Residential Segregation and the Black-White College Gap

Economic research shows that the neighborhood a child grows up in profoundly impacts a range of adult outcomes, including college attainment, employment, and intergenerational mobility.For examples of this research, see Raj Chetty and Nathaniel Hendren’s 2018 paper on intergenerational mobility and Eric Chyn’s 2018 paper on public housing. Hannah Rubinton and Maggie Isaacson also showed that, in cities across the country, there is a positive statistical relationship between neighborhood segregation and the gap between Black and white children in their level of college-degree attainment; that is, when segregation increases, the gap increases. Finally, several papers examine human-capital spillovers in quantitative spatial equilibrium models (e.g., Dionissi Aliprantis and Daniel Carroll’s 2018 paper, and Eric Chyn and Diego Daruich’s 2022 paper), and others examine the interaction between endogenous amenities and sorting by income but do not consider race (e.g., Milena Almagro and Tomas Dominguez-Iino’s 2022 article; Victor Couture, Cecile Gaubert, Jessie Handbury and Erik Hurst’s 2019 working paper; and Matthias Hoelzlein’s 2023 paper (PDF)). In the case of St. Louis, economist Raj Chetty and other researchers documented a Black-white gap in college attainment of 28 percentage points in their dataset; specifically, for children who grow up in the St. Louis commuting zone,A commuting zone is a geographic area designed to reflect part of the local economy where people live and work. Commuting zones generally include multiple counties, and their counties may not match all those found in related metropolitan statistical areas. 47% of white children will earn a college degree, while only 19% of Black children will.

At the same time, de facto residential segregation by race is a predominant feature of many American cities, and as a result, there is substantial racial inequality in exposure to advantageous neighborhoods [e.g., Patrick Bayer, Kerwin Kofi Charles and JoonYup Park’s 2021 article (PDF)]. According to a measure of neighborhood segregation—the dissimilarity index—the St. Louis commuting zone ranks as the eighth* most segregated in the country. A recent St. Louis Fed working paper by Victoria Gregory, Julian Kozlowski and Hannah Rubinton explores whether ongoing racial differences in neighborhood sorting can account for the Black-white gap in college attainment and residential segregation even after the end of de jure segregation.For further details, see their recently revised 2022 working paper, “The Impact of Racial Segregation on College Attainment in Spatial Equilibrium.”

Neighborhood Segregation in St. Louis

The maps below show the share of graduating high school students who enroll in college and the share of all Black students for school districts in St. Louis City and St. Louis County for 2020. There is a strong, negative correlation of -0.63 between the two variables. This suggests a strong relationship between segregation and the racial gap in college attainment in St. Louis.

Share of Students Enrolling in College, by St. Louis School District, 2020

Share of Black Students, by St. Louis School District, 2020

SOURCE FOR BOTH MAPS: Gregory, Kozlowski and Rubinton, 2022.

NOTES FOR BOTH MAPS: The first figure shows the share of graduating high school students in the school district who go on to attend a four-year college in 2020. The second figure shows the share of all Black students who attend the school district. The boundaries mark the city’s school district and the school districts that are at least partly within St. Louis County.

Modeling the Impacts

In their working paper, Gregory, Kozlowski and Rubinton built a model in which agents (i.e., individuals) live for two periods: childhood and adulthood. Adults have a child and decide how much to invest in their child’s education. The child chooses whether to go to college and then becomes an adult in the following period. The model incorporates household race, which can be either Black or white for the purposes of the study. Finally, agents choose a neighborhood to live in; these neighborhoods differ by rental prices, spillovers and racial composition. Neighborhood spillovers summarize a variety of factors that can influence young agents, such as peer effects, quality of local schools or networks.

The authors found that three key forces can account for the racial segregation and for about 80% of the Black-white gap in college attainment (graduation from a four-year college) in the data for the entire St. Louis metro area: (i) a Black-white wage gap, (ii) an amenity externality (described later in this post) and (iii) additional barriers to moving for Black households.

First, the Black-white wage gap is well documented in economic literature. Gregory, Kozlowski and Rubinton used the National Longitudinal Survey of Youth (NLSY) and ran a simple regression of earnings on race, college education, skill and gender. For the St. Louis area, they found a Black-white wage gap of about 8% both for workers with a four-year college degree and for those who did not graduate from a four-year college.

Second, an amenity externality can be understood in a few ways. Most obviously, a neighborhood may have better schools or parks. An important amenity the authors consider are the household preferences for the racial composition of the neighborhood—for example, fear of discrimination in an all-white neighborhood, white flight, or homophily.In the model, the amenity externality captures the interaction between: (i) local amenities like parks or riverfronts, and (ii) the racial composition of the neighborhood. For example, a Black household in an otherwise all-white neighborhood may fear discrimination and enjoy their neighborhood park less. Third, there are many examples of additional barriers to moving for Black households, such as real estate agents showing fewer houses to prospective minority renters or buyers.For more examples of racial discrimination in moving, see the report Housing Discrimination Against Racial and Ethnic Minorities 2012.

Possible Sources of Residential Segregation and College Attainment Gap

The authors also ran a number of counterfactuals and found that all three factors—the Black-white wage gap, the amenity externality and the additional barrier to moving for Black households—are quantitatively important for the model to match the college attainment gap and segregation in the data, but for different reasons. The wage gap means that Black and white households have different monetary resources. This impacts parents’ investments in their children's education, which is part of the spillover that creates positive externalities within the neighborhood. In contrast, the amenity externality has more of an impact on neighborhood choice. Without it, households choose to live in more integrated neighborhoods, which means that Black households live in neighborhoods with higher positive spillovers. The additional barriers to moving for Black households play a similar role to the amenity externality, but with smaller quantitative effects.

Exiting the Segregation Trap

The presence of spillovers and amenity externalities imply that the model may have multiple equilibria; that is, there are different configurations of neighborhoods that can be attained by the society. With multiple equilibria, the economy may be able to coordinate and transition from one equilibrium to another equilibrium and improve agents’ outcomes in doing so. The authors found two main equilibria in the model: a more segregated equilibrium, which is representative of St. Louis, and a more integrated equilibrium.

Gregory, Kozlowski and Rubinton found overall gains of moving from the segregated to the integrated equilibrium. However, there is heterogeneity, with some “winners” and some “losers.” For example, welfare gains are larger for Black households of any educational attainment and for white noncollege households, while white college households experience welfare losses. Using their model, the authors found that this transition, which could be achieved if agents were able to coordinate, would be relatively fast and the majority of changes would occur within one or two generations.

*Editor's Note: This post was updated on Feb. 5, 2024, to correct St. Louis’ ranking among commuting zones.

Notes

1. For examples of this research, see Raj Chetty and Nathaniel Hendren’s 2018 paper on intergenerational mobility and Eric Chyn’s 2018 paper on public housing. Hannah Rubinton and Maggie Isaacson also showed that, in cities across the country, there is a positive statistical relationship between neighborhood segregation and the gap between Black and white children in their level of college-degree attainment; that is, when segregation increases, the gap increases. Finally, several papers examine human-capital spillovers in quantitative spatial equilibrium models (e.g., Dionissi Aliprantis and Daniel Carroll’s 2018 paper, and Eric Chyn and Diego Daruich’s 2022 paper), and others examine the interaction between endogenous amenities and sorting by income but do not consider race (e.g., Milena Almagro and Tomas Dominguez-Iino’s 2022 article; Victor Couture, Cecile Gaubert, Jessie Handbury and Erik Hurst’s 2019 working paper; and Matthias Hoelzlein’s 2023 paper (PDF)).
2. A commuting zone is a geographic area designed to reflect part of the local economy where people live and work. Commuting zones generally include multiple counties, and their counties may not match all those found in related metropolitan statistical areas.
3. For further details, see their recently revised 2022 working paper, “The Impact of Racial Segregation on College Attainment in Spatial Equilibrium.”
4. In the model, the amenity externality captures the interaction between: (i) local amenities like parks or riverfronts, and (ii) the racial composition of the neighborhood. For example, a Black household in an otherwise all-white neighborhood may fear discrimination and enjoy their neighborhood park less.
5. For more examples of racial discrimination in moving, see the report Housing Discrimination Against Racial and Ethnic Minorities 2012.