The durability of gains from the Gautreaux Two residential mobility program: a qualitative analysis of who stays and who moves from low-poverty neighborhoods∗

Abstract

This paper examines mobility in the Gautreaux Two Housing Mobility Program, which attempted to alleviate poverty concentration by offering vouchers to residents of highly distressed Chicago public housing developments. In contrast to the original Gautreaux program, placement moves in Gautreaux Two have proven far less durable – most families quickly moved on from their placement neighborhoods to neighborhoods that were quite poor and very racially segregated.

Based on in-depth interviews with 58 Gautreaux Two participants and their children, we find that the primary factors motivating secondary moves included substandard unit quality and hassles with landlords. Other factors included feelings of social isolation due to poor integration into the new neighborhood, distance from kin, transportation difficulties, children's negative reaction to the new neighborhood, and financial difficulties. Policy implications include the need for further pre- and post-move housing counseling for families in mobility programs.

Keywords:

• low-income housing
• neighborhood
• vouchers
• policy

Acknowledgments

The authors acknowledge the generous research support provided by the John D. and Catherine T. MacArthur Foundation and the Ford Foundation. The authors thank the following researchers who collected and processed the data: Karen Burke, Gretchen Wright, Rechelle Paranal Higgins, Roi Lusk, Verity Decker, Kristin Turney, Katherine Hunt, Jennifer Pashup-Graham, Adam Williams, Micere Keels, Ruby Mendenhall, Heather Hill, Emily Snell, Anita Zuberi, and Joanna Reed. Aurie Pennick, Mary Davis and Dianne Prince of the Leadership Council of Greater Metropolitan Chicago and the Chicago Housing Authority generously assisted the research team, and the respondents graciously opened their homes and shared their stories.

Notes

¹See Pashup et al. (2005) for an analysis of the factors that made making an initial move through the Gautreaux Two program difficult.

²Some of these individuals were on a waiting list for a unit in a Chicago public housing project.

³Although Gautreaux counselors strove to place families in low-poverty, racially integrated neighborhoods, there were periods during its operation when it was very difficult to find housing in neighborhoods that met these criteria. In response, the program adjusted its definition of qualifying destinations to include neighborhoods that were quite poor and segregated but were judged to be improving. This practice began in 1981. While the program did not intend for these families to move to very segregated, high-poverty neighborhoods, Census-based analysis shows that some did (Mendenhall, DeLuca, and Duncan 2006). About one-fifth of Gautreaux families were placed in high-poverty, highly segregated neighborhoods, almost all of which were within the city limits of Chicago.

⁴Of the total respondent pool, 78% (n = 84) of the families were drawn from a random sample of program participants. As program take-up proceeded at a slower rate than anticipated, and it appeared that take-up would be less than 50%, we recruited an additional 23 respondents (22% of sample pool) from a list of “likely movers,” that is, participants who already had selected a unit and secured a verbal agreement with the landlord. Pashup et al. (2005) describe some of the difficulties participants faced in moving through the Gautreaux program. These difficulties included both external (tight rental market, landlord discrimination against housing vouchers and family size and bureaucratic delays) and internal (little exposure to life outside of Chicago's south side, poor understanding of program requirements, large household size and mental or physical health problems) factors.

⁵Due to the relatively small number of families in our sample, we provide a descriptive analysis of these demographic characteristics rather than a multivariate analysis.

⁶Despite their relatively small numbers of cases, qualitative studies such as Gautreaux Two provide enough statistical power for useful inference, provided probability sampling methods are employed. Duncan (2008) shows that for sample sizes of n = 75, the standard error of an estimated mean of 0.10 is 0.037, which provides substantial power to reject priors that the condition or event is quite prevalent in the population. In the case of a more evenly split proportion (say, 0.30), the standard error of the estimates is larger at 0.052. But even here, the implied confidence interval is small enough to be quite informative of population proportions.

⁷There were no notable differences between secondary movers who moved to highly segregated neighborhoods and those who moved on to less segregated neighborhoods in terms of the economic character of their baseline neighborhoods or the economic or racial characteristics of their Gautreaux Two placement neighborhoods. Nor were there differences in the proportion who had been placed in the suburbs versus the city or the proportion who had had recent residential experience outside of public housing.

⁸All respondents' names are pseudonyms.

⁹This is not to say that these families were impervious to the risks associated with moving back to poorer and more racially segregated neighborhoods. Five noted they were careful to choose units the second time around that were not in gang infested neighborhoods, and two talked specifically about the necessity of keeping their daughters close to home to avoid the added risk their secondary move posed.

¹⁰One such pending policy reform is the Section 8 Voucher Reform Act (SEVRA), which could have implications for a variety of aspects of voucher usage including streamlining unit inspections and decreasing barriers for voucher portability.