American neighborhoods are segregated by income. What does that mean for the future of income inequality, neighborhoods, and children?

As the saying goes, there are three things that matter in real estate: Location, location, location. Some neighborhoods offer more amenities than others—more trees, calmer streets, better schools, easier access to employment or transit.

It’s no surprise that neighborhoods with greater amenities become more expensive places to live. This puts the cost of housing in high-amenity neighborhoods out of reach for some households, leading to segregation by income.

Income segregation in America is not static, however. As Minneapolis Fed Assistant Director for Inequality Research and Monetary Advisor Alessandra Fogli and her co-authors observe in a recent Institute working paper, it has been increasing for 40 years. Wage inequality has also grown over this period. Fogli and her co-authors want to understand if there is a connection between these two trends. Does existing residential segregation cause further income inequality, which in turn feeds further segregation?

The answer matters because research shows that neighborhood amenities can influence more than quality of life. The connection between the neighborhood where a child grows up and their future economic outcomes has been documented by economists Raj Chetty and Nathaniel Hendren (a former and current Institute advisor, respectively). By comparing the outcomes of siblings who were different ages when their family moved to a better-off neighborhood, Chetty and Hendren show that each additional year spent living in the better-off neighborhood increases average income in adulthood.

The potential effect of neighborhoods on children’s future outcomes has spurred a lot of interest in understanding neighborhood income dynamics. “A really striking fact about U.S. cities is that some neighborhoods are very poor compared to others, and the neighborhoods that are poor tend to stay poor,” said Bryan Stuart, an economic advisor at the Philadelphia Fed. “If you go back in time several decades, in basically every city, poor neighborhoods have been poor for a long time. I wanted to understand why that happened.”

In a working paper that Stuart presented at the 2025 Institute Research Conference, he and co-authors Andrew Garin, Ethan Jenkins, and Evan Mast ask if people living in poor neighborhoods have limited opportunities that trap them in poverty. Or are people mobile while some neighborhoods remain trapped?

The research from Fogli, Stuart, and their co-authors sheds light on why Americans have become increasingly segregated by income and how the earnings of those living in America’s poorest neighborhoods evolve. These processes matter for policies that aim to change the trajectories of both people and neighborhoods.

Location as investment

In their paper “The End of the American Dream? Inequality and Segregation in U.S. Cities,” economists Alessandra Fogli, Veronica Guerrieri, Mark Ponder, and Marta Prato develop a theory to explain how residential income segregation contributes to income inequality.

First, parents care about their children’s future wages. This is an assumption, but one that resonates through history, from old tales of parents hoping for “a good match” to modern stories of parents investing thousands in SAT coaches, college admission coaches, and sorority coaches. Future income isn’t the only thing parents care about, but it may be one factor in their decision-making.

Second, there is a connection between parents’ wages and the future wages of their children. Parents with more means can invest in more education, training, and resources to help their children land well-paying jobs. They can also afford to live in neighborhoods that offer more of these opportunities. Data that match the income of children with the incomes of their parents show that children with parents in the top or bottom quintiles of the income distribution have about a 1 in 3 chance of ending up in the same quintile as adults.

Finally, the economists home in on the role of local neighborhood “spillovers,” the idea that investing in children’s education yields higher returns in neighborhoods where children are likely to earn higher incomes. This neighborhood spillover could come from a number of sources, the economists explain. In the U.S., neighborhood household income is linked to the funding of local public schools, which in turn affects school quality. Research by former Institute scholar Francesco Agostinelli shows how peers can influence children’s educational outcomes. And research from Institute advisor Jesse Rothstein points to access to local labor markets as a potentially important factor.

The existence of neighborhood spillovers means the neighborhood where a child lives matters. On average, children growing up in wealthy neighborhoods to rich parents who invest in their education will earn more in adulthood than children who grow up with similarly rich parents who make the same investment in their education but the family lives in a less wealthy neighborhood. The choice of where to live becomes a form of investment in a child’s education.

When more education leads to higher income

The feedback loop between income inequality and residential segregation is sparked when something causes income inequality to increase. In the United States, income inequality has been growing since at least the 1980s. One contributing cause: large wage growth for college-educated workers relative to wages of workers with less formal education. “A great economic divide has emerged between college-educated workers and those with less education,” Institute advisor David Autor and co-authors Claudia Goldin and Lawrence Katz wrote in 2020.

An increase in the college wage premium leads to greater income inequality, all else equal. It also jumpstarts a process of residential sorting. Parents recognize that obtaining a college degree has even more financial payoff than before, so they have even more incentive to locate in neighborhoods with larger local spillovers.

But the supply of housing often cannot expand quickly due to local regulations, construction time, and other factors. With demand increasing faster than supply, housing prices rise. This means only the richest households can afford the neighborhoods with the largest local spillovers. The result is even more residential segregation by income. As richer families concentrate in certain neighborhoods while poorer families concentrate in others, the gap between neighborhood spillovers increases, leading to even more future inequality and lower intergenerational mobility.

Another way to assess the relationship between inequality and segregation is with a counterfactual exercise: What would income inequality look like if households did not sort by income? The economists use their model of how households choose where to live to estimate income inequality after a change in the labor market causes the returns to college to increase, but now households are randomly relocated across neighborhoods. This means neighborhoods are no longer sorted by income, and the local spillover effect is the same in each neighborhood.

In this scenario, inequality still increases because the returns to college have gone up. And parents’ incomes still affect their children’s future wages via the channels described earlier, but location is no longer relevant for future earnings. Using this counterfactual, economists estimate that neighborhood segregation by income explains about 30 percent of the increase in inequality between 1980 and 2010.

How earnings grow across neighborhoods

Fogli’s research helps explain how neighborhood segregation by income can be reinforced when income inequality is rising. But it leaves open the question of what earnings dynamics look like within different neighborhoods. This is the question Stuart’s research tackles.

To define a neighborhood, Stuart and his co-authors use census tracts—relatively small areas within a single county with boundaries that follow visible features (such as roads or rivers.) Standard census tracts have around 1,600 housing units and 4,000 residents. The economists classify a neighborhood as “high poverty” if 30 percent or more of households fall below the federal poverty threshold. In 2024, the federal poverty threshold was $16,320 for a household of one and $32,130 for a household of four.

To study the earnings growth and movement patterns of individuals, Stuart and his co-authors build a dataset from both administrative and survey sources that allows them to identify where individuals live each year and how much they earn. Their analysis of this data reveals three important results.

1. Earnings growth is similar across neighborhoods.

While economists have data on earnings growth for people at different points in the income distribution, there wasn’t good data on earnings growth of people living in different neighborhoods. With their new dataset, Stuart and his co-authors find that earnings growth looks pretty similar for people living in different neighborhoods.

In all four of their neighborhood classifications, between 50 and 60 percent of people have income growth between $1 and $40,000 after 8 years. Around 10 percent experience even higher income growth, and around 35 percent see a decline in earnings. “Our results on earnings dynamics illustrate that there is significant individual earnings growth at all neighborhood poverty levels,” the economists write.

2. People are mobile.

While mobility in America may be declining, it is still relatively high, Stuart and his co-authors find: Approximately 43 percent of all adults move from their starting neighborhood over the next 10 years. This share is even higher, 51 percent, for people who start in a high-poverty neighborhood.

Importantly, mobility rates out of high-poverty neighborhoods vary with age and with family status. Among individuals who are 25 to 35 years old and live in a high-poverty neighborhood, about 70 percent of them will move to a lower-poverty neighborhood at some point in the next 10 years. For individuals who are 55 to 65 years old, only 33 percent will.

Households with children are also more likely to move out of a high-poverty neighborhood than households without children. “What that tells me is that while the effects of poorer neighborhood on children’s outcomes are very real, it’s also the group that is the most mobile,” Stuart said.

3. People with more income growth are more likely to move.

So individuals in both richer and poorer neighborhoods see earnings growth, and individuals in both richer and poorer neighborhoods move. As a third step, the economists look at the relationship between these two patterns. They find that people with larger increases in their earnings systematically move to higher-income neighborhoods. The relationship is stronger for people who begin in higher-poverty neighborhoods, as Figure 1 shows. In fact, the share of individuals who exit high-poverty neighborhoods increases sharply with the size of the earnings increase (Figure 2).

Poor people or poor neighborhoods?

The research from Fogli, Stuart, and their co-authors shows how neighborhoods can become trapped in poverty. Those with the means will move to wealthier neighborhoods, perhaps motivated to provide better educational opportunities to their children. Because people leave when their earnings increase, the median income in the neighborhood doesn’t go up.

This matters for how we think about policies to revitalize America’s poorest places. A policy that successfully raises resident incomes is not likely to revitalize the neighborhood unless it also entices higher-earning folks not to move.

That is the opposite of the goal of the U.S. Department of Housing and Urban Development (HUD) Moving to Opportunity pilot program, which offered families living in public housing projects in high-poverty neighborhoods a voucher to fund their housing in a wealthier neighborhood. While HUD’s analysis found very modest positive effects, subsequent analysis by Chetty and Hendren found that a move for a child below age 13 led to adult earnings that were 31 percent higher than those in the control group. Fogli and her co-authors estimate that scaling up this program would reduce neighborhood income segregation, though it would have only modest effects on income inequality.

Of course, the real answer to “are people or neighborhoods trapped in poverty?” is “both.” If 51 percent of people who start in a high-poverty neighborhood move within 10 years, that means 49 percent do not move.

“For me, this highlights that instead of doing one-off interventions where we pick one part of a city and make it better, we should be thinking about how we can raise the floor for all neighborhoods,” Stuart said. “Even if we do that, we’re going to have a gradient where people with more money are going to live in nicer neighborhoods. That’s not, I think, something that we’re going to be able to fix. But we can make it so that there are acceptable levels of safety and quality and environmental health in even the poorest neighborhoods.”