3 ms·
Yeah, I was thinking that despite the fact that the ultra wealthy use TFA's loophole, people who don't (i.e. net worth < $300M as the author explains) have a si
by hunter2_ 2y ago
Yeah, I was thinking that despite the fact that the ultra wealthy use TFA's loophole, people who don't (i.e. net worth < $300M as the author explains) have a situation where:
A - In a universe with cost basis step-up on death, they die with gains taxed at 0% and then pay 40% estate tax on everything.
B - In a world without cost basis step-up on death, they die with gains taxed at the 20% long term rate and then pay 40% estate tax on what remains.
Thus:
The step-up causes less tax revenue by percentage from the >$300M crowd who use the BBD strategy, but it causes more tax revenue by percentage from the $13M<crowd<$300M who do not use the BBD strategy. The latter pay more tax with option A! 20% on a chunk and 40% on the remaining chunk is less government revenue than just 40% unchunked, especially if the capital gains being realized on death are a majority of the net worth.
I wonder which crowd has more worth-at-death in aggregate (in the absence of BBD and the like -- if estate tax were to be paid by all, no loopholes), given that the less wealthy crowd is a much larger population.
- Veserv 2y agoNo, that is not how the math works. N is your cost basis. M is the gain. E is the estate tax. G is the gains tax. ((N + M) * E) is tax on the automatic step-up, option A. (M * G) + (N + M - (M * G)) * E is the tax on the non-automatic step-up, option B. Reorganized to ((N + M) * E) + (M * G) * (1 - E), it is clear that option B is strictly more taxes for any estate tax less than 100%.
- hunter2_ 2y ago:slaps_forehead: Of course, it would be long term (20%) and estate (40% but on slightly less), not one or the other. Mea culpa.