[Neal Soss' much-discussed chart of the United States' 20
largest "tax expenditures" tells us nothing about the value
in terms of costs and benefits to various constituencies of
each expenditure. Work place issues expert Ellen Dannin
provides an example of how we can look at employer-provided
A chart created by Credit Suisse's Neal Soss, showing the 20
largest tax expenditures for the United States, is making
the rounds. The term "tax expenditure" means any money spent
by the government through the tax code, such as tax credits,
deductions and exemptions.
The chart itself simply lists the 20 largest tax
expenditures and the cost of each to the budget. How the
list is understood is another matter.
Seven of the 20 tax expenditures on the list are primarily
directed toward investors. They include preferential
treatment of capital gains, exclusion of net imputed rental
income, deferral of income from controlled foreign
corporations, preferential treatment of qualified dividends,
step-up basis of capital gains at death, deduction of US
production activities and various energy and natural
resources credits/deductions. Together, they cost our
treasury $235 billion a year. The rest of the list is made
up of deductions that individual taxpayers or their
employers could take or benefits that go directly to people.
Low-income citizens generally do not take deductions when
they file their federal, state and local taxes, but they do
pay the payroll taxes that fund Social Security.
Many people will see the chart as a list of "loopholes" or
"entitlements" that must be eliminated if we are to put our
fiscal house in order. But, as one blogger says of the
chart, "Good luck to the politician who votes to eliminates
any of these." People would fight tooth and nail if they
There is an interesting human story lurking within tax/don't
tax and spend/don't spend choices. Our national discussion
of taxes sounds as if all tax expenditures are identical and
are a loss to the treasury and to the public. But, if we
take the time to follow the money, we will see that not all
money collected and spent by the government has the same
characteristics and effects.
Imagine a different chart that included the same tax
deductions, subsidies and benefits but showed the other part
of the equation - who all benefits from a particular
deduction, and how.
Take the most expensive single deduction on the chart - $171
billion for employer-sponsored health insurance - as an
example. It is possible to track the effects of tax support
for employers to provide health insurance and the incentives
the expenditure of tax money creates.
We know that many - but not all - job seekers prefer working
for an employer that offers health insurance.
We can also assess how subsidizing employers who provide
health insurance affects people's health and the nation's
well-being and productivity. For example, to the extent good
health care gives us a more productive workforce and
provides other personal and societal benefits, the benefits
may so exceed the costs as to make the loss of tax revenue a
On the other hand, we could also look into how much of that
government health insurance subsidy benefits investors in
private health insurance companies and does not lead to
wider social benefits.
We could also compare health outcomes and costs for those
with employer-provided health insurance compared to single-
payer systems or people who buy their own health insurance.
For each of these categories, we could ask how they differ
in terms of costs to achieve those outcomes. If there are
marked differences in costs and benefits, we might need to
adjust the size or even the existence of the tax
While we are asking these questions, we also ought to ask
why some people have no health coverage at all. Some might
have decided not to buy health insurance because they are
healthy and think they won't get sick. Others may not be
able to afford health insurance, but may have too much money
to qualify for Medicaid.
If these uninsured people become ill or injured, we, as a
society, may have a moral choice. Do we let them live with
the consequences of their choices or their situation in
life, or do we decide we are our brothers' keepers? If we
decide not to provide assistance, do we deny health care to
children whose parents have no coverage, either because they
are too poor or because they have made bad choices?
We might also want to consider whether and how those of us
who do have health insurance are affected by uninsured
people who survive illnesses or injuries but live on in a
compromised condition. It may be impossible to assess the
costs of lost human productivity and contributions to our
society caused by failure to provide adequate health care.
Many more questions might be asked about these and other
outcomes and inputs into our health system and the role that
providing a tax incentive to employers to provide health
insurance plays in promoting the well-being of the United
States as a whole.
The point is that merely looking at the name of a tax
subsidy or other tax expenditure tells us nothing about
whether we have made a wise decision to spend the money to
provide that benefit. Only by doing the hard work of digging
into the effects of an expenditure can we make a wise
An honest investigation into these and other tax
"expenditures" may lead to unexpected outcomes.
[Ellen Dannin is Fannie Weiss distinguished faculty scholar
and professor of law at Penn State Dickinson School of Law
and author of "Taking Back the Workers' Law - How to Fight
the Assault on Labor Rights."]
[Thanks to the author for sharing her analysis with