But framing the discussion in terms of some broader group is in this case deeply misleading. Here’s what the Piketty-Saez numbers tell us about the top 5 percent (incomes in 2012 dollars):
If you look at the bottom 4 percent of the top 5, you see good but not spectacular income gains. These are the kinds of gains that you might be able to explain in terms of skills, assortative mating, and so on. But the top 1 percent is in a different universe altogether. And in fact the gains within the top 1 percent are concentrated in an even smaller group: this is a Pareto distribution thing, in which the higher the income the greater the percentage gains.
The point is that using wider definitions than the one percent is, in effect, diluting the wolves of Wall Street by lumping them in with the upper middle class. Not the same story at all.