Assisted Housing Mobility and the Success of Low-Income Minority Families:
Lessons for Policy, Practice, and Future Research [1]

Xavier de Souza Briggs & Margery Austin Turner

Introduction 

In the social policy field, where complex goals and seemingly intractable problems often make it hard to generate useful answers about what works, there is an understandable tendency to label demonstration programs either “successes” or “failures.”  In the context of assisted housing mobility initiatives, such as the court-ordered Gautreaux desegregation program and the federal Moving to Opportunity (MTO) demonstration, the narrow question is:  Did they “prove” that using housing vouchers to relocate poor minority families “works” or not?  As housing researchers with experience in both policy development and evaluation, we care deeply about what works, but we think this narrow framing is the wrong way to think about research demonstrations and policy experimentation more generally.

There are two main reasons for this.  First, it is all too easy, particularly where interventions are complex and end up testing what no one expected, to learn the wrong lessons from simple success-or-failure diagnoses.  Policymaking settings are notorious for relying on simple stories about cause and effect success and failure, and so this risk is significant.  Second, success-or-failure dichotomies tend to ignore the real story:  Why outcomes are so varied across contexts, client groups, and of course approaches to implementation.  Both Gautreaux and MTO offer valuable lessons about how to design and implement effective mobility strategies and under what conditions these strategies are likely to produce benefits for particular kinds of participants.

Two recent events make a close examination of these lessons especially timely.  First, the fortieth anniversary of the filing of Gautreaux, a landmark in American housing and social policy as well as civil rights, provides an invaluable opportunity to consider these lessons and their implications for ongoing policy development and practice.  So does the massive forced relocation of families, including many poor African-American families from poor and segregated neighborhoods, in the wake of Hurricane Katrina last year—a “displacement” that triggered contentious debates about how to rebuild both places and lives without recreating severe disadvantage. [2]

As for the wisdom of mining lessons, consider that Gautreaux “succeeded” in ways that no one anticipated when it was launched, generating new hypotheses about the potential role of assisted housing mobility in helping black families escape poverty (not just live in racially diverse communities).  These new ideas were further tested in the five-metro MTO experiment that was directly inspired by Gautreaux, by other, non-experimental mobility programs, and, to some extent, by HOPE VI, the largest-ever federal commitment to replace and revitalize distressed public housing developments.  Considered together, these efforts represent a second round of experimentation in the legacy of Gautreaux, a round that is now generating important new results that offer lessons about how, where, and for whom to pursue the dream of expanding opportunity through wider housing choice.

This article focuses, then, not on whether Gautreaux, MTO, or HOPE VI  “succeeded” or “failed,” but on what their results teach us about how to make assisted housing mobility policies more effective in the future.  To a great extent, these lessons are grounded in a simple logic model for understanding the strengths and weaknesses of assisted housing mobility as a tool for (a) improving the quality of life of the poor or (b) helping them escape poverty.  Logic models, also known as “theories of change,” consist of a set of causes and effects, outlined in a sequence (or “chain”), to clarify the premises and contingencies on which success depends. [3]  In our view, the most basic logic model for assisted housing mobility is this:  The target families will want to move to promising locations, given the motivations they bring and the information about choices that can be offered; they will be able to move (relocate), given counseling, search assistance, and/or other supports; the families will be able to stay in new locations long enough to benefit from relocating, given their own resources and resilience as well as housing market conditions and post-move supports; the families will be able to take advantage of new locations, given family resilience and resources plus external supports if available; and there will be significant net benefits for participating families (even if specific effects are mixed for subgroups of movers). 

Drawing on a growing body of rigorous demonstration research, we elaborate on these causes and effects, as well as the threshold conditions and contingencies.  We are particularly concerned that policymakers, implementers, and researchers alike understand the latter.  That is, what must obtain for the potential of these initiatives to be realized, and where are the initiatives most vulnerable?

Grounded in our logic model, the body of this article discusses the following five lessons:

  1. Targeting people: Many poor families in high-risk neighborhoods want to move, but not all are able to relocate successfully or to take advantage of opportunities their new neighborhoods offer (that is, to leverage the value of relocation to healthier places).  The challenges are particularly daunting for families with serious health or mental health problems, and in some instances, living in dangerous and distressed places may have robbed parents and kids of the resilience and resourcefulness needed to take advantage of new opportunities in new places.
  2. Targeting places: As we review below, both common sense and a growing body of research evidence teach us that living in a racially isolated, high poverty community undermines a family’s well-being and life chances, yet conversely, we know much less about how to define the “opportunity rich” neighborhoods to which we should be helping families move.  We suggest that, instead of simple proxies, such as a neighborhood’s racial composition or poverty rate, destination neighborhoods should be targeted on the basis of concrete opportunities, such as community safety, quality schools, or access to skill-appropriate jobs.
  3. Staying there, not just getting there: Residential instability—moving frequently and often from one risky environment to another—is a serious and largely unrecognized problem for many low-income renters.  Their own economic insecurity, combined with tight housing market conditions and conflicts with landlords over unaddressed problems, result in frequent moves. In the MTO case, many who made courageous moves to dramatically different neighborhood environments were unable, or chose not, to stay longer than a year or two [4].  In part, this may reflect some families’ efforts to improve on initial neighborhood choices (first placements), but we read the emerging research evidence to suggest that many families need help staying in better neighborhoods, once they manage to get to them, if they are to garner significant, long-term benefits.
  4. Leveraging the value, and mitigating the risks, of new neighborhoods: Moving improves outcomes for many families, but moving is no panacea for the problems associated with persistent family poverty.  The latest research evidence offers new insight about how moving may affect health (including mental health), risky teen behaviors, education, and employment.  But some families need additional assistance in order to take full advantage of new opportunities or to cope, say, with the different adjustments that low-income minority girls and boys seem to face after they relocate.
  5. Low risk to receiving areas, untapped potential: There is little or no risk of damage to receiving communities if the programs are well managed, and housing assistance programs such as Housing Choice Vouchers (Section 8) have a largely untapped potential to target a wider geography of housing locations.  The failure to do so owes to stigmas associated with assisted housing, families’ lack of information and capacity to search more widely, landlord refusal to rent to assisted families, discrimination that is increasingly difficult to detect, and—distressingly, in the context of Gautreaux’s encouraging lessons—weak incentives in the conventional voucher program to help families look for housing in “non-traditional” locations.

After discussing the evidence for each of these lessons, the article outlines their implications for the next generation of policy and practice in the field of assisted housing mobility.  We recommend that future policy address “readiness to move” and explain its significance, drawing parallels to “readiness to work” in the area of employment and training policy.  We also recommend targeting places according to tangible measures of access to opportunity, such as school quality and proximity to skill-appropriate jobs, rather than proxies such as poverty rate or racial make-up.  And we underline the importance of performance management as a framework to guide implementation, explaining why assisted housing mobility is particularly vulnerable to the strong-idea-weakly-implemented trap.  Finally, we recommend strategies to help families stay in better places and make the most of them, including remedies for housing instability and more-than-housing strategies (“mobility plus”) that would help address disabling illnesses, social isolation, lack of childcare and transportation, low skills, and other barriers to getting ahead—barriers that many participants in mobility programs face before and after they relocate.

            To clarify our sources, the evidence presented here is drawn primarily from rigorous research on three mobility initiatives – interventions that have enabled low-income families to move from high-poverty to lower-poverty neighborhoods:

·        Gautreaux demonstration (non-experimental).  Research has been conducted over many years (primarily by scholars at Northwestern University) on low-income, minority families who received special-purpose housing vouchers, under court order, to move from poor, predominantly black neighborhoods in the city of Chicago to other city neighborhoods as well as racially integrated suburban communities. [5]

·        Moving to Opportunity (MTO) demonstration (randomized experiment).  Research has been conducted by researchers from a number of different institutions on a carefully controlled experiment to test the impacts of helping low-income families move from high-poverty assisted housing projects (in Baltimore, Boston, Chicago, New York, and Los Angeles) to low-poverty neighborhoods throughout their metropolitan regions.  Here, we rely not only on earlier qualitative and statistical studies but on new evidence emerging from qualitative interviews and ethnographic fieldwork conducted among MTO families in greater Boston, Los Angeles, and New York.

·        HOPE VI program – research conducted by the Urban Institute on what is happening to the original residents of five distressed public housing projects that are being demolished and replaced with mixed-income housing. Many of the residents moved out and stayed out of their original (project) neighborhoods, while others returned to enter the mixed-income developments.

LESSONS

1. Targeting people

The first key lesson from these initiatives, and step one in our logic model, is that assisted housing mobility is feasible, thanks to family motivation and support systems of counseling, search assistance, and other aid, which we know how to deliver more and more effectively.  Contrary to the skepticism that the minority poor strongly prefer to live among “their own,” many low-income families, including blacks, Hispanics, Asians and whites will volunteer for the opportunity to move from high poverty areas, typically in inner cities, to better neighborhoods in the same cities or in the surrounding suburbs.  For example, over 5300 families from assisted housing developments in five cities—about one-fourth of all eligible, assisted households in those cities—applied to participate in the MTO demonstration,[6] and during most of the period that the Gautreaux program was underway, the number of applicants vastly exceeded the available assistance. [7]

Many, though certainly not all, of those who receive assistance succeed in using the combination of a voucher and search assistance to find and “lease up” housing in lower-income and less racially segregated communities.  For example, the share of MTO families who were successful in moving ranged from a low of 34% in Chicago to a high of 61% in Los Angeles. [8]  Based on surveys, the families most likely to succeed were those that were more motivated about moving and more optimistic about their chances of success. [9]  In addition, they tended to own cars and to have fewer kids.[10]  In addition, families with strong social ties to their neighbors or with a disabled member were less likely to lease up in the private market, and Hispanic families were less likely than African Americans to be successful in moving, net of the other factors. [11]

Mobility counseling and search assistance make a difference, not only in families’ ability to find a house or apartment, but also in the types of neighborhoods to which they move.  Families who receive vouchers without mobility assistance are not as successful in moving to low-poverty neighborhoods.  In particular, a recent analysis of voucher locations in the fifty largest metro areas nationwide illustrates that minority and central city voucher recipients are not gaining access to the same opportunities as white and suburban recipients. [12]  Specifically, 25% of African American recipients and 28% of Hispanic recipients use the vouchers in high-poverty neighborhoods, compared to only 8% of whites. [13]  In the MTO experiment, some families were randomly assigned to receive conventional Section 8 vouchers, without any supplemental housing counseling or search assistance.  These families moved to neighborhoods with significantly higher poverty and crime rates than the families who received both vouchers and search assistance and who were restricted to using the vouchers in low poverty areas.[14]

Closely and logically related to the issue of who wants to relocate and who can is the issue of who benefits most from relocating. As we explore below with new fieldwork data on MTO, and while we still have much to learn about this, some of the factors that shape capacity to move also shape capacity to cope – and ultimately to thrive – in a new environment.  Both theory and empirical evidence strongly suggest that individual and family characteristics interact with neighborhood environment in complex ways and play a hugely important role in shaping social outcomes over time. [15]  Families with chronic health and mental health problems struggle to make positive connections with new service providers; they are unable to track down the child care single parents rely on to find and keep a job; and they may be unable to take other steps necessary to get by and get ahead.  In general, families that start with better information about school and job choices, who are working or have worked regularly before, and who show higher levels of optimism and resourcefulness, appear to be better poised to take advantage of new locations. [16] 

These findings highlight the critical importance of quality counseling – both in preparation for moving and after the move.  But they also raise difficult questions about screening out the least likely to succeed, about what kinds of families are or are not well suited to benefit from mobility.  Should scarce assistance resources be targeted to the families who appear most likely to succeed in moving and in leveraging the opportunities offered by new neighborhoods?  Would other families perhaps be better served by programs and services that address their immediate health problems, help them stabilize their lives, and strengthen their job skills?  These programs could enhance “readiness to move.”  Analogous to readiness to work, which highlights the need for preparatory interventions before traditional job training and placement, readiness to move highlights the importance of (a) understanding the heterogeneity of families we may want to serve with housing mobility programs; and (b) targeting family-level success conditions in addition to, and perhaps in advance of, conditions tied to subsidy levels, counseling approaches, location targets, and other features of programs that have absorbed mobility experts and advocates so far.

Would such a preparatory strategy unfairly deny opportunities available to otherwise deserving, not to mention highly motivated, families?  As researchers continue to monitor the use and effectiveness of assisted housing mobility programs, we need to learn more about the family attributes that contribute to success, and to think critically about the pros and cons of targeting in different ways.

2. Targeting places

A program model centered on moving to better locations naturally poses a threshold question:  What does “better” mean?  Ongoing research highlights the importance of the criteria used to identify suitable destination neighborhoods for participating families.  The Gautreaux demonstration – and subsequent court-ordered desegregation remedies – required minority families to move to majority-white neighborhoods.  Families that received special purpose vouchers and mobility counseling through MTO were required to use their vouchers in census tracts with poverty rates below 10%. [17]  Nationally, the vast majority of these tracts are located in majority white, stable, suburban communities.  But while many MTO families succeeded in moving to low-poverty neighborhoods, they differ from Gautreaux families in that most MTO movers remained in the same central-city jurisdiction and moved to neighborhoods that were majority minority. [18]  In many cases, these were what urban analysts call “transitional neighborhoods” that were becoming poorer over the course of the demonstration.

This set of initial location outcomes may reflect the tendency of MTO program counselors and the participants themselves to steer toward areas where landlords were known to accept federal Section 8 vouchers, or it may reflect the many challenges involved in moving to majority-white communities in the suburbs.  We return to these issues below, in the discussion of performance targets and careful implementation.  But for now, we underline the strong possibility that MTO’s specific failure to move a large number of families in the experimental group to more stable, racially diverse neighborhoods in higher performing suburban school districts may limit benefits for families over the long term.

            Future mobility programs should rethink the criteria used to define eligible destination neighborhoods as “opportunity areas.”  Both racial composition and poverty rate are in fact proxies for attributes that make a neighborhood a good place to live and, more specifically, a promising place for low-income parents to raise their children. One possibility would be to target neighborhoods that are far away from distressed, high-poverty (and majority-minority) neighborhoods, as Gautreaux’s suburban destinations were. [19]  This would eliminate many neighborhoods that are in the path of racial or economic transition.

But it would also make it more difficult for participating families, especially teens, to return to the old neighborhood on a regular basis.  This might isolate some families from key institutional resources (service providers or civic groups) and social resources (support networks of relatives and friends), as well as from key risks in the old neighborhood.  What families gain and lose depends very much on where the risks and resources in their lives were located at the outset. [20]  Contrary to the folk wisdom about cohesive neighborhoods of poor people, some families who participate in mobility programs report no support networks at all in their starting-point neighborhoods; their useful ties were elsewhere.  Also, some families had very weak links to institutions in those neighborhoods.  On the other hand, there is evidence that many families relocated as part of HOPE VI demolition have lost supportive ties, specifically ties from the projects left behind, and not replaced them in the first few years after moving. [21]  It is not yet clear to what degree MTO movers lost social support, or experienced a shift in informal support, based on relocation distance or other access factors; as noted above, there was no mean impact on social support for the experimental group as a whole. [22]

An alternative to the distant-moves approach would be to focus explicitly on identifying destination neighborhoods that provide access to specific child-rearing opportunities such as high performing schools or concentrations of entry-level jobs (which are increasingly suburban).  Researchers at the Kirwan Center are exploring strategies for mapping multiple indicators in order to identify neighborhoods that are rich with opportunities of different types. [23]  Tracking these indicators over time is crucial, since the geography of opportunity (and of risk) shifts with investment and disinvestment across neighborhoods, population migrations, and other dynamics.

Targeting neighborhoods based on the presence of specific opportunities will likely produce an array of destination communities that are predominantly white and lower poverty, even if race and poverty are not explicit selection factors.  Because of America’s legacy of discrimination and segregation, there are relatively few predominantly African-American neighborhoods that would qualify as “opportunity communities”. [24]  But not all white neighborhoods (or all low-poverty neighborhoods) are necessarily rich in opportunities.  Thus, focusing explicitly on the positive qualities that distressed, inner-city communities lack could significantly strengthen the performance of assisted housing mobility programs.

In addition to the challenge of defining desirable destination neighborhoods, important questions remain about the extent to which participating families must be widely dispersed in their new locations.  There are strong arguments against reconcentrating large numbers of families in just a few housing developments or census tracts, but there are also benefits to helping people sustain networks of friendship and support with people who live close by.  Some mobility counseling programs ask successful participants from previous years to host small gatherings of prospective movers, enabling families to get to know each other and to learn more about communities to which they might move.  Others encourage small numbers of participants to move together to the same building or subdivision in an opportunity-rich community, so that they feel less lonely and isolated. [25]  In effect, such mini-enclaves combine the strengths of “supportive housing” (where families with similar backgrounds receive multiple services) with the advantages of a healthy neighborhood environment. [26]  In sum, there are several promising alternatives for targeting place—defining better locations and relocating families to them—and future policy and research should reflect that fact.

3. Staying there, not just getting there

Following the logic model we outlined above, many of the benefits of relocating to so-called opportunity areas hinge on sustained exposure to the richer resources and lower levels of risk in such places, [27] as well as the positive life routines and relationships with service providers—health providers, schools, and more—that stable housing enables. Yet by the time of the interim evaluation some four to seven years after initial placements, many movers in the MTO experimental group had moved again (in some cases more than once)—and typically to poorer communities. [28]  And nationally, while black families are as likely as white families to exit poor neighborhoods over time, blacks are much more likely to fall back into poor areas through subsequent moves. [29]   This mobility from poor area to poor area, and from nonpoor ones back to poor ones, is particularly problematic for low-income renter households, whose mobility has increased in recent decades. [30]

This evidence, from the most carefully documented experiment in assisted housing mobility on one hand and from longitudinal national data on American families and housing markets on the other, underscores the importance of helping families stay out, not just get out, of risky places.  Still, the MTO evidence only sheds light on the medium term.  Long-run evidence, which comes from administrative data on Gautreaux participants, indicates that most relocatees to low-poverty, majority-white suburban communities have not moved back to poor, racially segregated neighborhoods, and furthermore that initial placement in a racially diverse, low poverty area is a good predictor of moves to similar areas later on. [31]

Given that about half of the MTO experimental group did not successfully lease up at the placement stage, why are so many who did struggling to stay in lower poverty areas?  Is it dislike of the new areas or feelings of isolation from relatives and friends left behind?  For the most part, no.  The major reasons are rent increases and problems with landlords over problems with the housing unit, factors the Census Bureau associates with “involuntary” moves. [32]  Though some movers found new neighborhoods uncomfortable socially, there is no evidence that dissatisfaction with the initial placement areas was generally a problem—to the contrary, rates of satisfaction were very high overall—or that movers suffered a loss of social support or socializing time with relatives or friends. [33]

These findings suggest, first, that interventions focused on the involuntary factors would help relocated families stay in healthier places once they get to them.  Pre-move counseling should seek the best-possible initial placements for families, not the quickest placements, which may be lowest cost in the near term and which may reflect the use-it-or-lose-it pressure that confronts voucher users who face time limits on search. [34]   Post-move counseling may help resolve some of the problems with landlords and their units that trigger tenant dissatisfaction and decisions to move.  Efforts to recruit and retain a broader pool of landlords are also important, since small pools tend to concentrate voucher-eligible units in a narrow geography and to relegate tenants, in effect, to a less competitive submarket of landlords.  And fostering institutional connections, such as to faith-based or secular institutions that act as “welcome wagons” is also promising, as we explore in the next section.

Third, where rent increases in particular markets are a significant factor, flexibility in the management of the voucher program (such as granting “exception rents”) can help and is well tested.  Also, supply-side strategies, which expand the supply of rental units that remain affordable over time and often place units under management by “social landlords” (nonprofits or socially responsible private firms), are an important structural solution, especially since many suburbs have little or no history of developing affordable rental housing.  Markets stratify by income (ability to pay), of course, but the local politics of land regulation acts, more specifically and directly, to exclude affordable, entry-level housing development from more opportunity-rich areas, especially in suburban jurisdictions and more desirable neighborhoods of central cities. [35] Khadduri, in the most comprehensive review yet done of the vouchers versus housing production (demand versus supply-side) policy question, concluded that supply-side subsidies, while often more costly than vouchers on a per-family-served basis (at least in terms of fiscal costs), are likely to be appropriate in several specific contexts:

·        For populations requiring special services, such as the homeless in transition, the low-income disabled, the elderly in assisted living settings, and other special-needs groups;

·        Where production subsidies are part of larger, place-based revitalization (area upgrading) strategies that go beyond housing to significantly improve services and improve quality of life;

·        Where production subsidies enable residents of subsidized housing to live in better neighborhoods; and

·        Where housing markets are tight (and affordable units scarce) or to preserve affordable housing where neighborhoods are gentrifying. [36]

In addition, production strategies are important wherever the supply of housing is not configured adequately for an area’s families, such as where larger-unit apartments are not being built to serve larger families with children.  In some instances, as in the public housing program and certain nonprofit-run subsidized housing developments, supply-side strategies also help less creditworthy families improve their credit without losing basic shelter and becoming homeless or “doubled up” (moving in with friends or relatives who may be struggling too).

In sum, while debates about housing mobility and its lessons have been dominated by voucher-based or “demand-side” strategies for the past decade or more—in part because both the Gautreaux and MTO programs were voucher based—we believe that it is time to address the adequacy of affordable housing supply, and so-called unit-based mobility strategies,[37] once again. This is especially important in the tight rental markets that many strong local economies have become, [38] and it was a central conclusion of the most important national policy statement in recent years, that of the Bipartisan Millennial Housing Commission appointed by Congress. [39] Yet the supply issue, a quiet crisis, remains unaddressed by national policymakers.

Third and finally, when subsequent moves are a must for particular families, Chicago’s Housing Opportunity Program (HOP) shows that second-move counseling can help families stay in lower poverty areas.  Rigorous analysis of location choices among families participating in HOP indicates that mobility assistance has a measurable impact on neighborhood outcomes over repeat moves. [40]

4. Making the most of new neighborhoods

Next in our logic chain, for low-income families who are willing and able to relocate to healthier neighborhoods, and then able to stay in them “long enough,” there is the question of making the most of new places—the final key to realizing significant benefits. While we are still learning much about how mover families actually make use of new locations, the dominant perceptions of those who move from high to low poverty neighborhoods are clearly positive. That is, research to date clearly establishes that assisted housing mobility yields dramatic improvements in perceived neighborhood quality for participating families.  In this section, we review the evidence for those perceptions, as well as on benefits realized, and we describe preliminary evidence on how families’ use of new places varies by type of family.

In particular, families who successfully move end up in dramatically safer neighborhoods.  MTO research finds that moving with an MTO voucher (to low-poverty neighborhoods) produced a 30.3 percentage point increase in perceptions of safety. [41]  Families moving with a regular voucher (generally to intermediate-poverty neighborhoods) also experienced significant – though smaller – gains in perceived safety, and similarly, eight out of ten HOPE VI families who moved with vouchers describe their new neighborhood as safer than their neighborhoods of origin. [42] 

Families place tremendous value on enhanced safety, telling interviewers what a relief it is not to worry constantly about the threat of violence, including indiscriminate or “random” violence in ghetto-poor neighborhoods left behind. [43]  Parents emphasize the freedom to let children play outside and to come and go from the home, free from fear.  For example, MTO participants reported:[44]

It gave me a better outlook on life, that there is a life outside of that housing . . . Overall I think I was more happy to be in this area because of my kids and I didn’t want them to grow up around seeing gangs.

 

. . . you can wake up every day and we’re not worried about seeing anybody getting shot and no gang members, nothing like that and it’s quiet and it’s cool and calm up here.  In the city there’s a lot of activities that’s going on that’s negative.  Here there’s a lot of positive.

 

I think if I had stayed in [the projects] I’d be a different person than what I am now.  I’d be a wild person; I’d probably be in a gang or something like that… Since I’ve moved out here, I think I got a better chance than I do out there.

 

These improvements in neighborhood environment have the potential to contribute to significant improvements in the well-being of both adults and children.  Specifically, research on families participating in the Gautreaux and MTO demonstrations provides evidence of gains in health, educational success, and employment and earnings.  This section briefly reviews the findings for each of these important domains.

Adult mental and physical health.  Among the strongest findings to date from the MTO demonstration are results showing substantial improvements in the health of women and girls who moved to lower poverty neighborhoods.  In particular, adult obesity is significantly lower among those who moved, a notewor­thy effect given the national attention now focused on the dangers of obesity for our long-term health.  MTO parents (who are mostly single mothers) and adolescent girls (ages twelve to nineteen) also enjoyed signifi­cant improvements in mental health, including reductions in psychological distress and depression, and increas­ing feelings of calm and peacefulness. [45]  These gains are statistically significant but also large in absolute magnitude—on par with mental health gains typical under the most effective psychotherapeutic treatments available.

Educational success.  Gautreaux research found striking benefits for children whose fami­lies moved to low poverty suburban school districts.  They were substantially more likely to complete high school, take college-track courses, attend college and enter the work force than children from simi­lar families who moved to neighbor­hoods within Chicago. [46]  To date, there is no evidence that MTO moves have led to better educational outcomes, possibly because so few children are attending significantly better schools in advantaged school districts or because it may be too early to detect benefits. [47]

However, new evidence emerging from in-depth interviews and ethnographic research is helping us understand these findings more clearly by offering insights on how parents think about school choices.  Some parents are unaware of the school choices available in their new neighborhoods, in part because most rely on limited information resources, such as word-of-mouth referrals from similarly situated relatives or friends.  Moreover, most parents emphasize perceived safety and convenience as indicators of a “good” school rather than reliable evidence on academic supports (such as small class sizes, strong counseling, and tutoring) and achievement.   These parents placed high priority on ensuring that their children would be safe at school, even if this meant staying at the school in the original neighborhood.  At least there the dangers were well known and informal “alliances” had already been negotiated with peers and problem youth.  Finally, some parents think their kids will benefit more from the stability of staying in the same school – and the same after-school care arrangements – than from moving to a new and unfamiliar school.  That is, parents recognized moving per se as disruptive and wanted school to serve as a source of social and emotional stability in their children’s lives. [48]

Delinquency and risky behavior.  Some of the early research on MTO families in individual sites suggested that young people whose families moved to low-poverty neighborhoods were engaging in less risky behavior and committing fewer crimes.  In Bal­timore, for example, moving to a low-poverty neighborhood was found to cut violent crime arrests among juveniles roughly in half. [49]  The most recent and com­prehensive data from MTO, however, suggests that moving to a lower-poverty envi­ronment is reducing crime, delinquency, and risky behavior among teen-aged girls but not boys. [50]

Qualitative research is currently under way to better understand why boys in the MTO experimental group do not seem to be enjoying the same ben­efits from mobility, at least within the same timeframes, and why boys may even suffer setbacks relative to counterparts in the control group.  One pos­sible explanation is that black and His­panic boys moving to integrated or pre­dominantly white neighborhoods are not engaging in any more criminal behavior but are being arrested more due to racial profiling or higher rates of detecting crime in low poverty areas.  Another pos­sibility is that some boys respond differently to the loneliness, fears, or boredom associated with relocation:  new peers and expectations, a loss of familiar activities, the felt need to act tough to gain respect, [51] and more.  Parents also tend to “manage” (monitor and discipline) boys and girls differently.  Finally, MTO results are tricky to interpret for another reason:  because boys and girls enter into and “age out” of specific risky behaviors—fighting, stealing, smoking, early sexual activity, and more—at different times, in different ways, for somewhat different reasons. [52]

Ongoing qualitative research also helps us better understand gains in mental health and reductions in risky behavior among adolescent girls.  In the distressed and violent communities from which they moved, many of these girls were sexual targets for older boys and men.  They suffered from sexual harassment, pressure to have sex, and even rape.  Escaping from these environments appears to offer a tremendous sense of relief and freedom for adolescent girls (and their parents), not only contributing to short-term gains in health and well-being but also potentially enabling them to stay in school and postpone child bearing over the longer-term. [53]

Employment.  The current evidence on how mo­bility affects adult employment and earnings is mixed and still somewhat inconclusive. [54]  Over the long term, Gautreaux families that moved out of segregated and distressed central city neighborhoods achieved greater employment success than their counterparts who stayed.  Specifically, employment rates were higher among Gautreaux participants who moved to the suburbs than for those who moved within the city of Chicago. [55]  And recent research using administrative data on wages and welfare receipt finds that Gautreaux women who moved to predominantly white neighborhoods with moderate to high resources spent significantly more time employed and less time on welfare. [56]

MTO results are not yet as clear, but some suggestive evidence is beginning to emerge.  The interim evaluation found no significant impacts on employment, earnings, or receipt of public assistance across the five demonstration sites. [57]  When interim evaluation results are stratified by site, however, we see significant increases in rates of employment in Los Angeles and in earnings in New York. [58]  In addition, exploratory analysis of variations in employment effects for different types of MTO participants yields suggestive evidence that women under forty may experience employment gains after the first year. [59]  Nonexperimental analysis also finds that, net of other factors, MTO adults who moved to low-poverty neighborhoods outside of the central city earned $55 per week more than those in control neighborhoods, with those who moved to a suburban neighborhood earning $75 more. [60] 

Here, too, we are using qualitative research methods to examine the reasons why relocation may contribute to better employment outcomes in some cities than others, for some categories of families and not others, only in some types of neighborhoods, etc.  Very few MTO families cited “getting a job” or “being near my job” as their most important reason for wanting to move; families were primarily motivated to escape unsafe areas.  Furthermore, it is by no means clear whether most new locations offer better access to jobs – despite their lower poverty rates and dramatically improved safety.  Finally, those not working clearly face multiple barriers that are not directly affected by location, such as disabling illnesses, limited skills, and a lack of reliable childcare or transportation.

Use of new places. The initial evidence from in-depth qualitative interviews and ethnographic fieldwork on MTO families in greater Boston, Los Angeles, and New York suggests that mover families’ use of new places varies according to the structure of their social relations, specific family needs that may or may not be met near the home, the nature of access to other places that are significant in their lives, and levels of trust toward new neighbors. [61]  Movers may remain quite isolated from new social ties, local service providers, and other institutions. For some movers, new neighborhoods are, at least for significant periods of time, residential locations rather than important social worlds.  For example, for a Latino family in the Los Angeles experimental group, life revolves around church, the private school run by church, and a socializing with relatives and friends who live throughout Southern California.  Neither the “old” (housing project) neighborhood nor the new one is a world of social influences; the new one is a convenient location with affordable homes, pleasant neighbors (casual ties at best), and a few playmates for the children.

Summary. Although the research literature provides strong evidence that neighborhood conditions have an important influence on people’s lives, they are obviously not the only and probably not the main source of influence. [62]  Some families and individuals can withstand the disadvantages of even the most distressed environment -- for example by “bounding” (isolating) family members from neighborhood risks and “bridging” to resources outside the neighborhood [63] -- while other families are likely to encounter serious problems regardless of the neighborhoods in which they live.  Likewise, the MTO evidence is that some domains (health, mental health) register significant benefits while others (employment, education) may not, or not everywhere, depending on the specifics of relocation outcomes and family choices over time. Girls may benefit more than boys; some boys may even experience risky setbacks, at least for a time, after relocating to lower poverty areas.  Though enhanced safety is a boon to families whether or not they take specific steps to make best use of new places, other benefits depend on family agency and resourcefulness.  The evidence is that some families are both more willing and more able to make useful connections (to service providers, neighbors, or others) after they move.  And starting points—a family’s initial strengths and vulnerabilities—matter, for straightforward reasons.  If a poorly treated chronic illness, for example, undermines a family’s capacity to make the most of a new place, or leads to even deeper isolation thanks to the disruption associated with moving per se, then the aggregate benefits of assisted housing mobility programs will unavoidably be a function, at least in part, of who is targeted for assistance, as we argued above.

5. Low risk, untapped potential

In the context of complexity—that of the programs and of family’s lives as well—it would be easy to miss one of the more striking features of assisted housing mobility initiatives: They are tiny, given the demand and promising results.  A final, broad lesson therefore is that these efforts represent significant untapped potential relative to the risk they pose to clients and communities. [64]

Consider location patterns for the federal rental voucher program on which the best-studied assisted mobility efforts are based.  As we noted above, minority voucher holders are especially likely to use their vouchers in racially segregated central-city neighborhoods with high poverty rates, despite the availability of units at affordable rents in many low poverty neighborhoods, including those in suburban areas.  HUD research on the location of Housing Choice Voucher households in the nation’s fifty largest metropolitan areas finds that the stock of rental housing in which vouchers can potentially be used is widely dispersed. [65]  Specifically, within the fifty largest metro areas, the voucher program utilizes only about 2% of all occupied housing units and 6% of all rental units with rents below the applicable Fair Market Rents defined by HUD. [66]  Virtually all census tracts contain at least some units below that threshold, and 83% have at least some voucher recipients living in them.  Nationwide, vouchers are generally not clustered geographically: In 90% of all tracts with any voucher recipients, the program accounts for less than 5% of all households. [67]  But where vouchers are clustered, the clustering is in high-poverty, mostly minority central-city neighborhoods. [68]   Thirty years since inception, and particularly with regard to minority clients, the federal voucher program falls far short of the policy vision of “a decent home and suitable living environment” for every American family, a vision enshrined in the Housing Act of 1949.

The main reasons for this untapped potential are persistent barriers: voucher holders’ lack of information and capacity to search more widely for housing, social stigmas held by landlords and community gatekeepers, oppositional politics, discrimination that is increasingly difficult to detect, and numerous bureaucratic impediments.  In this section, we briefly review the state of our knowledge about these important barriers and how they might be overcome, the nature of risks posed by these initiatives, a