Against a rising chorus of concern about increasing income inequality, some economists are pushing back, suggesting that it is not income inequality we should be concerned with but rather income mobility. Income mobility describes the ability of individuals to move up and down the income ladder over some period of time. As long as mobility is healthy, they argue, society can remain egalitarian in the face of inequality, because the poor can move up and the rich down.

Intuitively, some observers assume that higher income inequality should be correlated with decreased income mobility as the rich build a bigger lead on the rest of society. But there is little consensus about whether and how income mobility has changed. What little research does exist is inconsistent with regards to findings, methods, and data sources. Equitable Growth grantees Michael D. Carr and Emily E. Wiemers at the University of Massachusetts-Boston used a new dataset to revisit the measurement of earnings mobility, the part of income that comes from work. Their results suggest that lifetime earnings mobility has declined in recent years.

The authors construct a snapshot of earnings mobility in two time periods: 1981 to 1996 as well as 1993 to 2008. In each one, workers are observed at the beginning and end of the time period to capture mobility over a 15-year span. Workers are divided into ten income deciles (each representing 10 percent of workers, ordered from lowest to highest paid) and categorized at the beginning and end of the 15-year span. Use the interactive below to explore their results in each of these two time periods.


To see how mobility changed over these two time periods, Carr and Wiemers look at how the probability of moving up or down rungs on the earnings ladder has changed between the two periods. As the interactive below shows, they find that most workers are less likely to move up the income ladder now (1993-2008), and a bit more likely to fall down the income ladder than they were in the past (1981-1996). The exception is earners at the very top of the income distribution. By definition, these workers at the top rung cannot rise up any farther, but in recent years, members of the top group are less likely to slip down than they used to be.

Use the dropdown menu to look at different slices of the population. A particularly notable finding from the paper is that the largest declines in upward mobility are among workers with college degrees who start in the middle of the earnings distribution. It has become increasingly difficult for these college-educated, middle-class workers to ascend into high income jobs. In general, a worker’s starting position has become more predictive of their final position across the income spectrum.


Carr and Wiemers use a new data source that they contend has some advantages over data used in previous research. They use the Survey of Income and Program Participation, or SIPP, to obtain demographics on a large number of workers over a long period of time. The Census Bureau merged this survey with administrative records from the Internal Revenue Service and the Social Security Administration and has made the product available to researchers. This gives Carr and Wiemers accurate information on earnings and benefits for each worker. This approach yields a large, nationally representative sample with demographic information on each individual in the data set. Better still, the researchers are able to follow individuals for a long period of time.

Economics is undergoing a kind of data revolution as administrative datasets become more common. Studies like this one may shed new light on areas of research that were previously marked by inconsistent findings. It may be some time before we can say with certainty what is happening to economic mobility in America, but this research makes an important new contribution to the existing literature.