The key thing here is that most of these measures are measuring income, not wealth. The graphs in the article definitely show income. And what that means is t

The key thing here is that most of these measures are measuring income, not wealth. The graphs in the article definitely show income. And what that means is that "top 0.1%" or "top 0.001%" is not the same group year to year. Especially given the way capital income is handled: in the year when you sell a bunch of stock, or sell a house, or whatever, you suddenly move much higher in the income distribution, but just for that one year. To make this concrete, let's look at the "top 1%" of the population in the sense that their income is consistently in the top 1% of incomes (let's ignore lifecycle effects for now, which throw extra wrenches into the works). These people buy houses, then sell them at some point. I really doubt they average more than 20 years between sales, so in any given year at least 5% of them are selling houses. Any such house sale is likely accompanied by significant capital gains (from inflation, if nothing else, though the real estate markets in various places have been helping, of course). In what fraction of cases are those capital gains enough to move you into the 0.1% for that one year? What fraction of 0.1% incomes are then due to "top 1%" people selling their house? If we were tracking lifetime cumulative incomes here, rather than annual incomes, we might be able to say something more useful. But a significant part of what we are seeing in the data is that income at the top has become more volatile, so when you get taxable events those taxable events are larger than they used to be. Now of course the gap between someone who consistently makes a "0.1%" income and someone who consistently makes a "10%" income has in fact been widening. But I'd really like to see some data on how big those classes of people are. Consistent "0.1%" income accrues to _much_ less than 0.1% of the population, from everything I've seen.