Rich and poor kids should grow up next to each other

Affordable housing shouldn't mean segregation

Growing up in mixed-income communities can have profound benefits for poor children. (Illustration by The Argument using ChatGPT, photo by Hulton Archive/Getty Images)

Editor’s note: The first essay The Argument ever published asked, “How do we live with each other?” To mark our first anniversary on Aug. 17, we’re returning to that founding question with a series about what we owe people whose beliefs and choices we may find difficult to accept or even understand. Over the next six weeks, familiar voices at The Argument and new voices alike will approach that question from different angles.

Today, Milan Singh takes this question literally by exploring the way that affordable housing policy in the United States incentivizes economic segregation. How should we live with each other? Right next to each other, Singh argues.

To receive every essay in this series, subscribe today.

– Jerusalem Demsas

---

Poor kids in America tend to grow up in a totally different world from middle-class and rich kids. They often grow up in higher-crime neighborhoods, with friends whose parents are unemployed, and attending substandard schools.

Growing up in this separate world can make poverty harder to escape. A growing body of research suggests that where children live — and who they grow up around — can profoundly shape their prospects as adults.

That has massive implications for American affordable housing policy. Most affordable housing in America is old housing. New affordable housing is almost entirely built with the Low-Income Housing Tax Credit (LIHTC) and concentrates poverty in 100% affordable housing projects in high-poverty neighborhoods.

Projects with 100% affordable housing can sound good to some people. But in practice, they mean segregating low-income families away from the very neighborhoods that would give their children equality of opportunity.

To others, the idea of allowing low-income families to move into their neighborhood sounds like a threat. Don’t poor people bring crime? Won’t they undermine good public schools? Community cohesion?

I want to make the case for living right next to each other by building class-integrated communities. Currently, U.S. housing policy isn’t well designed to optimize for this goal. But the benefits of integration — for the rich in addition to the poor — are likely large enough that we should refocus housing policy toward promoting the development of mixed-income neighborhoods.

Location, location, location

When it comes to low-income housing, location is not just another amenity like in-unit laundry or plentiful natural light. That’s because research suggests that whether poor families can achieve upward mobility largely depends on where they live.

From 1994 to 1998, the Department of Housing and Urban Development ran an experiment called “Moving to Opportunity” (MTO), where families living in high-poverty neighborhoods were given vouchers to move into very low-poverty neighborhoods. MTO provided ideal conditions to study the effects of place-based low-income housing policy both because of the large number of participants (over 4,600 families) and because detailed tax data on these families allowed researchers to track children’s adult outcomes.

In a 2016 paper, economists Raj Chetty, Nathaniel Hendren, and Lawrence F. Katz studied the adult economic outcomes of poor children who moved as part of the program compared with those who didn’t. Children who moved into low-poverty neighborhoods before turning 13 went on to earn 34% more on average than children who were left behind. The former were also more likely to attend college (and to attend better colleges compared with the college-bound in the control group), more likely to live in high-income neighborhoods as adults, and less likely to become single parents.

In fact, the lifetime earnings increases for children in the MTO group were so large that the program paid for itself: The increase in tax revenues was greater than the initial voucher costs.

Follow-up research established that these beneficial effects were mainly driven by the children’s exposure to their new neighborhoods.

If a child from a poor neighborhood moved to a high-income neighborhood at birth, that would wipe out 80% of the expected earnings gap between them and the children born in the rich neighborhood. The benefits of moving to greener pastures scale with the amount of time spent in the new neighborhood.

But what exactly is going on here? What about these neighborhoods is changing the fortunes of these children so dramatically? It might be integration, but it could also just be better schools or low crime rates. If it’s the latter, then the solution might be educational reforms or better policing. We could just try to improve poor neighborhoods, in other words, without worrying about socioeconomic integration.

But in follow-up research published in 2022, Chetty and a different team of co-authors found that what really matters is something a lot harder to envision without thorough integration: cross-class friendships.

Using anonymized data from 21 billion Facebook friendships, Chetty and his team were able to measure the degree of cross-class friendships in locations across the country — basically, the share of a poor person’s friends who are rich.

They tested a bunch of things to see what was driving these positive benefits. Traditional measures of social capital like volunteering, the presence of civic organizations, and even the presence of tightly knit friendship networks mattered little for predicting whether poor kids would do better as adults. What mattered most was the prevalence of cross-class friendships.

The mechanism by which cross-class friendships generate greater upward mobility seems to be through exposure to adult role models. In a 2025 paper, Chetty found that the strongest predictor of children’s adult economic outcomes was the employment rates of the parents of other children their age in their neighborhoods — parents of the cohort they are most likely to make friends with. In places with increasing parental employment rates, adult outcomes for children improved; in places where parental employment was falling, children did worse as adults.

Integration is probably positive-sum

I grew up in Cambridge, Massachusetts. My high school is in the 87th percentile for cross-class friendships, using Chetty’s metric. Cambridge is an increasingly expensive place to live, but that’s largely a result of its economic success running up against the relatively inelastic supply of housing.¹

Growing up, all of my friends’ parents were employed (in a variety of occupations), and none of them used drugs or went to prison. The public school system was pretty good, and there were lots of adults around who had gone to Harvard, Yale, or MIT. This made me think that, since I was generally good at school, I could do this too, provided I applied myself.

Though I may not have actually needed this network — after all, my parents went to good colleges too — I still benefited.

I also grew up with friends who lived in public housing, whose parents had blue-collar jobs, or who chose to join the military instead of going to college after high school. I went to a fancy college where most of my classmates were from families like mine, but I’m very grateful that some of my closest friends are from families who aren’t — who have ambitions and dreams that are very different from my own. My perspective on life is all the richer for having grown up with them.

I share my own experience in part because it’s hard to imagine the socioeconomic integration of high-opportunity neighborhoods without the buy-in of the comfortable families who already live there — people who worry that an influx of poor families might change the character of their neighborhoods in ways they don’t want.

Concerns about the economic, fiscal, and criminal impacts of newcomers aren’t always the pure result of racism or classism. For example, an influx of new students into local public schools — students who may be below grade level in certain subjects — may strain teachers and reduce the amount of attention given to longtime residents’ children.

A 1976 U.S. Commission on Civil Rights report contains contemporaneous interviews with students, parents, and administrators going through integration. White students reported fear and anxiety about entering a different school with protesting adults causing chaos. They also complained of long and inconvenient transportation.

One Tampa student had to wake up at 4:45 a.m. to take the assigned bus. But, overall, reading through the report reveals that students were frustrated by media reports exaggerating the racial overtones of ordinary conflict and reported positive experiences with integration. One white Stamford student even called integration “the best thing that ever happened to me.”

But the research we have on both school and residential integration suggests that such concerns are probably overblown.

During Jim Crow, Black children were far worse off than white children on average: Their families were poorer, their life expectancies were lower, and their neighborhoods had much higher violent crime rates. There were widespread fears that the socioeconomic integration implied by racial desegregation would have negative effects on white students if their schools became more exposed to the problems of poverty.

Instead, research showed that court-ordered school desegregation “significantly increased both educational and occupational attainments, college quality and adult earnings, reduced the probability of incarceration, and improved adult health status” for Black people whose schools were integrated — while having “no effects on whites” who attended the same schools.

Of course, “no effects” just refers to the sorts of things economists can measure, like standardized test scores or earnings or health outcomes. Subjective experiences matter too. A 2004 qualitative study based on interviews with 242 graduates of desegregated high schools, including 136 white graduates, is useful here.

The white graduates described greater appreciation for other cultures, greater comfort and social competence in diverse settings, less fear of people of color, and less reliance on racial stereotypes. It’s good to take this with a grain of salt — few people in 2004 were going to admit to having hated integration — but this tracks with my experience as well: It’s nice to grow up and live with people who are different from you.

According to a different 2004 paper in the American Economic Review, there is “no evidence of an impact of [busing Black students to formerly white schools] on white students in receiving districts.” In particular, test scores for incumbent white students showed no decline.

Of course, the applicability of race-based school integration to class-based residential integration may be limited. But a recent paper from Chetty’s research lab, Opportunity Insights, provides further evidence that the positive effects of integration do not come at the advantaged group’s expense.

The paper analyzed HOPE VI — a federal program intended to turn distressed public housing projects into mixed-income communities — and found that, when wealthier people move into the revitalized public housing projects, the low-income neighborhoods hosting the projects become more socially integrated, and resident children earn more as adults.

And within wealthy neighborhoods, there’s no correlation between greater class integration and how much rich children earn as adults. In other words, the evidence suggests that integration is probably positive-sum overall.

To be clear, these findings are not dispositive evidence against every potential downside of neighborhood integration. I don’t think we can eliminate the possibility that, past a certain level, integration has costs for rich children. What we do know is that gradual efforts at integration — the kind tested in the MTO experiment — are unlikely to impose substantial costs on rich kids and will provide very large benefits for poor children.

Beyond public vs. private

There are two ways the government could promote economic integration: It could build and operate mixed-income housing itself, or it could provide vouchers to poor families so they can afford a place in a nicer neighborhood. Because vouchers are a more efficient use of public funds, and because of the benefits of economic integration, we should prefer the latter option.

But vouchers alone aren’t enough to achieve economic integration. After all, poor families might not be aware of which neighborhoods provide the greatest boost to economic mobility, or they may not have the time to tour apartments in faraway neighborhoods and fill out applications. Landlords may be unwilling to rent to voucher-holders.

In a 2018 to 2019 experiment, Chetty’s team at Opportunity Insights partnered with public housing authorities in Seattle and King County, Washington. In the first phase, one group of housing voucher recipients was provided with information about high-opportunity neighborhoods, financial assistance, customized support services during the moving process, and connections with landlords; the other group was just provided with housing vouchers. Among the group receiving the full support services, the share that moved to high-opportunity neighborhoods was 53%, compared with 15% in the control group.

The intervention in the experiment is estimated to have increased lifetime earnings by an average of $212,000. After accounting for increased tax revenue and higher voucher spending caused by families moving to areas with higher rents, the researchers found that each dollar of government spending on the program increased the incomes of voucher recipients by $1.35.

That sounds pretty good to me, but there is a third option for achieving integration: the government could simply dictate where people live. Singapore has been able to do it, but only by having 80% of the population live in public housing subject to strict quotas to force ethnic integration. That is, to put it mildly, unlikely to fly in the United States.

This is a free country, and people should have the freedom to live where they want to. But, to the extent that the government is in the business of subsidizing housing for the poor — and I think it should be — policy should encourage greater integration.

We are very far from that at the moment. The Low-Income Housing Tax Credit — the main way we build new affordable housing in this country — financed just under 4 million affordable housing units between 1986 and 2024. And it incentivizes building 100% low-income housing in neighborhoods that are already poor.

Without going too deep into the weeds, the basic reason why this happens is that developers must pay substantial fixed costs to participate in the LIHTC program; these costs include hiring attorneys to ensure that projects meet state-set standards when applying for the credits, arranging financing, and verifying the incomes of tenants. Crucially, these costs are essentially fixed regardless of the share of units in a project that are affordable (meaning leased at below-market rates), while the tax credits can only be applied toward the costs of building the affordable units.

This makes it unrealistic to build mixed-income projects with LIHTC money, since the developer would be paying the full costs of compliance while only receiving a partial subsidy.² There are also incentives to locate projects in high-poverty neighborhoods.

The structure of LIHTC does more than discourage mixed-income housing. It uses public money to reinforce the economic segregation that housing policy should be trying to overcome.

We do not need to dictate where people live. But we should stop making segregation the default. Affordable housing policy should give poor families a real chance to live in high-opportunity neighborhoods — and give poor, middle-class, and rich kids a chance to grow up in the same world.

Recommended reading:

1. This is because of zoning.

2. There are two other factors worth discussing briefly. First, because arranging LIHTC deals is incredibly complicated — take a look at Figure 1 on page 5 of this Urban Institute report — a highly specialized industry of financial institutions, law firms, and affordable housing developers has sprung up. These firms are specialized in building 100% affordable projects, so they may lack the specialized knowledge required to build mixed-income projects. Second, it is not clear whether there is that much market demand for mixed-income buildings. It is plausible that many rich people don’t want to live in a building with a high share of affordable units. This isn’t necessarily a problem; based on the Chetty research, what matters for upward mobility is integration at the neighborhood, not at the building level.