Imagine two children growing up in low-income families on opposite ends of the country. Both work hard, climb the income ladder, and land solid middle-class jobs as adults. By one common measure, both have escaped their parents’ circumstances. But one of them owns a home that has tripled in value, while the other rents and has almost nothing in savings. Are these two children really living the same version of the American Dream?
For years, researchers studying economic opportunity have leaned heavily on income to answer questions like this. A new working paper argues that income tells only part of the story, and that the geography of opportunity looks different when you measure wealth instead.
A gap in the map
A great deal of what Americans believe about where opportunity lives comes from studies tracking income across generations. These studies produced the now-familiar color-coded maps showing that children raised in some regions tend to out-earn their parents while children in others stay stuck. But income captures only money that flows in regularly, like paychecks, dividends, and interest. It misses the assets a family holds, especially the home, which for most households is the single biggest source of wealth.
That distinction matters, the authors note, because prominent economists have argued that wealth is a better gauge of a family’s true economic footing. In a paper distributed by the National Bureau of Economic Research, Ariel J. Binder and John L. Voorheis of the U.S. Census Bureau, along with Max Risch of Carnegie Mellon University, set out to fill what they describe as a “lacuna” at the center of the evidence: we know a lot about where poor children grow up to be rich, but much less about where they grow up to be wealthy.
Building a bigger picture from linked records
To study this, the researchers assembled what they describe as the most comprehensive dataset of its kind in the United States. Using the Census Bureau’s secure data-linking system, they connected responses from the 2000 Census long form, Internal Revenue Service tax records, and a nationwide database of property ownership and home values. The result was a sample of more than 3.4 million parent-child pairs, with children born between 1978 and 1986.
This linkage let them measure five different economic outcomes for the exact same families: total income, labor income (wages and salary), homeownership, housing wealth, and total wealth. Because all five came from the same parents and children, the team could compare them directly. They measured parents’ resources around the year 2000 and the children’s outcomes in 2019 through 2021, when the children were roughly 40 years old.
A quick note on how the researchers think about mobility helps here. They rank both parents and children from the bottom to the top of each distribution, then look at two things. “Absolute mobility” asks how well children from the lowest rungs end up doing on average. “Relative mobility” asks how tightly a child’s adult rank is tied to where their parents started; when that link is strong, the authors call it high intergenerational persistence, meaning your family’s position tends to stick to you. To produce reliable numbers even for sparsely populated counties, they used a statistical technique that borrows information from nearby areas. In the end, they generated estimates for 3,133 counties, nearly all of them.
Wealth and income don’t map onto each other
The first finding is that the geography of wealth mobility does not match the geography of income mobility. When the researchers compared counties, the link between income-based and wealth-based relative mobility was surprisingly loose, with correlations below 0.5. By contrast, the two income measures lined up closely with each other.
The familiar income map reappeared in their data: lower mobility across much of the Southeast, higher mobility in the Great Plains. But the wealth maps showed less of this regional clustering. Instead, wealth mobility tended to vary with population density. Many large metro areas and their suburbs, including New York, Los Angeles, Chicago, Houston, Boston, Seattle, and the San Francisco Bay Area, showed higher persistence in housing wealth than in income. In plain terms, in these dense, expensive places, a family’s housing wealth tends to pass down more reliably from one generation to the next than its income does.
The authors also found that some areas show more variation across counties than the income studies suggested. Labor income, housing wealth, and homeownership all varied more widely from place to place than total income did. This points, they argue, to even more geographic unevenness in opportunity than earlier work emphasized.
Different forces, different outcomes
Next, the team examined which local characteristics tend to go along with mobility. They pulled in 19 county-level traits, covering demographics, economic conditions, inequality, public spending, and social factors, and looked at how each related to the various mobility measures.
Some patterns held across the board. Income inequality, measured by the Gini coefficient, was consistently linked to lower mobility no matter which outcome they looked at, echoing what earlier research found for income alone. But other factors mattered for some outcomes and not others. The local marriage rate, for instance, was strongly associated with homeownership and wealth mobility but showed little connection to income mobility. Residential income segregation was tied to lower wealth mobility but not to labor income mobility, which the authors interpret as a sign that segregation may shape opportunity through channels other than access to good jobs.
Home prices produced a revealing split. In areas with expensive homes, children from poor families who managed to buy ended up with more housing wealth, but those same high prices were associated with lower rates of becoming a homeowner in the first place. Expensive housing, in other words, can reward the people who get in while shutting others out.
The long shadow of the Great Recession
The timing of the data let the researchers study one more question. The children in the sample were starting their careers around the 2008 financial crisis, and the team used differences in how hard the recession hit each county to look at long-run effects.
For income, the pattern was straightforward. Children from harder-hit counties earned less as adults, and this held regardless of whether their parents had been rich or poor. The recession appeared to shift everyone down by a similar amount.
Wealth behaved differently. Here the researchers found what they call a “pivot.” Children from poorer families in hard-hit areas ended up with less housing wealth and lower homeownership rates as adults. But children from wealthier families saw their wealth largely insulated from the local severity of the downturn. The authors suggest this may reflect the timing of parental help: parents often assist with home purchases later in life, through financial gifts or know-how, and wealthier families were better positioned to provide that cushion when it counted.
What to keep in mind
The researchers are careful to flag that these relationships are descriptive, not proof of cause and effect. They show which traits tend to travel together across counties, not that one produces the other. They also note real limits in their wealth measures: the figures cover assets but not debts, miss assets invisible to the tax and property systems, and capture only property held directly by individuals rather than through companies or trusts.
Even so, the central message is that the choice of yardstick shapes the conclusion. As the authors put it, anyone hoping to understand where opportunity lives should not rely on income alone. They have released their county-level estimates publicly, hoping other researchers will use them to keep filling in the map.
