159
Chapter 7
Tax Expenditures
The Silent Partner
GPRA’s focus on outcome measurement and performance budgeting seeks to
match up federal expenditures with the results they produce for citizens. But
sometimes the federal government sacrices revenue to provide tax breaks for
citizens who spend money in ways that accomplish public purposes. e lost
revenue is like an expenditure that should be justied based on the outcomes
it produces.
Accountability for the outcomes of such tax expenditures, however, is virtually
nonexistent on the federal level in the United States. Each year, the congressio-
nal Joint Tax Committee and an appendix to the president’s budget calculate the
revenues forgone from tax expenditures, but tax expenditures are not evaluated
like spending programs or compared with spending programs that aim to achieve
similar results. us, even a good performance budget that matches direct fed-
eral expenditures with outcomes will underestimate both the costs of government
and the outcomes it achieves. Tax expenditures usually fly under the radar screen,
receiving even less scrutiny than spending.
As policymakers seek to cope with the U.S. government’s scal problems, tax
expenditures will likely appear on the table as a source of budgetary savings. But
at the current time, the federal government is poorly equipped to make informed
decisions about tax expenditures, because it knows little about their actual effec-
tiveness in accomplishing desired outcomes. Until tax expenditures are subjected to
the type of analysis envisioned by GPRA, decision makers who want to eliminate
or reform tax expenditures will be flying blind.
160 ◾  Government Performance and Results
Defining Tax Expenditures
A tax expenditure occurs when a taxpayer receives a deduction, credit, or other tax
reduction in exchange for spending money or otherwise using resources in ways
the government has decided to encourage. Familiar examples of tax expenditures
include deductions for home mortgage interest and charitable contributions, as well
as tax credits for purchasing energy-efficient appliances. Sometimes the govern-
ment employs tax expenditures to increase compensation of its own employees,
instead of simply paying them more money. For example, military housing, meals,
and some types of combat pay are exempt from income tax; so are certain parking
and mass transit expenses federal agencies pay for their employees.
What Counts as a Tax Expenditure?
A fuzzy line separates some tax expenditures from more general tax policy. It seems
clear, for example, that the deduction for charitable contributions is a tax expendi-
ture that encourages people to donate money to charities. It also seems clear that
a decision to raise or lower income tax rates is a decision about general tax policy,
not a tax expenditure. But some ostensible tax expenditures might alternatively be
viewed as general tax policy. For example, OMB has categorized the revenue loss
from the child tax credit as a tax expenditure under the category of education,
training, employment, and social services (OMB 1998, 92; 2007a, 288).
Realistically, though, the child credit might be more accurately viewed simply
as a redistributive measure intended to allow parents to shield more income from
tax, rather than a tax expenditure that encourages parents to have more children or
educate them more extensively.*
Reduced tax rates on capital gains from the sale of securities or other assets
provide another timely and controversial example. e budgets from the two most
recent administrations reflect diering points of view. e Bush administration
argued that differential tax rates for dividends and capital gains on corporate equity
do not constitute tax expenditures, because they make the income tax rate on equity
capital (33% corporate rate plus individual income tax rate) more nearly equal to
the tax rate on other forms of income. Starting in 2005, OMB stopped counting
the forgone revenues from these tax rates on corporate equity as tax expenditures
(OMB 2007a, 299). e Obama administrations first budget referred to the capi-
tal gains and dividend rates as “preferentially low rates” and provided an alterna-
tive estimate of the tax expenditure calculated under the pre-2005 method (OMB
2009b, 311, 318).
*
However, we note that one author’s spouse did give birth to a child roughly nine months after
both parents attended a seminar with congressional staff in 1997, where it became quite appar-
ent that Congress would enact a substantial child tax credit that year. We leave it to the reader
to assess whether correlation implies causation.

Get Government Performance and Results now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.