3

Space and Place

Decades of research on intergenerational income mobility in the United States have focused on the challenge of estimating the degree to which economic status in one generation is passed on to the next. But newly available data suggest that any research efforts to understand economic mobility must consider the extent to which it varies by geography—a single, national estimate of economic mobility obscures the substantial heterogeneity across neighborhoods, cities, and regions of the country.

Using data from the Internal Revenue Service, a series of studies conducted by Raj Chetty along with several collaborators (Chetty & Hendren, 2018; Chetty et al., 2014a) has demonstrated that in some sections of the United States, such as the North Central states, spanning from Utah to Minnesota, children from low-income families have a reasonable chance of rising upward in the income distribution when they reach adulthood, while in other sections, including much of the Southeast, children from similar families are much more likely to remain poor as adults (Chetty et al., 2014a; Graham & Sharkey, 2013; The Pew Charitable Trusts, 2012). Even within the same commuting zones and counties, there is wide variation across Census tracts in children’s chances of upward mobility (Chetty et al., 2018). This research demonstrates that upward mobility varies substantially across America’s metropolitan areas and among neighborhoods within U.S. towns and cities.

This chapter examines the spatial dimension of intergenerational mobility in the United States. Space means the set of people (neighbors, extended family, romantic partners), institutions (schools, libraries, police departments, hospitals), processes (social interactions, community organization,

political activity), resources (public spending, community wealth), and hazards (crime, pollutants, water quality) in the environment outside the home. Although most of the relevant evidence is currently at the Census tract level, research is increasingly being produced at broader levels, such as counties, cities, commuting zones, states, and regions.

The focus on the link between space and intergenerational mobility is motivated by two basic claims. First, inequality across multiple domains is organized, to varying degrees, along spatial lines. Some of these lines were formed historically and demographically by forces such as immigrants from similar countries choosing to live in similar places in the United States or groups of people with similar interests, such as jobs, choosing to live together. But some lines were drawn intentionally to exclude people from housing markets on the basis of race or to maximize local resources in school districts. Some lines, such as zoning codes, are drawn to regulate what is built, which in turn affects who can live in a given community (Dreier et al., 2014; Jargowsky, 2015; Lens, 2022). Second, research across disciplines converges to conclude that the spatial organization of inequality has causal influences on the probability of upward or downward economic and social mobility. However, as discussed below, the exact underlying mechanisms remain unclear. Space may affect economic mobility through multiple pathways: (a) directly, through exposure to resources, opportunities, risks, and hazards in the individual’s environment (quality schools, social networks, labor market opportunities, violence), and (b) indirectly, through social interactions, preferences, norms, and perceptions that are conditioned by the residential environment (Galster & Sharkey, 2017). To date, most scholarship linking spatial factors with economic and social mobility outcomes considers absolute mobility (i.e., increase or decrease in economic well-being across generations). For that reason, the chapter will use the terms mobility and absolute mobility interchangeably and will clarify when the analysis refers to relative mobility.

The chapter begins with a review of the ways social scientists have studied the spatial dimension of economic and social mobility over time, before considering mechanisms that underlie the geography of mobility. It concludes by considering the implications of this evidence for public policy and for future directions for social science.

SPACE AND MOBILITY

The resurgence of interest in the role of residential environments in influencing economic and social mobility is often traced back to The Truly Disadvantaged, in which William Julius Wilson (1987) described how a set of economic, demographic, and social forces had led to a new form of concentrated poverty that was most pronounced in Black neighborhoods

across U.S. cities. Wilson argued that the concentration of poverty and the decline of employment opportunities in central city neighborhoods contributed to deteriorating social institutions and shifts in cultural norms, which amplified the effects of family poverty. In other words, it was difficult to grow up poor but worse to grow up in a neighborhood where most other families were also poor (see also Wilson, 1996).

The scientific turn toward understanding the role of urban neighborhoods in influencing the life chances of residents was also catalyzed by Massey and Denton (1993) on the persistence and consequences of racial and economic segregation for racial inequality, and later, research on the precarity of middle-class status for Black Americans (Pattillo, 1999), urban violence (Anderson, 1999), the spatial location of jobs (Kain, 1992), and the role of public policy in addressing the challenges of concentrated poverty (Goering & Feins, 2003).

While much of the early quantitative research on the impact of residential environments struggled to overcome the methodological challenges associated with nonrandom selection into neighborhoods and cities, recent evidence has led to something close to a consensus in support of the claim that residential environments can have a causal impact on economic mobility and on other intermediate outcomes related to mobility—most notably academic achievement, cognitive skill development, and health (Sharkey & Faber, 2014). Increased access to geocoded data, improved tools for spatial analysis, and the rise of quasi-experimental and experimental research designs undergird this view.

One prominent example of experimental research designs that demonstrates that neighborhoods matter comes from the Moving to Opportunity (MTO) program, a large-scale social experiment implemented by the U.S. Department of Housing and Urban Development (HUD), which randomly assigned housing vouchers with different restrictions to families in public housing in five cities (Goering & Feins, 2003). Results from the experiment were complex: take-up of the offer of a voucher was low in several sites; some families who used the voucher moved back into higher-poverty neighborhoods over time; and there was substantial heterogeneity in the program’s impact depending on origin neighborhood conditions, the outcome being studied, the gender of children in the family, and the age of children when the family moved (Burdick-Will et al., 2011; Chetty et al., 2016; Ludwig et al., 2012; Rosenblatt & DeLuca, 2012; Sampson, 2008; Sanbonmatsu et al., 2011). An influential reanalysis of the program’s impact on children’s long-term economic outcomes showed that children who moved to a low-poverty neighborhood at a young age (younger than 13) experienced meaningful increases in adult earnings (i.e., upward economic mobility), along with a host of other positive outcomes (Chetty et al., 2016). For this subgroup of families with young children, successfully

using a voucher to move to a low-poverty neighborhood leads to $3,477 higher earnings per year in early adulthood.

Other studies using quasi-experimental research designs have produced strong evidence that moving to a relatively low-poverty neighborhood has a causal effect on economic outcomes or intermediate outcomes, such as high school dropout rates. Chyn (2018), for instance, compared outcomes of children in Chicago public housing developments that were demolished to nearby children in developments that were not demolished and found positive impacts on high school dropout rates, employment, and labor market earnings. Similarly, a 40-year follow-up of the Chicago Gautreaux program—a desegregation remedy that helped Black families in public housing move to predominantly White neighborhoods across the Chicago metropolitan area—builds on and confirms earlier evidence (e.g., DeLuca et al., 2010) that the program increased children’s lifetime earnings, employment, and wealth, supporting upward mobility (Chyn et al., 2023).

A few studies demonstrate that exogenous changes in individuals’ environments can lead to shifts in their prospects for economic and social mobility. However, other research using experimental or quasi-experimental designs leads to somewhat different conclusions. Initial analyses from the full sample of MTO families found no effects on children’s test scores and later earnings (Sanbonmatsu et al., 2011); however, a subsequent analysis showed that the impact on cognitive skills varied widely across the five sites (Burdick-Will et al., 2011). Chyn (2018) found positive effects of moving out of highly disadvantaged public housing developments among both younger children and older children, while Chetty et al. (2016) found positive effects of moving out of high-poverty neighborhoods only for children who moved at a young age.

One of the limitations to the research on place-based factors affecting economic mobility is that most of the evidence comes from urban areas. There is much less evidence on the factors affecting economic mobility within or between suburban or rural areas. Evidence generated shortly after the initial work by Chetty et al. (2016) suggests that there may be different mechanisms at work in rural areas that affect economic mobility. For example, the racial makeup of a county, the quality of the schooling, and social capital—that is, the resources that come as a result of one’s social relationships in the community, particularly those from different backgrounds and income levels—appear to be even more important in rural areas than in urban areas (Weber et al., 2017). County-level labor market conditions and the outmigration of young adults between decennial censuses is also associated with positive upward mobility in rural areas (Krause & Reeves, 2017). Using measures such as the percent of the population with a high school degree and other factors of production, such as the per capita value

of capital in an area, Islam et al. (2015) showed that lower levels of human capital account for almost 80 percent of the income differential between persistently poor counties and the rest of the country, providing more evidence for the importance of quality schooling in these areas. Finally, a historical analysis by Edin et al. (2023) of extreme disadvantage in rural communities shows that a common set of factors tends to be found in places characterized by persistent poverty and limited upward mobility: segregated schools, violence, poor social capital, corruption, structural racism, and poor economic opportunities.

These heterogeneous findings suggest the need to go beyond the basic question of whether the neighborhood influences economic and social mobility and instead consider how residential environments and larger urban and rural settings provide opportunities for upward mobility, how mechanisms differ by environment, and who is most affected. The impact of the residential environment on individual life trajectories is a product of conditions in the immediate neighborhood and the larger area and region of the country, the timing and duration of individuals’ exposure to these environments, and individuals’ characteristics and unique vulnerabilities (Harding et al., 2011). In short, it is time to move toward questions of when, where, why, how, and for whom residential contexts matter (Sharkey & Faber, 2014).

MECHANISMS LINKING SPACE AND ECONOMIC AND SOCIAL MOBILITY

While scholarship on the “effects” of place has grown considerably over time, less attention has been paid to the mechanisms that undergird these effects and their heterogeneity—vital considerations for using research to improve policy. In this section we review evidence at the neighborhood level that points to schools, community violence, and local social networks as examples of central mechanisms linking the local residential environment with economic and social mobility. This review is not designed to be exhaustive, as there is evidence for numerous additional pathways by which the residential environment may affect upward mobility through, for instance, children’s exposure to environmental toxins (Mohai et al., 2009; Muller et al., 2018) or through exposure to institutions such as the jail and prison system (Western & Pettit, 2010). While we focus on key factors through which scholars believe place has effects on economic and social outcomes, we emphasize that more research is needed, specifically research that directly identifies how these mechanisms work to produce outcomes that foster upward mobility by improving human and social capital.

Schools

Because school quality and neighborhood advantage tend to be highly correlated (largely a function of public school catchment areas being zoned by residence), studies exploring schools as a key causal mechanism for economic and social mobility require designs that employ exogenous variation in school quality. H. Schwartz (2010) analyzed data on test performance among students from low-income backgrounds in Montgomery County, Maryland, one of the wealthiest urban school districts in the nation. Montgomery County is unique not only because of the quality of its public schools and the funds directed toward students from low-income backgrounds, but also because it features the nation’s oldest and most extensive inclusionary zoning program. As part of its zoning policy, the county’s housing authority is able to purchase up to one-third of the units set aside by developers to be rented or sold at below-market rates. The housing authority randomly assigns families selected for housing assistance to these units, which are scattered across all neighborhoods and school attendance zones throughout the county, affording some children from low-income backgrounds a chance to attend higher-performing and more diverse schools than those in their previous neighborhoods.

Exploiting this random assignment, Schwartz (2010) estimated the effect of attending elementary schools with relatively low levels of student poverty versus moderate or high levels of poverty. The study tracked academic performance among 850 students from low-income backgrounds over 5–7 years and found that students in low-poverty elementary schools performed 0.4 standard deviations higher in math and 0.2 standard deviations higher in reading than similar students assigned to schools with 20 percent or higher poverty rates. By the end of elementary school, the gap between students from low-income backgrounds assigned to low-poverty schools and their peers in the larger student body had been cut by half in math and by a third in English. Importantly, because of the unique ability to compare the outcomes of low-income students in relocated students’ original schools (which receive significantly more targeted resources than more affluent schools) to those that attended schools with more socioeconomic diversity (by virtue of the assignment to subsidized units through the program), the study suggests that one’s peers seem to matter more than high levels of school funding. Given the association between test scores, educational attainment, and later earnings, this study provides strong evidence that the socioeconomic composition of school peers contributes to upward mobility among children from low-income backgrounds.

Dobbie and Fryer (2011) analyzed the effect on academic performance of attending a charter school run by the Harlem Children’s Zone (HCZ). The HCZ is a community organization targeting a roughly 100-block area

of Harlem with high-quality social services, schools, and programs for youth and families. In order to identify the causal effect of attending an HCZ school, the authors exploited the fact that attendance at the HCZ Promise Academy Schools was based on a lottery among all applicants. As a second identification strategy, the researchers used variation in the probability of attending HCZ schools derived from the interaction of the student’s address and birth cohort. The study found that both older and younger students who attended a Promise Academy experienced substantial improvement in English and math performance and were less likely to be absent from school. Effect sizes ranged from one-quarter to four-fifths of a standard deviation, with larger gains in math.

Johnson and Nazaryan (2019) found that exposure to desegregated schooling—leveraging quasi-random variation in the timing of desegregation court orders—increased Black children’s long-term economic prospects, without reducing the prospects of White children. Through increases in school quality and the funding ushered in by desegregation, Black children had higher educational attainment, lower rates of incarceration, reduced risk of poverty, and higher lifetime earnings, signaling substantial upward mobility.

This literature provides support for the idea that schools are a central mechanism through which the residential environment affects children’s academic and developmental trajectories. It is also clear that the neighborhood and school settings likely have independent effects on children’s prospects for upward mobility (e.g., Rich & Owens, 2023; Wodtke et al., 2023).

Community Violence

A growing literature suggests that children’s exposure to community violence affects outcomes related to cognitive functioning and academic performance (Harding, 2009; Sharkey et al., 2014). The central challenge in estimating the impact of community violence on children’s outcomes is overcoming the problem of nonrandom selection into violent neighborhoods. To address this problem, Sharkey (2010) exploited exogenous variation in the timing of children’s interview assessments and local homicides to identify the effect of extreme local violence on children’s performance on cognitive skills assessments. He found that children perform substantially worse on cognitive skills assessments if they are given the assessments in the immediate period of 4–7 days following a local homicide.

Other studies have exploited exposure to specific incidents or episodes of violence, such as a school shooting or a series of random sniper shootings, and found that children in closer proximity to violence exhibit more extensive symptoms of posttraumatic stress and worse performance in school (Gershenson & Tekin, 2018; Sharkey, 2010; Sharkey et al., 2012, 2014). International research from Mexico and Brazil shows that school

failure rises and student test scores decline during periods of intense violence driven by gang warfare (Caudillo & Torche, 2014; Monteiro & Rocha, 2017).

While most of this research focuses on proximate outcomes, such as cognitive development and academic achievement, that are mediators of economic and social mobility, Sharkey and Torrats-Espinosa (2017) use shocks in the timing of federal law enforcement grants to localities to estimate the direct effect of changes in exposure to violence during young adulthood on upward mobility. They find that living in counties where violence was falling during the 1990s led to a meaningful increase in adult income for young people raised in low-income families.

Social Networks

Connections to nearby peers, friends, and neighbors is one of the fundamental mechanisms through which neighborhoods influence culture, norms, aspirations, information, and resources that may ultimately translate into upward or downward mobility. Much of the research focusing on networks and economic and social mobility is correlational, however, and identifying the causal impact of close connections has proven challenging (Anelli & Peri, 2019; Black et al., 2013; Carrell & Hoekstra, 2010; Lavy & Schlosser, 2011).

A recent set of papers (Chetty et al., 2022a,b) leveraged billions of Facebook friendships and found that children growing up in places with greater “economic connectedness”—that is, where more low-income people are friends with high-income people—tend to have greater economic mobility. Economic connectedness requires people with low income to be exposed to people with high income, as well as for people of both groups to be willing to forge those friendships. This finding affirms a century of social science in which social capital has been theorized as a key resource supporting economic advancement (Coleman, 1988; Du Bois, 1898; Granovetter, 1973).

Chetty et al. (2024) report that for children born between 1978 and 1992, earnings gaps decreased between Black and White people with low income, while earnings gaps increased between White people with high or low income. A proposed mechanism is the employment of same-race parents nearby, so that places with increases in parental employment also saw increases in children’s income ranks and vice versa. The paper shows that these diverging trends result from changes in childhood environments proxied by parental employment rates within communities defined by county, race, and class. Importantly, the study by Chetty et al. (2024) suggests that while one’s own parents matter, so do the other parents one grows up around, a finding consistent with an earlier paper pointing to the significance of two-parent families in neighborhoods as a key factor predicting

upward mobility for Black children—even for those who themselves have been raised by a single parent (Chetty et al., 2020).

Mechanisms at the Level of Cities, Urban Areas, and Commuting Zones

In their initial analysis exploring the features of commuting zones linked with economic mobility, Chetty et al. (2014a) found that rates of single parenthood and high school dropout, prevalence of violent crime, measures of social capital, overall economic inequality within commuting zones, and the degree of racial and economic segregation were linked with economic mobility. Although this study was not designed to establish causal relationships, it provides a starting point for considering likely mechanisms operating at levels wider than the immediate residential environment. The present report next focuses on two dimensions of the larger geographic environment: labor market conditions and racial and economic segregation.

Job loss is among the important “mechanisms that affect mobility and define opportunity structures” (Brand, 2015, p. 1). One strand of research exploits shocks in local economic conditions to examine the impact of employment on outcomes that are mediators of upward mobility, such as youths’ academic achievement and educational attainment. To identify the effect of local job losses on children’s academic performance, Ananat et al. (2011) exploited factory plant closings in North Carolina counties and found that changes in local economic conditions arising from plant closings have large effects on children’s reading and math scores throughout the area. Other research has found that large-scale transformation of the local economy, such as the opening of casinos on Native American reservation lands or the regional development investments made as part of the federal Tennessee Valley Authority, can have substantial positive effects on educational attainment, income, and employment (Akee et al., 2010; Kline & Moretti, 2014; Wolfe et al., 2012).

Local labor markets have been shaped in significant part by decades of macroeconomic transformation due to globalization and trade policies (see, e.g., Autor et al., 2021; Green & Sanchez, 2007). Seltzer (2024) finds that “declines in manufacturing have contributed to growing disparities in upward intergenerational income mobility” (p. 1223), with children living in counties that experienced large contractions in manufacturing throughout their adolescence facing larger economic penalties in adulthood. Similarly, Berger and Engzell (2022) use “data on local labor markets in the United States to document that automation significantly has reduced the chances for upward mobility among children born in low-income families in the early 1980s” (p. 11). The authors posit that automation affects children’s economic attainment by eroding the ability of families to invest in their human capital.

Turning to racial and economic segregation, Graham and Sharkey (2013) used data from three U.S. surveys that include at least two generations of family members and found a robust and consistent association between economic segregation and relative economic mobility (measured by the intergenerational income elasticity [see Chapter 1]) across all three samples. While this result is correlational, a small number of studies have made meaningful progress in generating causal evidence on the historical processes and policies leading to segregation and migration and their impact on economic mobility. For example, Aaronson et al. (2021a) analyzed the long-term effects of boundaries used by the federal Home Owners Loan Corporation to develop an appraisal process that used neighborhood-level characteristics to assess the risk of making home loans in cities across the country, a process that came to be known as “redlining.” Using multiple identification strategies, Aaronson et al. (2021b) found that the grades representing the riskiness of different neighborhoods had substantial causal effects on economic mobility for cohorts born many decades later (in the late 1970s and early 1980s).

These findings reflect the large body of evidence showing how active efforts to maintain and reinforce racial segregation have had long-term consequences on the trajectories of neighborhoods and the long-term educational and economic outcomes of different segments of the population, particularly Black Americans (Faber, 2020; Massey & Denton, 1993; Rothstein, 2017; Sharkey, 2013). Several studies using quasi-experimental designs support this conclusion. Chyn et al. (2022) leveraged variation in the historical placement of railroad lines in northern cities to instrument for (and identify the impacts of) racial segregation, showing a strong negative impact of segregation on upward mobility for all children and pronounced negative effects for Black children across the income distribution. Derenoncourt (2022) focused on the long-term effects of the Great Migration on economic mobility among Black Americans. Using a shift-share instrument for the scale of Black migration into non-Southern commuting zones,¹ Derenoncourt (2022) found that Black Americans born in the 1980s had lower rates of upward mobility in commuting zones that absorbed a larger share of Black Americans decades earlier. She points to spending on police, rates of incarceration, and rising violence as potential explanations for the findings.

___________________

¹ The author uses a shift-share approach to “[c]ombine information on pre-1940 Black southern migrants’ location choices with supply-side variation in county out-migration from 1940 to 1970, predicted from southern economic variables” (Derenencourt, 2022, p. 370).

The Role of Qualitative Research in Understanding How Space and Place Affect Economic and Social Mobility

Qualitative research has deepened the understanding of mechanisms that connect place to economic and social mobility. For example, recent work has focused on why the residential moves of low-income households often do not translate into gains in neighborhood quality. Ethnographic and interview-based fieldwork has revealed that, rather than select specific neighborhoods for moves, poor households are often pushed out of where they live, forced to combine decisions of whether to and where to move in one rushed step, frequently under duress (Carillo et al., 2016; DeLuca & Jang-Trettien, 2020; DeLuca et al., 2019; Harvey et al., 2020; Rosen, 2017). Housing instability—spurred by factors such as poor housing quality or violence—has profound consequences for the sorting of households and children into neighborhoods and schools (Cuddy et al., 2020; DeLuca et al., 2024). Understanding these processes and constraints informs effective policies aimed at reducing spatial inequality.

Another, longer-standing qualitative literature in cultural and urban sociology has focused on how people living in high-poverty, high-violence neighborhoods adapt to the threats and constrained resources. Bourgois (2003), for example, documents how drug dealers in East Harlem often have worked in the formal economy, but as a result of experiences of disrespect and discrimination, feel driven to the drug economy and adopt street culture as a “badge of pride.” MacLeod (1987) and Anderson (1999) similarly explore the adaptive behaviors adopted by some men with low incomes when opportunity to achieve their aspirations is blocked.

Other qualitative scholars have explored the challenges that young people growing up in high-poverty, high-violence communities face in coping with adverse events and in navigating their social networks, families, and communities (see Small et al., 2010). For example, social isolation and avoidance of risky peers may protect youth from violence in the near term but may also limit the building of human and social capital, which can in turn limit their upward mobility and risk locking them in disadvantaged communities over the long term (Chan Tack & Small, 2017; DeLuca et al., 2024; Edin et al., 2015).

A less-often explored but important additional mechanism through which place may affect children’s long-run prospects is through neighborhood effects on parent well-being and mental health. Earlier qualitative literature found that parents responded to living in low-income, high-crime communities through vigilance to protect their children, which can be additionally taxing to parents (Anderson, 1999; Furstenberg et al., 1999). Mixed methods combining qualitative and quantitative research on the MTO experiment found that moving to safer neighborhoods can lead to

substantial improvements in parent mental health (Ludwig et al., 2012; see also DeLuca et al., 2016; Turney et al., 2013). Such mechanisms can also help explain a puzzle that remains from Chetty et al. (2016): how is it that long-term economic mobility improved for young MTO children, despite small to no improvements in their schools after moving? Neighborhoods may have direct effects on children through safety and exposure to more advantaged peers, and they may also have indirect effects through improved parental mental health, which is particularly relevant for young children.

While much of this qualitative research focuses on urban areas, some recent work has shifted the focus to rural communities. For example, in their analysis of communities with persistent high poverty, poor health outcomes, and low economic mobility, Edin et al. (2023) highlight the history of state-sanctioned violence that pervades many of these places. They posit that community-level violence impedes upward mobility, and limited mobility can in turn spur violence (see also Mann et al., 2024).

IMPLICATIONS FOR SOCIAL POLICY

Evidence on the importance of neighborhoods and larger urban areas for economic and social mobility raises the question: how can neighborhood quality be improved for more households? Two common policy approaches include housing mobility policies that help move families to higher-opportunity areas and “place-conscious” policies (Pastor & Turner, 2010) aimed at bringing opportunity and investment into less-resourced communities. This section reviews selected evidence for each of these approaches. It is critical to acknowledge that many policies, including historical policies such as redlining and current policies such as residential zoning, have created or maintained spatial inequality and, likely, limited upward mobility. The section concludes by discussing a third approach to social policy, which involves curtailing or abolishing existing programs and policies that may be amplifying spatial inequality.

Evidence from Housing Mobility Programs

Policies for assisting low-income households in relocating to less-segregated, lower-poverty neighborhoods with housing voucher subsidies—known as housing mobility programs—have been implemented for nearly 50 years. Several of these interventions have included quasi-experimental or experimental changes in families’ neighborhoods and schools, increasing the validity of causal claims about neighborhood effects (DeLuca & Dayton, 2009; DeLuca et al., 2010; de Souza Briggs, 1997). James Rosenbaum’s pioneering research on Chicago’s Gautreaux desegregation program first provided evidence that housing policy that assisted low-income families

in relocating to communities with more resources, lower crime, and better schools could serve as a tool for economic and social mobility (Rubinowitz & Rosenbaum, 2000). The Gautreaux remedy required HUD and the Chicago Housing Authority to provide housing opportunities to Black families in public housing, allowing more than 7,000 Black households a chance to move to a wide range of neighborhoods, including the affluent and predominantly White Chicago suburbs.

Early evidence on Gautreaux showed that moving to Chicago’s mostly White and affluent suburban communities improved academic outcomes for children (Kaufman & Rosenbaum, 1992; Rosenbaum et al., 1991; Rubinowitz & Rosenbaum, 2000). The program had small to moderate effects on parental employment and earnings (Mendenhall et al., 2006). Strikingly, a recent 40-year follow-up of Gautreaux households with a more rigorous research design found that children who moved to mostly White, economically advantaged neighborhoods in the Chicago metro area earned more as adults, worked more years, were more likely to own a home, and were more likely to live in low-poverty and diverse neighborhoods decades later as adults compared with similar counterparts who moved to mostly Black segregated areas (Chyn et al., 2023). Thus, such moves led to durable and multigenerational benefits for families (see also Keels et al., 2005). A housing mobility program developed in Baltimore as a result of a different desegregation program also showed benefits to low-income Black families, including long-term moves to higher-opportunity, less-segregated communities and moderate gains in children’s test scores (DeLuca & Rosenblatt, 2017; DeLuca et al., 2016).

Motivated in part by the strong initial findings from Gautreaux, the MTO experiment was designed to test whether moving into low-poverty neighborhoods affected the social and economic outcomes of families living in areas of concentrated poverty (de Souza Briggs et al., 2010; Goering & Feins, 2003; Sanbonmatsu et al., 2011). As previously described, early MTO findings were complex: adults in treatment group households were no more likely to be employed nor less likely to be using public assistance than the control group; there were no improvements in children’s test scores and even an increase in delinquency among boys (Clampet-Lundquist et al., 2011; Sanbonmatsu et al., 2006, 2011). Yet, there were improvements in mothers’ mental health on par with those achieved through best practices in antidepressant medication (Ludwig et al., 2012). These outcomes are complemented by Chetty et al.’s (2016) influential reanalysis of the program’s positive long-run impact focused on children younger than age 13. Recent research also shows that children who received housing vouchers through MTO had lower hospitalization rates for asthma and psychiatric disorders, lower inpatient medical spending, and lower utilization of outpatient and clinic services (Pollack et al., 2019, 2023a). Findings on the

reduction of asthma attacks were replicated for children in the Baltimore mobility program (Pollack et al., 2023b). Yet the MTO findings discussed previously also suggest significant heterogeneity in the effects of residential mobility programs across subgroups of families.

In addition to Gautreaux and MTO, many other housing mobility programs have generated evidence suggesting that moving out of highly disadvantaged communities can lead to positive effects on children’s academic trajectories and economic outcomes. Evidence from the “Mt. Laurel” program in New Jersey, which followed families as they moved into a new mixed-income housing development (compared with those on a wait list), showed improvements in study habits but no change in grades (Massey et al., 2013). Evidence from families in public housing in Denver showed strong effects of various neighborhood characteristics, most notably violent crime rates, on high school academic success for low-income Latino and African American adolescents (Galster et al., 2016). Ludwig et al. (2010, as cited in Burdick-Will et al., 2011) analyzed data from housing assistance recipients in Chicago who had been randomly assigned a position on a wait list before the local housing authority opened this wait list for the first time in years. Exploiting variation in the timing of when families were offered housing in lower-poverty neighborhoods, the researchers found that children offered housing vouchers experienced improvements in both reading and math scores relative to children in the control group.

Such a convergence of powerful and rigorous evidence across cities and time has already led to an expansion of federal funding to scale housing mobility efforts, not only based on the quality of the evidence on impacts (Lubell et al., 2023), but also on the quality of the evidence on implementation and policy design (Bergman et al., 2024; DeLuca et al., 2023).

Evidence on Place-Conscious Investments

Place-conscious investments include policies and programs that direct resources to disadvantaged communities or urban areas, or to the families and individuals within disadvantaged places. A wide range of investments could fit under the umbrella of place-conscious investments, making it difficult to summarize the literature. However, some policies have been studied in great depth and provide insights into this approach.

One set of programs focuses on providing job experiences and training along with supplemental supports to residents of disadvantaged communities. The New Hope program, implemented in Milwaukee during the mid-1990s, stands out as a unique experiment in addressing low-wage labor market challenges and promoting economic mobility (Duncan et al., 2009). Unlike many contemporaneous welfare-to-work programs, New Hope adopted a place-conscious approach, specifically targeting individuals from low-income backgrounds residing in disadvantaged neighborhoods within the city.

The program randomly assigned applicants to take part, showing significantly higher employment and earnings while the program was operational (Huston et al., 2003). These positive employment outcomes can be partially attributed to the program’s guarantee of community service jobs for participants who were unable to secure private-sector employment. Beyond employment gains, the program demonstrated success in reducing poverty. By the fifth year, the poverty rate in the treatment group fell to 52 percent, compared with 60 percent in the control group. Furthermore, children of the adults in the treatment group experienced improvements in academic performance and behavior, indicating that the benefits were cross-generational (Duncan et al., 2009; Huston et al., 2001, 2003). Specifically, children in the treatment group scored significantly higher on reading and language assessments than their counterparts in the control group (Huston et al., 2003). The findings are notable because most previous studies of active labor market policies found that guaranteed or subsidized public-sector employment has no long-term impact on participants’ employment or earnings (Card et al., 2018).

The Jobs Plus program, implemented by HUD in the 1990s, stands as another noteworthy example of a place-conscious intervention designed to bolster employment prospects within specific communities. Unlike conventional welfare-to-work initiatives, Jobs Plus adopted a targeted approach, focusing on residents living in public housing developments across five diverse cities: Baltimore, Chattanooga, Dayton, Los Angeles, and St. Paul (Bloom & Riccio, 2005).

The program aimed to saturate these public housing developments with comprehensive employment-related services that were intended to equip residents with the necessary skills and support to secure and maintain employment over time. Additionally, Jobs Plus incorporated rent incentives as a means to financially encourage work participation among residents. Evaluations of the program revealed promising outcomes in three of the five sites that implemented the full spectrum of services and incentives offered (Bloom & Riccio, 2005). In these locations, residents experienced meaningful increases in both employment rates and annual earnings. Specifically, the program generated an estimated 10 percent rise in employment and an 8–19 percent increase in annual earnings (Bloom & Riccio, 2005).

Whereas New Hope and Jobs Plus targeted individuals within high-poverty areas, the Empowerment Zones/Enterprise Communities program in the mid-1990s offered tax breaks designed to encourage firms to expand, invest in, and improve local employment outcomes by hiring local residents (Tach & Wimer, 2017). The evidence base behind this program and others designed to stimulate demand in areas of disadvantage is mixed.

Busso et al. (2013) presented the most compelling evidence of positive impacts, estimating a 15 percent increase in jobs and an 8 percent wage

increase for residents within designated zones (compared with nonselected sites) between 1990 and 2000. These findings contrast with other studies that have reported minimal program effects in specific locations (Elvery, 2009; Oakley & Tsao, 2006). Additionally, concerns have been raised regarding the cost-effectiveness of such spatially targeted initiatives, particularly in the absence of additional investments (Glaeser & Gottlieb, 2008; Ladd, 1994). However, Austin et al. (2018) argued that place-based policies can be relatively cost-effective in improving economic outcomes, considering the high costs associated with moving to a new geographic area.

While the interventions discussed thus far focus primarily on poverty reduction, labor market prospects, and job creation in disadvantaged neighborhoods, it is crucial to acknowledge that economic opportunity is just one factor influencing economic mobility. A broader range of place-based interventions has been implemented to address various challenges within communities, including school quality, crime reduction, and community health (Braga, 2005; Dobbie & Fryer, 2011; Heller, 2014; Heller et al., 2013; H. Schwartz, 2010). These interventions, while not focused on labor market outcomes directly, may impact upward economic mobility indirectly by addressing factors such as exposure to violence, access to quality education, and overall community well-being, all of which are intricately linked to economic prospects.

The recognition that economic disadvantage tends to be concentrated in areas that face a range of associated challenges is the motivation for a set of interventions that have come to be identified as community change initiatives (CCIs). These interventions aim to “flood” such areas with resources in multiple domains, encompassing economic development, institutional support, physical infrastructure improvements, and social services (Kubisch et al., 2010). However, translating this vision into reality presents significant challenges. As Kubisch et al. (2010) highlighted, many CCIs have not received the sustained funding needed for transformative change. Coordinating services, building institutional capacity, and engaging residents are additional hurdles (Chaskin, 1997; Kubisch et al., 2010).

Evaluating the impact of CCIs is equally challenging. Programs often lack clear designs for assessment, and their reduced scale rarely allows for generating and sustaining tangible change (O’Connor, 1995, 1999; Sharkey, 2013). Those who live in communities targeted by such investments often move out, further complicating evaluation efforts (Theodos et al., 2015). One descriptive study of a high-profile CCI—the Sandtown Neighborhood Transformation Initiative in Baltimore—found that while relatively large investments in brick-and-mortar improvements led to higher rates of home ownership 20 years later (comparing Sandtown with other similarly situated neighborhoods in the city that did not receive such funds), it also likely led to higher rates of foreclosures during the 2008–2010 Recession

(Rosenblatt & DeLuca, 2017). Compared with similar neighborhoods, Sandtown saw no improvements in children’s test scores, violent crime, or life expectancy rates.

Infrastructure and development decisions by various levels of government also influence the availability of affordable housing, restricting or expanding opportunities for low-income families to live in different kinds of neighborhoods. Policies such as the low-income housing tax credit and the decisions of developers can significantly alter the supply of affordable housing (Owens & Smith, 2023). Furthermore, current practices of school assignment by residence can perpetuate spatial inequality by confining students to schools in their local, often low-income, neighborhoods. Other notable policy measures include inclusionary zoning, acquisition of housing in high-opportunity areas by public housing authorities, and efforts to eliminate vacant lots and abandoned homes in neighborhoods.

Although targeted interventions aimed at neighborhood revitalization have yielded mixed results, research on large-scale economic transformations suggest that substantial investments capable of altering the local economic landscape can have lasting positive effects on residents. For example, studies estimating the long-term impact of the Tennessee Valley Authority (TVA), a federal initiative established in the 1930s to revitalize the economically depressed Tennessee Valley region, demonstrate long-term economic benefits for residents (Kline & Moretti, 2014). Similarly, research on the introduction of casino gaming on American Indian reservations reveals positive economic impacts for these communities, including job creation and increased income levels (Akee et al., 2010; Wolfe et al., 2012); however, the benefits of casinos may be due in part to the fact that they are restricted to very specific areas. An evaluation of the Delta Regional Authority, which funds rural development projects in the Mississippi River Delta region, found positive earnings impacts in the health care and social services sectors but not on overall employment rates (Pender & Reeder, 2011). These examples suggest that although large-scale economic transformations have the potential to benefit residents in disadvantaged areas and support upward mobility, much more research is needed before concluding that large-scale infrastructure investments similar to the TVA would result in similar outcomes in other areas and historical contexts.

Interventions Amplifying Spatial Inequality

Programs that invest in disadvantaged areas and those that support moving out of poor neighborhoods remain the most prominent approaches to addressing the link between place and upward mobility. However, some policies and programs active in the United States have the opposite effect: they constrain economic mobility and amplify spatial inequality. Therefore,

a third approach to addressing concentrated disadvantage and supporting upward mobility involves scaling back or ending these interventions.

An extensive literature has documented the historical policies that exacerbated racial inequality in America’s neighborhoods and urban areas through the civil rights era, including racial zoning in such cities as Louisville, St. Louis, and Baltimore; explicit discrimination in new suburban housing developments; racially restrictive covenants that limited who could buy or lease homes; blockbusting and contract lending in cities absorbing growing Black populations leaving the South; the razing of Black neighborhoods during the period of interstate highway construction and urban renewal; and the federal provision of explicitly segregated public housing (Coates, 2014; Hirsch, 2009; Massey & Denton, 1993; Rothstein, 2013). New evidence has documented the long-term consequences of such policies, demonstrating their impact on segregation and, decades later, on economic mobility (Aaronson et al., 2021b; Faber, 2020).

Less attention has been given to policies and interventions that continue to generate or reinforce spatial inequality in the present by constraining residential mobility, subsidizing home ownership among the most affluent, and limiting the development of affordable housing. For example, the largest federal housing programs are tax subsidies such as the mortgage interest deduction, the property tax deduction, and the exclusion of capital gains on the sales of homes—all of these are highly regressive federal policies that provide strong incentives for the most affluent families to take on deep debt and purchase more expensive homes in exclusive areas (Fischer & Huang, 2013; Turner et al., 2013). In 2017, high-income households received four times the benefits of low-income households, half of which were due to the mortgage interest deduction (Fischer & Sard, 2017; see also Chapter 5). However, the reduction of the mortgage interest deduction through the 2017 Tax Cut and Jobs Act led to a more equitable distribution of taxes (Ambrose et al., 2022), as well as a more equitable geographic distribution (Blouri et al., 2023). Turning over the power to determine local land use regulations to policymakers in individual towns and cities limits the production of new housing and contributes to economic segregation (Glaeser & Gyourko, 2002; Gyourko et al., 2013; Rothwell & Massey, 2010). These are only a few examples of the active interventions in housing that continue to benefit high-income families and further divide American space and contribute to residential segregation by race and ethnicity and by income. Altering or ending existing housing and land use policies that exacerbate inequality, and instead implementing programs that confront inequality, would represent an initial step in an urban policy agenda designed to reduce neighborhood inequality.

A FORWARD-LOOKING SCIENCE OF SPACE AND MOBILITY

For a long period of time, quantitative academic research on the link between space and upward mobility centered on the question of whether neighborhoods matter (Sharkey & Faber, 2014). With improved data and methods, there now exists compelling evidence that the environments in which people spend their lives influence and constrain their decision-making, the opportunities available to them, and the hazards and risks that surround them. In building a forward-looking research agenda, the committee argues for a focus on a more complex set of questions that requires new forms of data and evidence.

The committee identified seven areas of research to help fill gaps in the existing research on space and economic and social mobility and to strengthen the connection between research and policy. Those seven areas focus on (a) improved evidence on mechanisms linking space and mobility; (b) evidence on heterogeneity in the relationship between space and mobility; (c) a greater reliance on qualitative methods to inform social policy; (d) rigorous research on the causal impact of various types of investments in places; (e) an expansion of policy discussions that consider the consequences of existing interventions that impact places; (f) consideration of the general equilibrium effects of social policies, as well as the feasibility and costs of different approaches; and (g) an expansion of research to disadvantaged places outside of metropolitan areas that dominate the literature.

First, more convincing evidence is needed on the mechanisms underlying the link between space and upward mobility. Existing research has demonstrated that schooling, public funding, community violence, social capital, and racial and economic segregation are all strongly associated with economic mobility. However, causal evidence on the importance of these and other factors is limited. More evidence is needed to understand the processes by which these factors lead to upward mobility. Identifying the central mechanisms underlying the link between space and mobility is crucial to moving the literature beyond the common focus on the poverty rate or other basic measures of disadvantage in places and instead considering the specific institutions, processes, or phenomena that impede or facilitate economic and social mobility. Such evidence could be qualitative research to identify hypotheses for why it is that violence affects children’s school performance or what specifically it is about mentors or local role models that helps promote educational or employment success; or it could be more archival or historical research focusing on how specific programs or policies have changed within places over time in ways that would directly impact economic and social mobility through institutional resources, labor market expansion, school redistricting, etc. A more refined focus on mechanisms is also essential to inform policy responses and to target investments more effectively.

Second, a more complete understanding of heterogeneity in the relationship between space and upward mobility is closely related to the goal of generating better evidence on mechanisms. As an example, early results from quantitative and qualitative analyses of MTO families revealed substantial heterogeneity in every aspect of the experiment, including take-up of the voucher, mental health improvements for mothers (Turney et al., 2013), destinations into which families moved (Rosenblatt & DeLuca, 2012), and the divergent impact on boys versus girls (Clampet-Lundquist et al., 2011; Sampson, 2008). Years later, reanalysis of the data from MTO showed that the characteristics, such as the level of violence, of communities in which families started and ended up made a substantial difference in whether the move led to improved outcomes for children (Burdick-Will et al., 2011). Reanalysis also showed that children who moved at young ages tended to see better outcomes than those who moved at older ages (Chetty et al., 2016). All of these examples are from MTO, but the focus on heterogeneity also needs to consider variation in findings across studies. For instance, while findings based in urban areas have shown that the benefits of MTO may decline as children age (Chetty et al., 2016), suggestive evidence for rural areas indicates that migration when older is still beneficial (Krause & Reeves, 2017). Foregrounding heterogeneity enables an understanding of which groups are most likely to take advantage of and benefit from social policies designed to reduce spatial inequality or increase economic and social mobility (Heckman & Landersø, 2022).

Third, more qualitative research is needed to strengthen the link between evidence and policy. Researchers can use survey and administrative data to powerfully illustrate national patterns in economic mobility with a scope that extends across geographies and generations, and identify factors associated with social mobility. These sources of data can often show where to look but cannot necessarily show how to make sense of what is seen there nor how to apply it to make more efficient policy investments. For this, a research approach is needed that combines these sources of data with rigorous qualitative methods. Social science has historically operated in methodological silos, with quantitative, statistical research on one side and qualitative, ethnographic research on the other. Yet both approaches are necessary to understand social opportunity and design effective policies for promoting upward mobility.

Qualitative research has already been enormously valuable for illuminating the mechanisms by which individuals’ environments influence their decision-making about where to live, how to stay safe, and how to get ahead (DeLuca et al., 2019; Edin et al., 2023; Harding, 2010; Harvey et al., 2020; Jones, 2009). Qualitative research also exposes the ways that local institutions become salient in the lives of families (Small, 2004) and the strategies residents deploy as they interact with peers; neighbors; employers;

and representatives of the state, such as case workers, teachers, or police officers.

While it is often difficult to peer inside the black box to understand how a large-scale social experiment such as MTO is experienced by the people targeted by the program, qualitative research can shed light on why a program is or is not working (Clampet-Lundquist et al., 2011; DeLuca & Rosenblatt, 2010; Kling et al., 2005). As an example, qualitative studies of housing programs in Baltimore and Seattle showed the challenges that families face as they attempt to find housing with poor credit and family instability, and revealed how intensive counseling and support gave families expanded choices and a greater chance to successfully transition into new, racially diverse communities with low rates of poverty and more abundant economic opportunities (Bergman et al., 2024; Darrah & DeLuca, 2014; DeLuca et al., 2023). This kind of research can then inform subsequent policy, such as the Seattle Creating Moves to Opportunity housing voucher experiment, which leveraged not only financial resources but also housing “navigators” to support families, resulting in significantly higher rates of opportunity moves for treatment families (Bergman et al., 2024). All of this work has led to substantial investments devoted toward expanded housing mobility programs in eight more sites.

Some of the most important questions that can be explored with qualitative data include the following:

• What are the mechanisms through which violence shapes academic outcomes?
• How do social ties (e.g., churches, local employment-based connections, nonprofit or socially minded community organizations) among Black families help explain recent gains in income mobility?
• Why do some places foster more social capital or economic connectedness than others?
• Why is the evidence inconclusive as to whether schools account for the effects of neighborhoods?

In particular, the committee believes there is great value in funding qualitative research—which was particularly fruitful in MTO—not only on its own, but also to accompany randomized controlled trials and quasi-experimental studies and studies relying on administrative data. The insights generated by qualitative data can also inform the collection of survey data and the hypotheses that are evaluated using survey data.

Fourth, while much of the research discussed is based on programs estimating the impact of residential moves, the literature on the impact of investing in places is much less developed. Comprehensive data on where public and private investment flows is a starting point for establishing the

relationship between funding and economic and social mobility. Looking beyond streams of funding, data on community-level social processes, institutions, political power, social networks, and other forms of social capital are crucial to developing a more complete accounting of the explanations for spatial inequality and economic and social mobility. A large literature points to dimensions of community life, such as collective efficacy—defined as neighborhood social cohesion and the willingness to act on behalf of the common good—as central to the capacity for residents to provide informal oversight of public spaces and achieve common goals such as the reduction of violence (Sampson et al., 1997), or to the array of institutions in a larger urban area that facilitates civic engagement and upward mobility (Chetty et al., 2022a,b; Putnam, 2000). While this research provides suggestive evidence that social capital is strongly linked with economic and social mobility, data limitations and a lack of convincing causal evidence have not allowed for strong causal claims on how social capital promotes mobility.

More broadly, the vast literature on investments in places—such as comprehensive community development initiatives—has not provided clear, causal evidence on the most effective ways to invest in communities or entire cities and regions, or how policymakers can or even whether they should intervene in places to improve economic and social mobility.

Fifth, while policy discussions on reducing spatial inequality typically focus on two approaches—moving people out of or investing in disadvantaged places—the committee argues for an expansion of this debate to include approaches to end or scale back existing interventions that amplify spatial inequality, as described above. For example, federal housing policies such as the mortgage interest deduction incentivize families to take on more debt to purchase more expensive homes in more exclusive areas; state occupational licensing requirements generate a constraint on geographic mobility that makes it more difficult for households to make moves across state lines in order to take advantage of economic opportunities; and land use policies implemented by local towns and cities can limit the development of affordable housing, creating extreme inequality across the localities within a given metropolitan area (and additional inequality in access to amenities such as high-performing schools and safety). In addition to developing and advocating for new interventions, it is important to draw attention to the ways that these and other existing policies exacerbate spatial inequality in ways that are often taken for granted.

Sixth, moving from theory and research into policy expands the type of knowledge needed. This is particularly true when considering the relative value of people- versus placed-based investments. As an example, improving educational opportunities for young people in highly disadvantaged areas may improve long-term prospects for economic mobility, and yet it also may raise the probability that they leave their communities for sections of the country offering greater economic opportunity.

The more general challenge of understanding the equilibrium effects of social policy is often neglected (Galster, 2012). Voucher programs designed to give children the chance to attend a high-quality school or to give families the chance to move to a community with lower poverty may, in fact, benefit those who receive the voucher—but if such efforts are scaled up, they may alter system-level dynamics in the sending and receiving schools or neighborhoods. Failing to consider the equilibrium impacts of social policy (i.e., the full range of effects on all potentially affected domains of society, including program participants and nonparticipants and focal and nonfocal communities) can lead to unanticipated challenges.

Additional research is needed to go beyond the consideration of what types of policies affect economic and social mobility to consider the system-level impacts of social policy as well as the magnitude, feasibility, and costs of different policy approaches. To move toward this goal, the committee argues for greater investment in research tools such as agent-based models designed to model the general equilibrium consequences of social policy.

Seventh, greater focus is needed on areas that have received relatively little attention in the literature on neighborhood effects and upward mobility, including “deeply disadvantaged” rural areas (Islam et al., 2015) and rural-adjacent small towns and suburbs. Examples include wide swaths of the Southern states that have extremely low rates of economic mobility, rural communities of the South and Appalachian regions, and Native American tribal lands that have been disconnected from sustained public investment and economic growth (Edin et al., 2023). Because much of the literature on neighborhood effects has focused on concentrated poverty in central cities, the study of space and economic and social mobility is heavily biased toward urban poverty. The most severe areas of disadvantage in the United States are not found in central cities, however, and the same attention and policy focus must expand beyond the nation’s urban areas. Similarly, the literature often misses rural-adjacent suburban towns and smaller cities.

SUMMARY, KEY CONCLUSIONS, AND RECOMMENDATIONS

While much of the early quantitative research on the impact of residential environments struggled to overcome the methodological challenges associated with nonrandom selection into neighborhoods and cities, recent evidence has led to something close to a consensus in support of the claim that residential environments can have a causal impact on economic and social mobility and intermediate outcomes related to mobility—most notably academic achievement, cognitive skill development, and physical and mental health. Increasing access to geocoded data, improved tools for spatial analysis, and the growth of convincing quasi-experimental and experimental

research designs has undergirded this view. The findings also suggest that there is considerable heterogeneity in the effects of space and place. The impact of the residential environment on individual life trajectories is a product of conditions in the immediate neighborhood and the larger area and region of the country, the timing and duration of individuals’ exposure to these environments outside the home, and individuals’ characteristics and unique vulnerability to the effects of the residential context. This perspective on the link between place and individual outcomes moves beyond the question of whether neighborhoods matter and toward the questions of when, where, why, and for whom residential contexts matter.

Conclusion 3-1: Recent evidence suggests that residential environments have a causal impact on economic and social mobility (measured by income, earnings, or occupation) and on intermediate outcomes related to mobility, such as cognitive development and educational attainment. The link between place and individual outcomes has now moved beyond the question of whether neighborhoods matter and toward questions of when, where, why, and for whom residential contexts matter.

While scholarship on the effects of place has grown considerably over time, less attention has been paid to the mechanisms that undergird these effects and the heterogeneity of these effects—vital considerations for using research to improve policy and understanding the potential scalability of existing interventions. Evidence at the neighborhood level points to schools, community violence, environmental exposures, and local social networks as examples of mechanisms linking the local residential environment with economic and social mobility. Evidence at larger levels of analysis—such as cities, counties, and commuting zones—points to segregation and local labor market conditions as examples of forces that influence economic and social mobility.

Research on school quality shows that attending schools in low-poverty areas leads to better academic outcomes for students from low-income backgrounds, with peer socioeconomic diversity potentially being more important than school funding. While the effects of exposure to phenomena such as community violence are difficult to identify, available causal evidence indicates significant negative effects vis-à-vis indirect outcomes such as cognitive functioning and academic performance. Evidence points to long-term positive impacts on income mobility in communities where violence decreased. Connections to peers, friends, and neighbors are hypothesized to influence local culture, norms, and resources, and thus economic mobility—although most evidence is correlational. Researchers have only recently employed quasi-experimental designs, concluding that economic mobility is positively influenced by low-income populations’ “exposure to”

and friendships with high-income people in their communities. Similarly, parental employment rates in the neighborhood are an important determinant of upward mobility, especially among Black boys from low-income backgrounds.

At larger levels of analysis, such as cities, counties, and commuting zones, available evidence highlights segregation and local economic and labor market conditions as key determinants of economic and social mobility. Research on variation in mobility across commuting zones specifically highlights negative associations with greater economic inequality and racial segregation. Work leveraging historical processes related to segregation provides causal evidence on long-term mobility impacts as well. Studies examining redlining and other segregationist spatial policies document lasting negative effects on intergenerational mobility in the most affected neighborhoods. While research exploiting shocks to local labor market conditions and their mobility consequences is limited, negative shocks such as localized job losses are associated with worse academic performance among children, while positive changes, such as casino openings in disadvantaged areas, show positive effects on education, income, and employment.

Conclusion 3-2: At the neighborhood level, the mechanisms linking the local residential environment with economic and social mobility include schools, community violence, and local social networks. At larger levels of analysis—such as cities, counties, and commuting zones—the forces that influence economic and social mobility include segregation and local labor market conditions.

This large and growing body of evidence on the importance of different geographic areas for economic and social mobility is consistent with research asking how to best employ spatial policy in a way that benefits the most households and facilitates upward intergenerational mobility. Broadly speaking, most spatial policy strategies are either housing mobility policies, which help relocate disadvantaged families to higher-opportunity areas, or place-conscious investments, which aim to bring opportunity and investment into disadvantaged communities. Housing mobility programs—such as the Gautreaux desegregation program and the MTO experiment—have shown positive impacts on academic outcomes, employment, and earnings, particularly for children moving to low-poverty neighborhoods. Place-conscious investment initiatives—such as the New Hope, Harlem Children’s Zone, and Jobs Plus programs—that target residents of disadvantaged communities with employment opportunities, training, and support services and/or provide resources to foster job creation and economic development have also shown success in improving employment, earnings, and academic outcomes for participants. However, the impacts of place-based programs

that have not targeted people in disadvantaged communities and provided a range of supports have been mixed.

A third approach emphasizes the importance of ending existing programs and policies that have historically amplified spatial and racial inequality and continue to do so. Zoning, discriminatory housing practices, and some regressive federal housing programs (e.g., mortgage interest deduction, property tax deduction, exclusion of capital gains on the sales of homes) reinforce residential segregation and constrain residential mobility. While federal homeowner policies exist that assist low-income families (e.g., affordable mortgage programs and credit programs), many homeownership policies disproportionately benefit the most affluent communities and households, further entrenching economic segregation and the concentration of economic mobility-relevant resources in higher-income areas.

Conclusion 3-3: Most spatial policy strategies are either housing mobility policies, which relocate families to higher-opportunity areas, or place-conscious investments, which aim to bring opportunity and investment into disadvantaged communities. Although both have the potential to boost economic and social mobility, evidence on the effectiveness of place-based programs that do not target people in disadvantaged communities and do not provide a range of supports is mixed.

Conclusion 3-4: In addition to new housing mobility initiatives and place-based investments, it is important to consider approaches that focus on ending existing programs and policies that have historically amplified spatial and racial inequality and continue to do so.

For a long period of time, quantitative academic research on the link between space and mobility centered on the question of whether neighborhoods matter. With improved data and methods, there now exists compelling evidence that the environments in which individuals spend their lives influence and constrain their decision-making, the opportunities available to them, and the hazards and risks that surround them. The committee identified seven areas of research to help fill gaps in the existing research on space and economic and social mobility and strengthen the connection between research and policy.

First, more convincing evidence is needed on the central mechanisms underlying the link between place and economic and social mobility. This is crucial to moving the literature beyond the common focus on the poverty rate or other basic measures of disadvantage in places and instead

considering the specific institutions, processes, or phenomena that impede or facilitate upward mobility.

Second, moving toward a more complete understanding of heterogeneity in the relationship between space and mobility is closely related to the goal of generating better evidence on mechanisms. By foregrounding heterogeneity, one is in a better position to understand which groups are most likely to take advantage of and benefit from social policies designed to reduce spatial inequality or increase economic and social mobility.

Third, there needs to be a larger role for qualitative research in strengthening the link between evidence and policy. Qualitative research can help illuminate the mechanisms by which individuals’ environments influence their decision-making about where to live, how to stay safe, and how to get ahead and the ways that local institutions become salient in the lives of families, as well as the strategies residents deploy as they interact with peers; neighbors; employers; and representatives of the state, such as case workers, teachers, or police officers.

Fourth, while much of the research discussed by the committee is based on programs estimating the impact of residential moves, the literature on the impact of investing in places is much less developed and has not provided clear, causal evidence on the most effective ways to invest in communities or entire cities and regions. This is another area where qualitative research can make a valuable contribution.

Fifth, policy discussions focused on how to reduce spatial inequality typically focus on two approaches: moving people out versus investing in places. However, this discussion should be expanded to include existing policies (e.g., federal housing policies, local land use policies) that exacerbate spatial inequality.

Sixth, consideration should be given to the general equilibrium effects of social policies, as well as the feasibility and costs of different approaches. For example, improving educational opportunities for young people in highly disadvantaged areas may lead to improvements in long-term prospects for economic mobility, and yet it also may raise the probability that they leave their communities for sections of the country offering greater economic opportunity.

Seventh, there needs to be a greater focus on regions and areas that have received relatively little attention in the literature on neighborhood effects and upward mobility, including deeply disadvantaged rural areas, rural-adjacent small towns, and suburbs. Although much of the literature on neighborhood effects has focused on concentrated poverty in central cities, the most severe areas of disadvantage in the United States are not found in central cities.

Recommendation 3-1: Researchers should strengthen the evidence base on space and place in the following areas: (1) causal impacts of the key mechanisms that link place with economic and social mobility; (2) heterogeneity in the relationship between place and economic and social mobility; (3) qualitative approaches to understanding the mechanisms that undergird the causal effects of place, as well as the mechanisms that explain evidence on effective policy; (4) causal impacts of various types of place-based investments; (5) existing policies and interventions that amplify spatial inequality; (6) general equilibrium effects of social policies and the feasibility and costs of different approaches; and (7) neighborhood contexts of disadvantaged rural, small town, and suburban areas.