«Moving to opportunity voluMe 14, nuMber 2 • 2012 U.S. Department of Housing and Urban Development | Office of Policy Development and Research ...»
• Families in the experimental group received Section 8 rental assistance certificates or vouchers that, initially, they could use only in low-poverty census tracts—those in which less than 10 percent of residents had incomes below the federal poverty threshold in 1990. Nonprofit organizations in each city provided mobility counseling to help families find and lease a unit in a low-poverty area. After 1 year in the low-poverty tract, families could use the voucher to live in any neighborhood. In addition to abiding by the geographic requirement, families had to abide by all of the regular rules of the Section 8 certificate and voucher programs.
• Families in the Section 8 group also received Section 8 certificates or vouchers to move into private-market housing but without any mobility counseling or additional locational constraints under the MTO program design.
• Families in the control group did not receive any housing vouchers or certificates through MTO but continued to be eligible for all programs and services to which they would be otherwise entitled.
Random assignment enables us to isolate the effects of different neighborhood and housing conditions on people’s well-being by comparing the average outcomes of groups of families who are similar on average in every way at baseline except that some were subsequently offered the opportunity to use a housing voucher to move to a new neighborhood with different characteristics.
In this article, we compare adults in the two treatment groups with the adults in the control group on their health and economic self-sufficiency 10 to 15 years after they enrolled in the program.
Before turning to the results of our analyses, we first discuss the hypothesized pathways through which moving to a lower poverty neighborhood might affect adult health and economic selfsufficiency. We then summarize the results of previous waves of MTO research and other studies.
We describe our sample and analytic strategy and then present estimates of the long-term effects of MTO on selected mental health, physical health, and economic outcomes. (Ludwig et al., 2011, and Sanbonmatsu et al., 2011, present additional results for other health and economic outcomes.) We find that, 10 to 15 years after random assignment, MTO moves had sizable effects on a number of important physical and mental health outcomes but no detectable effects on almost any measure of economic self-sufficiency. More specifically, in the area of mental health, we find that MTO moves led to lower levels of psychological distress and lower prevalence of depression.
In the area of physical health, we find beneficial effects on severe obesity, diabetes, and physical limitations, although we do not detect effects on hypertension or overall self-rated health. Earnings and employment rates have risen for all random assignment groups since they initially joined the program, and very few statistically significant differences emerged across the randomized groups on economic outcomes. We conclude this article with a discussion of some of the limitations of our findings and their implications for future research and policy design.
Hypothesized Pathways The MTO program offered families in the two treatment groups a housing voucher that they could use to relocate from a public housing project to an apartment (or house) in another neighborhood.
Families in both the experimental and Section 8 treatment groups could use the voucher to move to a low-poverty neighborhood, but for the experimental group, moving to a low-poverty area was an initial requirement of using the voucher at all. Exhibit 1 illustrates our hypothesis that moving to a lower poverty area might affect adults’ outcomes by changing their social environment, the resources of their community, and their physical environment.
Changes in the Social Environment One set of potential pathways involves the social environment—the environment created by the people living in the neighborhood. Moving to a more affluent area may reduce exposure to violence because of a safer social environment, expose movers to higher socioeconomic status (SES) peers, and offer new social networks. At the same time, moving may disrupt social ties such as relationships with families, friends, service providers, and faith communities.
A safer environment could improve physical health directly by reducing the likelihood of injuries from assaults and indirectly by providing a safer environment for outdoor activity such as exercise.
112 Moving to Opportunity The Long-Term Effects of Moving to Opportunity on Adult Health and Economic Self-Sufficiency Exhibit 1 Hypothesized Pathways Through Which MTO May Affect Adult Well-Being
MTO = Moving to Opportunity.
Note: This figure builds on the pathway diagrams shown in various chapters of the MTO interim impacts evaluation (see Orr et al., 2003).
Unsafe neighborhoods may discourage outdoor physical activity and exercise (Bennett et al., 2007;
Harrison, Gemmell, and Heller, 2007). Lower likelihoods of being the victim of violence and of witnessing violence could also improve mental health, because these types of exposures are associated with higher levels of psychological distress, depression, generalized anxiety disorder (GAD), and post-traumatic stress disorder (PTSD) (Ross and Mirowsky, 2001; Silver, Mulvey, and Swanson, 2002; Zapata et al., 1992). Neighborhoods with lower crime may also generate higher levels of trust and greater collective efficacy (the willingness of neighbors to work together and support shared values), which may help serve as a buffer against other types of stressors (Berkman and Glass, 2000; Cohen, 2004; Sampson, Raudenbush, and Earls, 1997).
Cityscape 113 Sanbonmatsu, Marvakov, Potter, Yang, Adam, Congdon, Duncan, Gennetian, Katz, Kling, Kessler, Lindau, Ludwig, and McDade Any MTO effects on mental health may amplify effects on physical health, given the link between chronic stress and the regulation of physiological systems and behavior (McEwen and Stellar, 1993; Merkin et al., 2009). Previous research has associated higher stress levels with hypertension (Kornitzer, Dramaix, and De Backer, 1999; NIH, 1997) and asthma attacks (Bloomberg and Chen, 2005; Wright, Rodriguez, and Cohen, 1998). Stress may lead people to increase their intake of dense-calorie foods (Torres and Nowson, 2007) and to turn to alcohol, tobacco, or other substances in an effort to cope. The new social environment may also expose movers to neighbors with higher education levels who, at least on average, may have healthier lifestyles (Darmon and Drewnowski, 2008; Pampel, Krueger, and Denney, 2010).
Higher SES neighbors and less stress may improve economic prospects. Living in a neighborhood with a greater share of employed adults may provide more social support for work (for example, Maurin and Moschion, 2009) or more referrals for job openings (Bayer, Ross, and Topa, 2008).
Moving out of a stressful environment may free up cognitive space for deliberative cognitive processes, such as financial planning (Porcelli and Delgado, 2009).
The degree to which having new neighbors translates into actual behavioral changes may depend on how much exposure adults have to their new neighbors and how well they connect with them (Christakis and Fowler, 2007). Similarly, new referral networks may not be helpful if new neighbors know mostly about jobs in occupations and industries that are not relevant for adults in the MTO program. Thus, moving to a more affluent area need not have any beneficial effects on behavior. Moving could even have negative effects if the move itself initially disrupted relationships and left adults feeling socially isolated, particularly if they do not have much in common with their new neighbors. MTO participants who move to lower poverty areas could also experience a decline in their socioeconomic standing relative to their new peers if their incomes do not rise with the move. Feelings of relative deprivation could negatively affect their physical and mental health (Stafford, 2003). In addition, although higher income people tend to lead healthier lifestyles on some dimensions, other health-risk behaviors, such as drinking, may be more prevalent among more affluent people, at least over part of the SES distribution (Pollack et al., 2005; SAMHSA, 2010). For immigrant families in MTO, moving to a more affluent area could mean greater acculturation and a less healthy lifestyle (Lara et al., 2005).
Changes in Community Resources Part of the hope of mobility programs such as MTO is that higher SES communities can offer lowincome families better access to community resources such as jobs, high-quality schools, highspeed broadband infrastructure, and supermarkets. Research studies generate mixed findings, however, as to whether more advantaged neighborhoods always offer low-income families greater access to resources.
Scholars hypothesize that geographic proximity to employers may be important for economic selfsufficiency. The spatial mismatch theory of John Kain (1968) suggests that a mismatch between the location of manufacturing and other jobs and the concentration of low-income families in urban areas might account for some of the lower employment levels observed. More recently, a series of studies in the late 1980s and early 1990s linked neighborhood poverty and job accessibility to labor force participation and wages for African-American and Hispanic workers (see Ihlanfeldt and 114 Moving to Opportunity The Long-Term Effects of Moving to Opportunity on Adult Health and Economic Self-Sufficiency Sjoquist, 1998, for a review). If MTO moves enable access to a broader range of employment opportunities, they may also lead to increased wages and fringe benefits. Lower poverty areas, however, are not always closer than higher poverty areas to the sort of job opportunities that may be relevant for lower skilled MTO families (Anil, Sjoquist, and Wallace, 2010). Higher SES neighborhoods may offer increased access to formal banks as opposed to high-cost alternatives such as check cashers and payday lenders, who seem to target high-poverty areas (see Graves, 2003, for details), and thus neighborhoods may improve economic welfare through better personal finance management.
The types of retail stores, restaurants, and advertisements in a new neighborhood might affect the health of low-income people who move there. The presence of more grocery stores could improve diet by reducing the purchase price or transportation cost of healthy food (Morland et al., 2002).
The presence of fewer liquor stores and fast-food restaurants could also affect diet and potentially reduce the consumption of high-caloric foods and alcohol (Dubowitz et al., 2008; Franco et al., 2008; Inagami et al., 2006; Zenk et al., 2005).
Living in a community with greater resources does not mean families will actually be able to take advantage of those resources or that those resources will be well suited to their needs. One reason low-income people may concentrate in urban cores is the availability of public transportation, which, in general, is more affordable than maintaining a private car (Glaeser, Kahn, and Rappaport, 2008). If new neighborhoods offer more limited public transportation options, MTO moves could reduce access to jobs, grocery stores, and other services. Furthermore, even if overall access to services remains the same or improves, access to certain types of social services could be worse if service providers for the poor are less common in higher income areas (Allard, 2004; Small and Stark, 2005).
For similar reasons, the effects of moves on healthcare access are hard to predict. The MTO demonstration took place in cities with some of the top-ranked hospitals in the country, many of them academic medical centers located, for historical reasons, near economically disadvantaged inner-city neighborhoods.1 These academic medical centers may be more experienced in meeting the linguistic and cultural needs and daily realities of low-income patients than are the healthcare facilities located in more affluent areas. Moreover, free or very low-cost primary healthcare services may be concentrated in high-poverty areas, and so access to low-cost healthcare services could decrease with moves to better neighborhoods.
Changes in the Physical Environment Finally, moving to a less impoverished neighborhood may influence well-being through changes in the physical environment of the neighborhood and the housing unit. Higher SES neighborhoods may have lower pollution, trash, and vandalism levels; higher quality buildings and outdoor spaces; and built environments more conducive to exercise. The built environment of a neighborhood—its buildings, parks, and streets—may also affect health behaviors such as exercise. Studies have associated greater walkability of a neighborhood with greater physical activity and lower Four of the five MTO cities are home to at least 1 of the top 14 honor-roll hospitals, as ranked by U.S. News & World Report on the basis of specialty rankings (Comarow, 2010).
rates of overweight and obesity (Lovasi et al., 2009; Mujahid et al., 2008; Sallis et al., 2009). Other studies have found green space buffers stress, especially in lower income socioeconomic groups (Maas et al., 2006), and have linked compromised outdoor air quality, which may be present in low-income neighborhoods, to coronary heart disease (Kan et al., 2008).
Poor housing quality can increase exposure to household dangers such as vermin, toxins, mold, and poor ventilation. These hazards may lead to a greater incidence of asthma (Bryant-Stephens, 2009). Poor-quality housing may also affect mental health: a study by Galea et al. (2005) links living in a neighborhood with dilapidated buildings to a greater incidence of depression after controlling for individual demographics and neighborhood median income. Poor-quality housing can also increase injuries from falls and fires and exposure to noise that can impair sleep. We expect to observe improvements in the health of participants who are able to move into better apartments using MTO vouchers.