Geography of Opportunity Rankings: What the Data Actually Shows

The dataset that changed the conversation

The modern study of US economic mobility runs on one body of work: the linked tax-return panels assembled by Opportunity Insights at Harvard, led by Raj Chetty and collaborators. The headline finding, published in the Quarterly Journal of Economics in 2014 and updated continuously since, is that the chance an American child raised in the bottom income quintile reaches the top income quintile in adulthood varies dramatically by the county where they grew up. The national average is roughly 7.5 percent. In some counties the rate exceeds 16 percent. In others it is below 4 percent.

That spread is the geography of opportunity. The county a child is raised in changes their adult earnings independently of family income, race, and education. Subsequent work isolated the causal effect using sibling comparisons and family moves (Chetty and Hendren, 2018), and showed that every additional year of childhood spent in a higher-opportunity county increases adult earnings by about 0.4 to 0.6 percent per year.

What gets counted, and how

Opportunity Insights builds its rankings from anonymized federal tax data linking parents to children across decades. The Census Bureau and IRS partnership behind the data is described in the agency's statistical research methodology documentation. For each county the team publishes:

• Absolute upward mobility (mean adult income rank for children raised at the 25th percentile of parental income).
• Relative mobility (the slope coefficient relating parent and child ranks).
• Race-specific mobility, broken out for white, Black, Hispanic, Asian and Native American children.
• Subgroup rankings by gender, by neighborhood density, and by parental household structure.

The data product most users encounter is the Opportunity Atlas, a county and census-tract map of these outcomes. The atlas reports on cohorts born between 1978 and 1983, observed in their early thirties, which means the current geography of opportunity reflects the conditions of the 1980s and 1990s in each place. The cohort lag is the most important caveat in the data.

What the rankings consistently find

Five patterns recur across every update of the data.

First, mobility is higher in the middle of the country than on the coasts. Counties in Utah, Iowa, Nebraska, the Dakotas, and parts of Minnesota produce the highest absolute upward mobility for children raised in low-income families. This was a counter-intuitive finding when it was first published and has held up across cohorts.

Second, mobility is lower in the post-industrial Midwest and the Deep South. The lowest-mobility counties cluster in the Mississippi Delta, in Appalachian Kentucky and West Virginia, and in legacy industrial centers that lost employment between 1980 and 2010. The contemporary income gap between these counties and the high-mobility West is partly a story of which counties were also low-mobility for the parental generation.

Third, race-specific gaps are large and persistent. The Race and Economic Opportunity project showed that Black boys, even those raised in high-income families, have substantially lower adult earnings than white boys raised in the same neighborhoods. The gap is not present for Black girls relative to white girls at the same parental income, an asymmetry that the authors attribute primarily to differences in employment outcomes for Black men.

Fourth, social capital matters as much as income. A 2022 Nature paper used 21 billion Facebook friendship ties to measure cross-class connection (what the authors call "economic connectedness") and found it is one of the strongest predictors of upward mobility, alongside school quality and family structure. The implication is that mixed-income neighborhoods and institutions, not just income transfers, build the rungs of the ladder.

Fifth, moves matter, and earlier moves matter more. A child who moves from a low-opportunity to a high-opportunity county at age 4 captures most of the difference between the two places by adulthood. A child who moves at 18 captures very little of it. This finding underpins the Creating Moves to Opportunity housing-voucher demonstration, which has informed federal policy on portable rental assistance.

Where the rankings can mislead

Three honest caveats.

The cohort lag means today's ranking reflects yesterday's conditions. A county that has had two decades of investment in early-childhood education, transit, and economic-connectedness institutions will show that improvement in the rankings only when the children who lived through those changes reach their thirties.

The income measure is taxable earnings. It misses non-cash benefits, informal-economy income, and the public-goods value of place. A county that delivers high upward mobility partly through subsidised housing or strong public services may rank lower on rules that count only earned income.

And the geography is administrative, not lived. A child raised in two adjacent census tracts that span a county line is treated as living in two different opportunity environments. Tract-level data (the atlas publishes both) is more granular but loses statistical precision for small populations.

How states and cities are using the data

Three classes of intervention have moved from the literature into practice.

• Housing-voucher mobility programs following the CMTO model. Seattle and Dallas have run the largest experiments; HUD authorised portable vouchers in select metros under the Housing Choice Voucher Mobility Demonstration in 2022.
• Early-childhood expansion in counties that score low on absolute mobility. Tennessee's pre-K expansion and Mississippi's third-grade reading reforms are the most-studied examples; both are still controversial in the evaluation literature.
• Economic-connectedness investments that target the social-capital channel from the 2022 Nature paper: cross-class school integration, mentoring programs, and mixed-income development near transit corridors. The Bridgespan Group and the Aspen Institute have published practical guides for foundation funders working on this channel.

These interventions interact with the questions covered in our companion pieces. The local labor market in which a child eventually enters adulthood is shaped by AI job displacement by county. The infrastructure and policy environment they inherit is shaped by AI readiness by state. The cost of building an adult life in the counties they migrate to is captured in the American Dream Affordability Index. And the social isolation that erodes the connectedness channel is mapped in America's Loneliness Map.

Reading list

The 2014 Land of Opportunity paper is the canonical starting point. The 2022 Nature paper on social capital is the most recent step-change. The Opportunity Atlas is the most useful interactive. The US Partnership on Mobility from Poverty published a policy translation of the findings that is more accessible than the underlying papers. For a critical perspective on what the data does not measure, the Roosevelt Institute and the Hamilton Project have published useful counterweights.

The geography of opportunity is one of the better-measured features of American life. The next decade's question is whether the places that have been low-mobility for fifty years can change their trajectory before another cohort comes of age inside them.

Sources

• Chetty, Hendren, Kline and Saez, *Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States*, QJE 2014.
• Chetty and Hendren, *The Impacts of Neighborhoods on Intergenerational Mobility I and II*, QJE 2018.
• Chetty et al., *Race and Economic Opportunity in the United States*, QJE 2020.
• Chetty et al., *Social capital I and II: determinants and effects of social mobility*, Nature 2022.
• Opportunity Insights, *The Opportunity Atlas*.
• US Department of Housing and Urban Development, *Housing Choice Voucher Mobility Demonstration*.