Class In America: Mobility, Measured

AMERICANS are deeply divided as to whether widening inequality is a problem, let alone what the government should do about it. Some are appalled that Bill Gates has so much money; others say good luck to him. But nearly everyone agrees that declining social mobility is a bad thing. Barack Obama's state-of-the-union speech on January 28th dwelt on how America's "ladders of opportunity" were failing (see "Barack Obama: The state of the president"). Paul Ryan and Marco Rubio, two leading Republicans, recently gave speeches decrying social immobility and demanding more effort to ensure poor people who work hard can better their lot.

Just as the two sides have found something to agree on, however, a new study suggests the conventional wisdom may be wrong. Despite huge increases in inequality, America may be no less mobile a society than it was 40 years ago.

The study, by a clutch of economists at Harvard University and the University of California, Berkeley, is far bigger than any previous effort to measure social mobility. The economists crunch numbers from over 40m tax returns of people born between 1971 and 1993 (with all identifying information removed). They focus on mobility between generations and use several ways to measure it, including the correlation of parents' and children's income, and the odds that a child born into the bottom fifth of the income distribution will climb all the way up to the top fifth.

They find that none of these measures has changed much (see chart). In 1971 a child from the poorest fifth had an 8.4% chance of making it to the top quintile. For a child born in 1986 the odds were 9%. The study confirms previous findings that America's social mobility is low compared with many European countries. (In Denmark, a poor child has twice as much chance of making it to the top quintile as in America.) But it challenges several smaller recent studies that concluded that America had become less socially mobile.

This result has caused a huge stir, not least because it runs counter to public perceptions. A recent Gallup poll found that only 52% of Americans think there is plenty of opportunity for the average Joe to get ahead, down from 81% in 1998. It also jars with other circumstantial evidence. Several studies point to widening gaps between rich and poor in the kinds of factors you would expect to influence mobility, such as the quality of schools or parents' investment of time and money in their children. Cross-country analyses also suggest there is an inverse relationship between income inequality and social mobility--a phenomenon that has become known as the "Great Gatsby" curve.

What is going on? One possibility is that social stratification takes time to become entrenched. In a new book, Gregory Clark, an economic historian at the University of California, Davis, who tracks mobility over hundreds of years by following surnames, reaches far more pessimistic conclusions (see "Social mobility: Have and have not"). Another, sunnier, explanation is that even as income gaps have widened over the past 30 years, other barriers to mobility, such as discrimination against women and blacks, have fallen.

Most likely, the answer lies in the nature of America's inequality, whose main characteristic is the soaring share of overall income going to the top 1% (from 10% in 1980 to 22% in 2012). The correlation between vast wealth accruing to a tiny elite and the ability of people to move between the rest of the rungs of the income ladder may be small--at least for now.

Whatever the explanation, it would be unwise to take much comfort from this study. For a start, since the gap between top and bottom has widened, the consequences of an accident of birth have become bigger. Second, if the gains of growth are going mostly to those at the top, that bodes ill for those whose skills are less in demand. Many economists worry that living standards for the non-elite will stagnate for a long time.

Is your town a launchpad or a swamp?

Third, although social mobility has not changed much over time, it varies widely from place to place. In a second paper, the economists crunch their tax statistics by region. They find that the probability of a child born into the poorest fifth of the population in San Jose, California making it to the top is 12.9%, not much lower than in Denmark. In Charlotte, North Carolina it is 4.4%, far lower than anywhere else in the rich world.

This geographic prism also offers some pointers on what influences mobility. The economists found five factors that were correlated with differences in social mobility in different parts of America: residential segregation (whether by income or race); the quality of schooling; family structure (eg, how many children live with only one parent); "social capital" (such as taking part in community groups); and inequality (particularly income gaps among those outside the top 1%). Social mobility is higher in integrated places with good schools, strong families, lots of community spirit and smaller income gaps within the broad middle class. Not a bad agenda for politicians to push, if only they knew how.

Click here to subscribe to The Economist