See chart below. I don't have all the necessary data to do what I did on the entire period 1960 through 2004 but I do have data on 1979 through 2005. I got it from a spreadsheet that can be downloaded from the web page at
http://www.cbo.gov/publication/20374" onclick="window.open(this.href);return false; (see "spreadsheets" to the right of the page). Effective total federal tax rates on the top 1% varied from 25.5% to 37% over the period. My first observation is that whatever has happened resulted in the US treasury doing better. I didn't do a full analysis of that because it's so obvious. In 2005 dollars (inflation adjusted terms), the average total federal "contribution" of households in the top 1% was $191,586 in 1979 and $486,252 in 2005.
The kicker is this: There is not sufficient evidence to conclude that, within the range of 25.5% to 37%, there is any association one way or another between the tax rate for the top 1% and what its per capita contribution to the treasury is. There is some suggestion that increasing revenue may increase per capita contribution if you just do a straight correlation. But if you remove the effects of "serial autocorrelation" as you're supposed to do with time series data you get p = 0.65 (by usual convention, p has to be less than or equal to 0.05 to indicate "significance," so it's not anywhere close). So there's essentially no evidence in the 1979 through 2005 data, at least in terms of just looking at the association between rates and per capita tax contribution, that raising rates increases revenue to the treasury. Of course there's also no evidence that cutting them does either.
What's clear is that the top 1% DOING better financially raised revenue to the treasury. Dramatically so.
