Tim Noah's been running a great series on the (already massive, and increasing) quantitative differences between the rich and the rest of us in Slate. I think these are probably the numbers to use (and over and over again, too):
People know we're living in a time of growing income inequality, Krugman told me, but "the ordinary person is not really aware of how big it is." The ignorance hypothesis gets a strong assist from a new paper for the journal Perspectives on Psychological Science: "Building a Better America—One Wealth Quintile at a Time." The authors are Michael I. Norton, a psychologist who teaches at Harvard Business School, and Dan Ariely, a behavioral economist (and blogger) at Duke. Norton and Ariely focus on the distribution of wealth, which is even more top-heavy than the distribution of income. The richest 1 percent account for 35 percent of the nation's net worth; subtract housing, and their share rises to 43 percent. The richest 20 percent (or "top quintile") account for 85 percent; subtract housing and their share rises to 93 percent. But when Norton and Ariely surveyed a group whose incomes, voting patterns, and geographic distribution approximated that of U.S. population, the respondents guessed that the top quintile accounted for only 59 percent of the nation's wealth.
"The top 1% own almost half the country when you leave out housing."
NOTE I suppose I'm trying to finesse a strategic debate -- 1% vs 20% -- with a choice of language. But it seems to me that one (and only one) reason for the success of the gay rights movement was "we are your sons and daughters, your aunts and uncles, your nieces and nephews, your sisters and brothers, your fathers and mothers." If the same argument is to be made in class terms, it makes sense to target as few families as possible, no?