Where birth buys the most
Where does a comfortable birth buy the most extra adult income — and is a small premium good news?
The adult income rank of children raised at the 25th and at the 75th parental percentile, county by county. Colour pairs the poor-kid outcome with the premium between the two.
Across all tracts, children of 75th-percentile parents reach an adult income rank 13.6 points above children of 25th-percentile parents. By county the gap runs from 6.4 to 24.0. A small premium is good news only where the floor is high — select a cell of the key to see where that holds.
Poor kids' rank →
County data · 1,581 rows
| County | State | FIPS | Tracts | Poor kids (rank) | Rich kids (rank) | Premium | Class |
|---|---|---|---|---|---|---|---|
| Autauga | AL | 01001 | 17 | 37.4 | 57.2 | 19.8 | Low floor, large premium |
| Baldwin | AL | 01003 | 43 | 39.6 | 55.6 | 16 | Low floor, middling premium |
| Barbour | AL | 01005 | 9 | 34.9 | 57 | 22.1 | Low floor, large premium |
| Bibb | AL | 01007 | 8 | 36.4 | 56.8 | 20.4 | Low floor, large premium |
| Blount | AL | 01009 | 16 | 39.1 | 56.8 | 17.7 | Low floor, large premium |
| Butler | AL | 01013 | 9 | 36.4 | 57.3 | 21 | Low floor, large premium |
| Calhoun | AL | 01015 | 35 | 37.4 | 53.8 | 16.4 | Low floor, middling premium |
| Chambers | AL | 01017 | 9 | 34.1 | 52.4 | 18.3 | Low floor, large premium |
| Cherokee | AL | 01019 | 8 | 36.7 | 54.9 | 18.2 | Low floor, large premium |
| Chilton | AL | 01021 | 12 | 39.1 | 56.4 | 17.2 | Low floor, large premium |
| Clarke | AL | 01025 | 9 | 38.4 | 57 | 18.6 | Low floor, large premium |
| Coffee | AL | 01031 | 16 | 39.8 | 57.4 | 17.5 | Low floor, large premium |
| Colbert | AL | 01033 | 15 | 37.7 | 56.8 | 19.2 | Low floor, large premium |
| Covington | AL | 01039 | 14 | 41.2 | 63.1 | 21.9 | Middle floor, large premium |
| Cullman | AL | 01043 | 22 | 40.6 | 57.4 | 16.7 | Middle floor, large premium |
| Dale | AL | 01045 | 14 | 38.7 | 55.3 | 16.6 | Low floor, large premium |
| Dallas | AL | 01047 | 17 | 34.8 | 51.7 | 16.9 | Low floor, large premium |
| DeKalb | AL | 01049 | 21 | 39.9 | 58.1 | 18.2 | Low floor, large premium |
| Elmore | AL | 01051 | 20 | 38.5 | 56.5 | 18 | Low floor, large premium |
| Escambia | AL | 01053 | 10 | 38.7 | 57.9 | 19.2 | Low floor, large premium |
Ranks are the mean adult household-income percentile, at ages 31–37, of children born 1978–83 whose parents were at the 25th or 75th percentile — “rich” here means comfortable, not wealthy. County values are tract means weighted by population; counties with fewer than 8 measured tracts are not estimated and stay grey. Connecticut’s nine 2022 planning regions do have estimates, but no shapes on this map, whose county boundaries predate them: they appear in the scatter and the table, and Connecticut stays grey. Both ranks come from the same local rank-rank regression, so the premium is a model-implied gap and noisier than either input. Classes are terciles across the 1,581 counties (a third in each, unweighted): the diagonals are the premium breaks, the verticals the floor breaks. The warm ramp marks a larger premium; the teal ramp a higher floor. These are descriptive averages over children who grew up decades ago, and they mix what a place does with who lives there.
Opportunity Atlas (Chetty et al.) · children born 1978–83 · 1,581 counties with at least 8 measured tracts
Reading notes
The premium is a model-implied gap from one rank-rank regression, noisier than either input. Tercile classes are derived (unweighted across counties). Connecticut’s 2022 planning regions have no shapes on this map. Descriptive: place effects are mixed with who lives there.
Source & method
Kept the 1,581 county rows (FIPS, name, state, measured-tract count, mean adult income rank for children of 25th- and 75th-percentile parents, and the premium between them), as published: population-weighted tract means over counties with at least 8 measured tracts; smaller counties are absent, not zero. Kept the national tract-level summary (means, SDs, correlation, the 13.6-point premium and its 5th–95th percentile range). Derived, and labelled as derived: tercile breaks of the county p25 outcome and of the county premium (unweighted, one third of counties per class), which define the 3×3 bivariate classes. Dropped the tract-level distributions and the 5,000-tract scatter sample — they carry no county identifier, so they cannot be linked to the map.
Capture provenance
- box:birth_death / articles/_build/data/s7_premium.json
- Modified 2026-06-12 · SHA-256 e2f95d34ee7c6c9bdae5a9b45ae8ca895062fdeeb4849918ed14f2d0e81617f2
- Captured
- 2026-09-12
- Fixture SHA-256
- 6fb0c533dc5093128db0880cd7f4333f636b63f391737e5fe8fe468b4eb8e8ab
- Reuse note
- Opportunity Atlas (Chetty, Friedman, Hendren, Jones & Porter 2018), tract outcomes kfr_p25 and kfr_p75 for children born 1978–83, aggregated to counties with ACS 2019–2023 population weights in the ZIP Code Destiny pipeline. Cite the Opportunity Atlas when reusing.
