The IRS has released yesterday the preliminary stats for year 2005 which I have used to extend my [and Thomas Piketty's] series [on the top income share by tax return unit] to 2005, posted at: http://elsa.berkeley.edu/~saez/TabFig2005prel.xls
2005 shows a very large increase in income concentration: the top 1% gains 14% in real terms from 2004 while the bottom 99% gains less than 1% (when including capital gains). The [previous] record peak of 2000 is surpassed even though 2005 is less of a high capital gains, high stock option year than 2000. By 2005, it looks like top incomes are showing strongly along all components: wages, business income, dividends, and capital gains.
The striking thing about 2003-2005 is the huge increase at the top with quasi-stagnation below the top 1%. In the late Clinton years, the top gained enormously but at least the bottom was also making progress (something you can see on Fig A2)...
Market income excluding capital gains