You left off the other side of the equation. He says the CB income data understates income on the high end because it exclude the value of capital gains, but it

You left off the other side of the equation. He says the CB income data understates income on the high end because it exclude the value of capital gains, but it ALSO understates income on the low end because it excludes the value of non-cash benefits. These effects work in opposite directions; it does thus seem plausible that the resulting Gini is in approximately the right ballpark. Furthermore, note that we don't actually care about the Gini AMOUNT here, we just care about the TREND. So unless the ratio of normal income to capital gains is changing quite a lot (and in the appropriate direction) it's not going to produce the result you want. > Frankly I could simply interpret his chart as showing that rich are increasingly moving their gains away from income and towards capital gains to avoid taxes. You are expressing a hunch. Got any data to back it up? > Second, for his graph of GINI for individuals, he's picked the GINI only for people that have an income. Isn't that the normal way of calculating it? > In that same paper, the one that includes everyone (not just income earners) shows increasing inequality in the last 30 years. Do you happen to have a link for that paper? (Or even just a chart showing that result?)