So, 42% of the wealth supplies 38% of the tax base? That's pretty close, of course, but when you're talking about trillions of dollars 4% is quite a big shortfall.

And it's 4% the other direction if you use the net worth number. He defines wealth as (net worth - home value) so I don't know that it's a good metric

The NTU distributions appear somewhat oddly presented in those tables, and odd portrayals of data make me curious. Odd displays can indicate the existence of information that a presenter intentionally wishes to avoid highlighting, or information that the presenter simply does not recognize is present, and that information can be as interesting as what a presenter does choose to highlight.

Based on the Slate data referenced elsewhere in the comments here, there looks to be around 84% of the aggregate income for the top 20% of the earners.

Then interpolating the data buckets shown within the NTU tables and particularly guessing that the 20% range is around 80% of the taxes paid, and matching that with the Slate income charts.

This implies that the folks with 80% of the income pay somewhere roughly around 80% of the taxes. Give or take.

I'd be interested in seeing a NTU-like breakdown matched with what data Slate has shown for relative incomes; matched with what looks to be the income and with slicing the income based on the different AGIs.