For almost a year on the campaign trail, John Edwards has been calling the more than 37 million American families living below the poverty line a “national shame.” Other presidential candidates have been talking about disadvantaged families, too, but Edwards alone has focused on racial and economic segregation as an obstacle to ending poverty. After all, black-white segregation remains extreme in more than half of the fifty largest metropolitan areas in the country. Concentrated poverty also tends to track racial segregation, creating a “geography of opportunity” that looks very different for black families relative to white families. To address the problem of geographic separation and concentrated poverty, he proposes to do away with public-housing projects and replace them with one million housing vouchers to give poor families a choice about where to live.

Edwards is not the first to focus on mobility as a solution to poverty. Urban sociologists have long described the horrors of public housing and drug-related violence, along with the high levels of racial isolation and segregation common in many American cities, and they have raised doubts about whether those problems can be addressed “on site.” In the summer of 2000, Owen Fiss made a forceful case in the pages of this magazine for the idea that justice requires us to support residential-mobility programs for the poor.

It makes intuitive sense to think that helping poor minority families leave dangerous city neighborhoods would bring about immediate improvements in their lives. We know that neighborhoods where white middle-class families live (often in the suburban outskirts of major cities) offer many resources, such as high-quality schools and better jobs. Unfortunately, it’s not so easy for poor black families to obtain access to such opportunity-rich communities, at least in part because of housing discrimination and lending practices that channel and trap them in undesirable neighborhoods. So it will take government intervention to help poor minority families find better places to live.

But before we invest resources in large-scale deconcentration programs, we need to know how and how well such an approach might work. Will helping poor families escape the ghetto break the cycle of poverty? If poor families move to suburban neighborhoods, will they get better jobs? Will their children’s academic achievement improve? The past few decades have seen an explosion of research devoted to answering just these questions. It makes sense to use some of this new knowledge as we think about new policy.

* * *

What do we know today about whether neighborhoods matter? We know that having affluent neighbors is linked to better educational outcomes, especially for younger children. We know that living in racially segregated, poor neighborhoods correlates with sharp increases in high-school-dropout and teenage-childbearing rates, thanks largely to an influential 1991 study by economist Jonathan Crane. Other scholars have found that young people growing up in high-crime, high-poverty neighborhoods are more likely to engage in delinquency, sex, and substance abuse than their peers in more affluent communities.

Why are poor neighborhoods so detrimental to youth? One argument is that such neighborhoods provide little exposure to positive, working-adult role models who can monitor young peoples’ deviant behavior. Economists put less weight on role models and more on rational responses to opportunities. While teenagers hardly make perfectly rational decisions about work, school, and delinquency, the choice to participate in risky behaviors depends in part on local labor market prospects. In inner cities, where youth have very little access to conventional means of achieving success, high-risk choices may make more rational sense.

Should we focus social progams mostly on parents and children themselves or invest in making sure they live in communities where they can prosper?

Poor neighborhoods may limit the employment prospects for adults as well. Researchers have found that forty to fifty percent of jobs are obtained through social networks that transmit information about positions or offer concrete help during job searches. Areas with higher family incomes and greater levels of education usually have more people working in living-wage and middle-income jobs who can serve as employment resources. The isolated inner city prevents the development of effective social networks among black adults that could lead to better jobs.

As rich as these studies are, however, social science research that compares families living in poor neighborhoods with families living in non-poor neighborhoods only gets us so far. Families choose neighborhoods and schools. This makes it hard to know whether neighborhoods themselves matter more than parents or children’s traits. What leads families to pick a certain neighborhood is probably also related to other aspects of the family that affect child development. For example, divorce or job loss might lead a family to move to a less desirable neighborhood. But divorce and job loss also have direct effects on kids—on how they perform in school and whether they get in trouble with the law. So less desirable neighborhoods might simply be home to concentrations of families that are, on average, having a harder time. To understand whether neighborhoods themselves have independent effects on families and youth, an experiment along the following lines might prove fruitful. Imagine tracking two families who have exactly the same characteristics and sending one to a very poor neighborhood and the other to a wealthy suburb. Such an experiment could tell us where to invest our resources if we want to improve family life. Should we focus social programs mostly on parents and children themselves (in the form of income support or job training) or invest in making sure they live in communities where they can prosper?

Of course, social experiments outside the laboratory setting are difficult. However, precisely such residential-mobility “experiments” have been implemented in recent decades. These residential-mobility programs differ in design, but allow poor families to move to better neighborhoods with targeted vouchers. Unlike the traditional Section 8 housing program, families cannot use the vouchers in just any neighborhood or with any landlord willing to take them. Instead, families are either assigned to units in more advantaged areas or they have to choose to live in communities that fall under a certain racial composition or poverty threshold (as defined by census data).

The first major residential-mobility program, the Gautreaux program, came as a result of a 1976 Supreme Court ruling in a lawsuit filed on behalf of public-housing residents against the Chicago Housing Authority and the U.S. Department of Housing and Urban Development. The suit charged these agencies with racially discriminatory practices in the administration of the Chicago’s low-rent housing programs. Between 1976 and 1990, the court remedy provided vouchers for more than seven thousand families in the Chicago metro area to move to non-segregated communities. About half moved to mostly white suburbs and half moved to non-public-housing city neighborhoods. While the choice to sign up with the program and relocate was voluntary, families did not choose the housing units to which they relocated. They were assigned to apartments in new neighborhoods by housing counselors (who were working with landlords) on the basis of their position on a waiting list, similar to a random-draw lottery.

Providing housing vouchers and housing support services, but not employment or transportation assistance, Gautreaux permitted low-income blacks to live in middle- and upper-income white suburbs. Participants moved to more than 115 suburbs surrounding Chicago, making it possible to compare families who moved to mostly white suburbs with those who moved to primarily minority city neighborhoods. Suburbs with black populations of more than thirty percent were excluded by the consent decree. A few very high-rent suburbs were excluded by funding limitations of Section 8 certificates.

In the late 1980s and early ’90s, research on Gautreaux (as summarized by James Rosenbaum in the aforementioned 2000 issue of Boston Review) showed large correlations between placement neighborhoods and gains in adult employment and children’s education. Low-income black children moving to middle-class white suburbs enjoyed better educational resources than those relocating to other city areas, and suburban mothers benefited from higher levels of post-move employment. The early evidence was powerful, suggesting the life chances of low-income families depend not just on who they are, but where they live. Critics questioned the findings, however, raising doubts about whether families who moved to suburbs and those who moved to other city neighborhoods were really initially comparable.

Recently, my colleagues and I have improved on previous research to explore Gautreaux’s long-term outcomes. We used administrative sources, such as state welfare records, as well as multiple neighborhood-level indicators, and we considered pre-program differences across families. We found that Gautreaux was indeed successful in helping public- housing families relocate to safer, more integrated neighborhoods. These families came from very poor neighborhoods, with census-tract poverty rates averaging forty to sixty percent, or three to five times the national poverty rate. Through the program, they moved to neighborhoods that were seventeen percent poor—less than half the original rate (the poverty rate for those who moved to the suburbs was even lower, at five percent). And by the late 1990s, fifteen to twenty years after relocating, Gautreaux mothers continued to live in neighborhoods with lower poverty rates. Neighborhood household incomes (in 2000 dollars) improved to more than double the original neighborhood-median income. Gautreaux also achieved striking success in moving low-income black families into more racially integrated neighborhoods. The origin communities were 83 percent black, while the program placed families in communities that averaged 28 percent black (most of the suburban moves were actually to areas that were more than ninety percent white). Some Gautreaux families later moved to neighborhoods that contained considerably more blacks—48 percent, on average—or a fairly even balance of blacks and individuals from other races (suburban movers were living in areas that were about 36 percent black). Despite the increase, these levels were less than half of what they had been in the origin neighborhoods.

Did the early socioeconomic gains in Gautreaux persist twenty years later? Based on our recent research using state and federal data on employment and welfare recipts, the answer seems to be: yes, to some extent. Women placed in more affluent, less segregated neighborhoods spent slightly less time on welfare and more time employed (with higher salaries) than women placed in areas with mostly black residents, more crime, and higher unemployment rates. Another striking finding is that there seems be a “second generation” of Gautreaux effects. Children of Gautreaux families who had relocated to less segregated neighborhoods are more likely to reside in such neighborhoods as adults. This suggests that the Gautreaux program may have helped poor black families become comfortable in more integrated communities.

* * *

The numbers tell one kind of story, but they do not give us much feel for how neighborhoods matter to these families. Fortunately there is also data from interviews with Gautreaux mothers that describe how moving to more affluent communities changed their lives. After the move, mothers described a new a sense of control over their lives and remarked that their new environments helped them to see that they had the ability to improve their circumstances. Specifically, women reported feeling better about not having to put down a public-housing address on job applications. Other women noted that they and their children got to know more white people and would feel less intimidated about interacting with whites in the future. Mothers also reported social responsiveness from their neighbors. Many said that they could count on neighbors’ help if their child misbehaved or seemed at risk of getting into trouble, if their child was sick and couldn’t attend school, or if there was some threat to their family or homes. Some reported a willingness to take jobs because they could count on a neighbor to watch their child if they were late getting home from work. Social and psychological factors seemed to play a large role for Gautreaux families in escaping persistent poverty and unsafe inner-city neighborhoods. (More recent interviews with Gautreaux mothers who relocated to moderate and low-poverty city neighborhoods suggest that while their immediate neighborhood was safe, the larger community to which their children had easy access continued to be dangerous.)

While the Gautreaux results had a profound effect on social scientists and policymakers, it wasn’t a perfect experiment. All families moved somewhere, and while their new communities helped us understand neighborhood effects, there was no control group of families who stayed in the old neighborhood. Some policymakers and academics were not convinced that the Gautreaux results really proved that neighborhoods mattered more than individual motivation or policies aimed at the economy.

In part to test the promise of Gautreaux, the Moving to Opportunity (MTO) program, legislated and funded in the 1990s, was designed as a rigorous social experiment. Beginning in 1994, MTO allowed public-housing residents in five cities (New York, Boston, Baltimore, Chicago, and Los Angeles) to apply for a chance to receive a housing voucher. Families were assigned at random to one of three groups. The experimental group received a Section 8 voucher that would allow them to rent an apartment in the private market, but they could only lease in census tracts with 1990 poverty rates of less than ten percent. (Unlike Gautreaux, there were no racial restrictions on the destination neighborhoods.) This group also received housing counseling to assist them in relocating. Another group received a Section 8 voucher with no geographical restrictions. Finally, the control group received no new housing assistance but could continue to live in public housing or apply for other housing assistance.

As with Gautreaux, families who moved with MTO vouchers relocated to neighborhoods with much lower poverty rates than their public-housing neighborhoods: the new neighborhoods were eleven percent poor on average, compared with about forty percent or more in the original communities. When they were contacted after four to seven years, families who had moved with low-poverty vouchers were still in neighborhoods that were significantly less poor, but many moved from their first MTO communities into more disadvantaged ones. MTO set no race-based limits on placement neighborhoods, and MTO families moving in conjunction with the program both began and ended up in neighborhoods with high minority concentrations.

Unsurprisingly, when families signed up for MTO, more than three-quarters reported that the most important reason for wanting to move was to escape inner-city gangs, drugs, and violence. One MTO mother told me, [It’s] bad, you know, when my oldest son saw this drug dealer on the corner and he said, ‘I want to be just like that when I get older.’ I said oh it’s time to go. It was a way of living but I knew there was something better. There had to be something better.” The movers reported much higher levels of neighborhood and housing quality than control group-families. Fewer experimental movers were victimized and they felt safer at night; they reported greater success getting police to respond to calls in their current neighborhoods and saw less drug-related loitering.

Therefore, in terms of what families wanted most—safety—MTO was a great success. MTO succeeded admirably in enabling families to achieve their stated goal of escaping violent neighborhoods and finding better-quality housing. The significance of improved well-being for the families who moved should not be ignored. When I asked one MTO mother how she felt when she got her new apartment in Northeast Baltimore, she exclaimed, “I was so happy. I couldn’t wait. . . . I never had a kitchen that big. I looked at it and all, and that week I could hardly sleep.” Similar results were ever present in the Gautreaux interviews carried out in the 1990s. Parents repeatedly referred to their original city neighborhoods as prisons and expressed relief when they were able to leave. One Gautreaux mover described “a freedom that I didn’t have over there in that concentration camp. It was very restrictive. I couldn’t take my kids outside. . . . My kids were in peril. It was my main objective to get out of there.”

Jeff Kling and colleagues also found significant reductions in MTO mothers’ psychological distress, suggesting the importance of environment for improving mental health. They also found significant reductions in obesity, healthier eating habits, and more exercise among the mothers who moved. Mothers worried less about violence and having to constantly monitor their children’s safety, and seemed thus freer to pursue other activities.

Teenage girls who moved with MTO also benefited from escaping high-poverty neighborhoods. Not only did they report significantly lower levels of depression and anxiety, they were also less likely to use drugs, drink, and smoke. In a recent study, Susan Popkin and colleagues reported that these young women experienced less anxiety in part because they no longer had to worry about the sexual harassment and pressure for sex they experienced in their city neighborhoods.

Unfortunately, the young men who relocated to low-poverty areas were actually more likely to engage in risky behavior and to be arrested for property crimes. Susan Clampet-Lundquist and colleagues found that MTO girls and boys socialize in different ways: boys were more likely to hang out with their friends on the corner or on a neighborhood basketball court, and girls were more likely to visit friends inside their homes or to go downtown to a mall. Boys may have been at higher risk of delinquency because these routines do not fit in as well in low-poverty neighborhoods, which may explain why they did not benefit from peers in their new neighborhoods as much as girls did.

Despite some significant successes, however, MTO’s results did not give policymakers what they were hoping for: increases in economic self-sufficiency for parents and better schooling outcomes for children. MTO mothers were no more likely to be employed, earned no more, and received welfare no less often than mothers assigned to the control group. Early MTO research had shown that moving to less poor neighborhoods helped children attend better schools and increased children’s test scores and school engagement, especially in Baltimore. However, the follow-up study showed virtually no educational benefits. In part, this was because almost seventy percent of the MTO children were attending schools in the same district where they signed up for the program. Some children did attend higher performing, less segregated schools in the suburbs, but the differences in outcome were small.

Researchers wanted to understand the disappointments with MTO, so in 2003-2004 they began to conduct in-depth interviews in the five MTO cities. As part of a fieldwork team, I conducted interviews in Baltimore. We hoped that asking parents and youth about their experiences changing neighborhoods and schools, as well as about their daily lives, would help us understand more about how neighborhood relocation might matter for success in school and work. We talked to almost 150 mothers, interviewed more than one hundred teenagers, visited several dozen schools, and interviewed many teachers.

The interviews examined some of the assumptions behind MTO—for example, that relocating to more affluent neighborhoods would put parents in touch with more employed adults who might become sources of information and connections to good jobs. We compared the stories from interviews with mothers who had moved to less poor neighborhoods to those of women who did not relocate. Both groups of women relied on informal networks to obtain employment in service-sector jobs, such as retail, corrections, and health care. While the experimental movers had more employed neighbors, few of them held those kinds of jobs. The control group mothers, by contrast, had fewer employed neighbors, but they were more likely to learn about job openings in the course of their daily lives. In general, mothers also lacked necessary educational credentials and work experience, and they suffered from health problems that made consistent employment difficult. Transportation was a concern too, especially for the women who moved far from the city. However, while the mothers who left the city reported fewer social connections, they were also less likely to be called upon to share resources with what sociologist Rebecca Kissane calls “needy ties,” and therefore in a better position to save money and time for themselves and their children.

The interviews also explored the educational assumptions behind MTO. Interviews with mothers suggested that there were social and structural processes that stood in the way of getting children into better schools and sustaining educational achievement after the move. Many families juggled severe problems: murder, drug addiction, suspicious landlords, diabetes, and depression took center stage in the lives of many, if not most, MTO families we interviewed. At times, schooling became a lower priority. Sometimes children were shuffled from one school to the next, depending on where a grandmother lived and where mom went to work—situations that made it hard to establish homework routines and discipline. Some parents left schooling decisions up to their children and didn’t want them to change schools if it was going to be disruptive or keep them from old friends. Other parents believed that if their children just worked hard, they could do well in any school. Such a position seems strange, considering just how low-performing many of their original neighborhood schools were. However, it is important to remember that these families never really had experience with better schools or the information that many middle-class parents use in making choices about their children’s teachers, courses, and schools. When MTO parents were asked what they liked about their child’s school, many didn’t consider academic rigor as important as a welcoming atmosphere. After the experience of violent and chaotic schools, many parents were satisfied with a school where they felt the teachers really cared.

Why did MTO’s results seem more disappointing than Gautreaux? One possibility is that in the case of Gautreaux, housing counselors worked with real estate agents and landlords to locate units for the participants, and either the landlords of these units or the neighborhoods in which they were located may have facilitated permanent relocations. In contrast, MTO families were allowed to choose their own units. We know from previous research on housing vouchers that poor families lack information about better housing options and will often choose neighborhoods (even within program-designated census tracts) that look similar to the areas they came from. Indeed, even though their vouchers enabled them to do so, few MTO families found housing in mostly white or integrated suburban communities. Therefore, the mix of housing counselor assistance and placement in high-resource, less racially segregated communities seemed to yield the greatest long-term benefits for families: both components appear to be important to the success of mobility programs. Perhaps it was the counseling and assignment to specific units and neighborhoods that helped Gautreaux families overcome landlord discrimination, lack of information about rental markets, or fears about more affluent white areas. Subsequently, these experiences might have helped families make successful second moves in the same or similar neighborhoods, thus contributing to residential stability and well-being.

It is also tempting to suspect that differences between the programs may be due to MTO’s focus on deconcentrating poverty rather than promoting racial integration. MTO’s criterion for a placement neighborhood was based on income class (a poverty rate under ten percent), while Gautreaux moved families to neighborhoods based on race (no more than thirty percent black neighbors). Partly as a result, MTO families did not move as far away from their original neighborhoods.

However, the programs differed in other important ways. First, MTO’s design provides comparisons between families who were and were not offered help in moving to low-poverty neighborhoods, while Gautreaux research can only be used to compare families placed in different kinds of neighborhoods. Second, MTO’s moves took place in the late 1990s, amid welfare reform and a booming economy, which provided a very different context for work and welfare. MTO families did indeed boost their employment and reduce their reliance on welfare, but these changes were no different, on average, from those experienced by the control group. Gautreaux welfare outcomes were measured in the early 1990s, well before welfare reforms were adopted by the state of Illinois. It is also important to note that many of the control group families were living in public-housing projects that were demolished during the time of the MTO program, and therefore many of these families also relocated, making it harder to know how they might have fared had they not been forced to move and how their outcomes would compare with those families who used the low-poverty voucher.

* * *

Drawing policy implications from Gautreaux and MTO requires us to think carefully about the possibilities and limits of residential-mobility strategies. The initial gains that many Gautreaux families achieved with vouchers and housing assistance persisted for at least one to two decades. This is extremely encouraging and suggests that it is possible for low-income black families to make permanent escapes from neighborhoods with concentrated racial segregation, crime, and poverty. We also know from patterns of residential concentration that in the absence of such a program, positive neighborhood changes like these are rare. However, we also find that it is difficult to translate the benefits of sustained residence in better neighborhoods into large economic and social gains for families and children. Housing mobility may be a necessary but insufficient lever for improving the lives of poor families trapped in inner-city neighborhoods.

In drawing conclusions for anti-poverty policy we also need to be mindful of whom these programs served. Most of the original Gautreaux families were first- or second-generation residents of Chicago public housing. Their backgrounds were considerably more favorable for long-term outcomes than those of families participating in MTO, many of whom had long family histories in public housing. And despite demographic similarities, families who volunteer for mobility vouchers clearly differ from other public-housing families. Thus, our findings generalize most readily to families voluntarily choosing to participate in residential-mobility programs. As the transformation of distressed public-housing projects continues, families often involuntarily move to new communities; involuntarily relocating families may not result in the same outcomes as those found for Gautreaux or MTO participants.

Both Gautreaux and MTO bet heavily that residential mobility and neighborhood change could alone promote families’ self-sufficiency; neither provided family-based employment support, transportation help, or educational assistance. That’s a tall order for a housing-only program to fill. Helping parents to acquire better jobs and transition off welfare may require coupling neighborhood change with services and supports tailored to individual families’ needs. Of course, such support can also be provided independent of mobility programs. Work by economist Greg Duncan and colleagues showed that a number of experimental work-support programs in the 1990s boosted work, family income, and children’s achievement. Some of these programs supplied poor parents with earnings supplements and center-based child-care assistance. Evidence from MTO also suggests that landlord problems figured prominently in distinguishing families that stayed in their placement units from those who moved on, often to higher poverty and more segregated neighborhoods. Providing tenants with assistance in securing units or dealing with difficult landlords might help address these problems and might ensure that families remain in opportunity-rich communities, even after a second move.

Many social policies also assume that low-income parents, if they had the resources, would approach opportunity the same way most middle-class families do. The MTO interviews in particular provide a reminder that poor families are not just wealthy families without a bank book. For example, poor parents often have less information about school-choice programs and may approach educational opportunities in different ways. It would be useful to test a program model in which mobility counselors were trained to inform parents about the new schooling choices in the area. While assisting with the sometimes disruptive effects of school transfers, counselors could help ensure that special needs are met, that receiving schools have information about the child, and that little instruction time is lost in the transition between schools.

Currently, we have the chance to further examine some of these questions and the future viability of mobility programs. Researchers are planning a ten-year follow-up to MTO to see whether some of the early improvements have more substantial long-term benefits. For example, the reduction in stress among the MTO movers might translate over time into stable employment prospects and better outcomes for their children. In Baltimore I am following families who are moving as part of a partial desegregation remedy to a court case filed in 1995—a case very similar to Gautreaux. In the Thompson case, a federal judge found the U.S. Department of Housing and Urban Development responsible for violating fair-housing laws by not looking beyond city limits for ways to house poor families, and awarded two thousand vouchers for use in high opportunity neighborhoods in the Baltimore region.

With funding from the Annie E. Casey Foundation, I have been following these “Thompson” families who, with the help of housing counselors and fair-housing lawyers, are relocating from public-housing projects to low-poverty, non-segregated neighborhoods all around the Baltimore metropolitan area. At the moment, over one thousand former public-housing families have successfully relocated to safer, more opportunity-rich communities. There are also extensive multi-partner efforts in place to help connect these families to employment and education resources in their new communities. For example, the Baltimore Regional Housing Coalition (BRHC) is trying to expand a city-based job-counseling program to include suburban employers and a subset of the Thompson movers. Another program, funded by the Abell Foundation and the Baltimore Housing Authority, provides cars and low-cost financing for Thompson families working in the suburbs. Additionally, the BRHC is proposing a way for housing counselors to assess families’ health needs and help them develop a plan for improvement. Time will tell whether these new programs and evaluations will make the implications of housing mobility programs clearer.

A separate policy issue left unaddressed in either Gautreaux or MTO research is the programs’ impact on receiving neighborhoods or the neighborhoods that get left behind in mobility programs. Designers of both programs worried that individuals living in the receiving neighborhoods might be subjected to higher crime rates and/or lower property values. To try to prevent clustering of poor families and negative attention, Gautreaux and MTO housing counselors tried to make sure that they did not place too many families in any one community. Another concern is whether focusing on a mobility strategy threatens to drain urban communities of the families most likely to help stabilize and develop these neighborhoods.

I am not aware of research that considers these tradeoffs in the context of mobility programs, but it is important not to view mobility as the only strategy for providing families with access to safer, high-resource communities. Other programs try to strengthen neighborhoods themselves. Community-development initiatives, while harder to evaluate, have shown promise in cities such as New York and Chicago. Federal, state, and city-level efforts, such as community-school partnerships, are trying to bridge local institutions, such as social-service agencies, schools, and health clinics to improve child and family well-being. Additional efforts target community “treatments”; instead of moving families between neighborhoods, the neighborhoods themselves are given interventions such as health-information campaigns or new policing strategies. Some combination of mobility and neighborhood development is most likely to help researchers better inform effective policy.

No one, regardless of political preference, wants to fund programs that don’t really work. We also want to distinguish good programs that have been implemented badly from poorly conceived programs. It takes time to understand what interventions work. Researchers want to get it right. However, we also need to ask ourselves whether individual families who want to escape dangerous neighborhoods and improve their life chances have to wait for us to figure it out. It’s hard to listen to the mothers who have left violent neighborhoods and not think that supporting mobility is worth something in itself, right now. As one of the MTO mothers explained on seeing some of the suburban neighborhoods for the first time, “This is so beautiful, the trees, the grass . . . where I was from there was none of that. If I had not had that opportunity to go into the MTO program, I would not have known what it would have been like to live in a house in a positive environment. I wouldn’t have had the opportunity to see that, but MTO gave me the opportunity to see how middle-class people live. It just made me want that.”

For decades, this nation has demonstrated a strong commitment to improving the education and employment outcomes of its citizens; providing opportunities to live in safe communities where families can prosper should also be part of that commitment. While more research is needed to understand how to make housing programs most effective, current research evidence suggests that well-designed residential-mobility programs can be important instruments for helping families escape poverty and improve the quality of their lives.