Negative after-tax income for lowest earners create misleading %change charts #1806
Comments
|
Perhaps we should ignore losses when calculating base income? Does base
income include the value of medicaid and other benefits?
dan
…On Tue, 2 Jan 2018, Max Ghenis wrote:
The bottom decile has negative aggregate after-tax income, according to both CPS and
PUF (per TaxBrain UI e.g. this; I don't have the data). This means that any reform
that changes after-tax income for this group will be reported as the opposite sign as
intended.
For example, this notebook considers a simple $100 UBI reform. Cells 10 and 11 show
that the bottom decile's after-tax income is -$5.3B under the baseline and -$2.7B
under the reform, respectively. Change to after-tax income is then technically +$2.6B
/ -$5.3B = -49%, as seen in the decile_graph plot in cell 14.
I'm not sure how best to address this, but it is pretty misleading. Most reforms don't
affect this group as significantly as UBI does, so it may have flown under the radar
until now. One solution could be to use a separate column which zeros out negative
after-tax incomes. I'm not sure how others handle it; TPC seems to mostly use
quintiles instead of deciles, which may help them avoid the issue.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the
thread.[AHvQVUfffdHcV7Q7E00EfA7DCVWkHbbZks5tGta1gaJpZM4RRQtz.gif]
|
|
@feenberg said:
Seems like a pretty ad hoc approach. @feenberg asked:
No. But capital losses and business losses that cause the bottom decile to have negative expanded income are so large that the bottom decile would probably still have negative expanded income even if benefits were added to expanded income. |
|
On Wed, 3 Jan 2018, Martin Holmer wrote:
@feenberg said:
Perhaps we should ignore losses when calculating base income?
Seems like a pretty ad hoc approach.
Do any other tax analysis models define expanded income in that way?
I don't know.
@feenberg asked:
Does [expanded] income include the value of medicaid and other benefits?
No. But capital losses and business losses that cause the bottom decile to have
negative expanded income are so large that the bottom decile would probably still have
negative expanded income even if benefits were added to expanded income.
Sometimes negative incomes are excluded from the lowest income category,
but included in the the total. I recall seeing that in several places.
Notice that if you tab by AGI capital losses are capped at $3000 and
presumably that solves the problem.
These losses don't come from the CPS, do they? We are imputing them from
the SOI file. I wonder if the imputation is going wrong and imputing an
unreasonable share of losses to poor taxpayers.
dan
… —
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the
thread.[AHvQVWo-0VYizXo7_oETywa8t85on67Zks5tG6m7gaJpZM4RRQtz.gif]
|
|
@MaxGhenis and @feenberg, |
|
Merge of pull request #1880 resolves issue #1806. |
The bottom decile has negative aggregate after-tax income, according to both CPS and PUF (per TaxBrain UI e.g. this; I don't have the data). This means that any reform that changes after-tax income for this group will be reported as the opposite sign as intended.
For example, this notebook considers a simple $100 UBI reform. Cells 10 and 11 show that the bottom decile's after-tax income is -$5.3B under the baseline and -$2.7B under the reform, respectively. Change to after-tax income is then technically +$2.6B / -$5.3B = -49%, as seen in the
decile_graphplot in cell 14.I'm not sure how best to address this, but it is pretty misleading. Most reforms don't affect this group as significantly as UBI does, so it may have flown under the radar until now. One solution could be to use a separate column which zeros out negative after-tax incomes. I'm not sure how others handle it; TPC seems to mostly use quintiles instead of deciles, which may help them avoid the issue.
The text was updated successfully, but these errors were encountered: