Downward Mobility in America

Pew has a new study out on downward mobility from the American middle class. The report focuses on people who were 14-17 years old in 1979, who were restudied in 2004 to 2006 when they were 39 to 44 years old.

The study looks at the problem two different ways. First, they arbitrarily define the middle class as the 30th to 70th percentiles on the national income distribution curve, that is, the middle 40% of the nation by wealth, and then ask who grew up in that zone but then fell below it as an adult. Second, they measure the percentage whose incomes fall 20 percentile points lower than those of their parents on that national scale. These numbers turn out to be pretty much the same.

Overall, they find, about 28% of people who grow up in the middle class fall out of it as adults. Since their definition of the lower class consists of the bottom 30% of the population, this is about what you would expect if growing up in the middle class conveyed no special advantage; that is, people raised at around the median income are no more likely to end up in the middle class than people who grow up in the 10th percentile.

The factors most strongly associated with falling more than 20 percentage points down the economic ladder are divorce, drug use, low education, and low scores on the on the Armed Forces Qualification Test (AFQT). Never getting married is also strongly associated with declining status.

Race is a factor for men: 38% of black men fall out of the middle, compared to 21% of white men. There is no difference between black and white women, both at around 30%. This means that among white Americans there is a major gender disparity: 30% of white women fall out of the middle class, but only 21 % of white men do. This is mainly because women are much more likely to fall out of the middle class after a divorce.

I find the basic pattern remarkable, because we like to think that middle class families are conveying certain values to their children that will help them "get ahead" in life. Apparently, this is not so. Being born in the middle class is no guarantee that you will end up there.

I wish the data showed in more detail how far people fell; falling from the 50th percentile of income to the 30th would meet both their definitions of falling out of the middle class, but it would not be nearly as dramatic a slide as falling from the 31st to the 11th.