Owen Fiss provides a compelling analysis of the problem of low-income black families. If he is right, then other reform efforts are unlikely to have much success until families can escape from poor neighborhoods. Fiss offers a compelling argument for residential mobility. While I believe we may find other promising approaches, no other approach has such strong evidence of successful outcomes. The available evidence suggests that residential mobility is one of the most promising solutions for the multiple problems associated with urban poverty.

I have been studying Chicago’s Gautreaux program for many years, and I have found that residential mobility can have truly impressive results in improving the lives of low-income families. In this program, low-income black families in public housing (or on the waitlist) were assigned to various neighborhoods in the city or suburbs by a quasi-random procedure. Participants circumvented the ordinary barriers to living in suburbs, not by their jobs, personal finances, or values, but by getting into the program and being randomly assigned to suburbs. The program gave them rent subsidies that permitted them to live in suburban apartments for the same cost as public housing. Participants moved to a wide variety of over 115 suburbs throughout the six counties around Chicago. Suburbs with more than 30 percent blacks were excluded by the consent decree, and a few very high-rent suburbs were excluded by funding limitations. Yet these constraints eliminated only a few suburbs.

The receiving suburbs ranged from thirty to ninety minutes driving time to their former homes. While all participants came from similar low-income, black city neighborhoods (usually public housing), some moved to mostly-white suburbs, while others moved to city neighborhoods, most of which were disproportionately black. In principle, participants had choices about where they moved, but, in actual practice, participants were assigned to city or suburb locations in a quasi-random manner. Clients were offered a unit according to their position on the waiting list, regardless of their preference. Although clients could refuse an offer, few did so, since they were unlikely to get another. As a result, participants’ preferences for city or suburbs had no effect on their placement location, and analyses indicate that the two groups were nearly identical.1

This program had amazing results. Housing policy is usually narrowly viewed as providing shelter, but housing policy can radically improve people’s lives. Studies of this program compared family outcomes in mostly white suburbs and mostly black city neighborhoods.2One study followed children who moved in this program and found that by the time that they were young adults, those moving to the suburbs were much more likely to graduate high school, attend college, attend four-year colleges, and (if they were not in college) to be employed and to have jobs with better pay and with benefits. A study of Gautreaux mothers found that suburban movers had higher employment rates than city movers, and the difference was especially large for adults who did not have jobs prior to the move. A recent study, using official records of AFDC receipt for all program participants, found strong neighborhood effects on AFDC receipt many years after moving.

Critics have said that most families would not remain in white suburbs. Yet recent research, which located 1,506 out of a sample of 1,507 families, found that over two-thirds of suburb-mover families remained in suburbs seven or more years after entering them. Others have argued that such a program cannot serve large numbers of families. In order to have low impact on receiving communities, the program avoided moving more than two or three families to any neighborhood, and in fact it succeeded in this goal. Still, the program could be greatly expanded without having large impact on any neighborhood. About four million people live in Chicago’s suburbs, and the vast majority of suburbs are over 80 percent white. Even if all Chicago’s public housing families were widely scattered among these suburbs, they would reduce the white proportion in any suburb by less than 2 percent.

If it is done poorly, however, residential mobility will not help families. Not all moves are beneficial. Under a federal mandate to tear down public housing, housing authorities across the country are moving thousands of families to other housing. In their haste to empty buildings, officials are not giving much thought to where families are moving. A recent study finds that families are being moved to low-income, mostly black areas, which are very similar to the neighborhoods they left. These moves are displacing families into equally bad neighborhoods that will have little benefit. In contrast, the Gautreaux program shows that a well administered program can move low-income black families to neighborhoods that have positive influences, but this must be done carefully, not in a willy-nilly fashion in a short period of time.

Another limitation is that some families may not be prepared to benefit. The Gautreauxprogram had three selection criteria that were intended to assure landlords they would get good tenants and make it more likely that participants would be able to remain in these apartments. The program tried to avoid overcrowding, late rent payments, and building damage by not admitting families with more than four children, large debts, or unacceptable housekeeping. None of these criteria were extremely selective. Because 95 percent of AFDC families have four or fewer children, the overcrowding restriction eliminates only a few eligible families. Moreover, Gautreaux administrators estimate that about 12 percent of applicants are rejected by the credit check or rental record and only 13 percent are rejected by counselors who find property damage on a home visit. Thus, all three criteria reduced the eligible pool by less than 30 percent. While these three conditions were not highly restrictive and they allowed a large proportion of low-income families to be eligible, some families were excluded.3

In contrast, a front-page article in the Chicago Tribune recently described a residential mobility program that included some families with histories of property vandalism and crime who were being moved into private apartments. Even though such families may not be typical of housing project residents, landlords will be reluctant to lease to families in that program. Failure to screen out families who are unprepared for the move, or failure to give them appropriate preparation, may doom many families to failure while stigmatizing the entire effort. Social policy cannot simply gloss over these difficulties. Families with poor housekeeping skills, poor rent-paying histories, large outstanding debts, destructive family members, and/or active criminal involvement will make poor tenants and will be evicted. Programs that go to great expense to move such families and compel landlords to accept them will waste money and waste political support. Residential programs must have appropriate selection criteria.4

These programs can be combined with other programs to address these problems: courses to teach housekeeping skills, credit management, etc. Moreover, if low-income families see incentives for meeting the selection criteria, they will have reason to alter their behavior. If such steps are taken, programs may not need to compel landlords to accept program participants. After a few years of operation, Cincinnati’s HOME program reports that landlords telephoned to request participants whenever they had a vacancy. The program became a preferred provider of tenants. While landlords knew that program participants were low-income blacks, they knew that they were screened on appropriate criteria and that prior participants had turned out to be good tenants. Landlords could not get such assurances from strangers who answered their newspaper ads.

The Gautreaux experience suggests that if residential mobility programs are careful to select (or prepare) families and to place them in appropriate areas, they can have truly impressive benefits. Just as Fiss indicates, residential moves lead to remarkable changes of life circumstances, and these changes have dramatic benefits on people’s behavior. Housing policy can do more than provide shelter–it can radically improve people’s lives.


1 A study of 330 families found the two groups were similar in age, education, marital status, long-term AFDC receipt, and second-generation AFDC receipt. See James E. Rosenbaum, “Housing Mobility Strategies for Changing the Geography of Opportunity,” Housing Policy Debate 6 (1995): 231-70. Another study found no correlation between mothers’ attributes (age and initial AFDC) and placement attributes (city/suburb, tract percent black, percent in poverty, percent unemployed, or percent low education).

2 James E. Rosenbaum, “Black Pioneers–Do Their Moves to Suburbs Increase Economic Opportunity?” Housing Policy Debate 2 (1991): 1179-1214. Leonard Rubinowitz and James E. Rosenbaum, Crossing the Class and Color Lines (Chicago: University of Chicago Press, 2000).

3 Gautreaux participants are similar to a random sample of Chicago AFDC recipients in their length of time on public assistance (about seven years) and their marital status (about 45 percent never married, 10 percent currently married). S. J. Popkin, “Welfare: A View from the Bottom,” unpublished dissertation, Northwestern University, 1988. But Gautreaux participants are less likely to be high school dropouts (39 percent pre-move vs. 50 percent), tend to be older (median age of 34 vs. 31), and have fewer children (mean of 2.5 vs. 3.0). Still, they are more likely to be second-generation AFDC recipients (44 percent vs. 32 percent). In sum, although Gautreaux participants may be of slightly higher socioeconomic status than the average public assistance recipient, most differences between them are not large.

4 James E. Rosenbaum and Shazia Miller, “Certifications and Warranties: Keys to Effective Residential Integration Programs.” Seton Hall Law Review 27 (1997):1426-49.