Twitter | Search | |
Wojtek Kopczuk
Economics conversation moves fast these days. David Splinter () has now posted a note () with his calculation of effective tax rates, building off his and Auten's alternative to PSZ inequality estimates
Reply Retweet Like More
Wojtek Kopczuk Oct 8
Replying to @wwwojtekk
This is a comparison of progressivity according to AS and PSZ
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @wwwojtekk
And here is 2018 that's labeled as "forecasted" because the details of how SZ imputed theirs are not documented (and we are still not even done with the 2018 filing season)
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @wwwojtekk
Where does this big difference in progressivity come from? It is primarily working off different inequality estimates. There is also a difference in how corporate tax is allocated (more standard 25% to wages in AS) but that works toward making PS more progressive.
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @wwwojtekk
In particular, refundable credits are not part of the difference. There is a legitimate controversy here about measurement of inequality (see AS paper). I don't take a strong side on all of it, beyond noting that some of these things are hard to know but turn out to matter a lot
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @gabriel_zucman
Finally, his paper doesn't have "top 400" estimates - I think by now I more or less understand how those were imputed, but that's based on Twitter conversation with and goes beyond what's been done previously in either SZ or AS.
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @wwwojtekk
And final comment - David Splinter worked on it for a couple of weeks, good foresight and relief that people don't write notes like this overnight
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @wwwojtekk
And I also should link back to the gif that I posted later:
Reply Retweet Like
Wojtek Kopczuk Oct 9
Replying to @wwwojtekk
Twitter also helps in better understanding of what is going on
Reply Retweet Like
David Gamage Oct 8
Replying to @wwwojtekk
Having read the Splinter note, I'm finding its explanations too sparse and ambiguous to follow. 1/
Reply Retweet Like
David Gamage Oct 8
Replying to @wwwojtekk
This seems to be key (2/):
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @davidsgamage
Yes, that's (shockingly to me when I first saw it) the biggest source of the difference between the papers. This particular one I find reasonably persuasive in terms of direction (though how much to move in that direction may be open for debate)
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @davidsgamage
One thing to keep in mind is that it's noncompliance from NIPA. Any offshoring is probably not part of that. So, you can't allocate this noncompliance to the top on account of offshoring - it would remove it from the bottom. You could go beyond NIPA (but none of them do that)
Reply Retweet Like
🍅 tomatoduck 🦆 Oct 8
Replying to @wwwojtekk
Can you explain what's happening here in easy to understand terms? for those of us not economists?
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @codesgood
The main thing is how much income different groups have. A lot is not on tax returns (retirement accounts, nondistributed earnings of corporations, tax evasion and some smaller things). No very good way to infer and people are battling over how it should be done.
Reply Retweet Like
Wojtek Kopczuk Oct 8
Replying to @codesgood
Some similar problems on tax side - who really pays corporate tax? Exactly how much property tax do the very rich pay? At the end of the day, it turns out that these things matter a lot and we end up with estimates that differ widely. I wouldn't present any of them as "the truth"
Reply Retweet Like
David Gamage Oct 8
I've now reread the Auten+Splinter paper along with the new Splinter note. If I am understanding correctly (not sure that I am, so someone please correct me if I am confused), AS do not account for unrealized gains except for retained corporate earnings. 1/ ()
Reply Retweet Like
Wojtek Kopczuk Oct 8
That's identical to Saez-Zucman. They both distribute national accounts.
Reply Retweet Like
Oliver Pardo Oct 8
Reply Retweet Like