Winterspeak links to a graph of income growth at the top end of the income spectrum from the Wall Street Journal. The graph itself just presents Emmanuel Saez and Thomas Piketty's research, which is well known. In short, Saez and Piketty have shown that only incomes in the top 10% rose in real terms from 2000-2006, and only incomes in the top 1% saw any significant income growth. But the graph reminds me of one of the most grating conservative arguments against raising taxes on the wealthy. The conservative argument generally goes something like this: "Yes, middle-class incomes have stagnated, but incomes in the top 10% haven't been growing much either, so it's unfair to raise taxes on the top 1%." The fallacy is obvious to anyone who thinks about it for measurable time (unlike, say, journalists). On the issue of whether to raise taxes on the top 1%, real income growth in the top 10%, as a group, is irrelevant. What's relevant (if you consider income growth relevant to such a discussion at all) is income growth in the top 1%. Comparatively small increases in real incomes in the top 10%-2% could be—and as Saez and Piketty have shown, is—masking substantial increases in real incomes in the top 1%. Real incomes in the top 1% have increased at an 11% annual rate in the Bush era; real incomes in the top 0.01% have increased at a whopping 22.2% annual rate in the Bush era. The fact that real incomes in the top 5%-1% have increased at an annual rate of only 1% in Bush era doesn't matter, because both candidates want to extend the Bush tax cuts for people in the top 5%-1%. Obama is only proposing raising taxes on incomes in the top 1%. If you want to argue against Obama's plan on fairness or efficiency grounds, then by all means, go ahead. There are certainly arguments to be made. But please, please don't use statistics on all incomes in the top 10%, or the top 5%.