less.

Figure 6 highlights tax legislation enacted since 1974 that likely
influenced the aggregate revenue losses due to tax expenditures. The sum
of estimated revenue losses declined following the Tax Reform Act of
1986,35 primarily because of individual and corporate marginal tax rate
reductions which indirectly scaled back the value of all but a few tax
credits. The 1986 act, which created the last major tax reform, also
eliminated or limited the scope of various tax expenditures directly, for
example, by repealing the investment tax credit, phasing out the interest
deduction for consumer credit over 5 years, and limiting the expensing of
the intangible drilling costs for oil and gas to successful, domestic
wells. While materially reducing the number and scope of tax expenditures
broadened the tax base, the act resulted in no net change in federal
revenue because of the lower tax rates. In contrast, the sum of estimated
revenue losses increased following the Omnibus Budget Reconciliation Act
of 1993,36 which directly increased several tax expenditures-for example,
extending the EITC to single workers with no children earning $9,000 or
less-and indirectly increased the value of other tax expenditures by
increasing the top individual income tax rates and adding a third rate.
The sum of estimated revenue losses accelerated following the Taxpayer
Relief Act of 1997, which expanded several tax expenditures-for example,
increasing eligibility for traditional individual retirement accounts-and
created an assortment of new tax expenditures, including the child tax
credit and

34The effective tax rate is the ratio of taxes paid to a taxpayer's total
income.
35Pub. L. No. 99-514, October 21, 1986.
36Pub. L. No. 103-66, August 10, 1993.

postsecondary education tax incentives. The Economic Growth and Tax Relief
Reconciliation Act of 200137 reduced tax rates again and also increased
the individual AMT exemption. The influence on the aggregate trend is less
apparent for legislation expanding or adding tax expenditures while also
reducing tax rates.

37Pub. L. No. 107-16, June 7, 2001.

Figure 6: Tax Legislation Enacted From 1974-2004 That May Have Influenced
the Sum of Revenue Loss Estimates for Tax Expenditures

900 Billions of dollars (in constant 2004 dollars)

7419197519

    76197719781979198019811982198319841985198619871988198919901991199219931

99419

                                      951

99619

                          9719981999200020012002200320

40Fiscal year

The Revenue Act of 1978: P.L. 95-600, November 6, 1978, made the EITC a
permanent provision of the Internal Revenue Code and reduced individual
and corporate tax rates.

The Economic Recovery Tax Act of 1981: P.L. 9734, August 13, 1981, enacted
several new tax expenditures, expanded existing tax expenditures for
individuals and businesses, and phased in a 23 percent cut in individual
tax rates.

The Economic Growth and Tax Relief Reconciliation Act of 2001: P.L.
107-16, June 7, 2001, expanded several tax expenditures, lowered tax
rates, and created an increase in the individual alternative minimum tax
exemption amount.

individuals.	August 5, 1997, expanded The Jobs and Growth Tax Relief
several tax expenditures Reconciliation Act of 2003: P.L. 108-27,and
created many new May 28, 2003, increased the child tax credit,

The Omnibus Budget provisions, such as the and accelerated reductions in
the regular
Reconciliation Act of child, HOPE, and Lifetime income tax rates that were
scheduled for
1990: P.L. 101-508, Learning education 2004 through 2006.

November 5, 1990, credits.
expanded family eligibility
for the EITC and replaced American Jobs Creation Act of 2004: P.L.
108-357, October 22, 2004,
the top three marginal tax extended several tax provisions that expired in
2003 or were scheduled
rate brackets of the Tax to expire in 2004, including the research and
development tax credit;
Reform Act of 1986 with two simplified the rules for utilizing foreign tax
credits; and reduced the
brackets. corporate tax rate from 35 percent for domestic manufacturers,

                  producers, farmers, and small corporations.

Source: GAO analysis of OMB budget reports on tax expenditures, fiscal
years 1976-2005.

Note: The effects of legislative changes on tax expenditure estimates
might not have occurred within the same year that the legislation was
enacted. Summing the revenue loss estimates does not take into account
possible interaction effects among the tax expenditures that we mentioned
earlier in the report. Changes in economic conditions and estimation
techniques, can affect revenue loss estimates

for tax expenditures, making them differ from year to year. Changes to the
number of tax expenditures reported by Treasury would also affect the
amount of revenue loss reported if some tax expenditures were eliminated
or added. Finally, revenue loss estimates include the effect of certain
tax credits on receipts only and not the effect of the credits on outlays.

Changes in economic conditions and in the baseline tax system can also
affect revenue loss estimates for tax expenditures, making them differ
from year to year. For example, rising housing prices may cause the
estimated cost of the mortgage interest deduction to increase as
homeowners finance larger mortgages or take out equity with home equity
loans. In addition, changes in tax expenditure baselines could also cause
estimates to differ from year to year. For example, for fiscal years 2003
and 2004, Treasury redefined accelerated depreciation tax expenditures so
that they are calculated relative to a replacement cost basis baseline
rather than the historic cost basis previously used. This redefinition had
the effect of reducing the estimated size of the accelerated depreciation
tax expenditures.38

    Tax Expenditures for Individual Taxpayers Accounted for Most of the Sum of
    Tax Expenditure Revenue Losses

The sum of estimated revenue losses due to tax expenditures for individual
income taxpayers accounted for substantially more of the revenue loss
between 1974 and 2004 than corporate tax expenditures, as shown in figure

7. The sum of revenue loss estimates for tax expenditures that arise under
the individual income tax increased from approximately $187 billion for
1974 to $487 billion for 1987 (in 2004 dollars). After decreasing to
approximately $363 billion for 1988, the sum gradually increased to a high
of approximately $688 billion for 2002 and then declined in 2003 and 2004.
On average over the entire period, revenue loss estimates for individual
income taxpayers accounted for about 83 percent of the sum of revenue loss
estimates per year. While estimated revenue losses for all tax
expenditures tripled, the sum of revenue loss estimates for corporate tax
expenditures increased from approximately $57 billion for fiscal year 1974
to a high of about $116 billion in 1984 (in 2004 dollars). After 1984, the
sum dropped back to approximately $57 billion in 1992 and increased
slightly over the rest of the period, with some fluctuation between years.
In 2004, revenue loss estimates for tax expenditures that arise under the
corporate income tax accounted for 11 percent of the sum of revenue losses
due to all

38This is one of the reasons why the sum of tax expenditure revenue loss
estimates by Treasury and JCT, which had been tracking each other rather
closely since 1987, diverged in 2003 and 2004. While Treasury has changed
its tax expenditure baselines over time, JCT's tax expenditure baseline
has changed little over time. See app. III for additional information on
how JCT and Treasury measure tax expenditures.

tax expenditures. At about 10 percent of total federal receipts, corporate
income taxes also accounted for a smaller share than individual income
taxes.

      Figure 7: Sum of Revenue Loss Estimates by Taxpayer Group, 1974-2004

                 Dollars in billions (in constant 2004 dollars)

                                    N/A N/A

197419751976 1977 197819791980 1981 1982 1983198419851986 198719881989 1990 1991
    199219931994 1995 1996 19971998 1999 2000 20012002 20032004 Fiscal year

Individual Corporate

Source: GAO analysis of OMB budget reports on tax expenditures, fiscal
years 1976-2006.

Note: Treasury did not report separate estimates for the individual and
corporate income tax expenditures for fiscal years 1981 and 1982. Total
revenue loss estimates for fiscal years 1981 and 1982 were approximately
$410 and $448 billion (in 2004 dollars). The location of tax expenditure
estimates under the individual and corporate tax expenditure headings does
not imply that these categories of filers benefit from the special tax
provisions in proportion to the respective tax expenditure amounts shown.
For instance, the ultimate beneficiaries of corporate tax expenditures
could be shareholders, employees, customers, or other providers of
capital, depending on economic forces. In addition, summing the revenue
loss estimates does not take into account possible interaction effects
among the tax expenditures that we mentioned earlier in the report.
Changes in economic conditions and estimation techniques, can affect
revenue loss estimates for tax expenditures, making them differ from year
to year. Changes to the number of tax expenditures reported by Treasury
would also affect the amount of revenue loss reported if some tax
expenditures were eliminated or added. Finally, revenue loss estimates
include the effect of certain tax credits on receipts only and not the
effect of the credits on outlays.

The sum of revenue loss estimates due to individual income tax
expenditures is primarily attributable to a small number of large tax
expenditures. The fourteen tax expenditures listed in table 2-each with an
annual revenue loss estimated at $20 billion or more-accounted for

about 75 percent of the sum of revenue losses for fiscal year 2004. Ten of
the 14 largest tax expenditures focused entirely on individual taxpayers,
and the other 4 were available for individuals and corporations. Most of
the largest tax expenditures are long-standing ones, and only 2 of the 14
were added to the tax code since 1986.39 The child tax credit, enacted in
1997, is among the largest tax expenditures based on its estimated revenue
losses alone, not counting associated outlays of $8.9 billion in fiscal
year 2004. With revenue losses estimated at $4.9 billion, the EITC does
not appear on this list; if $33.1 billion in associated outlays were
included, this refundable credit ranks among the largest tax expenditures.

39The exclusion of net imputed rental income on owner-occupied homes is a
long-standing feature of the income tax system. It was first listed as a
tax expenditure by Treasury in the 2006 budget; JCT does not list this tax
expenditure.

Table 2: Revenue Loss Estimates for the Largest Tax Expenditures Reported
for Fiscal Year 2004, with Taxpayer Group and Budget Function

                                      Revenue loss             
                         Revenue loss estimate as              
                                      a                        
                             estimate   percentage             
      Tax expenditure    (Billions $)       of sum Taxpayer        Budget     
                                                   group          function    
Income tax exclusion        $102.3        14.0% Individual      Health     
        of employer                                            
     contributions to                                          
     medical insurance                                         
         premiums                                              
     and medical carea                                         
     Deductibility of                                                         
mortgage interest on          61.5         8.4% Individual    Commerce and 
           owner                                                      housing
      occupied homesa                                              credit     
     Net exclusion of                                                         
pension contributions         47.7         6.6% Individual      Income     
            and                                                   security
     earnings: 401(k)                                          
     Net exclusion of                                                         
pension contributions         47.0         6.5% Individual      Income     
            and                                                   security
    earnings: employer                                         
          plansa                                               
     Deductibility of                                                         
nonbusiness state and         45.3         6.2% Individual  General        
           local                                               purpose fiscal
taxes (other than on                                                       
      owner-occupied                                             assistance
          homes)a                                              
        Accelerated                                                           
      depreciation of            44.7         6.1%  Corporate    Commerce and 
       machinery and                                   and            housing
         equipment                                 Individual      credit     
       Capital gains                                                          
     exclusion on home           29.7         4.1% Individual    Commerce and 
           sales                                                      housing
                                                                   credit     
     Deductibility of                                                         
        charitable               27.4         3.8%  Corporate    Education,   
    contributions other                                and       training,
    than education and                             Individual   employment,   
          healtha                                                  social     
                                                                  services    
Exclusion of interest                                                      
     on public purpose           26.2         3.6%  Corporate  General        
           state                                       and     purpose fiscal
     and local bondsa                              Individual    assistance   
Capital gains (other                                                       
     than agriculture,           25.2         3.5% Individual    Commerce and 
          timber,                                                     housing
iron ore, and coal)a                                            credit     
     Exclusion of net                                                         
imputed rental income         24.6         3.4% Individual    Commerce and 
            on                                                        housing
owner-occupied homes                                            credit     
     Step-up basis of                                                         
     capital gains at            24.2         3.3% Individual    Commerce and 
           death                                                      housing
                                                                   credit     
Child credit (effect          22.4         3.1% Individual    Education,   
     on receipts only)                                           training,    
                                                                  employment, 
                                                                   and social 
                                                                  services    
Exclusion of interest                                                      
     on life insurance           20.1        2.98%  Corporate    Commerce and 
         savingsa                                      and            housing
                                                   Individual      credit     

Source: GAO analysis of OMB budget report on tax expenditures, fiscal year
2006.

aDenotes tax expenditures that have been reported by Treasury since 1974.

Note: Some tax expenditures split into additional listings to reflect
legislation expanding existing tax expenditures. For example, Treasury
began listing the net exclusion of pension contributions and earnings with
separate estimates for employer-defined benefits and 401(k) pension plans
following 2001 legislation increasing the contribution limits for 401(k)
accounts. From year to year, revenue loss estimates may change because
Treasury updates their estimates for each new budget to reflect
legislation enacted, prevailing economic conditions, and the latest
taxpayer data available. Although there are substantial revenues forgone
for these 14 large tax expenditures, the estimated amount of

federal spending that would be required to provide equivalent assistance
is frequently larger than the revenue forgone because this spending could
be subject to income tax. For example, the outlayequivalent estimate for
the income tax exclusion of employer contributions for medical insurance
premiums and medical care is $126.7 billion for fiscal year 2004.
Outlay-equivalent estimates for tax expenditures are discussed in more
detail in app. III.

Tax expenditure revenue loss estimates reflect federal income tax revenue
forgone and do not account for provisions that exclude certain earnings
from payroll taxes. For example, the income tax exclusion for health care
not only permits the value of health insurance premiums to be excluded
from the calculation of employees' taxable earnings for income taxes but
also excludes the value of the premiums from the calculation of Social
Security and Medicare payroll taxes for both employees and employers.40
Some researchers have estimated that these payroll tax revenue losses
amount to more than half of the income tax revenue losses. 41 If payroll
tax revenue losses were 50 percent of the $102.3 billion in income tax
revenue loss estimated by Treasury, the combined revenue loss associated
with the exclusion of employer contributions for health insurance premiums
would be $153.5 billion in 2004.

    While Tax Expenditures Have Exceeded Discretionary Spending in Some Years,
    They Have Remained Relatively Stable as a Share of the U.S. Economy

The sum of tax expenditure outlay-equivalent estimates exceeded the amount
of discretionary spending for most years during the last decade, as shown
in figure 8.42 Outlay-equivalent estimates, introduced by Treasury in
1981, allow the value of a tax expenditure to be compared with a direct
federal outlay. The sum of the outlay-equivalent estimates reported by
Treasury was approximately $853 billion in 2004.43 Until 1987, the sum of
outlay-equivalent estimates for tax expenditures was roughly the same
magnitude as discretionary spending. From 1988 through 1995, the sum of
tax expenditure outlay-equivalent estimates averaged about $104 billion
(in

40Employers may deduct their premium payments as a business expense in
calculating their net taxable income; this deduction, like those for other
labor costs, is not a tax benefit for employers.

41John Sheils and Randall Haught, "The Cost of Tax-Exempt Health Benefits
in 2004," Health Affairs (Feb. 25, 2004 ); and Leonard E. Burman and
Jonathan Gruber, "Tax Credits for Health Insurance," Tax Policy Center
Discussion Paper No. 19 (Washington, D.C.: The Tax Policy Center, June
2005).

42When measured in terms of sum of revenue loss estimates, tax
expenditures also exceeded discretionary spending in some years.

43The sum of tax expenditure outlay-equivalent estimates plus outlays
associated with refundable credits amounted to $896 billion in fiscal year
2004.

2004 dollars) less than annual discretionary spending. Beginning in 1996,
the sum of tax expenditure outlay-equivalent estimates surpassed
discretionary spending and averaged about $114 billion (in 2004 dollars)
more than annual discretionary spending through 2003. However, in 2003,
the sum of Treasury's tax expenditure estimates declined markedly, and the
sum of tax expenditure outlays fell below discretionary spending in fiscal
year 2004. This decline may be due, at least in part, to changes in the
way Treasury defined and measured several tax expenditures in these years.
Just as the sum of tax expenditure outlay-equivalent estimates increased
since the late 1990s, discretionary spending also increased over this
period. Between 1996 and 2002, the sum of tax expenditure estimates
increased by an average of approximately $46 billion annually, while
discretionary spending increased by an average of $21 billion annually (in
2004 dollars). Mandatory spending-larger than the sum of tax expenditure
estimates or discretionary spending-consistently rose over the period
shown by an average of $43 billion annually (in 2004 dollars).

Figure 8: Sum of Tax Expenditure Outlay-equivalent Estimates Compared with
Total Mandatory and Total Discretionary Outlays,

1981-2004

Dollars in billions (in constant 2004 dollars)

1,500 1,400

1,300

1,200

1,100

1,000

900

800

700

600

500 400 300 200 100

0 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Fiscal year

                               Mandatory spending

                             Discretionary spending

                  Tax expenditure outlay equivalent estimates

  Source: GAO analysis of OMB budget reports on tax expenditures, fiscal years
                                   1983-2006

Note: Mandatory spending includes net interest. Summing the
outlay-equivalent estimates does not take into account possible
interaction effects among the tax expenditures. In addition, tax
expenditure estimates developed in different years generally use different
economic, demographic, and other assumptions. Finally, changes to the
number of tax expenditures reported by Treasury would also affect the
amount of outlay-equivalent estimates reported if some tax expenditures
were eliminated or added.

Figure 9 compares tax expenditures and federal outlays as a share of GDP
as a way to measure the amount of federal spending through the tax code
and other programs relative to the economy.44 As a share of the U.S.
economy, the sum of tax expenditure outlay-equivalent estimates peaked at
10.9 percent of GDP in 1986. Since 1988, the sum of tax expenditure
outlays has remained relatively stable at about 7.5 percent of GDP. Over
the period shown, mandatory spending also was fairly constant as a share
of the economy, at an average of 12.7 percent of GDP. As a share of the
economy, discretionary spending declined from 10.1 percent of GDP in 1981
to 6.3 percent in 1999 and 2000, with some fluctuation between the years.
In recent years, discretionary spending has grown faster than the economy,
increasing to 7.8 percent of GDP in fiscal year 2004. Averaging about 18.0
percent of GDP in the 1980s through the early 1990s, federal receipts
steadily rose to 20.9 percent of GDP in 2000 and since declined to 16.3
percent of GDP in fiscal year 2004. With total federal outlays- including
mandatory and discretionary spending plus net interest- reaching 19.9
percent of GDP, the federal unified budget deficit amounted to 3.6 percent
of GDP ($412 billion) in fiscal year 2004. The on-budget deficit in fiscal
year 2004 amounted to 4.9 percent of GDP ($567 billion).45

44Expressing tax expenditures as a share of the nation's economy provides
the context for assessing trends in federal revenue and spending. GDP is a
commonly used measure of domestic national income. GDP is the value of all
goods and services produced within the United States in a given year and
is conceptually equivalent to incomes earned in production. It is a rough
indicator of the economic earnings base from which the government draws
its revenues.

45Whereas the unified budget is a consolidated measure of federal
activity, the on-budget deficit excludes Social Security and the Postal
Service which are off-budget under current law. In fiscal year 2004, the
off-budget surplus included a $151 billion Social Security surplus and a
$4 billion surplus for the Postal Service.

Figure 9: Sum of Tax Expenditure Outlay-Equivalent Estimates Compared to
Total Mandatory and Discretionary Outlays and Receipts as a Percentage of
GDP, 1981-2004

Percent of GDP

                                       25

                                       20

                                       15

                                       10

                                       5

                                       0

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
1996 1997 1998 1999 2000 2001 2002 2003 2004 Fiscal year

Receipts

Mandatory spending

Discretionary spending

Tax expenditure outlay equivalent estimates

Source: GAO analysis of OMB budget reports on tax expenditures, fiscal
years 1983-2006.

Note: Mandatory spending includes net interest. Whereas the mandatory and
discretionary numbers represent actual money that was spent by the federal
government, tax expenditure figures are estimates. In addition, summing
the outlay-equivalent estimates does not take into account possible
interaction effects among the tax expenditures that we mentioned earlier
in the report. Tax expenditure estimates developed in different years
generally use different economic, demographic, and other assumptions.
Changes to the number of tax expenditures reported by Treasury would also
affect the amount of outlay-equivalent estimates reported if some tax
expenditures were eliminated or added.

Tax expenditures span almost all federal mission areas, but their relative
size differs across budget functions. To gauge the relative role of tax
expenditures, the sum of tax expenditure outlay-equivalent estimates and
federal outlays can be compared to total spending by budget function. For
2004, Treasury reported tax expenditures for 16 of 20 budget functions.
Five of the functions accounted for 91 percent of the sum of the tax
expenditure outlay-equivalent estimated dollar amounts in 2004- commerce
and housing credit; education, training, employment and social services;
income security; health; and general government, as shown in figure 10.
(See app. III for a list of tax expenditures reported for 2004 by budget
function.) For the most part, these same five budget functions accounted
for the largest percentage of total outlay-equivalent estimates

over time, although the relative size of the estimated outlay-equivalent
dollar amounts for the five budget functions varied somewhat over the
period shown. For example, the health and the education, training,
employment and social services budget functions more than doubled between
1986 and 2002 (in 2004 dollars).

Figure 10: Size of Tax Expenditure Outlay-Equivalent Estimates by Budget
Function, 1981-2004 Dollars in billions (in constant 2004 dollars)

1000 900 800 700 600 500 400 300 200 100 0 1981 1982 1983 1984 1985 1986
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
2002 2003 2004

Other

Education, training, employment, and social services

General government

Health

Income security

Commerce and housing

Source: GAO Analysis of OMB Budget Reports on Tax Expenditures, Fiscal
Years 1983-2006.

Note: The other budget functions that Treasury lists tax expenditures
under are national defense; international affairs; general science, space,
and technology; energy; natural resources and environment; agriculture;
transportation; community and regional development; social security;
veterans' benefits and services; and net interest. The general government
budget function includes tax expenditures listed by Treasury under the
general purpose fiscal assistance budget subfunction. For two budget
functions-commerce and housing credit; and education, training,
employment, and social services-tax expenditures are listed by
subfunction. In addition, summing the outlayequivalent estimates by budget
function does not take into account possible interaction effects among the
expenditures. Tax expenditure estimates developed in different years
generally use different economic, demographic, and other assumptions.
Changes to the number of tax expenditures reported by Treasury would also
affect the amount of outlay-equivalent estimates reported if some tax
expenditures were eliminated or added.

The sum of the tax expenditure outlay-equivalent estimates was greater
than what the federal government spends in discretionary and mandatory
spending for some budget functions. As shown in figure 11, the sum of the
tax expenditure outlay-equivalent estimates exceeded federal outlays for
three budget functions: energy, 46 commerce and housing credit, and
general government. Outlay-equivalent estimates for tax expenditures in
the commerce and housing credit budget function totaled $300 billion for
2004, while budget outlays for that function totaled $5 billion.47 Seven
of the 14 largest tax expenditures, listed in table 2 with revenue losses
exceeding $20 billion in 2004, were reported under the commerce and
housing credit budget function. The mortgage interest deduction-the second
largest single tax expenditure in fiscal year 2004-had an outlayequivalent
estimate of $61.5 billion, compared to $45 billion in outlays for the
Department of Housing and Urban Development, which is responsible for,
among other things, mortgage credit and housing assistance programs.48
Various tax expenditures for accelerated depreciation and capital gains
listed under the commerce and housing credit budget function also provide
incentives for a wide range of different investments that can affect other
federal mission areas. The general government budget function included two
of the largest tax expenditures-the deduction of state and local income
and sales tax, and the exclusion of interest on public purpose state and
local bonds-which together accounted for about $71.5 billion in tax
expenditures outlays.49

46Total outlays for the energy budget function were negative in fiscal
year 2004; energy discretionary outlays were $3.4 billion and energy
mandatory outlays were negative $3.6 billion. For more information on the
major federal energy-related tax and spending programs, see GAO, National
Energy Policy: Inventory of Major Federal Energy Programs and Status of
Policy Recommendations, GAO-05-379 (Washington, D.C.: June 10, 2005).

47Within the commerce and housing credit budget function, outlays in
fiscal year 2004 by subfunction were: $ 2.7 billion for mortgage credit,
negative $4.1 billion for the Postal Service, negative $2 billion for
deposit insurance, and $8.7 billion for other advancement of commerce.

48Housing assistance is a subfunction under the income security budget
function.

49The deduction for property taxes on owner-occupied homes is listed under
the commerce and housing credit budget function. Estimated at $19.9
billion in outlay-equivalent value, this deduction was one of the 15
largest tax expenditures in fiscal year 2004. Including this deduction,
the sum of the outlay-equivalent estimates for tax expenditures related to
state and local governments amounted to $102.7 billion in fiscal year
2004. The outlay-equivalent value of the exclusion for state-and local
private purpose bonds totaled $9.4 billion in fiscal year 2004. These
amounts are reflected across the related budget functions.

Figure 11: Tax Expenditure Outlay-equivalent Estimates Compared with
Federal Outlays by Budget Function, Fiscal Year 2004 Dollars in billions
National defense

International affairs

General science, space and technology

Energy

Natural resources and environment

Agriculture

Commerce and housing credit

Transportation

Community and regional development

                         Education, training, employment, and social services

Health

Medicare

Income security

Social security

Veterans benefits and services

Administration of justice

General government

Net interest

0 50 100 150 200 250 300 350 400 450 500

Budget function

Tax expenditure outlay equivalent estimates Federal outlays

  Source: GAO analysis of OMB budget reports on tax expenditures, fiscal year
                                     2006.

Note: Federal outlays for the energy budget function were negative because
revenues for this function exceeded spending in 2004. Summing the
outlay-equivalent estimates by budget function does not

take into account possible interaction effects among the expenditures.
Outlays associated with the refundable child credit and EITC are shown in
the outlay bars.

As figure 11 shows, the sum of outlay-equivalent estimates for tax
expenditures was nearly the same magnitude as outlays in two budget
functions: international affairs and education, training, employment, and
social services. Within the education, training, employment, and social
services budget function, the sum of outlay-equivalent estimates of the
tax expenditures represented 49 percent of the total federal support.50
This budget function includes two of the largest tax expenditures-the
child tax credit and charitable contributions other than for health. The
sum of the outlay-equivalent estimates for tax expenditures was
substantially less than total outlays in the health and income security
budget functions.51 The income tax exclusion for employer-provided health
care-the largest single tax expenditure-accounted for 12 percent of the
sum of tax expenditure outlay-equivalent estimates and represented about
27 percent of total federal support in the health function, which includes
Medicaid. Outlays in the income security function include mandatory
outlays refunded under the EITC and child tax credit. No tax expenditures
are reported by Treasury for two budget functions: administration of
justice and Medicare.52

50Total federal support includes federal outlays and tax expenditure
outlay-equivalent amounts within the same budget function.

51Treasury and JCT list the EITC tax expenditure under the income security
budget function and the child tax credit under the education, training,
employment and social services budget function.

52JCT does not list any tax expenditures under the administration of
justice budget function but does list exclusions for Medicare benefits as
a tax expenditure under the Medicare budget function.

  Systematic Review of Tax Expenditures Is Integral to Reexamining the Federal
  Base, but Little Progress Has Been Made Since 1994 to Increase Scrutiny

Although tax expenditures represent a substantial federal commitment of
resources, little progress has been made in the Executive Branch to
increase the transparency of and accountability for tax expenditures. The
entire set of tools the federal government can use to address national
objectives-including discretionary and mandatory spending, tax provisions,
loans and loan guarantees-should be subject to periodic reviews and
reexamination to ensure that they are achieving their intended purposes
and designed in the most efficient and effective manner. The nation's
current and projected fiscal imbalance provides an additional impetus for
engaging in such a review and reassessment. Tax expenditures may not
always be efficient, effective, or equitable, and consequently,
information on these attributes can help policymakers make more informed
decisions as they adapt current policies in light of our fiscal challenges
and other overarching trends. In addition, some tax expenditures, at least
as currently designed, may serve to exacerbate other key private sector
and public policy challenges (e.g., controlling health care costs). To
review tax expenditures, information is needed to assess economic
efficiency, effectiveness, distributional equity, and administration and
compliance costs, although data and methodological challenges may impede
studies of some tax expenditures. Over the past decade, the Executive
Branch made little progress to integrate tax expenditures in the budget
presentation and review processes that apply to spending programs, as we
recommended in 1994.

    Long-Term Fiscal Challenge Provides Additional Impetus to Reexamine Federal
    Spending and Tax Policies, Including Tax Expenditures

Simply put, our nation's fiscal policy is on an unsustainable course.
Longterm simulations by GAO, the Congressional Budget Office (CBO), and
others show that over the long term we face large, escalating, and
persistent deficits due primarily to known demographic trends and rising
health care costs.53 This unsustainable fiscal path will gradually erode
the nation's economy and increasingly constrain the federal government's
capacity to address emerging challenges and opportunities. The long-term
fiscal challenge is too big to be solved by economic growth alone or by
making modest changes to existing spending and tax policies, including tax

53For additional information on our long-term fiscal modeling, see GAO,
Our Nation's Fiscal Outlook: The Federal Government's Long-Term Budget
Imbalance, http://www.gao.gov/special.pubs/longterm/. See also U.S.
Congressional Budget Office, The Long-Term Budget Outlook (Washington,
D.C.: December 2003), and the Office of Management and Budget, Analytical
Perspectives, Budget of the United States Government, Fiscal Year 2006
(Washington, D.C.: 2005).

expenditures. In addition, the long-term fiscal challenge makes it all the
more important to ensure all major federal spending and tax programs and
policies-including tax expenditures-are efficient, effective, and
relevant. The revenues forgone through tax expenditures either reduce
resources available to fund other federal activities or require higher tax
rates to raise a given amount of revenue.

Our long-term simulations illustrate the magnitude of fiscal challenges we
will face in the future.54 Figures 12 and 13 present these simulations
under two different sets of assumptions. In figure 12, we begin with CBO's
August 2005 baseline-constructed according to the statutory requirements
for that baseline.55 Consistent with these requirements, this simulation
assumes that discretionary spending grows with inflation for the first 10
years, and that tax cuts which are currently scheduled to expire will
expire. After 2015, discretionary spending is assumed to grow with the
economy, and revenue is held constant as a share of GDP at the 2015 level.
In figure 13, only two assumptions are changed: (1) discretionary spending
is assumed to grow with the economy rather than merely with inflation for
the entire period (not just after 2015), and (2) all tax cuts which are
currently scheduled to expire are made permanent. For both simulations,
Social Security and Medicare spending is based on the 2005 Trustees'
intermediate projections, and we assume that benefits continue to be paid
in full after the trust funds are exhausted. Medicaid spending is based on
CBO's December 2003 long-term projections under mid-range assumptions.

54Long-term simulations illustrate the relative fiscal and economic
outcomes associated with alternative policy paths and should not be viewed
as precise forecasts.

55 Congressional Budget Office, The Budget and Economic Outlook: An Update
(Washington, D.C.: August 2005).

Figure 12: Composition of Spending as a Share of GDP under Baseline
Extended Percent of GDP 50

40

30

20

10

0 2004 2015 2030 2040 Fiscal year

All other spending
Medicare & Medicaid
Social Security
Net interest
Source: GAO's August 2005 analysis.

Notes: In addition to the expiration of tax cuts, revenue as a share of
GDP increases through 2015 due to (1) real bracket creep, (2) more
taxpayers becoming subject to the AMT, and (3) increased revenue from
tax-deferred retirement accounts. After 2015, revenue as a share of GDP is
held constant.

Figure 13: Composition of Spending as a Share of GDP Assuming
Discretionary Spending Grows with GDP after 2005 and All Expiring Tax
Provisions Are Extended

Percent of GDP 50

40

30

20

10

0 2004 2015 2030 2040 Fiscal year

All other spending Medicare & Medicaid Social Security

Net interest

Source: GAO's August 2005 analysis.

Notes: Although expiring tax provisions are extended, revenue as a share
of GDP increases through 2015 due to (1) real bracket creep, (2) more
taxpayers becoming subject to the AMT, and (3) increased revenue from
tax-deferred retirement accounts. After 2015, revenue as a share of GDP is
held constant.

Both of these simulations illustrate that, absent policy changes on the
spending or revenue side of the budget, the growth in federal retirement
and health entitlements will encumber an escalating share of the
government's resources. Indeed, when we assume that recent tax reductions
are made permanent and discretionary spending keeps pace with the economy,
our long-term simulations suggest that by 2040 federal revenue may be
adequate to pay little more than interest on the federal debt. Neither
slowing the growth in discretionary spending nor allowing the tax
provisions to expire-nor both combined-would eliminate the imbalance.
Although revenues will be part of the debate about our fiscal future,
making no changes to Social Security, Medicare, Medicaid, and

other drivers of the long-term fiscal gap would require at least a
doubling of federal taxes in the future and that seems both unrealistic
and inappropriate. Accordingly, substantive reform of Social Security and
the major health programs remains critical to recapturing our fiscal
flexibility.

While Social Security and Medicare dominate the long-term outlook, they
are not the only federal programs or activities that bind the future. The
federal government undertakes a wide range of programs, responsibilities,
and activities that may explicitly or implicitly expose it to future
spending. These "fiscal exposures" range from explicit liabilities, such
as environmental cleanup and disposal, to the implicit promises embedded
in current policy or public expectations, such as assistance following a
major disaster.56 Policymakers may benefit from a better understanding of
the long-term costs of decisions when they are made. For large and
significant spending programs and tax provisions, consideration of
estimates of present values for the long-term commitments implied could
facilitate analysis and decisionmaking.57 While the fiscal exposure
concept focuses only on items that may expose the government to future
spending, some new or existing tax expenditures may have uncertain or
accelerating future growth paths with long-term implications. These would
need to be considered concurrently with long-term spending exposures in
addressing long-term fiscal sustainability.

Confronting the nation's fiscal challenge will require a fundamental
reexamination and reprioritization of the entire set of tools the federal
government can use to address national objectives, including major
spending and tax policies and programs. To effectively respond to social,
economic, and security changes and challenges emerging in the 21st
century, the federal government cannot accept what it does, how it does
it, who does it, and how it is financed as "givens." To assist Congress in
reexamining the base of government, we issued a report that provides
examples of the kinds of difficult choices the nation faces with regard to
discretionary spending; mandatory spending, including entitlements; as

56GAO, Fiscal Exposures: Improving the Budgetary Focus on Long-Term Costs
and Uncertainties, GAO-03-213 (Washington, D.C.: Jan. 24, 2003).

57See app. III for information about Treasury's supplemental discounted
present value estimates for select tax expenditures that involve deferrals
or other long-term revenue effects.

well as tax policies and compliance activities.58 The tax policies and
programs financing the federal budget can be reviewed with an eye toward
the overall level of revenue needed to fund federal operations and
commitments, the mix of taxes that should be used, and the extent to which
the tax code is being used to promote certain societal objectives.59

    Some Tax Expenditures May Not Be Efficient, Effective, or Equitable

Some tax expenditures may not always be efficient, effective, or
equitable, and consequently, information on these attributes can help
policymakers make more informed decisions as they adapt current policies
in light of our fiscal challenges and other overarching trends. Periodic
reviews of tax expenditures could help to establish whether these programs
are relevant to today's needs; if so, how well tax expenditures have
worked to achieve their objectives; and whether the benefits from
particular tax expenditures are greater than their costs. To measure
benefits and costs, information is needed concerning their effects on
economic efficiency, effectiveness, distributional equity, and
administration and compliance costs. To the extent that periodic reviews
show that specific tax expenditures are not effective, efficient, or
equitable, those tax expenditures might be eliminated or redesigned,
perhaps at a lower cost in revenue forgone. Coordinated reviews of tax
expenditures with related federal spending programs could assess the
relationships and interactions of programs within similar mission areas
and identify which strategies are effective. Policymakers could use such
evaluations to reduce overlap and inconsistencies and direct scarce
resources to the most effective or least costly methods to deliver federal
support.

Tax expenditures, if well designed and effectively implemented, can be an
effective tool and appropriate to further some federal goals and
objectives. For those activities that merit a subsidy (where too little of
the activity would otherwise be undertaken), subsidies through the tax
code are one option. For example, a tax expenditure for medical insurance
would improve economic efficiency if, absent a subsidy, too few workers
would purchase insurance and the tax expenditure encouraged workers to
insure

58GAO-05-325SP.

59For background information, criteria, and key questions for assessing
the pros and cons of tax reform proposals, both proposals for a major
overhaul of the current federal tax system and incremental changes to the
system, see GAO, Understanding the Tax Reform Debate: Background,
Criteria, and Questions, GAO-05-1009SP (Washington, D.C.: September 2005).

in a cost-effective manner. Because the benefits from research may not
fully accrue to the firms that bear the costs of research, a tax
expenditure aimed at spurring private-sector investment in research and
development may be an appropriate response assuming it stimulates
additional research whose benefits exceed the social costs associated with
the forgone revenues.

However, studies we and others have done raise concerns about the
efficiency, effectiveness, or equity of some tax expenditures and about
how tax expenditures relate to other federal activities aimed at the same
mission area.

o 	While tax expenditures may be intended to improve economic efficiency,
poor targeting or design may introduce additional economic inefficiencies.
For example, the income tax exclusion of employer-paid health insurance
premiums, by shifting a portion of the costs to all taxpayers, reduces the
after-tax cost of insurance for the beneficiary.60 The income tax
exclusion is credited with increasing health care coverage for employees,
and the risk pooling under group health insurance generally allows
employees to obtain insurance at lower costs than in the individual
insurance market. However, this tax benefit also leads people to obtain
more coverage than they would otherwise and increases the demand for
health care by enabling those insured to obtain services at discounted
prices. Some researchers believe that the unlimited availability of the
exclusion for employer-provided health insurance has led to excessive use
of health care services, which, in turn, has helped to drive up health
care prices faster than the overall price level.61 Capping the exclusion
at the average premium cost has been suggested as one option to improve
the economic efficiency of this tax expenditure and reduce the associated
revenue loss; another option suggested is replacing the tax exclusion with
a tax credit to improve equity since the tax savings per dollar of premium
would be the same for all taxpayers. In another example, the mortgage
interest deduction encourages home ownership by lowering the costs of
borrowing for taxpayers who itemize their deductions. However, by doing
so, the deduction encourages households to invest more in housing and less
in

60This tax-free benefit is not reported on the employee's W-2 form as the
amount is not considered wages subject to federal and state income taxes
or federal payroll taxes.

61Bob Lyke, Tax Benefits for Health Insurance and Expenses: Current
Legislation. Congressional Research Service Issue Brief IB98037
(Washington, D.C.: February 2005).

other assets that might contribute more to the nation's productivity and
economic capacity. According to CBO's Budget Options, limiting the
deductibility of interest to $500,000 of mortgage debt might still provide
taxpayers with a sizable incentive to become homeowners and could boost
investment in businesses and education.62

o 	Tax expenditures may not be an effective way to achieve federal goals
if targeting them to entities or activities meant to receive the benefits
is difficult, if they subsidize activities that would have been undertaken
without their stimulus, or if they serve to exacerbate other key private
sector and public policy challenges. For example, the income tax exclusion
of employer-paid health insurance premiums reduces the after-tax cost of
insurance for the beneficiary. However, the exclusion offers no benefit to
workers whose employers do not offer health benefits or who purchase their
own insurance. Further, this tax benefit also leads people to obtain more
comprehensive coverage than they would otherwise and could increase the
demand for health care to the extent that it shields those insured from
the full costs of health care, complicating efforts to moderate health
care spending. The exclusion also tends to favor higher-income workers
more likely to have employersponsored coverage. In another example,
individual retirement accounts (IRAs) also receive preferential tax
treatment with $7.5 billion in estimated revenue losses in fiscal year
2004. Contributions may be tax-deductible depending on the IRA type, and
earnings generally are not taxable until distribution and not taxable at
all in some cases.63 Although the tax benefits indeed seem to encourage
individuals to contribute to these kinds of accounts, the amounts
contributed may not be totally new saving. Some contributions may
represent amounts that would have occurred without the tax incentives or
amounts shifted from taxable assets or financed by borrowing. In a 1996
symposium examining universal deductible IRAs available in the early
1980s, researchers reached three widely divergent conclusions: (1) yes,
most contributions represented new saving, (2) no, most IRAs contributions
were not new saving; and (3) maybe, about 26 cents of each dollar

62Congressional Budget Office, Budget Options (Washington, D.C.: February
2005).

63For more information on the types of IRAs as well as the rules and limit
on contributions and distributions, see Department of the Treasury,
Internal Revenue Service, Individual Retirement Accounts (IRAs),
Publication 590.

contributed may have represented new saving.64 More recent research
examining the universal IRA experience estimated that at most 9 cents of
each dollar contributed represented new saving.65 Since 1986, Congress has
restricted IRA eligibility for higher-income taxpayers and increased the
contribution limits, and the overall effect of IRAs on personal saving
remains subject to considerable debate.

o 	Although tax expenditures, by design, result in individuals with
similar incomes and expenses paying differing amounts of tax depending on
whether they engage in tax-subsidized activities, tax expenditures still
may raise equity concerns. Some tax expenditures benefit mainly
upperincome taxpayers because they are most likely to itemize and because
the value of tax expenditures is generally greatest for those in higher
tax brackets.

Tax expenditures also can contribute to mission fragmentation and program
overlap,66 and this, in turn, creates the potential for duplication and
service gaps. Though sometimes necessary to meet federal priorities,
mission fragmentation and program overlap can create an environment in
which programs do not serve participants as efficiently and effectively as
possible. Like spending programs, tax expenditures may reduce government
effectiveness to the extent that they duplicate or interfere with other
federal programs. For example, in the higher education mission area, the
federal government helps students and families save and pay for the costs
of postsecondary education through tax expenditures and longerstanding
federal financial aid programs, consisting of grants, loans, and
work-study income. Since the 1990s, the federal government has offered
multiple tax incentives to help families pay for post-secondary education,
including the nonrefundable Lifetime Learning and HOPE tax credits,
deductions for qualifying post-secondary expenses and interest on student
loans, and two tax-preferred ways to save for future education expenses.
The tax-preferred saving vehicles interact with the traditional federal
aid

64GAO, National Saving: Answers to Key Questions, GAO-01-591SP
(Washington, D.C.: June 2001).

65Orazio P. Attanasio and Thomas DeLeire, "The Effect of Individual
Retirement Accounts on Household Consumption and National Saving," The
Economic Journal, Vol. 112, July 2002, pp. 504-538.

66We define mission fragmentation as the involvement of multiple agencies
in a similar programmatic area and program overlap as providing the same
services to the same target groups.

system and can affect the net federal assistance received. Further, some
tax filers do not appear to make the most effective use of certain
educationrelated tax incentives, and we have found that some people who
appear eligible for the tuition deduction and/or the tax credits did not
claim them.67 One reason may be that the differing income phaseouts and
interactions among the tax credits and deductions are difficult for
taxpayers to understand; CBO, JCT, IRS's National Taxpayer Advocate,
Treasury, and others have suggested ways to consolidate the education tax
credits and deductions.

Others have also questioned the efficiency, effectiveness, and equity of
other tax expenditures and suggested ways to design and better target
specific provisions.

o 	In December 2004, the IRS National Taxpayer Advocate designated the
complexity of the Internal Revenue Code, including the complexity of
reporting requirements related to tax expenditures, as the most serious
problem facing taxpayers and the IRS.68 The IRS National Taxpayer Advocate
also recommended consolidating the various types of retirement saving
vehicles and creating uniform rules regarding early withdrawals, plan
loans, and portability.

o 	In its January 2005 report to the Senate Finance Committee, JCT staff
presented various options to improve tax compliance and reform tax
expenditures.69 Options include repealing some tax expenditures and
restructuring others to simplify the law or achieve the intended purpose
in a more fair or efficient way.

o 	In its February 2005 budget options compendium prepared for the House
and Senate Budget Committees, CBO listed several options to

67GAO, Student Aid and Postsecondary Tax Preferences: Limited Research
Exists on Effectiveness of Tools to Assist Students and Families Through
Title IV Student Aid and Tax Preferences, GAO-05-684 (Washington, D.C.:
July 29, 2005).

68IRS, Taxpayer Advocate Service, National Taxpayer Advocate 2004 Annual
Report to Congress, (Washington, D.C.: December 2004).

69Joint Committee on Taxation, Options to Improve Tax Compliance and
Reform Tax Expenditures, JCS-2-05. Prepared by the staff of the Joint
Committee on Taxation. (Washington, D.C.: Jan. 27, 2005).

eliminate or restructure tax expenditures.70 Options include further
limiting the tax benefit of itemized deductions to the 15 percent rate for
higher-bracket taxpayers and capping itemized deductions for state and
local taxes and charitable contributions to the amount exceeding 2 percent
of adjusted gross income.

o 	Finally, in December 2004 for the Senate Budget Committee, CRS updated
its biennial compendium on tax expenditures. 71 This volume includes for
each tax expenditure: JCT's revenue loss estimate, the legal
authorization, a description of the tax provision, its impact including
distribution of benefits when available, the rationale at the time of
adoption, assessment summarizing the arguments for and against the
provision, citations to relevant research. According to CRS, congressional
budget decisions will take into account the full spectrum of federal
programs only when tax expenditures are considered in conjunction with
direct spending programs.

    Assessing the Efficiency, Effectiveness, or Equity of Tax Expenditures Can
    Be Challenging

Inadequate or missing data and difficulties in quantifying the benefits of
some tax expenditures can impede studies of their efficiency,
effectiveness, and equity. 72 A key challenge is that data necessary to
assess how often a tax expenditure is used and by whom generally would not
be collected on tax returns unless IRS needs the information to know the
correct amount of taxes owed or is legislatively mandated to collect or
report the information. For example, tax exclusions-including those for
employerprovided health insurance and pensions which are among the largest
tax expenditures-generally are not reported on individual taxpayers'
returns. In some cases, IRS may combine reporting requirements to minimize
its workload and taxpayer burden, and as a result, the information
collected may not identify specific beneficiaries or activities targeted
by a tax expenditure. For example:

70Congressional Budget Office, Budget Options. (Washington, D.C.: February
2005).

71U.S. Congress, Senate Committee on the Budget, Tax Expenditures:
Compendium of Background Material on Individual Provisions, S. Prt.
108-54. Prepared by the Congressional Research Service (Washington, D.C.:
December 2004).

72We have also noted limitations in the quality of agency performance and
evaluation information and agency capacity to produce rigorous evaluations
of effectiveness for federal spending programs. See GAO, Performance
Budgeting: Current Developments and Future Prospects, GAO-03-595T
(Washington, D.C.: Apr. 1, 2003).

o 	In our 2002 report on three tax expenditures meant to encourage
employment of the disabled among other economically disadvantaged workers,
we could not determine the amounts used to hire, retain, and accommodate
workers with disabilities.73 We found that information on the work
opportunity and disabled access credits was not available from tax data
because tax returns provided only the total amount of credits reported,
and employers could claim the work opportunity credit for employing other
types of workers and claim the disabled access credit for expenditures
made to accommodate customers with disabilities. Also, information
regarding use of the barrier removal deduction for providing
transportation or architectural accommodations was not available in IRS
databases.

o 	As we reported in 2003, for one of the seven Liberty Zone tax benefits,
the business employee credit, IRS was in the process of collecting but was
not planning to report information about the number of taxpayers claiming
the credit and the amount of credit claimed.74 IRS was also not planning
to collect or report information about the use of the other six benefits,
and taxpayers do not report these benefits as separate items on the
existing returns. For example, taxpayers added the amount of depreciation
they are allowed under the Liberty Zone special depreciation allowance
benefit to other depreciation expenses and report their total depreciation
expenses on their returns. IRS officials said that they do not need
information on each specific benefit claimed to properly target their
enforcement efforts.

Further, IRS's financial management system does not currently have cost
accounting capabilities. As a result, comparisons of the costs of
administering existing or proposed tax expenditures with similar
administrative costs for spending programs may be impossible. Regarding
taxpayer compliance costs, although IRS is working to develop improved
estimates of taxpayer compliance burden, it is not yet clear whether this
modeling effort will provide estimates of additional compliance costs that
may result from particular tax expenditures.

73GAO, Business Tax Incentives: Incentives to Employ Workers with
Disabilities Receive Limited Use and Have an Uncertain Impact, GAO-03-39
(Washington, D.C.: Dec. 11, 2002).

74GAO, Tax Administration: Information Is Not Available to Determine
Whether $5 Billion in Liberty Zone Tax Benefits Will Be Realized,
GAO-03-1102 (Washington, D.C.: Sept. 30, 2003).

According to IRS officials, IRS seeks to collect information necessary to
determine whether taxpayers have accurately reported their income and
calculated the correct amount of tax liability. By focusing on information
essential to administering the tax code, IRS aims to ensure that taxpayers
are not burdened unnecessarily by record keeping and reporting, and IRS
can minimize its own administrative costs for data collection and
processing. For tax expenditures recorded on particular lines on tax
forms, such as deductions and credits for individual taxpayers, data on
the use of these tax expenditures are available. IRS Statistics of Income
Division publications detail the number of individual tax returns on which
taxpayers claimed each deduction or credit, the total amounts claimed, and
the distribution of claims among taxpayers by income level.

If policymakers conclude that additional data would facilitate reexamining
a particular tax expenditure, decisions would be required on what data are
needed, who should provide the data, who should collect the data, how to
collect the data, what it would cost to collect the data, and whether the
benefits of collecting additional data warrant the cost of doing so.
Another factor to consider is how to facilitate data sharing and
collaborative evaluation efforts. For example:

o 	Limited data are available on the prevalence and use of business-owned
life insurance, and GAO has reported that more comprehensive data could be
useful in assessing the tax-favored treatment of this investment.75 Data
on the amount of tax-free income that businesses received from death
benefits could help explain the potential effect of changes to the tax
treatment of policies on tax revenues. Businesses holding the policies or
insurance companies that sold them could provide this and other data.
Several agencies, including Treasury and the Securities and Exchange
Commission, already collect some financial information from businesses and
insurers and could be tasked to collect additional data for tax policy
purposes.

o 	In the higher education area, the Department of Education (Education)
is unable to analyze the use of higher education tax credits or their
effects because it lacks access to individual taxpayer data needed to
identify users of the credits. Treasury has access to taxpayer data but
has not used these data for evaluating the education tax credits since

75GAO, Business-Owned Life Insurance: More Data Could Be Useful in Making
Tax Policy Decisions, GAO-04-303, (Washington, D.C.: May 13, 2004).

their implementation in 1998. In 2002, GAO recommended that Education and
Treasury collaborate in studying the impact of tax credits and student aid
programs on postsecondary attendance, choice, completion, and costs.76 A
key first step would be identifying opportunities for, and limits to, data
sharing and develop a plan to address data needs, but little action has
been taken.

o 	In the case of the empowerment zone, enterprise community, and renewal
community programs, the lack of tax benefit data limits the ability of the
Department of Housing and Urban Development (HUD) and the Department of
Agriculture (USDA) to administer and evaluate the overall programs.77 We
recommended that HUD, USDA, and IRS collaborate to (1) identify the data
needed to assess the use of the tax benefits and the various means of
collecting such data; (2) determine the cost-effectiveness of collecting
these data, including the potential impact on taxpayers and other program
participants; (3) document the findings of their analysis; and, if
necessary, (4) seek the authority to collect the data, if a cost-effective
means is available.

When data on the cost and use of tax expenditures are available or can be
reasonably estimated and other relevant data are available, economic
analysis can be useful in evaluating whether a tax expenditure is
efficient, effective, or equitable. Econometric modeling analysis can
estimate how a tax expenditure affects the prices and quantities of
targeted goods and services and determine how taxpayers' incomes are
affected. Although isolating and quantifying the outcomes associated with
tax expenditures is challenging-just as it is for spending programs,
research results are useful in demonstrating how particular tax
expenditures work or providing insight on ways to refine their design. For
example, research has generally shown that the EITC effectively increases
recipients' participation in the labor force, particularly for single
parents, and lifts millions of recipients out of poverty. Some tax
expenditures are enacted on a temporary basis, specifically to provide an
opportunity for evaluating their effects before they are extended. For
example, the research tax credit, enacted on a

76 GAO, Student Aid and Tax Benefits: Better Research and Guidance Will
Facilitate Comparison of Effectiveness and Student Use, GAO-02-751
(Washington, D.C.: Sept. 13, 2002).

77GAO, Community Development: Federal Revitalization Programs are Being
Implemented, but Data on the Use of Tax Benefits Are Limited, GAO-04-306
(Washington, D.C.: Mar. 5, 2004).

temporary basis in 1981 and extended 11 times as of 2004, was
substantially modified in 1989 after researchers showed the original
credit formula undercut the incentive it was intended to provide to
undertake additional research spending.

In some cases, economic research has not yielded definitive results or was
limited by data and methodological issues. For example, although the
various tax expenditures aimed at encouraging saving for, among other
things, retirement, education, and health care have resulted in
substantial sums being placed in these tax preferred accounts, economists
disagree about whether tax incentives, such as for IRAs, are effective in
increasing the overall level of personal saving. In the case of the
research credit, GAO reported in 1996 that studies done at that time
provided mixed evidence on the amount of spending stimulated and used
publicly available data that were not a suitable proxy for tax return
data.78 To fully assess the value to society of the research tax credit,
researchers need to look at more than just the amount of spending
stimulated per dollar of revenue cost. Comparisons should include (1) the
total benefits gained by society from research stimulated by the credit
and (2) the estimated costs to society resulting from the collection of
taxes required to fund the credit. The social benefits of the research
conducted by individual companies include any new products, productivity
increases, or cost reductions that benefit other companies and consumers
throughout the economy. Although most economists agree that research
spending can generate social benefits, the effects of the research on
other companies and consumers are difficult to measure.

Ultimately, evaluation results could be used to identify how well tax
expenditures are working, to both identify ways to better manage
individual tax expenditures and decide how best to ensure prudent
stewardship of taxpayers' resources. Whether in time of deficit or
surplus, reexamining both the spending and tax sides of the budget is
essential to ensure the reasonableness, relevancy, and sustainability of
existing programs and position the nation for the future. In the case of
the EITC, Treasury and IRS are using evaluation results to identify ways
of reducing erroneous claims, while maintaining participation among
eligible claimants and minimizing taxpayer and IRS's administrative
burden. Additional evaluations of other tax expenditures may identify
opportunities to retarget

78GAO, Tax Policy and Administration: Review of Studies of the
Effectiveness of the Research Tax Credit, GAO/GGD-96-43 (Washington, D.C.:
May 21, 1996).

or eliminate ineffective or outdated tax expenditures. Tax expenditures,
unless well designed to correct market failures, can distort economic
decisions in ways that reduce economic performance from what it otherwise
could be and thereby lower our future economic well-being. If a tax
expenditure or group of tax expenditures is reduced or eliminated, any
resulting increase in tax revenues could be offset if policymakers deem
that to be appropriate fiscal policy. In any event, in order to raise a
given amount of federal revenue, tax rates must be raised higher than they
otherwise need to be due to revenue losses from tax expenditures. Thus,
the net change after tax rate adjustments could, depending on overall
congressional priorities and preferences, result in tax reductions for
many taxpayers in place of the preferential treatment for some taxpayers.
According to a recent estimate, a broad-based income tax system-
eliminating basically all credits, deductions, special rates, exclusions
for employer-provided fringe benefits and employee contributions to
retirement account as well as eliminating the AMT--could raise about the
same amount of revenue as the current income tax system while lowering tax
rates by about one-third.79

    The Executive Branch Has Made Little Progress Since 1994 to Improve Scrutiny
    of Tax Expenditures

Although OMB and Treasury in 1994 supported expanding federal reviews of
tax expenditures, the Executive Branch made little progress over the past
decade to integrate tax expenditures in the budget presentation and to
incorporate tax expenditures under review processes that apply to spending
programs, as we recommended in 1994. Even though the sum of tax
expenditure outlay-equivalent estimates is about the same magnitude as
discretionary spending overall and greater than outlays in some budget
functions, this is not readily visible to policymakers and the public
because tax expenditures are not integrated in the budget presentation.
Since their initial efforts to outline a framework for evaluating tax
expenditures and preliminary performance measures, OMB and Treasury have
largely ceased to make progress and have retreated from setting a schedule
for evaluating tax expenditures. One of the key impediments to moving
forward in conducting reviews of tax expenditures' performance is the
continuing lack

79See the President's Advisory Panel on Federal Tax Reform, "Understanding
Tax Bases: Staff Presentation," (presentation before the Panel's public
meeting, Washington, D.C., July 20, 2005,
http://taxreformpanel.gov/meetings/docs/understanding_tax_bases.ppt
(downloaded September 13, 2005). The calculations for the presentation
were produced by the U.S. Department of the Treasury, Office of Tax
Analysis, at the request of the Panel. The broad income tax base scenario
assumes integration of individual and corporate income taxes with the
corporate tax rate set equal to the top individual rate.

of clarity about the roles of OMB, Treasury, IRS, and departments or
agencies with outlay program responsibilities. So far, GPRA plans and
reports are underutilized as a way to provide more information about the
performance of tax expenditures and their contributions relative to
spending programs. Tax expenditures are not subject to annual budget
reviews, and OMB has not generally subjected them to scrutiny under PART
in tandem with spending programs sharing common, crosscutting goals.

Tax Expenditures Are Not Integrating tax expenditure costs in the annual
budget presentation is Integrated in the Annual Budget crucial to
providing a comprehensive picture of federal resources to

Presentation or Financial Statement Reporting

facilitate reexamining the base. As a start in acting on our 1994
recommendation, OMB began presenting revenue loss sums for tax
expenditures alongside outlays and credit activity for each budget
function in the fiscal year 1998 budget. These summary tables were a
useful starting point in highlighting the relative magnitude of tax
expenditures across mission areas. However, OMB discontinued the reporting
practice after the fiscal year 2002 budget, and instead, the Analytical
Perspectives contains Treasury's list of tax expenditures with associated
revenue loss estimates for each one. Isolating tax expenditure cost
information in a supplemental volume, however, provides a less
comprehensive picture for policymakers and the public to compare all of
the policy tools used within a mission area, such as health care or
energy, because all the tools are not displayed together in the budget.
OMB has demonstrated that it is feasible to display tax expenditure totals
alongside spending programs in each budget function. Such a display is a
first step in providing the public and policymakers with a more useful and
accurate picture of the extent of federal support and activities.

GAO also recommended in 1994 that the budget presentation include, to the
extent possible, information to highlight for policymakers and the public
the effectiveness, distributional equity, and economic efficiency for all
federal resources allocated in a mission area. In the tax expenditure
chapter in Analytical Perspectives, OMB added a section outlining possible
performance measures developed by Treasury, which could be used to present
information about the performance of tax expenditures. Although this
overview was initially introduced in the 1997 budget and expanded in the
1999 budget, no performance information is actually displayed. OMB states
that the measure examples provided are "illustrative" in nature,
acknowledges that the performance measure discussion "although broad, is
nonetheless incomplete," and noted that many tax expenditures are not
explicitly cited.

The Chief Financial Officers Act,80 as expanded by the Government
Management Reform Act of 1994,81 required federal agencies to prepare
annual audited financial statements beginning in fiscal year 1996. OMB
Circular A-136 Financial Reporting Requirements requires agencies to
combine the annual GPRA program performance report with the financial
statements and other information in a combined performance and
accountability report. In accordance with generally accepted accounting
principles, the basis on which federal agencies are required to prepare
their financial statements, tax expenditures may be presented as other
accompanying information. The Federal Accounting Standards Advisory Board
(FASAB), which promulgates federal accounting standards, recognized that
tax expenditures, which can be large in relation to spending programs that
are measured under federal accounting standards, may not be fully
considered in entity reporting. FASAB based its views, in part, on the
fact that, in some cases, the association of tax expenditures with
particular programs is not clear and the information is available
elsewhere. The Board agreed to permit reporting entities to present, as
other accompanying information, information on tax expenditures that the
reporting entity considers relevant to its programs, if suitable
explanations and qualifications are provided. As a result, tax expenditure
amounts, which in some cases are larger than similar spending programs,
are not required to be disclosed to the public as part of federal
agencies' financial statements nor are they disclosed in the consolidated
financial statements of the federal government. Similarly, OMB's guidance
for the performance and accountability reports does not require reporting
of tax expenditure information in agencies' reports. Reporting such
information would ensure greater transparency of and accountability for
tax expenditures.

Progress on Developing OMB has not designed and implemented a structure
for conducting reviews

Structure for Reviewing Tax of tax expenditures' performance, as we
recommended in 1994. Our

Expenditures' Performance Has recommendation was consistent with language
in the Senate Committee on

Stalled Government Affairs' Report82 on GPRA, which specified that the
Director of OMB was to establish an appropriate framework for periodic
analyses of the effects of tax expenditures in achieving performance
goals. To

80Pub. L. No. 101-576, Nov. 15, 1990. 81Pub. L. No. 103-356, Oct. 13,
1994. 82S. Rep. No. 103-58, p. 29 (1993).

significantly increase the oversight and analysis of tax expenditures, the
committee report also called for a schedule for periodic tax expenditure
evaluations.

The ultimate goal of designing a structure for conducting performance
reviews of tax expenditures was to begin developing and presenting
performance information in the federal budget that would help demonstrate
the relative effectiveness, efficiency, and equity of federal outlays and
tax expenditure efforts within a mission area. In our 1994 report, we
emphasized that in designing the structure for tax expenditure performance
reviews, OMB should consider

o 	the roles of OMB, Treasury, and departments or agencies with outlay
program responsibilities in assessing the performance of tax expenditures
and their relationship and interaction with related spending programs; and

o 	which tax expenditures and outlay programs are related or interact and
should be jointly considered.

GAO recommended that OMB and Treasury conduct case studies of the proposed
review structure to identify (1) successful methods agencies devise for
reviewing tax expenditures' performance, (2) how best to report the
results of these reviews, and (3) how to ensure that adequate resources
are available for such reviews.

Although OMB, working with Treasury, took a number of steps consistent
with our recommendation, it has not resolved the roles of OMB, Treasury,
and departments or agencies with outlay program responsibilities;
established a schedule for reviewing tax expenditures; or addressed
lessons learned from tax expenditure case study reviews that Treasury
performed. If the Executive Branch cannot define roles and set firm plans,
it will continue to face additional challenges in developing objective,
measurable, and quantifiable performance measures for tax expenditures
that support federal missions and goals.

Defining roles of agencies. One of the key impediments to moving forward
in conducting reviews of tax expenditures' performance is the continuing
lack of clarity about the roles of OMB, Treasury, IRS, and departments or
agencies with outlay program responsibilities. According to officials at
OMB, it is difficult to determine which agencies in addition to Treasury
and IRS have jurisdiction over particular tax expenditures. For

example, one OMB official noted that tax expenditures meant to encourage
savings were not the purview of any single agency. OMB officials also
stated that OMB does not have the expertise or resources to conduct its
own comprehensive analyses of tax expenditures, so individual agencies
should take responsibility for identifying tax expenditures that affect
their missions, with Treasury's Office of Tax Analysis leading efforts to
evaluate tax expenditures.

Without clarification on the roles of federal agencies, inaction, overlap
or inconsistency in evaluating tax expenditures can occur. For example, in
2002 we reported that gaps existed in monitoring the relative
effectiveness of Title IV grants and loans and the HOPE and Lifetime
Learning tax credits in promoting postsecondary education.83 The lack of
collaboration between the Department of Education and the Treasury left
little information available to help Congress weigh the relative
effectiveness of grants, loans, and tax credits. Although data and
methodological challenges make it difficult to isolate the impact of these
tools, some academic researchers have used statistical techniques and
research designs to mitigate these challenges. We recommended in 2002 that
the departments develop a plan to share data and collaborate to provide
Congress with evidence about the impact of higher education tax credits
and student aid, but little action has been taken to implement the
recommendation.

To define the roles of federal agencies in reviewing tax expenditures,
OMB, working with Treasury and other federal agencies, will need to
exercise judgment in resolving how to address tax expenditures spanning
mission areas. In some cases, Treasury could take the lead, such as in
evaluating tax expenditures that broadly support investment and saving, or
other agencies could work with Treasury to evaluate tax expenditures that
directly affect their mission areas. For example, an evaluation of the
various energy supply tax expenditures might involve both Treasury and the
Department of Energy in assessing their effects on increasing production
as well as on energy security and the environment.

Establishing a schedule for evaluations. Periodic reviews of tax
expenditures are also impeded because OMB has not developed a schedule for
such reviews. In its 1997 GPRA report and again in the fiscal year 1999
budget, OMB set the expectation that the Executive Branch would lay out a

83GAO-02-751.

schedule for tax expenditure evaluations. Beyond three initial pilot
studies in 1997, however, no schedule has been set for further evaluations
or case studies to explore methods and resource needs for measuring and
reporting tax expenditure performance.

As the roles of federal agencies are clearly defined, OMB and Treasury,
working with other agencies, would be positioned to establish a schedule
for tax expenditure evaluations. Opportunities exist to develop a
strategic approach to the selection and prioritization of areas in
allocating scarce evaluation resources. In our January 2004 report on
OMB's PART, we recommended that OMB target PART assessments based on such
factors as the relative priorities, costs, and risks associated with
related clusters of programs and activities and that OMB select similar
programs for review in the same year to facilitate comparisons and
tradeoffs.84 Similar considerations would be useful in setting a schedule
for tax expenditure evaluations.

Testing the evaluation framework. Although OMB outlined an initial
framework for tax expenditure analysis in its May 1997 GPRA report to the
President and Congress, OMB has not taken steps to address lessons learned
from tax expenditure case study reviews that Treasury performed. OMB's
framework focused on the methodology that could be used to assess the
performance of tax expenditures. OMB emphasized that developing a
framework that is comprehensive, accurate, and flexible to reflect the
objectives and effects of the wide range of tax expenditures would be a
significant challenge. The initial framework for evaluating tax
expenditures was expected to follow the basic structure for performance
measurement-inputs, outputs, and outcomes. For tax expenditures, the
primary input is the revenue loss. The outputs are the quantitative or
qualitative measures of goods and services, or changes in investment and
income, produced by the tax expenditures. Outcomes, in turn, were defined
as the changes in the economy, society, or environment that the tax
expenditures aim to accomplish.

In 1997, Treasury did three pilot evaluations of selected tax expenditures
to test the evaluation methods that OMB had described in its framework for
tax expenditure analysis. In addition to seeking to learn lessons about
applying the framework, the pilots were also intended to help identify

84GAO, Performance Budgeting: Observations on the Use of OMB's Program
Assessment Rating Tool for the Fiscal Year 2004 Budget, GAO-04-174
(Washington, D.C.: Jan. 30, 2004).

resource needs for evaluating tax expenditures. Treasury selected one
pilot each to be done by the individual, corporate, and international
units within its Office of Tax Analysis. Results from the three tax
expenditure pilots-the exclusion for worker's compensation benefits, the
tax credit for non-conventional fuels, and the tax exclusion for certain
amounts of income earned by Americans living abroad85-were summarized
alongside each tax expenditure's description in the tax expenditure
chapter of the Analytical Perspective volume of the fiscal year 1999
budget. Although OMB originally expected to complete additional
evaluations to refine the tax expenditure framework and improve
performance measures, no further pilot evaluations have been completed. In
reporting the results of these pilots, Treasury said that much of the data
needed for thorough analysis was not available and that in at least one
case, it was difficult to identify a clear purpose for the tax
expenditure. Treasury did not discuss the resources that would be needed
to continue doing such evaluations.

However, OMB officials we interviewed reiterated that the data
availability issues raised in the 1997 pilots remain a major challenge,
and data constraints limit the assessment of the effectiveness of many tax
expenditures. To improve the data available to assess the effects of some
major tax expenditures, principally those aimed at personal savings,
Treasury and IRS are developing a data set that is to follow a sample of
individual income taxpayers over at least 10 years, beginning with tax
year 1999. The new data set aims to capture the changing demographic and
economic circumstances of individual taxpayers for use in analyzing the
effects of changes in tax law over time. In addition to the panel sample,
OMB reported in the fiscal year 2006 budget that it is working with
Treasury's Office of Tax Analysis and other agencies to improve data
available for assessment of saving-related tax expenditures. No time frame
was given in the 2006 budget for when any results will be reported.

The challenges in producing credible performance information and the
ability of federal agencies to produce evaluations of their programs'
effectiveness are not unique to tax expenditures. As our work on GPRA and
PART implementation shows, the credibility of performance data has

85Revenue loss estimates for the three tax expenditures included in
Treasury's first pilot study amounted to approximately $7.8 billion (in
2004 dollars) or 1.4 percent of total revenue loss in fiscal year 1997;
the three amounted to $9.2 billion or 1.3 percent of total revenue loss
for fiscal year 2004.

been a long-standing weakness.86 Developing and reporting credible
information on outcomes achieved through federal programs remains a work
in progress. In past reports, we have identified several promising ways
agencies can maximize their evaluation capacity. For example, careful
targeting of federal evaluation resources on key policy or performance
questions and leveraging federal and nonfederal resources show promise for
addressing key questions about program results. Other ways agencies might
leverage their current evaluation resources include adapting existing
information systems to yield data on program results, drawing on the
findings of a wide array of evaluations and audits, making multiple use of
an evaluation's findings, mining existing databases, and collaborating
with state and local program partners to develop mutually useful
performance data.

Statutory Impetus for Tax Congressional expectations for reviews of tax
expenditures in connection Expenditure Reviews Is with agencies' reviews
of related outlay and other programs generally have

                                 Underutilized

not been met. Enacted in 1993, GPRA is designed to inform congressional
and executive decisionmaking by providing objective information on the
effectiveness and efficiency of federal programs and spending.87 GPRA
requires agencies to measure performance toward the achievement of annual
goals and report on their progress in annual program performance reports.
Through the strategic planning requirement, GPRA requires federal agencies
to consult with the Congress and key stakeholders to regularly reassess
their missions and strategic goals as well as the strategies and resources
they will need to achieve their goals. Although GPRA offers a promising
opportunity for the Executive Branch to develop useful information about
the results of tax expenditures, agencies are not using their GPRA
strategic plans and annual performance plans and reports to assess tax
expenditures and their performance relative to spending programs
contributing to the same strategic goals and objectives. Without

86GAO has suggested various approaches to addressing this and other
challenges. See GAO, The Government Performance and Results Act: 1997
Governmentwide Implementation Will Be Uneven, GAO/GGD-97-109 (Washington,
D.C.: June 2, 1997); and GAO, Results-Oriented Government: GPRA Has
Established a Solid Foundation for Achieving Greater Results, GAO-04-38
(Washington, D.C.: Mar. 10, 2004).

87GPRA is the centerpiece of a statutory framework for helping resolve
long-standing management problems that have undermined the federal
government's efficiency and effectiveness. The framework also includes the
Chief Financial Officers Act of 1990 as amended by the Government
Management Reform Act of 1994 and other reform legislation such as the
Paperwork Reduction Act of 1995 and the Clinger-Cohen Act of 1996. For
more information on the Government Performance and Results Act, see
GAO-04-38.

integrating tax expenditures that have a direct bearing on federal
missions and goals, policymakers may not have complete information to
fully evaluate whether the government is achieving results or how the
performance of tax expenditures interact with or compare to related
spending programs.

The Senate Governmental Affairs Committee Report on GPRA stated that tax
expenditures should be taken into consideration in a comprehensive
examination of government performance. The report stated that a schedule
for periodically assessing the effects of specific tax expenditures in
achieving performance goals should be included in the annual performance
plans and that annual performance reports would subsequently be used to
report on these tax expenditure assessments. In addition, the report noted
that these assessments should consider the relationship and interactions
between spending programs and tax expenditures and the effects of tax
expenditures in achieving federal performance goals.

Although GPRA expanded the supply of performance information generated by
federal agencies, evaluating crosscutting federal efforts continues to be
a challenge. GPRA requires the President to include in his annual budget
submission a federal government performance plan. Congress intended that
this plan provide a single cohesive picture of the annual performance
goals for the fiscal year.88 The governmentwide performance plan could
help Congress and the Executive Branch address critical federal
performance and management issues, including redundancy and other
inefficiencies in how we do business. However, this provision has not been
fully implemented, and the current agency-by-agency focus of the budget
does not have a broad, integrated perspective of planned performance on
governmentwide outcomes. As envisioned by Congress, the governmentwide
plan could relate and address the contributions of alternative federal
strategies, including tax expenditures, to governmentwide goals. Agencies'
annual performance plans and reports could highlight crosscutting program
efforts and provide evidence of the coordination of those efforts. We have
previously recommended that OMB fully implement GPRA's requirement to
develop a governmentwide plan to

88S. Rep. No. 103-58, p. 29 (1993).

provide a more cohesive picture of the federal government's goals and
strategies.89

Prior to a 2003 revision, OMB's Circular A-11 guidance on GPRA reporting
stated that descriptions should be provided for use of tax expenditures in
annual performance plans when achievement of program or policy goals is
dependent upon these governmental actions and annual performance reports
must include the results of any assessment of how specific tax
expenditures affect achieving its performance goals.90 However, the
circular also stated that few agencies were responsible for such analyses.
In addition, as part of a broader A-11 revision update, OMB streamlined
its GPRA guidance in 2003 and no longer describes tax expenditures as part
of guidance on performance plans and performance reports in the circular.
According to OMB, it is up to individual agencies to decide whether to
address tax expenditures in their GPRA reports and that many agencies
focus on outlay programs over which they have more direct control. OMB
officials told us that some agencies see tax expenditures as closely
related to what they do and some do not, or agencies might not have enough
knowledge about tax expenditures to consider them carefully. Our review of
selected GPRA Performance and Accountability reports indicated the
acknowledgement of tax expenditures in achieving federal performance goals
varied by agency. For example:

o 	The Department of Energy (DOE) and HUD both acknowledged tax
expenditures or tax policy as factors that affect agency goals. However,
the DOE's fiscal year 2004 report provided no further discussion on how
the tax expenditures contributed to achieving the agencies' performance
goals. HUD's fiscal year 2004 report acknowledged the tax incentives for
renewal communities, empowerment zones, and enterprise communities as
helping to achieve its objective of providing capital resources to improve
economic conditions in distressed communities. As discussed previously,
the outlay-equivalent value for tax expenditures amounts to more than
other spending in the energy as well as the commerce and housing credit
mission areas.

89GAO-04-38.

90OMB, "Preparation and Submission of Strategic Plans, Annual Performance
Plans, and Annual Program Performance Report", Part 6 of Circular A-11,
Preparation, Submission, and Execution of the Budget (Washington, D.C.:
2002).

o 	The fiscal year 2004 reports released by the Department of Commerce
(Commerce), the Department of Veterans Affairs, and the Department of
Health and Human Services (HHS) do not mention tax expenditures at all,
even though tax expenditures exist under the different mission areas
related to these departments. For instance, several large tax
expenditures, such as capital gains and accelerated depreciation, are
listed by Treasury as related to the Commerce mission area, but it is
unclear how, if at all, these tax expenditures relate to Commerce's
performance goals. Also, the income tax exclusion for employerprovided
health care, the largest single tax expenditure, clearly intersects with
HHS's mission to assure access to health care.

o 	Treasury's fiscal year 2004 report explicitly identified a few tax
expenditures-the New Markets Tax Credit and a new health coverage tax
credit-as related to achieving its strategic objective to stimulate U.S.
economic growth. In the context of its strategic objective to improve and
simplify the tax code, Treasury reported on its efforts to, among other
things, simplify the EITC and consolidate the higher education tax
benefits. Treasury also reported on its efforts to improve determination
of EITC eligibility and educate taxpayers about this provision. Treasury
did not include information about tax expenditures as other accompanying
information in the financial statement in its 2004 report.

Tax Expenditures Are Not Tax expenditures have not been incorporated into
Executive Branch Subject to Annual Budget budget reviews, as we
recommended in 1994. We recommended that OMB

Reviews

use information on outlay programs and tax expenditures to make
recommendations to the President and Congress about the most effective
methods for accomplishing federal objectives. We concluded that better
targeting by Congress and the Executive Branch of all federal spending and
subsidy programs could save resources and increase economic efficiency
through (1) better coordination of spending programs with tax
expenditures; (2) reduction of overlap and inconsistencies among all
federal subsidy programs; and (3) encouragement of trade-offs among tax
expenditures, outlays, and loans.

The congressional budget process is the annual vehicle through which
Congress articulates both an overall fiscal stance-overall targets for
spending and revenue-and its priorities across various broad categories.
The process provides the overall constraints for spending and revenue
actions by Congress for each year and the rules of procedure that can be
used to constrain new entitlement and tax legislation not assumed in the

annual budget resolution. The conflicts and uncertainties entailed in
budgeting and policymaking are often mitigated by focusing decisions on
incremental changes in resources each year. As a result, this incremental
process focuses disproportionate attention on proposed changes to existing
programs and proposals for new programs, with the base of programs often
being taken as "given." Moreover, the process routinely examines only the
one-third of federal spending subject to the annual appropriations
process. Unlike discretionary spending programs, which are subject to
periodic reauthorization and annual appropriation, tax expenditures-like
entitlement programs-are permanent law and are generally not subject to a
legislative process that would ensure systematic annual or periodic
review. In addition, the budget rules that were grounded in
statute-including discretionary spending caps, pay-as-you-go (PAYGO)
limits on mandatory spending and tax cuts-and enforced by executive
actions if violated, expired at the end of fiscal year 2002. Before their
expiration, PAYGO procedures restricted Congress' ability to add new tax
expenditures or expand existing ones unless offsetting funds could be
raised. Because tax provisions are not as visible in the budget as
spending programs, there is an incentive for policymakers to use tax
provisions rather than spending programs to accomplish programmatic ends.
However, both have a negative effect on the government's "bottom-line."
Reinstituting budget enforcement mechanisms, such as discretionary
spending caps, PAYGO discipline on both the spending and tax side, and
fiscal benchmarks, could help the President and Congress sort out the many
claims on the federal budget, including tax expenditures.

Within the Executive Branch, OMB has not used its PART process, which is
central to the Executive Branch's budget and performance integration
initiative,91 to systematically review tax expenditures and promote joint
and integrated reviews of tax and spending programs sharing common,
crosscutting goals. OMB describes PART as a diagnostic tool meant to
provide a consistent approach to assessing federal programs as part of the
executive budget formulation process. It applies 25 questions to all

91GAO, 21st Century Challenges: Performance Budgeting Could Help Promote
Necessary Reexamination, GAO-05-709T (Washington, D.C.: June 14, 2005).
For a detailed examination of PART, see GAO-04-174. Another significant
element of the performance and budget integration initiative is efforts to
restructure budgets. See GAO, Performance Budgeting: Efforts to
Restructure Budgets to Better Align Resources with Performance,
GAO-05-117SP (Washington, D.C.: February 2005).

"programs"92 under four broad topics: (1) program purpose and design, (2)
strategic planning, (3) program management, and (4) program results (i.e.,
whether a program is meeting its long-term and annual goals) as well as
additional questions that are specific to one of seven mechanisms or
approaches used to deliver the program.93 PART is designed to be
evidence-based, drawing on a wide array of information, including
authorizing legislation, GPRA strategic plans and performance plans and
reports, financial statements, inspectors general and GAO reports, and
independent program evaluations. Drawing on available performance and
evaluation information, the PART questionnaire attempts to determine the
strengths and weaknesses of federal programs with a particular focus on
individual program results and improving outcome measures.

Since the fiscal year 2004 budget cycle, OMB has applied PART to 607
programs (about 60 percent of the federal budget), and given each program
one of four overall ratings: (1) "effective," (2) "moderately effective,"
(3) "adequate," or (4) "ineffective" based on program design, strategic
planning, management, and results. A fifth rating, "results not
demonstrated," was given-independent of a program's numerical score- if
OMB decided that a program's performance information, performance
measures, or both were insufficient or inadequate. Over the next 2 years,
OMB plans to assess nearly all remaining Executive Branch spending

94

programs.

Whereas OMB, through its development and use of PART, has provided
agencies with a powerful incentive for improving data quality and
availability on the spending side, relatively little progress has been
made in evaluating the effectiveness of tax expenditures. So far, OMB has
used

92 There is no standard definition for the term "program." For purposes of
PART, OMB described the unit of analysis (program) as (1) an activity or
set of activities clearly recognized as a program by the public, OMB,
and/or Congress; (2) having a discrete level of funding clearly associated
with it; and (3) corresponding to the level at which budget decisions are
made.

93The seven major categories are competitive grants, block/formula grants,
capital assets and service acquisition programs, credit programs,
regulatory-based programs, direct federal programs, and research and
development programs.

94For the limited exceptions, the Administration is considering
alternative methods and timelines for assessment of programs with limited
impact and large activities where it is difficult to determine an
appropriate unit of analysis.

PART to address tax expenditures in only two cases-the EITC compliance
initiative and the New Markets Tax Credit (NMTC).95

o 	For the EITC, which has outlays for the refundable portion, the direct
federal spending PART instrument was used to evaluate IRS' initiative to
improve the payment accuracy rate for the EITC-and not the refundable EITC
itself. OMB rated the compliance initiative as "ineffective" in the fiscal
year 2004 budget because data showed no progress in reducing the high
rates of erroneous payments. The review did not evaluate the effects of
the EITC on workforce participation or examine its contribution relative
to other federal programs aimed at reducing poverty.

o 	The NMTC, which is administered like a grant by CDFI, was evaluated as
part of OMB's crosscutting review of community and economic development
programs. OMB rated the NMTC as "adequate" and reported in 2005 that CDFI
had established meaningful long-term and annual performance measures but
that data were not yet available to evaluate the effectiveness of the NMTC
or establish baselines for the performance measures.

We have urged a more comprehensive, consistent, and integrated approach to
evaluating all programs relevant to common goals-encompassing spending,
tax expenditures, and regulatory programs-using a common framework. Such
an analysis is necessary to capture whether a program complements and
supports other related programs, whether it is duplicative and redundant,
or whether it actually works at cross-purposes to other initiatives. OMB
officials we interviewed said that OMB would need Treasury's assistance to
determine what information or criteria to include in a PART instrument
tailored to examine tax expenditures. As of July 2005, OMB said that it
was planning to review the health insurance tax credit program next year
but that it has not decided whether the PART review will be limited to
administration or will also cover the program's tax policy purpose.

95The NMTC program issues federal tax credits to private sector entities
in return for investments in low-income communities, such as development
or rehabilitation of real estate projects.

Conclusions	As we move forward in shaping government for this century, the
federal government cannot accept all of its existing programs, policies,
functions, and activities as "givens." Outmoded commitments and operations
constitute an encumbrance on the future that can erode the capacity of the
nation to better align its government with the needs and demands of a
changing world and society. Reexamining the base of all major existing
federal spending and tax programs, policies, functions, and activities by
reviewing their results and testing their continued relevance and relative
priority for our changing society is an important step in recapturing our
fiscal flexibility and bringing the panoply of federal activities into
line with 21st century trends and challenges. The decisions we face
involve difficult choices about the appropriate size and role of the
federal government and how to finance the federal government. The revenues
forgone through tax expenditures reduce resources available to fund other
federal activities or they require higher tax rates to raise a given
amount of revenue. Reviewing their results and testing their continued
relevance and relative priority is an important step in the process
towards fiscal responsibility and national renewal. Such a fundamental
review of major programs, policies, and activities, including tax
expenditures, can serve the vital function of updating the federal
government's approach to meet current and future challenges.

Regular and systematic evaluation will be necessary to inform policy
decisions about the efficiency, effectiveness, and equity of tax
expenditures or whether they are the best tool for accomplishing federal
objectives within a functional area. Beginning the governmentwide
reexamination process now would enable decisionmakers to be more strategic
and selective in choosing areas for review over a period of years.
Reexamining selected parts of the budget base over time may make the
reviews more feasible and less burdensome, and it would allow
decisionmakers to focus on all federal efforts-discretionary spending,
mandatory spending, and tax expenditures-sharing common goals.

Unfortunately, over a decade has passed since Congress encouraged
systematic reviews of tax expenditures and since we made recommendations
to facilitate such reviews and to display information on tax expenditures
in the federal budget in a manner that enables policymakers to look at
resource commitments across related outlays and tax expenditures. Although
specific tax expenditures, such as the EITC and Liberty Zone tax benefits,
have received varying degrees of scrutiny, efforts to date have not
provided the Congress and others with an

integrated perspective on the extent to which programs and tools-
including tax expenditures-contribute to national goals and position the
government to successfully meet 21st century demands. In addition, the
lack of a requirement to disclose tax expenditures in agencies' annual
performance and accountability reports may result in important performance
and cost related data not being fully considered with other federal
resources allocated to achieve similar objectives. Although challenges
must be overcome to provide systematic reviews of tax expenditures, these
challenges cannot be addressed absent effective leadership within the
Executive Branch. Accordingly, we are making several recommendations to
OMB.

  Recommendations for Executive Action

To ensure that policymakers and the public have the necessary information
to make informed decisions and to improve the progress toward exercising
greater scrutiny of tax expenditures, we recommend that the Director of
OMB, in consultation with the Secretary of the Treasury, take the
following four actions:

o 	resume presenting tax expenditures in the budget together with related
outlay programs to show a truer picture of the federal support within a
mission area;

o 	develop and implement a framework for conducting performance reviews of
tax expenditures. In developing the framework, (1) determine which
agencies will have leadership responsibilities to review tax expenditures,
how reviews will be coordinated among agencies with related
responsibilities, and how to address the lack of credible performance
information on tax expenditures; (2) set a schedule for conducting tax
expenditure evaluations; (3) re-establish appropriate methods to test the
overall evaluation framework and make improvements as experience is
gained; and (4) to identify any additional resources that may be needed
for tax expenditure reviews.

o 	develop clear and consistent guidance to Executive Branch agencies on
how to incorporate tax expenditures in strategic plans, annual performance
plans, and performance and accountability reports, to provide a broader
perspective and more cohesive picture of the federal government's goals
and strategies to address issues that cut across Executive Branch
agencies; and

o 	require that tax expenditures be included in the PART process and any
future such budget and performance review processes so that tax
expenditures are considered along with related outlay programs in
determining the adequacy of federal efforts to achieve national
objectives.

  Agency Comments and Our Evaluation

We provided a draft of this report to OMB, Treasury, and IRS for their
review and comments. We received written comments from OMB's Associate
Director for Economic Policy in a letter dated September 2, 2005. These
comments are reprinted in app. II along with our analysis of certain
issues raised by OMB. OMB disagreed with our recommendations and several
of our findings, and also raised concerns about our use of Treasury's tax
expenditure estimates. Where appropriate, we made changes in our report in
response to these comments. The Secretary of the Treasury did not submit
comments, instead deferring to OMB. IRS staff provided a technical
correction that we incorporated.

In commenting on our report, OMB raised concerns about our use of tax
expenditure estimates developed by Treasury and reported in the annual
federal budget. For example, OMB commented that we accepted uncritically
the concept of tax expenditures first advanced in the 1960s and said that
we ignored limitations about the "volume" of total tax expenditures. To
the contrary, the background section of our draft report, as well as
several pages in app. III, clearly identified issues related to the tax
expenditure concept, including that characterizing individual provisions
as tax expenditures is a matter of judgment, and that disagreements exist
about classifying what should be included in the income tax base. Pursuant
to the Congressional Budget Act of 1974, the term tax expenditure, as our
draft stated, has been used in the federal budget for three decades, and
the tax expenditure concept-while not precisely defined-is nevertheless a
valid representation of one tool that the federal government uses to
allocate resources. Regarding the "volume" of tax expenditures, we
acknowledged throughout our draft report limitations in the methodology of
summing the individual tax expenditures. To provide an example of the
extent that interaction effects among tax expenditure estimates can affect
summing them, at our request, Treasury calculated total tax expenditures
for five itemized deductions that took these effects into account; we
included this information in our draft report. As our report stated, tax
expenditure estimates-both those published in the budget as well as those
produced by JCT-are the best and only data available to measure the value
of tax expenditures and make comparisons

to other spending programs. In our opinion, summing the estimates provides
perspective on the use of tax expenditures as a policy tool and represents
a useful gauge of the general magnitude of government subsidies carried
out through the tax code.

OMB also stated that we reported that more attention should be given to
tax expenditures due to the severity of the nation's long-term fiscal
imbalance and stated that the Administration rejects any attempt to
address the long-term fiscal imbalance with tax increases. To the
contrary, we believe that tax expenditures, like other federal programs
and activities, should be reevaluated as to their effectiveness and
continued relevance as part of a periodic reexamination of what the
federal government does and how it does business. Although the long-term
fiscal gap heightens the need to ensure resources are not wasted, this
reexamination would be appropriate regardless of the fiscal position.
Further, OMB's implication that focusing more attention on tax
expenditures would automatically increase taxes is unfounded. As our
report clearly stated, for any given level of revenue, the revenues
forgone through tax expenditures require higher tax rates to obtain a
given amount of revenue. Thus, if the evaluations of tax expenditures we
call for lead to reducing or eliminating some tax expenditures, the net
change after rate adjustments could, depending on overall congressional
priorities and preferences, result in tax reductions for many taxpayers.
We adjusted sections of our report to reinforce the point that reviewing
tax expenditures is consistent with good stewardship of taxpayers'
resources and does not automatically result in tax increases depending on
other related changes. At the same time, our current and projected fiscal
imbalance serves to reinforce the need for reassessing all activities. We
also added a recent estimate calculated by the Department of the Treasury
for the President's Advisory Panel on Federal Tax Reform which showed that
a tax system where basically all tax expenditures were eliminated could
raise the same amount of revenue as the current tax system while lowering
tax rates by about a third.

OMB also stated that information on tax expenditures is not useful for
budgeting and that tax expenditures have never been included in the
congressional budget process. To the contrary, the tax expenditure list is
legally required under the 1974 Congressional Budget Act and, before the
expiration of the Budget Enforcement Act in 2002, PAYGO procedures
restricted Congress' ability to add new tax expenditures or expand
existing ones unless offsetting funds could be raised. Whereas OMB favors
reporting tax expenditures separately from the rest of the budget, we
believe an integrated presentation is also useful to show the relative

magnitude of tax expenditures compared to spending and credit programs
across mission areas. This is not a recommendation to equate tax
expenditures with outlays. We are recommending that OMB focus on
integrating tax expenditures in the President's budget presentation to
show a truer picture of federal support in a mission area and on including
tax expenditures under budget and performance review processes that apply
to related spending programs. As our report stated, OMB began presenting
tax expenditure sums alongside outlays and credit activity for each budget
function in the federal budget from fiscal year 1998 through fiscal year
2002, but has discontinued the practice.

Finally, OMB commented that it would be unwise to follow our
recommendations for the conceptual and methodological reasons mentioned
above, as well as for other practical reasons. We address OMB's comments
on our recommendation to resume including tax expenditures in the budget
together with related outlay programs in the paragraph above.

o 	Regarding our recommendation to develop a framework for conducting
performance reviews of tax expenditures, OMB stated that it has some
potential promise but it is clearly a job for Treasury because no other
agency has access to the data that would be needed to conduct such an
analysis. However, we are not recommending that OMB be responsible for
conducting the actual reviews, just for developing and overseeing the
implementation of a framework for conducting the performance reviews. OMB
would not need to have access to taxpayer data to manage the process. In
addition, we recognize the challenges in using taxpayer data, which is the
reason we recommend that OMB work in consultation with Treasury to develop
and implement the framework. Also, our report recognizes the scarcity of
evaluation resources, and we suggest factors that would be useful in
taking a strategic approach to selecting and prioritizing tax expenditure
evaluations. To make this point more apparent in our report, we added a
fourth requirement to our recommendation to identify any additional
resources that may be needed for tax expenditure reviews.

o 	OMB said that our recommendation to develop clear and consistent
guidance to Executive Branch agencies on how to incorporate tax
expenditures in GPRA reports would be counterproductive because agencies
do not administer the tax code, and they should not be saddled with
responsibility for something they do not control. OMB misstated our
recommendation; this report does not recommend that agencies be
responsible for administering parts of the tax code. As the tax

expenditure chapter in OMB's Analytical Perspectives volume of the fiscal
year 2006 budget states, tax expenditures may also contribute to achieving
goals identified in Federal agencies annual and strategic plans for their
programs and activities. The aim of our recommendation was to provide a
more cohesive perspective of the government's programs and
strategies-including tax expenditures-to address common federal goals. As
our report states, in passing the Government Performance and Results Act,
the Senate Governmental Affairs Committee called for inclusion of tax
expenditures in the GPRA process so that more and better information would
be available on the performance of tax expenditures themselves and the
effects of tax expenditures would be considered in achieving federal
performance goals. Our recommendation is consistent with this intent.

o 	Regarding our recommendation to require tax expenditures to be included
in the PART process and any future such budget and performance review
processes, OMB stated that it has no current plans to implement any of the
recommendations in this report, but stated that other tax expenditures may
be evaluated with the PART in the future. OMB also stated that the
Department of the Treasury manages the tax code, so any new PARTs for tax
expenditures would generally mean more PARTs for Treasury. Within the
Executive Branch, major responsibility for management of the tax code was
given to the Department of the Treasury. Given that the Administration is
aiming to assess nearly 100 percent of federal outlay programs under PART,
Treasury would be facing less scrutiny than other agencies to the extent
that tax expenditures are not similarly evaluated under PART. Our
recommendation merely calls for bringing tax expenditures in line with the
performance management attention PART gives to outlay programs. Further,
if our second recommendation to develop an evaluation framework for tax
expenditures is implemented, OMB would be better positioned to target
crosscutting reviews of related clusters of programs and activities.

We are sending copies of this report to the relevant congressional
committees and other interested parties. Copies of this report will also
be made available to others upon request. In addition, the report will
also be available at no charge on GAO's Web site at http://www.gao.gov.

If you or your staff have any questions about this report, please contact
Mike Brostek at (202) 512-9110 or [email protected]. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on the
last page of this report. GAO staff who made major contributions to this
report are listed in appendix VI.

David M. Walker Comptroller General of the United States

Appendix I

Scope and Methodology

A decade ago, we examined the growth in tax expenditures and examined
opportunities to focus policymakers' attention on tax expenditures.1 To
assist the Congress in reexamining the base of federal programs and
policies critical to achieving fiscal discipline in the budget as a whole,
this report updates our 1994 work. Specifically, this report describes (1)
how tax expenditures have changed over the past three decades in reported
number and aggregate size and in comparison to federal spending, revenue,
and the economy; and (2) the progress that has been made since 1994 in how
the Executive Branch scrutinizes tax expenditures. To meet each of our
objectives, we relied on past GAO work, agency and congressional reports,
and relevant tax expenditure literature. In addition, we interviewed
agency officials from the Department of the Treasury's Office of Tax
Analysis; the Internal Revenue Service's Office of Research, Analysis, and
Statistics; the Office of Management and Budget (OMB); congressional staff
from the Joint Committee on Taxation (JCT); and experts on tax policy to
obtain a greater understanding of information gained through our
literature review and to corroborate findings.

To identify how tax expenditures have changed over the past three decades
in number and size in terms of aggregate revenue loss and
outlayequivalents, we analyzed tax expenditure estimates developed by
Treasury and reported by OMB in the Federal Budget's Special Analyses,
Appendixes, and Analytical Perspectives for fiscal years 1974 to 2004. Tax
expenditure estimates are reported for individual and corporate taxpayer
groups and categorized by budget function. We chose the tax expenditure
estimates reported in the budget for our analysis because Treasury
develops (1) revised estimates based on changes in tax policy and economic
activity for the year prior to the reported fiscal budget year (i.e.,
retrospective estimates), and (2) outlay-equivalent estimates that
facilitate comparison to federal spending. Even though Treasury's
estimates are retrospective, the final reported numbers are still
estimates and may not reflect additional policy changes. Although the tax
expenditure concept can also be applied to other kinds of taxes, such as
excise taxes, this report only covers tax expenditures for the federal
income tax system.

We determined the number of tax expenditures for each fiscal year by
adding the number of items in the list of tax expenditures reported by
Treasury for each fiscal year. In certain fiscal years, Treasury reported

1GAO, Tax Policy: Tax Expenditures Deserve More Scrutiny,
GAO/GGD/AIMD-94-122 (Washington, D.C.: June 3, 1994).

Appendix I Scope and Methodology

estimates for select tax expenditures as two line items on their list,
such as the expensing of exploration and development costs, which was
split out as two tax expenditures, one pertaining to oil and gas and one
for other fuels between fiscal years 1980 and 1995. To be consistent with
Treasury's reporting of these tax expenditures in years when they were
listed as only one item, we summed the revenue loss estimates in the years
they were listed as two tax expenditures and counted them as one. To
determine the number of distinct tax expenditures across fiscal years, we
reviewed the names and descriptions for each tax expenditure reported by
OMB in the Budget's Special Analyses, Appendixes, and Analytical
Perspectives for fiscal years 1974 to 2004. We conducted two independent
reviews to verify that our list contained only distinct tax expenditures
across fiscal years. To assist in our review, we also relied on the
descriptions reported in the Congressional Research Service's compendiums
on tax expenditures and legislative histories of certain tax expenditures,
as needed. App. IV contains our compilation of all tax expenditures
reported by Treasury between 1974 and 2004.

We aggregated tax expenditure revenue loss estimates to measure growth
over time. We also summed the revenue loss estimates by their reported
corporate and individual basis to see how the amounts differed between the
two taxpayer groups. We converted all sums for each fiscal year into
constant dollars to adjust for inflation using the chain price indexes
reported in the fiscal year 2006 federal budget. While summing tax
expenditure estimates provides a useful perspective, aggregate numbers
should be interpreted carefully due to interactive effects between tax
expenditures and potential behavioral changes.

To identify how tax expenditures have changed over the past three decades
in comparison to federal spending, revenue, and the economy, we summed the
outlay-equivalent estimates for each fiscal year and compared them to the
federal budget position in aggregate. We used historical data on spending
drawn from OMB historical tables and compared them to the sums for tax
expenditure outlay-equivalent estimates in dollar value and as a
percentage of GDP. We also used outlay-equivalent estimates to compare tax
expenditure trends over time by budget function. Finally, we used
historical data on spending by budget function from OMB historical tables
and compared them to the sum of tax expenditures by budget functions for
fiscal year 2004. We worked with Treasury officials to verify any
discrepancies we found in using the tax expenditure estimates and modified
our data accordingly.

Appendix I Scope and Methodology

To determine the amount of progress since 1994 in how the federal
government scrutinizes tax expenditures, we examined actions taken to
implement our earlier recommendations to OMB intended to encourage more
informed policy debate about tax expenditures and to stimulate joint
review of related tax and spending programs. We recommended (1) developing
a structure for conducting reviews of tax expenditures' performance (2)
conducting case studies to assess performance review structure (3)
presenting tax expenditures in the annual budget, and (4) incorporating
tax expenditures into the annual budgetary review process. We reviewed
relevant literature, interviewed relevant agency officials and tax policy
experts, and relied on previous GAO work to determine the progress that
has been made in implementing our recommendations. We reviewed efforts to
include tax expenditures under the Government Performance and Results
Act's statutory framework for strategic planning, performance measurement,
and program evaluation.2 We also considered activities to include tax
expenditures under OMB's Program Assessment Rating Tool process.

To describe how tax expenditures are measured and reported, we reviewed,
but did not verify, the procedures used by the Joint Committee on Taxation
and Treasury to estimate the magnitude of revenues forgone through tax
expenditures or, in Treasury's case, their outlay-equivalent values as
well. As described in app. III, JCT and Treasury use different conceptual
approaches to identify the provisions of the tax code they label as tax
expenditures. In addition, their estimating models, macroeconomic
assumptions, and choice of data cause their revenue loss estimates to
differ somewhat.

We conducted our work between August 2003 and July 2005 in accordance with
generally accepted government auditing standards.

2 Pub. L. No. 103-62.

Appendix II

Comments from the Office of Management and Budget

Note: GAO comments supplementing those in the report text appear at the
end of this appendix.

See comment 1.

See comment 2.

See comment 3.

See comment 1.

Appendix II
Comments from the Office of Management
and Budget

                                 See comment 1.

                                 See comment 4.

Appendix II
Comments from the Office of Management
and Budget

See comment 5.

                             See comments 1 and 6.

Appendix II
Comments from the Office of Management
and Budget

                                 See comment 1.

                                 See comment 7.

Appendix II
Comments from the Office of Management
and Budget

                                 See comment 8.

                                 See comment 9.

Appendix II
Comments from the Office of Management
and Budget

                                See comment 10.

                                See comment 11.

                                  Appendix II
                     Comments from the Office of Management
                                   and Budget

The agency comments and evaluation section of this report discusses our
overall comments on the Office of Management and Budget's letter dated
September 2, 2005. The following are our additional comments on issues
raised by OMB.

  GAO Comments 1.

2.

3.

4.

5.

6.

See the agency comments and evaluation section of this report.

While we believe that the nation's current and projected fiscal imbalance
provides an additional impetus for engaging in such a review and
reassessment, we believe tax expenditures should be reviewed and evaluated
for efficiency and effectiveness even if there were no fiscal imbalance.
We did not suggest that extra attention to tax expenditures would
eliminate the long-term fiscal imbalance. As our report stated,
substantive reform of Social Security and the major health programs
remains critical to recapturing our fiscal flexibility.

Our report cites several examples of changes in the presentation of tax
expenditures over time. For example, starting with the fiscal year 1999
budget, OMB began including a section outlining possible performance
measures and issues in evaluating tax expenditures. This section was a
first step in responding to congressional expectations for the Executive
Branch to provide information about how tax expenditures meet their
objectives and affect the performance of other federal programs.

We do not take for granted that tax expenditures are similar to spending
programs. We devote a section of our background to describing how tax
expenditures differ from, may substitute for, and work in conjunction with
other spending programs to achieve policy objectives. Also, see the agency
comments and evaluation section of this report.

In our report, we recommend adding useful comparisons to spending programs
to the budget document, while not detracting from or changing in any way
how the tax expenditure lists can be used to think about tax policies.

To the contrary, throughout our draft report we note and even emphasize
the limitations in the methodology of summing the individual tax
expenditures. In fact, to ensure that summing limitations of tax
expenditures were clearly acknowledged, we discussed the limitations in
(1) the introduction of our methodology, (2) a footnote in

Appendix II
Comments from the Office of Management
and Budget

the Results In Brief section, (3) the section devoted to explaining the
limitations which precedes our presentation of the trends in tax
expenditures over time, and (4) a footnote for all 10 figures where we
summed the tax expenditure estimates. In addition, we report a
quantitatively significant example of interaction effects of tax
expenditure estimates, which was developed by Treasury at our request. The
example shows that the revenue loss calculation assuming the simultaneous
elimination of several itemized deductions would be less than the sum of
the revenue loss estimates for each itemized deduction, each calculated
assuming the rest of the tax code was unchanged. As our report stated, tax
expenditure estimates produced by Treasury and JCT are the best and only
available data to measure the value of tax expenditures and make
comparisons to other spending programs. In our view, summing the estimates
provides perspective on the use of tax expenditures as a policy tool and
represents a useful gauge of the general magnitude of government subsidies
carried out through the tax code. Our report also cites several other
researchers who have summed tax expenditure estimates to help gain
perspective on the use of this policy tool and examine trends in the
aggregate growth of tax expenditure estimates over time.

7.	In this report, we provide a number of examples of studies we and
others have done of tax expenditures; our reviews often are at the request
of Congress, and OMB examined two tax expenditures under the
Administration's PART initiative. We also provide illustrations of the
major legislation that has affected tax expenditures since the late 1970s.
However, we stand by our statement that tax expenditures are not subject,
or effectively subject, to several major processes that apply to outlay
programs that increase the likelihood of reviews and, perhaps more
importantly, increase the quantity and quality of information available to
policymakers in determining whether and how to modify tax expenditures.
Developing such enhanced information for policymakers and displaying it in
a manner that facilitates their understanding of the total federal effort
to address functionally related issues, e.g., ensuring adequate housing or
stimulating economic development, is the thrust and intent of our report
and recommendations.

8.	We disagree with OMB's characterization that the current tax
expenditure presentation is "more than adequate" for the public and
policymakers. We realize that the current budget volume is not organized
by separate budget functions; however, OMB had previously

Appendix II
Comments from the Office of Management
and Budget

presented revenue loss sums for tax expenditures alongside outlays and
credit activity for each budget function in the federal budget from fiscal
year 1998 through fiscal year 2002. As we state in our report, these
summary tables were a useful starting point in highlighting the relative
magnitude of tax expenditures and related outlay programs across mission
areas. In addition, our current recommendation gives OMB latitude on how
to present them with other outlay programs with similar purposes. Further,
Congress has shown significant interest in reviewing all tools within a
mission area. For example, recent congressionally-requested studies we
conducted have reviewed all tools-including tax expenditures-used in the
post-secondary education and energy areas.1 Also, see comment 2 as well as
the agency comments and evaluation section of this report.

9.	We disagree with the opinion that OMB has implicitly expressed that it
would not have a leadership role regarding our second recommendation.
First, we are not recommending that OMB be responsible for conducting the
actual reviews, just to develop and oversee the implementation of a
framework for conducting the performance reviews. OMB would not need to
have access to taxpayer data to manage the process. Secondly, we recognize
the challenges in using taxpayer data, which is the reason we recommend
that OMB work in consultation with Treasury to develop and implement the
framework. Third, taxpayer data may not be the only source of performance
information on tax expenditures, which is why we recommend that the
framework address the lack of credible performance information on tax
expenditures. Finally, our report recognizes the scarcity of evaluation
resources and we suggest taking a strategic approach to select and
prioritize tax expenditure evaluations based on such factors as the
relative priorities, costs, and risks associated with related clusters of
programs and activities and that OMB select similar programs for review in
the same year to facilitate comparisons and tradeoffs. To make this point
more apparent in our report, we added a fourth element to our
recommendation that OMB and Treasury in developing a framework for
evaluating tax

1GAO, Student Aid and Postsecondary Tax Preferences: Limited Research
Exists on Effectiveness of Tools to Assist Students and Families Through
Title IV Student Aid and Tax Preferences, GAO-05-684 (Washington, D.C.:
Jul. 29, 2005), and National Energy Policy: Inventory of Major Federal
Energy Programs and Status of Policy Recommendations, GAO-05-379
(Washington, D.C.: Jun. 10, 2005).

Appendix II
Comments from the Office of Management
and Budget

expenditures are to identify any additional resources that may be needed
for tax expenditure reviews.

10. OMB misstated our recommendation. This report does not recommend that
agencies be responsible for administering parts of the tax code. As we
state in our report, in passing the Government Performance and Results
Act, the Senate Governmental Affairs Committee called for inclusion of tax
expenditures in the GPRA process so that more and better information would
be available on the performance of tax expenditures themselves and the
effects of tax expenditures would be considered in achieving federal
performance goals. Our recommendation is consistent with this intent.
Also, see the agency comments and evaluation section of this report.

11. Our recommendation aims to bring tax expenditures in line with the
performance management attention PART gives to outlay programs. Our report
discussed the two cases where OMB has applied PART to tax expenditures-the
EITC compliance initiative and the New Markets Tax Credit. Within the
executive branch, the Department of the Treasury has major responsibility
for managing programs implemented through the tax system. Given that over
the next 2 years the Administration plans to assess nearly all remaining
executive branch outlay programs, Treasury would be facing relative less
scrutiny than other agencies to the extent that the tax expenditure tool
is not similarly evaluated under PART. Although OMB disagreed with our
recommendations as a whole, we are encouraged that OMB is still
considering how other tax expenditures could be evaluated with PART in the
future. In moving forward, PART reviews of tax expenditures in isolation
might be revealing, but we would urge a more comprehensive and
crosscutting approach to assessing all tools-including tax
expenditures-related to common goals.

Appendix III

How Tax Expenditures Are Measured and Reported

To understand the trends in the size of tax expenditures, it is helpful to
understand how tax expenditures are measured and reported annually. This
appendix explains the baselines used to distinguish tax expenditures from
other provisions in the tax code and provides an explanation of the
different methods that are used to measure tax expenditures.

  Tax Expenditures Are Reported Annually by Law and Measurements Depend on
  Baselines Used

The Congressional Budget Act of 1974 defines tax expenditures as "those
revenue losses attributable to provisions of the federal tax laws which
allow a special exclusion, exemption, or deduction from gross income or
which provide a special credit, a preferential rate of tax, or a deferral
of tax liability. Both the congressional Joint Committee on Taxation (JCT)
and the Department of the Treasury's Office of Tax Analysis annually
compile a list of tax expenditures and estimates of their cost each year.
The Department of the Treasury's (Treasury) tax expenditure estimates are
included in the annual federal budget by the Office of Management and
Budget (OMB).

While, in general, the tax expenditure lists published annually by JCT and
Treasury are similar, they differ somewhat in the number of tax
expenditures reported and the estimated revenue loss for particular
expenditures. Part of this difference arises because the organizations use
different income tax baselines to determine tax expenditures. To determine
the tax code provisions that satisfy the definition of a tax expenditure,
the existing tax law must be compared or measured against an alternative
set of tax rules that represent a baseline. The Congressional Budget Act
did not define a specific baseline tax structure. As a result, the
Treasury and the staff of the JCT have used judgment to define the
different baselines that they use to develop lists of tax expenditures.
Before the fiscal year 1983 budget, there were few differences between the
Treasury and JCT tax expenditure lists because both organizations used a
baseline patterned on a comprehensive income tax, which was deemed the
"normal" baseline. JCT has used this baseline consistently over time in
producing its tax expenditure list, while Treasury has modified its normal
baseline over time and provided alternative baselines. In general, the
normal income tax law baseline developed by both Treasury and JCT
represents a broad-based income tax on individuals and a separate income
tax on corporations. The normal baseline includes income from all sources,
including wages and salaries, fringe benefits and other forms of employee
compensation, interest income, dividends, realized capital gains, and net
income from noncorporate businesses such as sole proprietorships and
partnerships. The

Appendix III How Tax Expenditures Are Measured and Reported

normal baseline generally allows for personal exemptions, deductions for
costs incurred to earn income, and a standard deduction.

Currently, the normal baselines used by both Treasury and JCT differ
somewhat. Treasury's normal baseline excludes several tax expenditures
that are included in the normal baseline used by JCT and leads to several
tax expenditures being reported by JCT only. For instance, the exclusion
of Medicare hospital insurance benefits is included in the JCT list but
this provision is not included in the federal budget tax expenditure list
because Treasury views the exclusion of government benefits received in
kind as part of its normal baseline. Additional examples of specific tax
expenditures reported by only JCT or Treasury can be found at the end of
this appendix in table 2.

In the fiscal year 1983 budget, Treasury introduced the concept of a
reference baseline. The reference baseline used by Treasury is also
patterned on a broad-based income tax, but it is closer to existing law
because tax expenditures by definition are limited to special exceptions
that serve programmatic functions, such as national defense, income
security, and education. Under Treasury's reference baseline, two
conditions are necessary for a provision to qualify as a tax expenditure:
(1) The provision must be "special" in that it applies to a narrow class
of transactions or taxpayers and (2) There must be a general provision to
which the special provision is a clear exception. The set of general tax
rules in the existing tax code is used as the standard by which various
provisions are determined to be special. Whereas accelerated depreciation
was considered a special rule exception under the normal baseline, these
provisions were not considered tax expenditures under the reference
baseline, because accelerated depreciation was considered to be the
general treatment for the depreciation of business assets. The
preferential tax rate for capital gains was included in Treasury's tax
expenditure list based on the general tax code rule that income from any
source is considered taxable. For fiscal year 1983, Treasury began to
report estimates using the reference baseline for some tax expenditures
and then reinstituted reporting estimates for the normal baseline in
fiscal year 1985. This reporting practice has continued to the present.

In recent years, Treasury modified treatment of certain provisions under
its normal and reference baselines and introduced two supplemental
baselines. In the 2005 and 2006 budgets, Treasury excluded the reduced tax
rate on dividends and capital gains that have already been taxed under the
corporate income tax from the reference law baseline because it believes

Appendix III How Tax Expenditures Are Measured and Reported

that since current law taxes these forms of corporate income twice, it is
an inappropriate baseline to use. Also, in the 2004, 2005, and 2006
budgets, Treasury changed how it computed the accelerated depreciation tax
expenditure under the normal baseline by using a measure of economic
depreciation rather than straight-line depreciation as the baseline
depreciation method, which was used in prior years. The measure of
economic depreciation is generally faster than the straight-line method,
so the tax expenditure estimates for accelerated depreciation for fiscal
years 2002, 2003, and 2004 (from the 2004, 2005, and 2006 budgets) are
smaller than what they would have been if the straight-line depreciation
method were used. In addition, in the 2004 budget, Treasury began
reporting two supplemental baselines, as discussed in figure 14.

Figure 14: Treasury's Supplemental Reporting for Comprehensive Income and
Consumption Tax Baselines

In the 2004 budget, Treasury began reporting estimates for the 30 largest
tax expenditures using comprehensive income and consumption tax baselines.

o  Treasury defines its comprehensive income baseline as the
real-inflation adjusted- accrual of wealth arising between the beginning
and end of the year. The comprehensive income baseline includes all
accrual of wealth, whether or not realized, whether or not related to a
market transaction, and whether it is a return to capital or labor.
Inflation adjusted capital gains would be included in comprehensive income
as they accrue. According to Treasury's reporting, 13 large tax
expenditures under its normal and reference baselines, such as capital
gains on home sales, would continue to be considered tax expenditures
under a comprehensive baseline. Treasury was uncertain about whether 6
would still be considered tax expenditures and concluded 4 would probably
not be tax expenditures under the comprehensive income tax baseline. The
tax exemption of in-kind benefits from government programs such as food
stamps, public housing, and Medicaid would be added to Treasury's tax
expenditure list under the comprehensive income tax baseline.

o  Treasury defines its consumption baseline as a combination of an income
tax plus a deduction for net saving. The major difference between
Treasury's comprehensive income and consumption baselines is the treatment
of tax expenditures related to saving. According to Treasury, 4 tax
expenditures under its normal or reference baseline would still be
considered tax expenditures under its consumption baseline, and another 12
would probably still be considered tax expenditures as well, such as the
child tax credit. The capital gains exclusion on home sales would not be
considered a tax expenditure under the consumption tax baseline. However,
tax expenditures unrelated to broad based saving incentives would remain
tax expenditures under a consumption baseline.

Source: Office of Management and Budget. Analytical Perspectives, Budget
of the United States Government, Fiscal Year 2006, (Washington, D.C.:
2005).

Appendix III How Tax Expenditures Are Measured and Reported

Both Treasury and JCT provide estimates of revenue loss, which is the
amount of revenue that the government forgoes as the result of each
special provision in the tax code. Revenue loss is estimated for each tax
expenditure separately by comparing the revenue raised under current law
with the revenue that would have been raised if the single provision did
not exist, assuming that taxpayer behavior and all other tax and spending
provisions remain constant. A revenue loss estimate does not represent the
amount of revenue that would be gained if a particular tax expenditure
were repealed, since repeal of the expenditure would probably change
taxpayer behavior in some way that would affect revenue.

Treasury and JCT tax expenditure lists will also differ because each
organization uses a different de minimis amount, which is the minimum
amount of revenue loss threshold for Treasury and JCT to report a tax
expenditure. JCT excludes tax expenditure estimates that result in revenue
losses that are less than $50 million over its 5-year projected period.
For instance, the tax exemption for certain small insurance companies was
not included in JCT's January 2005 list of tax expenditures because the
estimated revenue loss was below its de minimis amount. Treasury rounds
all yearly estimates to the nearest $10 million and excludes tax
expenditures with estimates that round to zero in each of the 7 years that
it reports tax expenditure estimates.

JCT and Treasury estimates of revenue loss also differ somewhat due to
different economic and technical assumptions. For instance, JCT and
Treasury use different sources for macroeconomic assumptions incorporated
in their revenue loss estimates. JCT uses CBO macroeconomic assumptions in
its tax expenditure projections and Treasury uses assumptions based on
consultations with OMB, and the Council of Economic Advisers,1 the same
assumptions used for the President's budget. In addition to projecting
future revenue losses, Treasury also reports re-estimates for the past
fiscal year, which incorporate changes in tax policy and reflect more
up-to-date economic and taxpayer data. Table 3 compares tax expenditure
reporting by JCT and Treasury.

1 At OMB, the economic forecast is produced by the Troika: economists
drawn from the Council of Economic Advisers, the Department of the
Treasury, and OMB.

          Appendix III How Tax Expenditures Are Measured and Reported

      Table 3: Comparison of Tax Expenditure Reporting by JCT and Treasury

                                  Joint Committee on  U.S. Department of the  
                Report Elements       Taxation (JCT)                 Treasury 
                 Period covered Current fiscal year         Last fiscal year, 
                                and 4                          current fiscal 
                                 future fiscal years       year, and 5 future 
                                                                 fiscal years 
                  Baseline used               Normal Reference (since 1983),  
                                                                       normal 
          Measurement estimates         Revenue loss    Revenue loss, outlay- 
                       produced                                    equivalent 

Macroeconomic assumptions

Congressional Budget Office (CBO) (mid-year update)

OMB, Treasury, and Council of Economic Advisersa

De minimis rule	Excludes provisions with estimates of less than $50
million over the 5-year period Rounds all yearlyestimates to the nearest
$10 million and excludes provisions with estimates that round to zero in
each of the 7 years

Categorized by	Budget function, taxpayer Budget function, taxpayer group
(i.e., individual or group (i.e., individual or corporate) corporate)

Supplemental information	Distributional estimates by Present value
estimates (for income class (for 9 deferral expenditures); expenditures);
summary of comprehensive and recent legislation regarding consumption
baselines used tax expenditures; list of (for select expenditures)
expiring tax expenditure provisions; and summary of differences between
Treasury and JCT lists of tax expenditures

Sources: Office of Management and Budget, Analytical Perspectives, Budget
of the United States Government, Fiscal Year 2006 (Washington, D.C.:
2005); Joint Committee on Taxation, Estimates of Federal Tax Expenditures
for Fiscal Years 2005-2009, JCS-1-05 (Washington, D.C.: Jan. 12, 2005);
and Polackova Brixi, Hana, and Christian M.A. Valenduc, and Zhicheng Li
Swift, Tax Expenditures- Shedding Light on Government Spending through the
Tax System: Lessons from Developed and Transition Economies (Washington,
D.C.: The World Bank, 2004).

a Estimates are based on mid-session economic assumptions; exceptions are
the earned income tax credit and the child tax credit, which involve
outlay components and hence are updated to reflect the economic
assumptions used elsewhere in the budget. At OMB, the economic forecast is
produced by the Troika: economists drawn from the Council of Economic
Advisers, the Department of the Treasury, and OMB. See Analytical
Perspectives, "Economic Assumptions and Analyses" for a discussion of the
Troika assumptions.

Outlay-Equivalent Estimates In addition to revenue loss estimates,
Treasury also measures tax Facilitate Comparison to expenditures in terms
of their outlay-equivalent value, which allows the Direct Spending
Programs cost of a tax expenditure to be compared with a direct federal
outlay, were

each to provide the same benefit to the taxpayer. JCT does not produce
outlay-equivalent estimates. The underlying economic assumptions used

Appendix III How Tax Expenditures Are Measured and Reported

for the outlay-equivalent and revenue loss estimates are the same.
However, to estimate outlay-equivalents, Treasury will increase-"gross
up"-the revenue loss estimate by the average marginal tax rate that
applies to the relevant taxpayers (the taxpayers that take the particular
credit or deduction or earn the income that is excluded from tax).2 The
result is an estimate of the amount of direct spending that would be
needed to leave the relevant taxpayer with the same amount of benefit,
after he or she paid tax on the amount received through the spending, as
the taxpayer would get from the tax provision itself. For example, the
outlay-equivalent estimate for the housing and meal allowances for
military personnel tax expenditure reflects the additional pre-tax income
that military personnel would have to be paid to raise their income after
federal taxes by the amount of the benefits, so that it can be compared
with other defense outlays on a consistent basis. An exception to this
general rule of increasing the revenue loss estimate is made for tax
expenditures that are believed to reduce the price of particular goods and
services. In this case no gross up is made because a spending program that
led to the same price reduction would not increase the tax liability of
the taxpayer. For instance, revenue loss estimates for accelerated
depreciation on rental housing and state prepaid tuition do not differ
from the outlay-equivalent estimates for these tax expenditures.

Outlay-equivalents can also differ from revenue loss estimates because
they are calculated based on an even flow of virtual payments over the
year to make the estimates comparable to actual outlay programs. Even for
those tax expenditures that do not require a calculated adjustment,
differences between the revenue losses and outlay-equivalents can occur
solely because of differences in timing factors. Although revenue loss
estimates can be affected by the collection patterns of the corporate and
personal income taxes, the cash flow of direct spending programs can
differ widely from the annual tax collection cycle. Of the 146 tax
expenditures reported in the fiscal year 2006 budget, 91 were "grossed up"
for the outlay-equivalent estimate with the implied rate varying across
different provisions. Just as there is debate over which tax provisions
should be listed as tax expenditures, tax experts do not always agree on
whether specific tax expenditures should be grossed up or not. It may not

2The net revenue loss for the federal government would be unchanged.
Assuming the comparable outlay program would be taxable, the recipients
would pay taxes on the higher outlay amount, and federal revenue also
would be higher-resulting in no net change in the federal budget position.

          Appendix III How Tax Expenditures Are Measured and Reported

be apparent to observers why the outlay-equivalent and revenue loss
estimates are the same for some tax expenditures and why they differ for
other tax expenditures. 3

Other estimates of tax expenditures produced by JCT and Treasury also may
differ from revenue loss estimates. These supplemental estimates are
discussed in figure 15.

Figure 15: Supplemental Estimates Developed by Treasury and JCT

o  Since the 1995 fiscal year budget, Treasury has produced present value
estimates for the approximately 20 tax expenditures that involve tax
deferrals or other long-term revenue effects. Revenue loss estimates,
which are cash based, can overstate the real effect on receipts to the
government for tax deferrals because deferred taxes will ultimately be
paid. To produce present value estimates Treasury must make certain
assumptions. Assumptions also may be specific to individual tax
expenditures, such as the time frame when people will retire and begin to
collect funds from retirement accounts (and pay income taxes on them).
Treasury uses the simplifying assumption that interest rates and tax rates
remain constant over time.

o  JCT also presents a distributional analysis of several tax
expenditures. This analysis estimates the amount of benefits by income
class for the deductions for medical expenses, real estate taxes,
charitable contributions, and the child care and earned income credits.
JCT does not report all individual tax expenditures because of the
difficulty in making reliable estimates of the income distribution of
items that do not appear on tax returns under present law.

Sources: Office of Management and Budget, Analytical Perspectives, Budget
of the United States Government, Fiscal Year 2006 (Washington, D.C.:
2005); and Joint Committee on Taxation, Estimates of Federal Tax
Expenditures for Fiscal Years 2004-2008, JCS-803 (Washington, D.C.: Dec.
22, 2003).

  Comparison of JCT With Treasury Tax Expenditure Lists

Although there are differences between how Treasury and JCT develop and
measure tax expenditures, the sum of revenue loss estimates from each list
has followed relatively the same trend in the past. Figure 16 compares the
sum of revenue loss estimates for JCT and Treasury since the last
comprehensive tax reform, when the Tax Reform Act of 1986 was adopted.
Since fiscal year 2002, the trends in the sums of the two sets of revenue
loss estimates have diverged. Since the fiscal year 2004 Budget,
Treasury's

3For more information on how outlay-equivalent estimates are measured, see
the Office of Management and Budget, Special Analysis G-Fiscal Year 1983
(Washington, D.C.: 1982); and the U.S. Congressional Budget Office, Tax
Expenditures: Current Issues and 5-Year Budget Projections for Fiscal
Years 1984-1988 (Washington, D.C.: October 1983).

          Appendix III How Tax Expenditures Are Measured and Reported

estimates of dividends and capital gains tax expenditures are lower than
JCT's, at least in part, because Treasury changed its definition of the
tax expenditures to reflect the reduced tax rates only on dividends and
capital gains from sources other than corporate equity. Treasury also
redefined the accelerated depreciation tax expenditures under the normal
baseline to reflect depreciation relative to a replacement cost basis,
rather than the historic cost basis previously used.

Figure 16: Sum of Revenue Loss Estimates Reported by the Joint Committee
on Taxation and the U.S. Department of the Treasury (1987-2004)

Dollars in billions (in constant 2004 dollars) 900 850 800 750 700 650 600
550 500 450 400 350 300 250 200 150 100

50 0

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
2002 2003 2004

Fiscal year

                        U.S. Department of the Treasury

                          Joint Committee on Taxation

Source: GAO analysis of OMB Fiscal Year Budget and Joint Committee on
Taxation Reports on Tax Expenditures, fiscal years 1989-2005.

Notes: Summing the revenue loss estimates does not take into account
possible interaction effects among the tax expenditures. Changes in
economic conditions and estimation techniques, can affect revenue loss
estimates for tax expenditures, making them differ from year to year.
Changes to the number of tax expenditures reported by Treasury would also
affect the amount of revenue loss reported if some tax expenditures were
eliminated or added. Finally, revenue loss estimates include the effect of
certain tax credits on receipts only and not the effect of the credits on
outlays.

Table 4 lists the tax expenditures and their associated revenue loss
estimates that were reported by both Treasury and JCT for fiscal year
2004. The table details the number and size of tax expenditure estimates
between the two lists. For example, in the National Defense budget
function, the revenue loss estimate for the exclusion of benefits and
allowances to

          Appendix III How Tax Expenditures Are Measured and Reported

armed forces personnel was estimated at $2.5 billion by Treasury and $2.7
billion by JCT. In the same function, JCT also reported revenue losses for
two tax expenditures not listed by Treasury.

Table 4: List of Tax Expenditures Reported by the U.S. Department of the
Treasury and the Joint Committee on Taxation for Fiscal Year 2004

Dollars in millions

                                                      Treasury  JCT estimates 
                                                     estimates  
Budget function       Tax expenditure name         Corporate     Corporate 
                                                     Individual    Individual 
      National         Exclusion of benefits and                              
       Defense            allowances to Armed             2,460         2,700
                           Forces personnel                     

Exclusion of military disability benefitsb

Deduction for overnight-travel expenses of National Guard and Reserve
Membersb

     International     Exclusion of income earned           2,680       3,400 
        Affairs          abroad by U.S. citizens                        
                     Exclusion of certain allowances         850        
                          for federal employees                         
                                 abroad                                 
                         Extraterritorial income     5,500        5,200 
                                exclusion                               
                     Inventory property sales source 1,500        5,400 
                             rule exception                             
                          Deferral of income of      7,240        4,600 
                     controlled foreign corporations                    
                      Deferred taxes for financial                      
                         firms on certain income     2,130        1,900 
                             earned overseas                            
                        Expensing of research and            -50        
General Science,           experimental           -2,280       3,500 
      Space, and              expenditures                              
      Technology                                                        
                     Credit for increasing research  4,630   50   3,900     a 
                               activities                               
                      Expensing of exploration and    230    30    500      a 
        Energy             development costs d                          

aa

Expensing of exploration and development costs: other fuels b

a

Excess of percentage over cost depletione 1,210 110 400

aa

Excess of percentage over cost depletion: other fuels

b

            Alternative fuel production credit           1,000  40  500   100 
    Exception from passive loss limitation for working          20       
          interests in oil and gas properties c                          
      Capital gains treatment of royalties on coal c            70       

a

Exclusion of interest on energy facility bonds 20 80 100 Enhanced oil
recovery credit 300 30 200 100

aa

New technology credit 330

          Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions

                                                      Treasury  JCT estimates 
                                                     estimates  
    Budget function       Tax expenditure name       Corporate      Corporate 
                                                     Individual    Individual 
                          Alcohol fuel credit f           20 10       a       
                      Tax credit and deduction for        20 50 
                           clean fuel-burning                   
                               vehicles c                       
                        Exclusion from income of                            a 
                         conservation subsidies             100 
                      provided by public utilities              
                       Tax credit for electricity                        a200 
                          production from wind,                 
                      biomass, and poultry waste b              
Natural Resources  Expensing of exploration and       210 20           a a 
                           development costs,                   
    and Environment         nonfuel minerals                    
                     Excess of percentage over cost                  100      
                           depletion, nonfuel                   
                                minerals                        
                     Exclusion of interest on bonds     110 390      200      
                     for water, sewage, and                     
                       hazardous waste facilities               
                       Capital gains treatment of            70 
                         certain timber income c                
                        Expensing of multiperiod        230 110          a200 
                          timber-growing costs                  
                     Tax incentives for preservation     230 70      400      
                         of historic structures                 
                     (under Commerce and housing                
                     credit budget function                     
                                for JCT)                        
                        Special rules for mining                          a a 
                         reclamation reserves b                 
                     Special tax rate for nuclear                    300      
                     decommissioning reserve                    
                                 fund b                         

a

Exclusion of contributions in aid of construction for water and sewer
utilities b

Expensing of capital costs with respect to complying with EPA sulfur
regulations b

Exclusion of gain or loss on sale or exchange of certain brownfield sites
b

a

Agriculture Expensing of certain capital outlays 20 80 300

Expensing of fertilizer and soil conditioner costsb	a 100 aa

Expensing of soil and water conservation expenditures b

Expensing of certain multiperiod production costs c 10 40

Treatment of loans forgiven for solvent farmers 10 100

Capital gains treatment of certain income c 670

a

Income averaging for farmers 40
Deferral of gain on sales of farm refiners c 10
Bio-Diesel tax credit b
Exclusion of cost-sharing payments b a a

          Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions

Treasury estimates JCT estimates Budget function Tax expenditure name
Corporate Individual Corporate Individual

a

Expensing of the costs of raising dairy and breeding cattle b

aa

Five-year carryback period for net operating losses attributable to
farming b

    Commerce and  Exemption of credit union income  1,270        1,200 
Housing Credit                                                      
                     Excess bad debt reserves of      -20              
                      financial institutions c                         
                    Exclusion of interest on life   2,010 18,820 1,400 24,700 
                          insurance savings                            
                  Special alternative tax on small     10              
                  property and casualty                                
                        insurance companies c                          
                  Deduction of unpaid property loss                    
                            reserves for                         1,500 
                   property and casualty insurance                     
                             companies b                               
                  Tax exemption of certain           180               
                  insurance companies owned                            
                    by tax-exempt organizations c                      
                    Small life insurance company       80         100  
                              deduction                                
                      Special treatment of life                        
                          insurance company                      1,700 
                              reservesb                                
                      Exclusion of interest on       220   800    300  
                       owner-occupied mortgage                         
                            subsidy bonds                              
                   Exclusion of interest on rental     80  280    100  
                            housing bonds                              
                  Deductibility of mortgage                                   
                  interest on owner-occupied              61,450       61,400
                                homes                                  
                  Deductibility of state and local                            
                  property tax on owner-                  19,930       18,700
                           occupied homes                              
                   Capital gains exclusion on home        29,730       17,900 
                                sales                                  
                   Exclusion of net imputed rental                     
                          income on owner-                24,590       
                          occupied homes c                             
                  Exception from passive loss rules                    
                           for $25,000 of                 5,030        
                            rental loss c                              
                    Credit for low-income housing   2,930  730   3,000  1,300 
                             investments                               
                     Accelerated depreciation on      -10  760    300   3,000 
                           rental housing                              
                  Deferral of income from post-1987  290   810    600     400 
                          installment sales                            
                   Cancellation of indebtedness c           30         
                  Exceptions from imputed interest          50       a    300 
                                rules                                  
                  Capital gains (except                                       
                  agriculture, timber, iron ore,                       
                  and                                     25,150       66,100
                                coal)                                  
                  Capital gains exclusion of small         160         
                         corporation stock c                           
                  Step-up basis of capital gains at                           
                                death                     24,200       35,900

          Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions

                                                   Treasury    JCT estimates  
                                                   estimates   
Budget function     Tax expenditure name          Corporate      Corporate 
                                                    Individual     Individual 
                    Carryover basis of capital             210          4,300 
                          gains on gifts                       
                   Ordinary income treatment of             50 
                          loss from small                      
                    business corporation stock                 
                              sale c                           
                    Accelerated depreciation of                               
                       buildings other than        -2,980 -280    1,800 1,900
                          rental housing                       
                    Accelerated depreciation of                               
                           machinery and          37,080 7,610  52,900 16,100
                             equipment                         
                    Expensing of certain small         680 840 
                            investments                        
                   (JCT did not report a revenue               
                          loss estimate)                       
                     Amortization of start-up            70 10       a        
                               costs                           

Deduction for U.S. production activities b

               Special rules for certain film and TV productionb

Graduated corporation income tax rate 2,450 3,300

Exclusion of interest on small issue bonds 100 350 100

Deferral of gain on like-kind exchanges b 1,200

a

Deferral of gain on involuntary conversions resulting from
Presidentially-declared disastersb

Expensing of magazine circulation expenditures b a a

aa

Special rules for magazine, paperback book, and record returns b

Completed contract rules b 200 a

          Cash accounting, other than agriculture b                a      700 
      Exception from net operating loss limitations for          700 
          corporations in bankruptcy proceedings b                   
      Tax credit for employer-paid FICA taxes on tipsb           200      300 

Transportation Deferral of tax on shipping companies 20 100

Deduction for clean fuel vehicles and refueling 200                  
property b                                                           
Exclusion of reimbursed employee parking expenses 2,470              
c                                                                    

a

Exclusion of employer-provided transit passes 410 3,800

Tax credit for certain expenditures for maintaining railroad tracks b a

Community and Investment credit for rehabilitation of structures, other 20
20 100
Regional than historic
Development

Exclusion of interest for airport, dock, and similar 180 670 200 600 bonds

Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions

                                                      Treasury  JCT estimates 
                                                     estimates  
Budget function       Tax expenditure name         Corporate     Corporate 
                                                     Individual    Individual 
                   Exemption of certain mutuals and      60     
                             cooperatives                       
                               income c                         
                   Empowerment zones, enterprise        280 800 
                   communities, and                             
                         renewal communities c                  
                   Empowerment zone tax incentives b                 300      
                   Renewal community tax incentives                  100      
                                   b                            
                        New markets tax credit           70 220      100      

aa

Expensing of environmental remediation costs 70 10 Deferral of capital
gains with respect of dispositions of transmission property b

New York City Liberty Zone tax incentives b 100

aa

District of Columbia tax incentives b

                         Wage credit for Indian reservation          a      a 
                                     employmentb                        
                         Accelerated depreciation for Indian        100 
                                     reservation                        
                                    investments b                       
                            Exclusion of scholarship and                      
Education, Training,           fellowship income           1,320     1,500
     Employment, and                                                    
     Social Services                                                    
                                   HOPE tax credit            3,320     4,300 
                           Lifetime Learning tax credit c     2,190     
                           Education Individual Retirement              
                                      Accounts                  110     

          Deductibility of student loan interest               760        700 
          Deduction for higher education expenses             1,280     2,700 
               State pre-paid tuition plans                    210        500 
        Exclusion of interest on student loan bonds        60  230  100   300 
Exclusion of interest on bonds for private nonprofit  210   760  300   800 
                  educational facilities                                
         Credit for holders of zone academy bonds          90       100 

Exclusion of interest on savings bonds redeemed to 10                     
finance educational expenses                                              
Parental personal exemption for students age 19 or 3,200 1,500            
over                                                                      
Deductibility for charitable contributions (education) 510 3,180 1,100    
5,200                                                                     
Exclusion of employer-provided education assistance 530 800               
Special deduction for teacher expenses 150 100                            

a

a

Work opportunity tax credit 240 40 200

Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions

                                                    Treasury    JCT estimates 
                                                    estimates   
    Budget          Tax expenditure name          Corporate         Corporate 
function                                       Individual       Individual 
                 Welfare-to-work tax credit               50 10          a100 
                Employer-provided child care                600 
                         exclusion g                            
             Employer-provided child care credit                         a100 
                Assistance for adopted foster               290 
                         children c                             
                Adoption credit and exclusion               450 
            Exclusion of employee meals and                 810 
            lodging (other than                                 
                          military)                             
                       Child credit h                    22,400        44,100 
             Credit for child and dependent care          2,990         3,100 
                          expenses                              
                 Credit for disabled access               10 20       a       
                        expenditures                            
                Deductibility for charitable                                  
                  contributions, other than        1,170 26,200  1,800 27,900
                    education and health                        
              Exclusion of certain foster care              440 
                          payments                              
              Exclusion of parsonage allowances             430 
            Exclusion of benefits provided under                       16,900 
            cafeteria plans b i                                 
              Exclusion of miscellaneous fringe                               
                         benefits b                                     5,800

Exclusion of employee awards b

       Exclusion of income earned by voluntary employees'               3,200 
                   beneficiary associations b                  

          Deferral of taxation on spread on                               400 
          acquisition of stock                                         
             under incentive stock option plans and                    
                            employee                                   
                    stock purchase plans b j                           
Health   Exclusion of employer contributions for                           
                             medial                        102,250     96,000
              insurance premiums and medical carek                     
             Deductibility of self-employed medical                           
                           insurance                        3,300       3,300
                            premiums                                   
            Medical savings accounts/health savings          620          300 
                            accounts                                   
               Deductibility of medical expenses            7,380       5,900 
          Exclusion of interest on hospital            400  1,470  500  1,200 
          construction bonds                                           
          Exclusion of workers' compensation benefits                         
          (medical                                                      3,700
                           benefits)b                                  
           Deductibility of charitable contributions   150  2,940  900  3,500 
                            (health)                                   
              Tax credit for orphan drug research      180         200 
          Special Blue Cross/Blue Shield deduction (in 400         500 

Commerce and housing credit budget function for JCT)

Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions Exclusion of untaxed Medicare benefits: prescription
drug insurance (Part D)b

                                                      Treasury  JCT estimates 
                                                     estimates  
    Budget            Tax expenditure name            Corporate     Corporate 
function                                          Individual    Individual 
             Tax credit for health insurance                 50           a a 
             purchased by certain                               
               displaced and retired individuals l              
             Exclusion of medical care and                                    
             CHAMPUS/TRICARE                                            1,700
             medical insurance for military                     
             dependents, retirees,                              
                    and retiree dependents b                    
Medicare  Exclusion of untaxed Medicare benefits:                          
                            hospital                                   16,800
                       insurance (Part A)b                      
             Exclusion of untaxed Medicare benefits:                   11,000 
              supplementary medical insurance (Part             
                               B)b                              

Exclusion of certain subsidies to employers who maintain prescription drug
plans for Medicare b

Income Security	Exclusion of railroad retirement system benefitsc (For 400
JCT this is combined with the exclusion of Social Security benefits:
retired workers)

               Exclusion of worker's compensation benefits    5,490     4,800 
                   Exclusion of public assistance benefits     410      3,200 
    Exclusion of special benefits for disabled coal miners      60     
               Exclusion of military disability pensions c     100     
               Exclusion of damages on account of personal              1,400 
                  physical injuries or physical sickness b             
      Net exclusion of pension contributions and earnings:    46,970   94,600 
                                            employer plans             
      Net exclusion of pension contributions and earnings:    47,730   
                                            401(k) plans c             
      Net exclusion of pension contributions and earnings:    7,450    13,000 
                               individual retirement plans             
      Net exclusion of pension contributions and earnings:     970     
                   low and moderate income savers credit c             
      Net exclusion of pension contributions and earnings:    8,830     6,200 
                                               Keogh plans             
         Exclusion of other employee benefits: premiums on    2,070     2,400 
                                 group term life insurance             
         Exclusion of other employee benefits: premiums on     260      2,400 
                         accident and disability insurance             
                     Small business retirement plan credit 40   40   a      a 
                 Income of trusts to finance supplementary      20     

                            unemployment benefits c

          Appendix III How Tax Expenditures Are Measured and Reported

                         (Continued From Previous Page)

Dollars in millions

                                                      Treasury  JCT estimates 
                                                     estimates  
Budget function       Tax expenditure name         Corporate     Corporate 
                                                     Individual    Individual 
                   Special employee stock ownership                  800      
                             plans (ESOPs)            1,600 320 
                       rules (Located under the                 
                       Education budget function                
                               for JCT)                         
                     Additional deduction for the            30 
                                blind c                         
                     Additional deduction for the         1,700 
                               elderly c                        
                   Additional standard deduction for                          
                           the blind and the                            2,000
                               elderly b                        
                    Tax credit for the elderly and           20             a 
                               disabled                         
                     Deduction of casualty losses           550 
                   Earned income tax credit (EITC)m       4,893        34,100 
                   Tax credit for certain                                     
                   individuals for elective                     
                   deferrals                                            2,500
                        and IRA contributions b                 
                   Exclusion of Social Security                               
Social Security benefits: retired workers             19,200        20,000
                   (JCT included railroad retirement            
                               benefits)                        
                     Exclusion of Social Security         3,580 
                         benefits: disabled c                   
                     Exclusion of Social Security               
                         benefits: dependents             4,140 
                            and survivors c                     
      Veterans'       Exclusion of veterans death                             
      Benefits          benefits and disability           3,300         3,100
    and Services             compensation                       
                    Exclusion of veterans' pensions         110 
                     Exclusion of GI bill benefits          130 

aa

Exclusion of interest on veterans' housing bonds 10 40

    General Purpose     Exclusion of interest on    6,210 19,940 7,100 18,200 
                        public purpose state and                       
Fiscal Assistance          local bonds                              
                      Deductibility of nonbusiness                            
                         state and local taxes            45,290       44,300
                       other than owner-occupied                       
                                 homes                                 
                      Tax credit for corporations                      
                         receiving income from      1,000        1,400 
                         doing business in U.S.                        
                              possessions                              
       Interest       Deferral of interest on U.S.          50                
                             savings bonds                              1,700

Source: OMB. Analytical Perspectives, Budget of the United States
Government, Fiscal Year 2006. (Washington, D.C.: 2005; and JCT, Estimates
of Federal Tax Expenditures for Fiscal Years 2004-2008, JCS-8-03
(Washington, D.C.: December 22, 2003).

Notes: Treasury estimates were rounded to the nearest $10 million.
Provisions with estimates that rounded to zero in each year were not
reported by Treasury.

a An estimate was not reported by JCT because the positive tax expenditure
was less than $50 million.

bTax expenditure reported by JCT only. The exclusion of untaxed Medicare
benefits: prescription drug insurance (Part D) and the exclusion of
certain subsidies to employers who maintain prescription drug plans for
Medcare were listed by JCT as tax expenditures, but revenue loss estimates
were reported only for future years.

cTax expenditure reported by Treasury only. Seven additional tax
expenditures were reported on Treasury's list in the fiscal year 2006
budget, but estimates for these tax expenditures were reported

Appendix III How Tax Expenditures Are Measured and Reported

only for future years. The seven tax expenditures are: (1) expensing of
capital costs with respect to complying with EPA sulfur regulations, (2)
exclusion of gain or loss on sale or exchange of certain Brownfield sites,
(3) Bio-diesel tax credit, (4) deduction for U.S. production activities,
(5) special rules for certain film and TV production, (6) tax credit for
certain expenditures for maintaining railroad tracks, and (7) deferral of
capital gains with respect of dispositions of transmission property.

dJCT reported this tax expenditure as "expensing of exploration and
development costs, oil and gas."

eJCT reported this tax expenditure as "excess of percentage over cost
depletion, oil and gas."

fFor the Treasury estimates, the partial exemption from the excise tax for
alcohol fuels results in a reduction in excise tax receipts of $1.4
billion in 2004. For the JCT estimates, the exemption from the excise tax
for alcohol fuels results in a reduction in excise tax receipts, net of
income tax effect, of $1.1 billion in each of the fiscal years 2004
through 2006, and $1.2 billion per year in fiscal years 2007 and 2008.

gThe JCT estimate includes employer-provided child care purchased through
dependent care flexible spending accounts.

hThe Treasury estimate in the table indicates the effect of the child tax
credit on receipts. The effect of the credit on outlays is $8.9 billion in
2004. The JCT estimate includes refundable amounts, amounts used to offset
income taxes, and amounts used to offset other taxes. The amount of
refundable child tax credit and earned income tax credit used to offset
taxes other than income tax or paid out as refunds is $44.3 billion in
2004.

IThe estimate includes amounts of employer-provided health insurance
purchased through cafeteria plans and employer-provided child care
purchased through dependent care flexible spending accounts.

jThe estimate does not include offsetting denial of corporate deduction
for qualified stock option compensation.

kThe JCT estimate includes employer-provided health insurance purchased
through cafeteria plans.

lIn addition to the receipts shown, there are outlays of $70 million in
2004.

mThe Treasury estimate indicates the effect of the earned income tax
credit on receipts. The effect of the credit on outlays is $33.1 billion
in 2004. The JCT estimate includes refundable amounts, amounts used to
offset income taxes, and amounts used to offset other taxes. The amount of
refundable child tax credit and earned income tax credit used to offset
taxes other than income tax or paid out as refunds is $44.3 billion in
2004.

For fiscal year 2004, table 5 lists the aggregate revenue loss estimates
reported by both Treasury and JCT for each budget function. The table
permits a comparison of the number and size of Treasury's versus JCT's tax
expenditure estimates when summed by budget function.

          Appendix III How Tax Expenditures Are Measured and Reported

Table 5: Sum of Revenue Loss Estimates for Tax Expenditures by Budget
Function Reported by the U.S. Department of the Treasury and Joint
Committee on Taxation, Fiscal Year 2004

Dollars in millions General purpose fiscal assistance 72,440 3 71,000 3
Interest 50 1 1,700 1

                                     Treasury               JCT 
                                  Sum of Revenue Sum of Revenue Number of Tax 
                                  Number of Tax                 
          Budget Function         Loss Estimates Loss Estimates  Expenditures 
                                    Expenditures                
          National defense              $2,460 1         $2,900 
       International affairs            19,900 6         20,900 
    General science, space, and         2,350 12          7,500 
             technology                                         
               Energy                   3,670 11          2,100 
       Natural resources and             1,440 5          1,900 
            environment                                         
            Agriculture                    880 6            600 
    Commerce and housing credit       265,750 29        325,900 
           Transportation                2,900 3          4,100 
       Community and regional            2,400 6          2,700 
            development                                         
        Education, training,                                    
       employment, and social          74,270 27        128,000 
              services                                          
               Health                  119,140 9        117,700 
              Medicare                   N/A N/A         27,800 
          Income security             129,953 20        168,000 
          Social security               26,920 3         20,000 
       Veterans' benefits and                                   
              services                   3,590 4          3,400 

Source: GAO Analysis.

Note: Summing the revenue loss estimates by budget fuction does not take
into account possible interaction effects among the tax expenditures.

Appendix IV

Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

To identify how tax expenditures have changed over the past three decades
in number and size, in terms of aggregate revenue loss, we analyzed the
list of tax expenditures reported by the Department of the Treasury
(Treasury) in the Budget's Special Analyses, Appendixes, and Analytical
Perspectives for fiscal years 1974 to 2004. The tax expenditures reported
by Treasury during this period are listed in table 6.

Table 6: Tax Expenditures Reported by the U.S. Department of the Treasury
(1974 to 2004)

First Year Last Year Name of Tax Expenditure Reported by Reported by
(Organized by Budget Function) Treasury Treasury Type Taxpayer Group
National defense

Exclusion of benefits and allowances to armed forces 1974 2004 Exclusion
Individual personnel

                             International affairs

       Exclusion of gross-up on     1974 1976    Exclusion       Corporate    
    dividends of LDC corporations                              
       Special rate for Western     1974 1979 Preferential tax   Corporate    
    Hemisphere trade corporations                              
                                                    rate       
    Deferral of income of domestic  1974 1985     Deferral                    
international sales corporations                              Corporate
                (DISC)                                         
      Exclusion of income earned    1974 2004    Exclusion       Individual   
       abroad by U.S. citizens                                 
       Deferral of income from                                                
controlled foreign corporations  1977 2004     Deferral     
               (normal                                           Corporate
             tax method)                                       
    Exclusion of income of foreign  1984 2000    Exclusion       Corporate    
       sales corporations (FSC)                                
      Interest allocation rules     1986 1995    Deduction                    
exception for certain financial                               Corporate
              operations                                       
Inventory property sales source  1986 2004      Credit        Corporate    
           rules exception                                     
     Deferred taxes for financial   1998 2004     Deferral       Corporate    
    firms on certain income earned                             
               overseas                                        
Exclusion of certain allowances  1999 2004    Exclusion       Individual   
     for federal employees abroad                              
       Extraterritorial income      2000 2004    Exclusion       Corporate    
              exclusion                                        
     General science, space, and                               
              technology                                       
      Expensing of research and     1974 2004     Deferral     Individual and 
     experimentation expenditures                              
         (normal tax method)                                     Corporate    
    Credit for increasing research  1981 2004      Credit      Individual and 
              activities                                       
                                                                 Corporate    
Suspension of the allocation of  1983 1995    Deduction                    
     research and experimentation                                Corporate
             expenditures                                      

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function) Energy

      Capital gains treatment of    1974 2004 Preferential tax   Individual   
          royalties on coal                                    
                                                    rate       
    Excess of percentage over cost  1974 2004    Deduction     Individual and 
           depletion, fuels                                    
                                                                 Corporate    
     Expensing of exploration and   1974 2004     Deferral     Individual and 
       development costs, fuels                                
                                                                 Corporate    
      Residential energy credits    1978 1987      Credit        Individual   
        New technology credit       1978 2004      Credit        Corporate    
     Energy credit for intercity    1980 1989      Credit        Corporate    
                buses                                          
         Alcohol fuel credits       1980 2004      Credit      Individual and 
                                                                 Corporate    
     Alternative fuel production    1980 2004      Credit      Individual and 
                credit                                         
                                                                 Corporate    
Exclusion of interest on energy  1980 2004    Exclusion     Individual and 
            facility bonds                                     
                                                                 Corporate    
     Exception from passive loss                                              
limitation for working interests 1988 2004    Deduction       Individual
                  in                                           
        oil and gas properties                                 
     Tax credit and deduction for   1992 2004                  Individual and 
     clean-fuel burning vehicles                  Credit/      
                                                 Deduction       Corporate    
      Exclusion of conservation                                               
     subsidies provided by public   1993 2004    Exclusion       Individual
              utilities                                        
     Enhanced oil recovery credit   1994 2004      Credit      Individual and 
                                                                 Corporate    

                       Natural resources and environment

      Pollution control: 5-year     1974 1980     Deferral       Corporate    
             amortization                                      
      Capital gains treatment of    1974 2004 Preferential tax   Individual   
        certain timber income                                  
                                                    rate       
    Exclusion of interest on bonds  1975 2004    Exclusion     Individual and 
        for water, sewage, and                                 
      hazardous waste facilities                                 Corporate    
Exclusion of payments in aid of  1976 1981    Exclusion                    
    construction of water, sewage,                               Corporate
      gas and electric utilities                               
Capital gains treatment of iron  1977 1997 Preferential tax   Individual   
                 ore                                           
                                                    rate       
Tax incentives for preservation  1977 2004      Credit      Individual and 
        of historic structures                                 
                                                                 Corporate    
     Investment credit and 7-year   1980 2000                    Individual   
    amortization for reforestation            Credit/ Deferral 
             expenditures                                      
    Excess of percentage over cost  1980 2004     Deferral     Individual and 
     depletion, nonfuel minerals                               
                                                                 Corporate    
     Expensing of exploration and   1980 2004     Deferral                    
      development costs, nonfuel                                 Corporate

minerals

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function) Agriculture

    Special rules for mining reclamation   1984 1997 Deduction Individual and 
                  reserves                                     
                                                                 Corporate    
Expensing of multiperiod timber growing 1986 2004 Deferral  Individual and 
                    costs                                      
                                                                 Corporate    
Expensing of capital costs with respect 2004 2004 Deferral                 
            to complying with EPA                                Corporate
             sulfur regulations                                
    Exclusion of gain or loss on sale or   2004 2004 Exclusion Individual and 
             exchange of certain                               
              brownfield sites                                   Corporate    

      Capital gain treatment of     1974 2004 Preferential tax   Individual   
            certain income                                     
                                                    rate       
     Expensing of certain capital   1974 2004     Deferral     Individual and 
               outlays                                         
                                                                 Corporate    
       Deductibility of noncash                                               
patronage dividends and certain  1977 1981    Deduction     Individual and
                other                                          
        items of cooperatives                                    Corporate    
         Exclusion of certain       1978 1981    Exclusion       Individual   
        cost-sharing payments                                  
    Special investment tax credit   1986 1987      Credit        Individual   
     carryback rules for farming                               
Expensing of certain multiperiod 1986 2004     Deferral     Individual and 
           production costs                                    
                                                                 Corporate    
Treatment of loans forgiven for  1986 2004    Exclusion       Individual   
           solvent farmers                                     
Deferral of 1988 drought-related 1988 1990     Deferral       Individual   
               payments                                        
     Income averaging for farmers   1997 2004 Preferential tax   Individual   
                                                    rate       
Deferral of gain on sale of farm 1998 2004     Deferral       Corporate    
               refiners                                        
        Bio-diesel tax credit       2004 2004      Credit        Individual   

                          Commerce and housing credit

Surtax exemption (through 1978)  1974 1980 Preferential tax   Corporate    
                                                    rate       
          Dividend exclusion        1974 1987    Exclusion       Individual   
     Deductibility of interest on   1974 1990    Deduction       Individual   
           consumer credit                                     
Deferral of capital gain on home 1974 1997     Deferral       Individual   
                sales                                          
        Capital gains (except                                                 
    agriculture, timber, iron ore,  1974 2004 Preferential tax   Individual
              and coal)                                        
                                                    rate       
      Deductibility of mortgage     1974 2004    Deduction       Individual   
interest on owner-occupied homes                            
Deductibility of state and local 1974 2004    Deduction       Individual   
    property tax on owner-occupied                             
                homes                                          
     Accelerated depreciation of                                              
     buildings other than rental    1974 2004     Deferral     Individual and
               housing                                         
         (normal tax method)                                     Corporate    
     Accelerated depreciation on    1974 2004     Deferral     Individual and 
      rental housing (normal tax                               
               method)                                           Corporate    

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function)

     Excess bad debt reserves of    1974 2004    Deduction       Corporate    
        financial institutions                                 
    Exclusion of interest on life   1974 2004    Exclusion     Individual and 
          insurance savings                                    
                                                                 Corporate    
Exemption of credit union income 1974 2004    Exemption       Corporate    
Credit for purchase of new home  1975 1977      Credit        Individual   
    Excess first year depreciation  1975 1980     Deferral     Individual and 
                                                                 Corporate    
Expensing of construction period 1975 1981     Deferral     Individual and 
          interest and taxes                                   
                                                                 Corporate    
    Exclusion of interest on small  1975 2004    Exclusion     Individual and 
             issue bonds                                       
                                                                 Corporate    
    Exclusion of certain income of  1976 1976    Exclusion     Individual and 
             cooperatives                                      
                                                                 Corporate    
     Accelerated depreciation of    1977 2004     Deferral     Individual and 
       machinery and equipment                                 
                                                                 Corporate    
    Step-up basis of capital gains  1977 2004    Exclusion       Individual   
               at death                                        
    Investment credit, other than   1978 1990      Credit                     
    employee stock ownership plans                               Corporate
    (ESOPs) and rehabilitation of                              
structures, energy property, and                            
      reforestation expenditures                               
    Exclusion of capital gains on   1978 1997    Exclusion       Individual   
    home sales for persons age 55                              
               and over                                        
       Exclusion of interest on     1978 2004    Exclusion     Individual and 
owner-occupied mortgage subsidy                             
                bonds                                            Corporate    
Graduated corporation income tax 1978 2004 Preferential tax   Corporate    
       rate (normal tax method)                                
                                                    rate       
    Amortization of start-up costs  1980 2004     Deferral     Individual and 
                                                                 Corporate    
Exclusion of interest on rental  1980 2004    Exclusion     Individual and 
            housing bonds                                      
                                                                 Corporate    
Exclusion of interest on certain 1981 1984    Exclusion       Individual   
         savings certificates                                  
     Reinvestment of dividends in   1981 1986    Exclusion       Individual   
         public utility stock                                  
      Safe harbor leasing rules     1981 1990    Deduction/      Corporate    
                                                   Credit      
        Net interest exclusion      1983 1983    Exclusion       Individual   
     Small life insurance company   1984 2004    Deduction       Corporate    
              deduction                                        
    Special investment tax credit   1986 1987      Credit                     
      carryback rules for steel                                  Corporate
              companies                                        
    Credit for low-income housing   1986 2004      Credit      Individual and 
             investments                                       
                                                                 Corporate    
    Exemption of RIC expenses from  1987 1993    Deduction       Individual   
      the 2% floor miscellaneous                               

itemized deduction

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function)

     Deferral of income from post   1987 2004     Deferral     Individual and 
        1987 installment sales                                 
                                                                 Corporate    
     Exception from passive loss                                              
     rules for $25,000 of rental    1987 2004    Deduction       Individual
                losses                                         
     Small property and casualty    1988 1988 Preferential tax                
     insurance company deduction                                 Corporate
                                                    rate       
       Special merger rules for     1988 1990    Deduction       Corporate    
        financial institutions                                 
      Treatment of Alaska Native    1988 1996    Exemption       Corporate    
             Corporation                                       
Carryover basis of capital gains 1988 2004     Deferral       Individual   
               on gifts                                        
Exceptions from imputed interest 1988 2004    Exclusion       Individual   
                rules                                          
Special alternative tax on small 1988 2004                                 
        property and casualty                    Exemption       Corporate
         insurance companies                                   
       Tax exemption of certain                                               
     insurance companies owned by   1988 2004                  
                 tax-                            Exemption       Corporate
         exempt organizations                                  
     Ordinary income treatment of   1989 2004    Deduction       Individual   
       loss from small business                                
        corporation stock sale                                 
    Deferral of gains from sale of                                            
      broadcasting facilities to    1990 1995     Deferral     
               minority                                          Corporate
            owned business                                     
      Cancellation of indebtness    1993 2004    Exclusion       Individual   
      Expensing of certain small    1993 2004     Deferral     Individual and 
             investments                                       
                                                                 Corporate    
Capital gains exclusion of small 1994 2004    Exclusion       Individual   
          corporation stock                                    
Capital gains exclusion on home  1997 2004    Exclusion       Individual   
                sales                                          
Exclusion of net imputed rental  2004 2004    Exclusion       Individual   
       income on owner-occupied                                
                homes                                          
    Deduction for U.S. production   2004 2004    Deduction     Individual and 
              activities                                       
                                                                 Corporate    
    Special rules for certain film  2004 2004    Deduction     Individual and 
          and TV production                                    
                                                                 Corporate    
            Transportation                                     
     Deductibility of nonbusiness   1974 1979    Deduction       Individual   
         state gasoline taxes                                  
      Five-year amortization on     1974 1980     Deferral       Corporate    
        railroad rolling stock                                 
     Deferral of tax on shipping    1974 2004     Deferral       Corporate    
              companies                                        
     Deduction for motor carrier    1981 1986    Deduction       Corporate    
           operating rights                                    
    Exclusion of interest on state  1981 1990    Exclusion                    
    and local government bonds for                               Corporate
       mass commuting vehicles                                 
Exclusion for employer-provided  1993 2004    Exclusion       Individual   
            transit passes                                     
Exclusion of reimbursed employee 1993 2004    Exclusion       Individual   
           parking expenses                                    
        Tax credit for certain                                                
     expenditures for maintaining   2004 2004      Credit      
               railroad                                          Corporate

tracks

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function)

 Community and regional development Education, training, employment, and social
                                    services

     Five-year amortization for housing    1974 1990 Deferral  Individual and 
               rehabilitation                                  
                                                                 Corporate    
Investment credit for rehabilitation of 1974 2004  Credit   Individual and 
           structures (other than                              
                  historic)                                      Corporate    
     Exclusion of interest for airport,    1983 2004 Exclusion Individual and 
           dock, and similar bonds                             
                                                                 Corporate    
      Exemption of certain mutuals and     1989 2004 Exemption   Corporate    
             cooperatives income                               
        Empowerment zones, enterprise      1993 2004  Credit   Individual and 
          communities, and renewal                             
                 communities                                     Corporate    
Expensing of environmental remediation  1997 2004 Deferral  Individual and 
                    costs                                      
                                                                 Corporate    
            New market tax credit          2000 2004  Credit   Individual and 
                                                                 Corporate    
Deferral of capital gains with respect  2004 2004 Deferral                 
             of dispositions of                                  Corporate
            transmission property                              

        Child care facilities: 5-year      1974 1975 Deferral    Corporate    
                amortization                                   
    Credit for employing AFDC recipients   1974 1982  Credit   Individual and 
            and public assistance                              
recipients under work incentive program                       Corporate    
     Credit for child and dependent care   1974 2004  Credit     Individual   
                  expenses                                     
         Deductibility of charitable       1974 2004 Deduction Individual and 
          contributions (education)                            
                                                                 Corporate    
         Deductibility of charitable       1974 2004 Deduction Individual and 
     contributions, other than education                       
                 and health                                      Corporate    
Exclusion of employee meals and lodging 1974 2004 Exclusion   Individual   
            (other than military)                              
Exclusion of scholarship and fellowship 1974 2004 Exclusion   Individual   
                   income                                      
       Parental personal exemption for     1974 2004 Exemption   Individual   
           students age 19 or over                             
Maximum tax on personal service income  1975 1980 Exclusion   Individual   
         Investment credit for ESOPs       1976 1990  Credit     Corporate    
    Exclusion of other employee benefits:  1976 1992 Exclusion   Individual   
           employer contributions                              
        to prepaid legal expense plan                          
Expensing of costs of removing certain  1976 1999 Deferral                 
          architectural barriers to                              Corporate
               the handicapped                                 
             General jobs credit           1977 1984  Credit     Corporate    
       Exclusion of employer-provided      1978 2004 Exclusion   Individual   
           educational assistance                              
         Work opportunity tax credit       1978 2004  Credit   Individual and 
                                                                 Corporate    
    Exclusion of interest on student loan  1980 2004 Exclusion Individual and 
                    bonds                                      
                                                                 Corporate    

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function)

      Deduction for certain adoption    1981 1987    Deduction Individual     
                 expenses                         
     Deduction for two earner married   1983 1987    Deduction Individual     
                 couples                          
       Employer-provided child care     1983 2004    Exclusion Individual     
                exclusion                         
    Exclusion of interest on bonds for  1983 2004    Exclusion Individual and 
      private nonprofit educational               
                facilities                                          Corporate 
    Exclusion of parsonage allowances   1984 2004    Exclusion Individual     
     Exclusion of certain foster care   1988 2004    Exclusion Individual     
                 payments                         
     Exclusion of interest on savings   1988 2004    Exclusion Individual     
        bonds redeemed to finance                 
           educational expenses                   
        Credit for disabled access      1990 2004       Credit Individual and 
               expenditures                       
                                                                    Corporate 
      Adoption credit and exclusion     1996 2004 Credit/Exclusion Individual 
               Child credit             1997 2004      Credit Individual      
    Credit for holders of zone academy  1997 2004      Credit Corporate       
                  bonds                           
      Deductibility of student loan     1997 2004    Deduction Individual     
                 interest                         
       State prepaid tuition plans      1997 2004    Exclusion Individual     
     Education individual retirement    1997 2004    Exclusion Individual     
                 accounts                         
        Welfare-to-work tax credit      1997 2004       Credit Individual and 
                                                                    Corporate 
             HOPE tax credit            1997 2004      Credit Individual      
       Lifetime Learning tax credit     1997 2004      Credit Individual      
      Assistance for adopted foster     2000 2004    Exclusion Individual     
                 children                         
      Deduction for higher education    2001 2004    Deduction Individual     
                 expenses                         
Employer-provided child care credit  2001 2004      Credit Corporate       
      Special deduction for teacher     2003 2004    Deduction Individual     
                 expenses                         
        Discharge of student loan       2004 2004    Exclusion Individual     
               indebtedness                       
                  Health                          
    Deductibility of medical expenses   1974 2004    Deduction Individual     
Exclusion of employer contributions  1974 2004    Exclusion Individual     
          for medical insurance                   
        premiums and medical care                 
       Deductibility of charitable      1977 2004    Deduction Individual and 
          contributions (health)                  
                                                                    Corporate 
    Exclusion of interest on hospital   1980 2004    Exclusion Individual and 
            construction bonds                    
                                                                    Corporate 
Tax credit for orphan drug research  1982 2004      Credit Corporate       
      Exclusion of employer share of    1988 1992    Exclusion Individual     
          hospital insurance tax                  
      Special Blue Cross/Blue Shield    1988 2004     Deduction Corporate     
                deduction                         
    Credit for child medical insurance  1990 1993      Credit Individual      
                 premiums                         

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function) Income security

     Medical Savings Accounts/Health Savings   1996 2004 Deduction Individual 
                    Accounts                                       
    Workers' compensation insurance premiums   1998 2002 Deduction Individual 
    Self-employed medical insurance premiums   1998 2004 Deduction Individual 
    Tax credit for health insurance purchased  2002 2004  Credit   Individual 
              by certain displaced                                 
             and retired individuals                               

     Excess of percentage standard deduction   1974 1977 Deduction Individual 
                  over minimum                                     
               standard deduction                                  
    Exclusion on capital gains on home sales   1974 1979 Exclusion Individual 
               for persons age 65                                  
                    and over                                       
           Exclusion of disability pay         1974 1984 Exclusion Individual 
        Additional exemption for elderly       1974 1987 Exemption Individual 
       Additional exemption for the blind      1974 1987 Exemption Individual 
Exclusion of untaxed unemployment insurance 1974 1987 Exclusion Individual 
                    benefits                                       
        Deductibility of casualty losses       1974 2004 Deduction Individual 
    Exclusion of military disability pensions  1974 2004 Exclusion Individual 
      Exclusion of other employee benefits:    1974 2004 Exclusion Individual 
              premiums on accident                                 
            and disability insurance                               
      Exclusion of other employee benefits:    1974 2004 Exclusion Individual 
             premiums on group term                                
                 life insurance                                    
    Income of trusts to finance supplementary  1974 2004 Exclusion Individual 
                  unemployment                                     
                    benefits                                       
     Exclusion of public assistance benefits   1974 2004 Exclusion Individual 
     Exclusion of railroad retirement system   1974 2004 Exclusion Individual 
                    benefits                                       
Exclusion of worker's compensation benefits 1974 2004 Exclusion Individual 
Net exclusion of pension contributions and  1974 2004 Exclusion Individual 
               earnings: employer                                  
                      plans                                        
Net exclusion of pension contributions and  1974 2004 Exclusion Individual 
              earnings: individual                                 
               retirement accounts                                 
     Tax credit for the elderly and disabled   1974 2004  Credit   Individual 
            Earned income tax credit           1975 2004  Credit   Individual 
Exclusion of special benefits for disabled  1975 2004 Exclusion Individual 
                   coal miners                                     
Net exclusion of pension contributions and  1983 2004 Exclusion Individual 
                 earnings: Keogh                                   
                      plans                                        
       Additional deduction for the blind      1986 2004 Deduction Individual 
      Additional deduction for the elderly     1986 2004 Deduction Individual 
    Extending tax-exempt organizations status  1988 1990           Individual 
                  to voluntary                           Exemption 
employee beneficiary and other associations                     
      Exclusion of employer provided death     1988 1997 Exclusion Individual 
                    benefits                                       

Appendix IV Compilation of Tax Expenditures Reported by Treasury (1974 to 2004)

                         (Continued From Previous Page)

                                              First Year Reported by Treasury

Last Year Reported by

Treasury Type Taxpayer Group Name of Tax Expenditure (Organized by Budget
Function) Social Security

             Special ESOP rules            1988 2004 Exemption Individual and 
                                                                 Corporate    
    Small business retirement plan credit  2001 2004  Credit   Individual and 
                                                                 Corporate    
Net exclusion of pension contributions  2001 2004 Exclusion   Individual   
            and earnings: 401(k)                               
                    plans                                      
Net exclusion of pension contributions  2001 2004 Exclusion   Individual   
            and earnings: low and                              
        moderate income savers credit                          

     Exclusion of Social Security benefits:    1974 2004 Exclusion Individual 
                 dependents and                                    
                    survivors                                      
     Exclusion of Social Security benefits:    1974 2004 Exclusion Individual 
                    disabled                                       
     Exclusion of Social Security benefits:    1974 2004 Exclusion Individual 
                 retired workers                                   

       Veterans' benefits and services General Purpose Fiscal Assistance

        Exclusion of GI bill benefits      1974 2004 Exclusion   Individual   
    Exclusion of veterans' death benefits  1974 2004 Exclusion   Individual   
               and disability                                  
                compensation                                   
       Exclusion of veterans' pensions     1974 2004 Exclusion   Individual   
     Exclusion of interest on veterans'    1983 2004 Exclusion Individual and 
                housing bonds                                  
                                                                 Corporate    

       Credit for corporations in U.S.     1974 1976  Credit     Corporate    
                 possessions                                   
Deductibility of nonbusiness state and  1974 2004 Deduction   Individual   
           local taxes (other than                             
    on owner-occupied homes and gasoline)                      
Exclusion of interest on public purpose 1974 2004 Exclusion Individual and 
            state and local bonds                              
                                                                 Corporate    

      Tax credit for corporations receiving    1977 2004  Credit    Corporate 
                income from doing                                  
          business in U.S. possessions                             
                    Interest                                       
Deferral of interest on U.S. savings bonds  1975 2004 Deferral  Individual 
               General government                                  
      Credits and deductions for political     1974 1987           Individual 
                  contributions                           Credit/  
                                                         Deduction 

Source: GAO analysis of OMB budget reports on tax expenditures, fiscal
years 1976-2006.

Note: Names of tax expenditures may have changed over time. Our list
includes the most recent name reported by Treasury for each tax
expenditure. The list of tax expenditures reflects all provisions reported
by Treasury, including those enacted but effective for future fiscal
years.

Appendix V

Glossary

Adjusted gross income (AGI): All income subject to taxation under the
individual income tax after subtracting above-the-line deductions, such as
certain contributions for individual retirement accounts and alimony
payments. Personal exemptions and the standard or itemized deductions are
subtracted from AGI to determine taxable income.

Alternative Minimum Tax (AMT): A separate tax system that applies to both
individual and corporate taxpayers. It parallels the regular individual
income tax system but with different rules for determining taxable income,
different tax rates for computing tax liability, and different rules for
allowing the use of tax credits.

Baseline: A benchmark for measuring the budgetary effects of proposed
changes in federal revenues or spending. Or, a benchmark for identifying
and measuring exceptions to the basic provisions of the tax structure.

o 	CBO baseline: CBO's estimate of spending, revenue, the deficit or
surplus, and debt held by the public during a fiscal year under current
laws and current policy. For revenues and mandatory spending, CBO projects
the baseline under the assumption that present laws continue without
change. For discretionary spending subject to annual appropriations, CBO
is required to adjust the current year's discretionary budget authority to
reflect inflation, among other factors.

o 	Comprehensive income tax baseline: This baseline, also called
Haig-Simons income, is the real, inflation adjusted, accretion to wealth
arising between the beginning and ending of the year. It includes all
accretions to wealth, whether or not realized, whether or not related to a
market transaction, and whether a return to capital or labor. Inflation
adjusted capital gains would be included in comprehensive income as they
accrue.

o 	Consumption tax baseline: A broad-based consumption tax is a
combination of an income tax plus a deduction for net saving. Many current
tax expenditures related to preferential taxation of capital income and
savings would not be considered tax preferences under a consumption tax
(e.g., capital gains), but preferences unrelated to broad-based saving or
investment incentives would remain tax preferences under a consumption
baseline.

o 	Normal income tax baseline: The Budget Act did not specify the baseline
income tax against which tax preference provisions should be

Appendix V Glossary

measured, and deciding whether provisions are exceptions from the normal
baseline is a matter of judgment. The normal income tax baseline is meant
to represent a practical and broad-based income tax that reflects the
general and widely applicable provisions of the current federal income
tax. For the individual income tax, the Joint Committee on Taxation's
(JCT) normal tax baseline includes one personal exemption for each
taxpayer, one for each dependent, the standard deduction, the existing tax
rate schedule, and deductions for investment and employee business
expenses. Itemized deductions that are not necessary for the generation of
income but exceed the standard deduction level are classified as tax
expenditures. Very similar in scope to JCT's normal income tax baseline,
Treasury's baseline is patterned on, but allows several major departures
from, a comprehensive income tax, where income is defined as the sum of
consumption and the change in net wealth during a given period.

o 	Reference tax law baseline: The reference baseline is closer to
existing tax law and is also patterned on, but still allows several major
departures from, a comprehensive income tax. Thus fewer tax provisions are
considered tax preferences under the reference tax baseline than under the
normal tax baseline. These include the lower tax rate for certain
corporations, preferential rates on capital gains, accelerated
depreciation, deferral of tax on income from controlled foreign
corporations, etc.

Budget function: One of 20 broad categories into which budgetary resources
are grouped so that all budget authority and outlays can be presented
according to the national interests being addressed. There are 17 broad
budget functions, including national defense, international affairs,
energy, agriculture, health, income security, and general government.
Three other functions-net interest, allowances, and undistributed
offsetting receipts-are included to complete the budget.

De minimis rule: The level of revenue loss below which a revenue loss
estimate is not reported for a tax preference.

Direct loans: A disbursement of funds by the government to a nonfederal
borrower under a contract that requires the repayment of such funds with
or without interest.

Discretionary spending: Outlays controlled by appropriation acts, other
than those that fund mandatory programs.

Appendix V Glossary

Entitlement authority: Authority to make payments (including loans and
grants) for which budget authority is provided in advance by
appropriations acts to any person or government if, under the provisions
of the law containing such authority, the U.S. government is obligated to
make the payments to persons or governments who meet the requirements
established by law.

Government Performance and Results Act (GPRA): Enacted in 1993, GPRA, also
known as the Results Act, intends to improve the efficiency and
effectiveness of federal programs by requiring federal agencies to develop
strategic plans, annual performance plans, and annual program performance
reports.

Grants: A federal financial assistance award making payment in cash or in
kind for a specified purpose. The federal government is not expected to
have substantial involvement with the state or local government or other
recipient while the contemplated activity is being performed.

Gross domestic product (GDP): The value of all final goods and services
produced within the borders of a country such as the United States during
a given period. The components of GDP are consumption expenditures (both
personal and government), gross investment (both private and government),
and net exports.

Mandatory spending: Also known as direct spending. Mandatory spending
includes outlays for entitlement authority (for example, the food stamp,
Medicare, and veterans' pension programs), payment of interest on the
public debt, and nonentitlements such as payments to the states from
Forest Service receipts. By defining eligibility and setting the benefit
or payment rules, the Congress controls spending for these programs
indirectly rather than directly through appropriations acts.

Tax expenditure: A revenue loss attributable to a provision of the federal
tax laws that grants special tax relief designed to encourage certain
kinds of behavior by taxpayers or to aid taxpayers in special
circumstances. The Congressional Budget and Impoundment Control Act of
19741 lists six types of tax expenditures: exclusions, exemptions,
deductions, credits, preferential tax rates, and deferral of tax
liability.

1Pub. L. No. 93-344, July 12, 1974.

Appendix V Glossary

o 	Preferential tax rates: A reduction of the tax rate on some forms of
income, such as capital gains.

o 	Tax credit: An amount that offsets or reduces tax liability. When the
allowable tax credit amount exceeds the tax liability, and the difference
is paid to the taxpayer, the credit is considered refundable. Otherwise,
the difference can be (1) allowed as a carryforward against future tax
liability, (2) allowed as a carryback against past taxes paid, or (3) lost
as a tax benefit.

o 	Tax deduction: An amount that is subtracted from the tax base before
tax liability is calculated. Deductions claimed before and after the
adjusted gross income line on the Form 1040 are sometimes called
"above-the-line" and "below-the-line" deductions, respectively.

o 	Tax deferral: A provision allowing taxpayers to reduce current tax
liability by delaying recognition of some income or accelerating some
deductions otherwise attributable to future years. This can increase the
taxpayer's future tax liability, as the deferred income is eventually
recognized, or reduce the deductions available on future income.

o 	Tax exclusion: An item of income that would otherwise constitute a part
of the taxpayer's gross income, but is excluded under a specific provision
of the tax code. Exclusions generally do not appear on the U.S. Individual
Income Tax Return (Form 1040), and excluded income is not reflected in
total reported income.

o 	Tax exemption: A reduction in taxable income offered to taxpayers
because of their status or circumstances.

Tax expenditure revenue loss estimate: The measure of the revenue cost of
each tax expenditure. The revenue cost is the difference between tax
liability under current law and the tax liability that would result if
taxes were recomputed without that tax expenditure. Revenue cost estimates
assume (1) economic behavior does not change, and (2) all other tax
expenditures remain in the code unchanged.

Tax expenditure outlay-equivalent estimate: The amount of budget outlays
that would be required to provide the taxpayer the same after-tax income
as would be received through the tax provision. The outlayequivalent
measure allows the cost of a tax preference to be compared with

Appendix V Glossary

a direct federal outlay, were each to provide the same benefit to the
taxpayer.

Unified budget: A comprehensive budget in which receipts and outlays from
federal funds and trust funds are consolidated; generally a cash or cash
equivalent measure in which receipts are recorded when received and
expenditures are recorded when paid, regardless of the accounting period
in which the receipts are earned or the costs incurred.

Appendix VI

                        GAO Contact and Acknowledgments

GAO Contact Michael Brostek, (202) 512-9110

Acknowledgments	In addition to the individual named above, MaryLynn
Sergent, Assistant Director, as well as Eric Gorman, Edward Nannenhorn,
Anne Stevens, and Lynn Wasielewski made key contributions to this report.
Other individuals also contributing to this report included Ellen Grady,
Susan Irving, Shirley Jones, Donna Miller, Amy Rosewarne, and William
Trancucci.

Selected Bibliography

Attanasio, Orazio P. and Thomas DeLeire. "The Effect of Individual
Retirement Accounts on Household Consumption and National Saving," The
Economic Journal, Vol. 112 (July 2002): 504-538.

Brixi, Hana Polackova, Christian M.A. Valenduc, and Zhicheng Li Swift. Tax
Expenditures-Shedding Light on Government Spending through the Tax System:
Lessons from Developed and Transition Economies. Washington, D.C.: The
World Bank, October 2003.

Burman, Leonard E. "Is the Tax Expenditure Concept Still Relevant?" The
National Tax Journal. vol. LVI, No. 3, (September 2003).

Burman, Leonard E. and Jonathan Gruber. "Tax Credits for Health
Insurance," Tax Policy Center Discussion Paper No. 19. Washington, D.C.:
The Tax Policy Center, June 2005.

Carasso, Adam and Eugene Steuerle. "Tax Expenditures: Revenue Loss Versus
Outlay Equivalents." Tax Notes, Vol. 7, (October 13, 2003).

Datta, Lois-ellin and Patrick G. Gasso. Evaluating Tax Expenditures: Tools
and Techniques for Assessing Outcomes. San Francisco, Calif.: American
Evaluation Association, 1998.

National Taxpayer Advocate. 2004 Annual Report to Congress. Washington,
D.C.: December 31, 2004.

President's Advisory Panel on Federal Tax Reform. "Understanding Tax
Bases: Staff Presentation," (presentation before the Panel's public
meeting, Washington, D.C., July 20, 2005,
http://taxreformpanel.gov/meetings/docs/understanding_tax_bases.ppt
(downloaded September 13, 2005).

Rivlin, Alice M. and Isabel Sawhill. Restoring Fiscal Sanity 2005: Meeting
the Long-Run Challenge. Washington, D.C.: The Brookings Institution, 2005.

Salamon, Lester M. The Tools of Government: A Guide to the New Governance.
New York, N.Y.: Oxford University Press, Inc., 2002.

Sheils, John and Randall Haught. "The Cost of Tax-Exempt Health Benefits
in 2004." Health Affairs, (February 25, 2004).

Selected Bibliography

Slemrod, Joel and Jon Bakija. Taxing Ourselves: A Citizen's Guide to the
Debate Over Taxes, 3rd Edition. Cambridge, Mass.: The MIT Press, 2004.

Toder, Eric J. "Tax Cuts or Spending-Does It Make a Difference?" National
Tax Journal, Vol. LIII, No. 3, Part 1. (September 2000).

Toder, Eric J. The Changing Composition of Tax Incentives: 1980-99.
Washington, D.C.: (The Urban Institute), March 1, 1999.

U.S. Congressional Budget Office. Budget Options. Washington, D.C.:
February 2005.

U.S. Congressional Budget Office. The Budget and Economic Outlook: An
Update. Washington, D.C.: August 2005.

U.S. Congressional Budget Office. The Long-Term Budget Outlook.
Washington, D.C.: December 2003.

U.S. Congress. Joint Committee on Taxation. Estimates of Federal Tax
Expenditures For Fiscal Years 2005-2009. JCS-1-05. Washington, D.C.:
January 12, 2005.

U.S. Congress. Joint Committee on Taxation. Options to Improve Tax
Compliance and Reform Tax Expenditures. JCS-2-05. Washington, D.C.:
January 27, 2005.

U.S. Congress. Joint Economic Committee. Tax Expenditures: A Review and
Analysis. Washington, D.C.: August 1999.

U.S. Congress. Senate Committee on the Budget. Tax Expenditures:
Compendium of Background Material on Individual Provisions. S. Prt.
108-54. Washington, D.C.: December 2004.

U.S. Congress. Senate Committee on Governmental Affairs. Government
Performance and Results Act of 1993 (P.L. 103-58). Washington, D.C.: June
15, 1993.

U.S. Office of Management and Budget. "Tax Expenditures," Analytical
Perspectives, Budget of the United States Government, Fiscal Year 2006

Washington, D.C.: 2005.

Related GAO Products

  Tax Policy and IRS Administration

Understanding the Tax Reform Debate: Background, Criteria, and Questions.
GAO-05-1009SP. Washington, D.C.: September 13, 2005.

Internal Revenue Service: Status of Recommendations from Financial Audits
and Related Financial Management Reports. GAO-05-393. Washington, D.C.:
April 29, 2005.

Financial Audit: IRS's Fiscal Years 2004 and 2003 Financial Statements.
GAO-05-103. Washington, D.C.: November 10, 2004.

Tax Administration: IRS Should Reassess the Level of Resources for Testing
Forms and Instructions. GAO-03-486. Washington, D.C.: April 11, 2003.

Tax Administration: IRS Should Continue to Expand Reporting on Its
Enforcement Efforts. GAO-03-378. Washington, D.C.: January 31, 2003.

Tax Administration: Impact of Compliance and Collection Program Declines
on Taxpayers. GAO-02-674. Washington, D.C.: May 22, 2002.

Tax Deductions: Further Estimates of Taxpayers Who May Have Overpaid
Federal Taxes by Not Itemizing. GAO-02-509. Washington, D.C.: March 29,
2002.

Tax Policy: Tax Expenditures Deserve More Scrutiny. GAO/GGD/AIMD94-122.
Washington, D.C.: June 3, 1994.

Specific Tax Expenditures	Climate Change: Federal Reports on Climate
Change Funding Should Be Clearer and More Complete. GAO-05-461.
Washington, D.C.: August 25, 2005.

Student Aid and Postsecondary Tax Preferences: Limited Research Exists on
Effectiveness of Tools to Assist Students and Families Through Title IV
Student Aid and Tax Preferences. GAO-05-684. Washington, D.C.: July 29,
2005.

Earned Income Tax Credit: Implementation of Three New Tests Proceeded
Smoothly, But Tests and Evaluation Plans Were Not Fully Documented.

GAO-05-92. Washington, D.C.: December 30, 2004.

                              Related GAO Products

Community Development: Federal Revitalization Programs are Being
Implemented, but Data on the Use of Tax Benefits Are Limited. GAO-04306.
Washington, D.C.: March 5, 2004.

Tax Administration: IRS Issued Advance Child Tax Credit Payments on Time
but Should Study Lessons Learned. GAO-04-372. Washington, D.C.: February
17, 2004.

New Markets Tax Credit Program: Progress Made in Implementation, but
Further Actions Needed to Monitor Compliance. GAO-04-326. Washington,
D.C.: January 30, 2004.

Tax Administration: Information Is Not Available to Determine Whether $5
Billion in Liberty Zone Tax Benefits Will Be Realized. GAO-03-1102.
Washington, D.C.: Sept. 30, 2003.

Business Tax Incentives: Incentives to Employ Workers with Disabilities
Receive Limited Use and Have an Uncertain Impact. GAO-03-39. Washington,
D.C.: December 11, 2002.

New Markets Tax Credit: Status of Implementation and Issues Related to
GAO's Mandated Reports. GAO-03-223R. Washington, D.C.: December 6, 2002.

Public Housing: HOPE VI Leveraging Has Increased, but HUD Has Not Met
Annual Reporting Requirement. GAO-03-91. Washington, D.C.: November 15,
2002.

Student Aid and Tax Benefits: Better Research and Guidance Will Facilitate
Comparison of Effectiveness and Student Use. GAO-02-751. Washington, D.C.:
Sept. 13, 2002.

Tax Policy and Administration: Review of Studies of the Effectiveness of
the Research Tax Credit. GAO/GGD-96-43. Washington, D.C.: May 21, 1996.

Federal Budget	Our Nation's Fiscal Outlook: The Federal Government's
Long-Term Budget Imbalance. http://www.gao.gov/special.pubs/longterm/.

21st Century Challenges: Performance Budgeting Could Help Promote
Necessary Reexamination. GAO-05-709T. Washington, D.C.: June 14, 2005.

Related GAO Products

Management Reform: Assessing the President's Management Agenda.

GAO-05-574T. Washington, D.C.: April 21, 2005.

Long-Term Fiscal Issues: Increasing Transparency and Reexamining the Base
of the Federal Budget. GAO-05-317T. Washington, D.C.: February 8, 2005.

21st Century Challenges: Reexamining the Base of the Federal Government.
GAO-05-325SP. Washington, D.C.: February 2005.

Performance Budgeting: Efforts to Restructure Budgets to Better Align
Resources with Performance. GAO-05-117SP. Washington, D.C.: February 2005.

Opportunities for Congressional Oversight and Improved Use of Taxpayer
Funds: Budgetary Implications of Selected GAO Work. GAO-04649. Washington,
D.C.: May 7, 2004.

Budget Process: Long-term Focus Is Critical. GAO-04-585T. Washington,
D.C.: March 23, 2004.

Results-Oriented Government: GPRA Has Established a Solid Foundation for
Achieving Greater Results. GAO-04-38. Washington, D.C.: March 10, 2004.

Performance Budgeting: Observations on the Use of OMB's Program Assessment
Rating Tool for the Fiscal Year 2004 Budget. GAO-04-174. Washington, D.C.:
January 30, 2004.

Fiscal Exposures: Improving the Budgetary Focus on Long-Term Costs and
Uncertainties. GAO-03-213. Washington, D.C.: January 24, 2003.

Federal Budget: Opportunities for Oversight and Improved Use of Taxpayer
Funds. GAO-03-1030T. Washington, D.C.: July 17, 2003.

Performance Budgeting: Current Developments and Future Prospects.
GAO-03-595T. Washington, D.C.: April 1, 2003.

National Saving: Answers to Key Questions. GAO-01-591SP. Washington, D.C.:
June 2001.

GAO's Mission	The Government Accountability Office, the audit, evaluation
and investigative arm of Congress, exists to support Congress in meeting
its constitutional responsibilities and to help improve the performance
and accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

Obtaining Copies of The fastest and easiest way to obtain copies of GAO
documents at no cost

is through GAO's Web site (www.gao.gov). Each weekday, GAO postsGAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To Testimony have GAO e-mail you a list of newly posted products
every afternoon, go to

www.gao.gov and select "Subscribe to Updates."

Order by Mail or Phone	The first copy of each printed report is free.
Additional copies are $2 each. A check or money order should be made out
to the Superintendent of Documents. GAO also accepts VISA and Mastercard.
Orders for 100 or more copies mailed to a single address are discounted 25
percent. Orders should be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone:	Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

  To Report Fraud, Contact:
  Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: [email protected] Programs Automated answering system: (800)
424-5454 or (202) 512-7470

Congressional	Gloria Jarmon, Managing Director, [email protected] (202)
512-4400 U.S. Government Accountability Office, 441 G Street NW, Room 7125

Relations Washington, D.C. 20548

Public Affairs	Paul Anderson, Managing Director, [email protected] (202)
512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
*** End of document. ***