Russel Sage Foundation

The U.S. is widely viewed as a land of opportunity in which, regardless of origins, anyone can succeed by virtue of working hard. The available empirical evidence suggests that, contrary to this conventional view, the U.S. has the lowest rate of intergenerational mobility among the advanced countries for which relevant data are available.

The research within this area has typically proceeded by estimating intergenerational elasticities (IGE) that measure the strength of the relationship between the income (or earnings) of parents and the income (or earnings) of their grown children. In the 1980s, the prevailing view was that the earnings IGE for fathers and sons was in the range of .2, thus implying that origins and destinations were but weakly related and that the U.S. was a very mobile society. However, subsequent research based on better data and methods produced IGEs as high as .6, and many scholars now believe that the U.S. has much less mobility than was previously thought.

But even these most recent estimates are potentially problematic. In almost all cases, previous estimates of intergenerational economic mobility in the U.S. have relied on analyses of survey data with relatively small samples, possible recall problems, significant attrition, and other limitations. Although a variety of compelling statistical fixes have been developed to address these problems, there is a limit to what sophisticated statistical methods alone can accomplish.

To address the many issues posed by survey data, most scholars in the field agree that administrative data, similar to the registry data used in Scandinavian countries, should be employed. As an initial step in this direction, Professor David Grusky, Dr. Pablo Mitnik, and Dr. Christopher Wimer will conduct, in collaboration with the Internal Revenue Service’s Statistics of Income (SOI) division, the first studies of intergenerational mobility in the U.S. using a large panel of income tax returns. These analyses will serve as a template for future studies based on full-population IRS data and will provide new estimates of intergenerational mobility that are more precise and reliable than those currently available. Because the IRS sample is relatively large, there will be sufficient power to estimate mobility rates at different tiers of the income and earnings distributions, and to compare the extent of economic mobility using pre-tax and after-tax measures of income.