Some Nice Curves

March 28th, 2012 at 8:05 am

I recently cited this important new paper from the CRS on tax expenditures.  In my earlier post, I emphasized the “reality-check” findings of the piece regarding how much you can realistically pay for lower tax rates by closing tax expenditures (answer: less than you think, and a ton less than Rep Ryan claims in the new House budget).

Here, I’d like to briefly highlight some nice analysis in the paper regarding the distributional effects of various tax expenditures, i.e., who benefits from different tax breaks?

In general, as shown in the figure here (see second figure), tax expenditures flow upstream; 66% of their benefits go the top 20%; 24% of their benefits accrue to just the top 1%.

But the picture below shows some very different outcomes within that aggregate result.  The figure plots the share of tax benefits on the Y-axis and the share of incomes of the households who get them on the X-axis.  If a particular tax expenditure were proportionally distributed such that the poorest 10% of households got 10% of the tax benefits, the bottom 50% got 50%, etc., the plot of that benefit would be coincident with the diagonal line.

Thus, measures above the line are progressive—more of their benefits go to lower income families—and vise-versa.

The EITC—a wage subsidy for low-income, working families—and the special break for realized capital gains provide good examples of the opposite ends of the distributional continuum.  As the figure shows, the benefits of the EITC are exhausted before you hit the 20th percentile of the income scale (the sample here is all taxpayers).  Conversely, the benefits of the lower rate on cap gains don’t start rising until you get above the 60th percentile of taxpayers, and about half of the benefits go to the top fifth.

Check out the doc (appendix  B) for more of these types of graphs, but the table below provides a summary measure—the Suits index—of a bunch of tax expenditures.   The numbers range from -1 to +1, with values close to 1 indicative of a highly progressive expenditure (the EITC scores 0.937) and those close to -1 concentrated at the top.


Print Friendly, PDF & Email

6 comments in reply to "Some Nice Curves"

  1. pjr says:

    The CRC report includes the following footnote: Tax expenditures are defined under the Congressional Budget and Impoundment Control Act of 1974 (P.L. 93-344) as “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.” However, the definition is not applied to businesses, which exclude/deduct most of their gross income, including questionable expenses. Working people know and see this but economists who address tax expenditures simply ignore it.

  2. Tyler says:

    Dr. Bernstein,

    I thought I should pass this article along to you:–but-no-one-wants-them.html

    Perhaps unemployment benefits are just too generous.


  3. Misaki says:

    I think many people tend to forget that while there are differences in opinion on who “deserves” to benefit from increases in the government’s budget deficit, the CBO and Moody’s have already calculated what the actual effect of taxes and so on is on unemployment and GDP.

    So it just comes down to the question of whether government spending to reduce unemployment is something that the voting population of the US wants. (And the unemployed don’t vote very much.) People sometimes assume that the economic profession does not have an answer to whether tax breaks are more ‘effective’ than aid to the poor and so try to argue for or against a certain action based on its supposed effectiveness, when in fact that analysis is not what prevents unemployment from being fixed.

  4. Ezra says:

    Important to note: The nice EITC curves you show are for reduced tax liabilities, not total expenditures. But the vast majority of EITC costs are for the refundable portion of the credit. As this recent Eissa and Hoynes paper shows, more than a third of EITC expenditures went to the 4th, 5th, and 6th income deciles (see pg 700):

    Yes, it’s still a very progressive credit, but EITC is certainly not “exhausted before you hit the 20th percentile.”