Flat tax post - Githubissues

UBICenter / blog

UBI Center blog, hosted via Jupyterbook.

http://blog.ubicenter.org

MIT License

2 stars 7 forks source link

Flat tax post #47

Closed MaxGhenis closed 3 years ago

MaxGhenis commented 3 years ago

By @ngpsu22

Three remaining issues:

As I was cleaning up the code, I introduced an error and the numbers differ from Nate's original (which is from the first commit)
Poverty gap needs to be calculated at the SPM unit level
Add economic context around flat taxes (marriage, intertemporal distortion, efficiency)

MaxGhenis commented 3 years ago

I think this looks right now, but the winners chart differs from the original and I'm not sure why. @ngpsu22 ideas?

Also, the other graphs are slightly different because I calculated budget-neutral amounts by SPM unit rather than person.

Original

Current

MaxGhenis commented 3 years ago

The previously described issue was identified as assigning new (flat) tax liabilities to persons based on their personal AGI rather than the SPM unit's AGI, i.e. these lines:

target_persons['new_tax'] = target_persons.agi * flat_tax
# ...
target_persons['new_spm_resources'] = (target_persons.spm_resources + 
                                         target_persons.hh_current_tax +
                                         target_persons.hh_fica - 
                                         target_persons.new_tax +  # HERE
                                         target_persons.hh_ubi)

This explains the new results looking less favorable to the UBI scenarios (tax was underestimated).

@ngpsu22 wdyt of using tax after credit rather than before? This could be simpler to explain (I'm also still not entirely clear if EITC and ACTC are the only differences; IPUMS documentation on FEDTAX and FEDTAXAC and forum responses don't leave me certain).

MaxGhenis commented 3 years ago

Actually I think it has to be after credits, since credits aren't only EITC and ACTC, they're also nonrefundable credits which can't be separated from the rest of the tax code. This means slightly higher rates are required to achieve poverty- and inequality-neutrality. Updated accordingly in the latest commit.

Some to-dos before publishing:

Add UBI Center logo to charts
Add studies from OECD around US tax/transfer progressivity vs other countries, as I cited in https://twitter.com/MaxGhenis/status/1261393049050968066 and tweets in comments from that thread
[Nice-to-have] Change hover labels to say "UBI" or "Head tax" depending on whether the value is positive or negative, rather than "UBI (head tax)" in each

MaxGhenis commented 3 years ago

Couple other things to mention:

Economists favor eliminating deductions https://www.igmchicago.org/surveys/tax-reform/
Cool charts on marriage bonus/penalty in https://taxfoundation.org/understanding-marriage-penalty-and-marriage-bonus/#_ftn3

MaxGhenis commented 3 years ago

@ngpsu22 think this is just about ready, only thing left is making the hover labels look nice on the new charts and adding some detail on the OECD data. LMK if you have other feedback before I post it.

https://nbviewer.jupyter.org/github/UBICenter/blog/blob/909c8543a052ffe02b704d64c83b58990d9bf63b/jb/20201223/us-flat-tax.ipynb