This is a misleading take on the data. A much better analysis would look at the size of the population who was middle class or upper class. If there is an equal increase in upper class population for every loss of middle class population, that just means that our incomes as a population are increasing. All things equal, both the mean and median would increase in that circumstance.<p>And that's what the data looks like to a large degree. The middle class has shrunk from 61% to 50%, but the upper class has increased by nearly the same amount, 14% to 21%. Yes, that is a decrease in the middle class-or-better bucket, but it's nowhere near dire enough to be described as a tipping point. I agree that the shrinking middle class is something that needs to be discussed, but this article does the discussion a giant disservice.