New research shows rising class, falling race gaps in economic mobility

The question of how best to enhance upward mobility has endured as a public policy theme in Rochester. The Rochester-Monroe Anti-Poverty Initiative drives efforts to improve financial stability, while the city’s Office of Financial Empowerment focuses on banking access, education, and business development for low-income residents. Another initiative, the Commission on Racial and Structural Equity, has aimed to dismantle structural inequity in the city and county.

Although these efforts are laudable, it is instructive to comprehend the striking changes one sees in intergenerational economic mobility in the United States by both race and class over a relatively short period.

Thought-provoking new research sheds valuable light on this and related matters. Specifically, the research focuses on children born between 1978 and 1992 and demonstrates that class-based inequality widened sharply among white families, while racial inequality between white and Black children—especially among those from low-income families—narrowed greatly. These divergent trends extend beyond income to educational attainment, test scores, employment, incarceration, and mortality, pointing to broad changes in life chances rather than labor-market outcomes alone.

Using linked administrative data covering 57 million children including federal tax records, Census and American Community Survey data, and SAT/ACT scores, the authors—researchers at Harvard University, Cornell University and the U.S. Census Bureau—measure children’s outcomes primarily at age 27 and compare them across birth cohorts. Among white children, earnings rose for those from high-income families but fell for those from low-income families, increasing the white class gap by about 30 percent. In contrast, Black children experienced rising earnings across the parental income distribution, reducing the white–Black earnings gap for children from low-income families by approximately 30 percent. Significantly, this racial convergence occurred largely through improved chances for Black children to escape poverty rather than increased access to the very top of the income distribution.

This research shows that standard explanations centered on family-level characteristics—such as parental education, wealth, occupation, or marital status—account for little of these trends. In fact, observable family factors explain only about 7 percent of the growing white class gap and about 10 percent of the shrinking white–Black gap. Similarly, neighborhood-level differences common to all residents of an area do not explain the divergence: controlling for childhood county or census tract still leaves the main patterns intact. These findings indicate that the key drivers operate within neighborhoods, affecting race and class groups differently.

The research identifies community-level changes in childhood environments, proxied primarily by parental employment rates, as the central mechanism. The authors define communities as groups sharing the same race, class, and county. Across counties and subgroups, changes in children’s outcomes are strongly correlated with changes in parental employment rates in their communities. Where parental employment rose, children’s later outcomes improved; where it fell, outcomes deteriorated. Interestingly, the relationship between parental employment trends and child outcomes is nearly identical across race and class groups, implying that differences in employment trends across groups can explain both the widening white class gap and the narrowing white-Black gap.

To establish causality, the authors use a quasi-experimental research design based on children who moved across counties at different ages. Children who moved early in childhood to communities where parental employment was improving experienced significantly better outcomes than those who moved later—while late movers showed little benefit. This age-at-move pattern rules out explanations based solely on adult labor demand or selection. Sibling comparisons further strengthen the case: Younger siblings, who spent more years exposed to improving environments, fare better than older siblings when families move to such places. The estimates obtained imply that roughly 90 percent of the observed association between parental employment changes and child outcomes reflects a causal childhood exposure effect.

This noteworthy research also distinguishes between economic resource explanations and social interaction mechanisms. When it comes to this dichotomy, several pieces of evidence favor social interaction. Children’s outcomes are most strongly associated with the parental employment rates of peers they are most likely to interact with—those in their own birth cohort and their own race and class group. The effects decay sharply for adjacent cohorts and are stronger within racial groups, consistent with homophily in social networks. Moreover, cross-race effects are larger in counties with more interracial interaction, such as places with higher interracial marriage rates.

Overall, the findings from this research provide plenty of food for thought. In particular, the outcomes for children tend to mimic the economic trajectories of adults in their social communities. In addition, we learn that intergenerational mobility is malleable even over a decade and that the social environments during one’s childhood are very salient determinants of outcomes later in life.

From a policy perspective, this research points to interventions that strengthen community employment, reduce segregation, and build social capital—alongside traditional investments in education and income support—as promising avenues for enhancing upward economic mobility.

Amitrajeet A. Batabyal is a Distinguished Professor, the Arthur J. Gosnell professor of economics, and the Head of the Sustainability Department, all at the Rochester Institute of Technology in Rochester, New York. These views are his own.