Different Cities, Different Ladders

"Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States." Raj Chetty, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez. National Bureau of Economic Research Working Paper No. 19843, January 2014.

Although the United States is often called the "land of opportunity," recent research has suggested that Americans enjoy less economic mobility across generations than historically assumed. But measuring economic mobility for the United States as a whole masks significant regional differences, as discussed in a recent working paper by Raj Chetty and Nathaniel Hendren of Harvard University and Patrick Kline and Emmanuel Saez of the University of California, Berkeley. They find that the United States is best described as a “collection of societies,” some of which display high levels of economic mobility and some of which do not.

A large body of research is devoted to measuring the intergenerational elasticity of earnings (IGE), which describes how differences in earnings persist from one generation to the next. The higher the IGE, the lower the mobility. Estimates of the IGE in the United States range from about 0.3 to nearly 0.6, well above many European countries.

But the IGE has several shortcomings, according to Chetty and his co-authors, such as being very sensitive to differences in how income at the bottom of the distribution is measured. The authors thus focus instead on a “rank-rank” measure of mobility, which compares children’s ranks in the national income distribution to their parents’ ranks. To create this measure, they link the tax records of about 10 million individuals born between 1980 and 1982 (the children) to the tax records of the people who first claimed them as dependents (the parents).

Chetty and his co-authors are primarily interested in within-country variation, which they study by analyzing mobility in 741 “commuting zones” across the United States. Commuting zones are aggregations of counties, similar to metropolitan areas, but they include both rural and urban areas.

The authors calculate both relative mobility, or how children fare compared to each other, and absolute mobility, or how children fare compared to their parents. Studying both is important because a high degree of relative mobility might indicate worse outcomes for the children of rich parents rather than better outcomes for the children of poor parents.

Of the 50 largest commuting zones by population, the highest relative mobility is in Los Angeles, Calif., where children from the poorest families end up only about 23 percentage points lower in the income distribution than children from the richest. The lowest relative mobility is in Cincinnati, where they end up almost 43 percentage points further down the ladder. The highest absolute upward mobility, which the authors define as the average rank of children born to parents at the 25th percentile, is in Salt Lake City, where those children rise to the 46th percentile on average. The lowest is in Charlotte, N.C., where they rise to the 35th percentile. Overall, mobility tends to be lowest in the Southeast, somewhat higher on the West Coast and in the Northeast, and highest in the Great Plains.

What accounts for these regional differences? One clue is that the authors find a strong positive correlation between parent income and college attendance rates and a negative correlation with teen pregnancy rates. In their view, this suggests that the forces influencing a child’s mobility are at work long before the child actually enters the labor market. This view is supported by their finding that the structure of a local labor market — such as the number of manufacturing jobs, which traditionally have offered relatively high wages to lower-skilled workers — has little bearing on mobility. School quality, however, as measured by test scores and dropout rates, does have a large effect.

Another factor that is highly predictive of both relative and absolute mobility is race; in general, there is less mobility for both black children and white children in areas with large black populations. The underlying mechanism appears to be segregation: Areas with large black populations tend to be more segregated by both race and income, which means that low-income children of all races are likely to live in neighborhoods with less school funding and fewer successful role models.

Family structure, particularly the fraction of children living in single-parent households, also is strongly correlated with mobility. As with race, the effect is at the community rather than at the individual level; children from both single- and two-parent families in areas with a large proportion of single-parent families have relatively worse outcomes as adults. The authors propose that family structure indicates the stability of the social environment more broadly and might capture variation in other attributes correlated with mobility, such as income inequality or the level of community engagement.

Of course, correlation is not causation, and the authors are careful to note that their research cannot say what actually causes differences in economic mobility. But it may suggest avenues for both parents and policymakers to improve outcomes for children born to low-income families.

Related Links

Chetty, Raj, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez. "Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States." National Bureau of Economic Research Working Paper No. 19843, January 2014. (Paper available online with subscription.)

Equality of Opportunity Project website.