Association for Public Policy Analysis & Management

Previous studies have examined the sorts of neighborhood conditions to which families living in site-based subsidized housing are exposed and how these conditions compare to those experienced by low-income families in the private rental sector (Massey and Kanaiaupuni, 1993; Hogan, 1996; Newman and Schnare, 1997; Rohe and Freeman, 2001; Freeman, 2004; McClure, 2006, 2008; Talen and Koschinsky, 2011).  Though these studies have provided valuable insights, they have taken only a cross-sectional, comparative approach.  They have not addressed longitudinal questions about how families in public housing fare geographically after they leave subsidized housing and whether their long-term trajectories of residential mobility compares favorably to those experienced by similar families who never lived in subsidized housing.  We think that such a longitudinal perspective is crucial if we are to paint a clearer, more holistic portrait of the impact of site-based subsidized housing on families’ residential quality of life and opportunities.

In this paper we undertake a longitudinal investigation of the residential mobility patterns of low income American families that employs sequence analysis.  Sequence analysis distinguishes sets of low-income households according to their trajectories of a given neighborhood indicator experienced over time.  We focus upon three sets of low-income, predominantly black American individuals in PSID who formed their first independent household during the period 1988 to 1992 (soon after the significant welfare reform laws were implemented) who resided: (1) for a spell in traditional, first-generation public housing; (2) for a spell in second-generation, privately owned but publicly subsidized developments; and (3) solely in the private, unsubsidized housing market but were otherwise comparable to the first set of families.  We examine what neighborhoods these households inhabit during the subsequent two-decade period when they were most likely raising children and trying to enter and advance within the labor force.  We address the following research questions for these groups: 

1. 1.    What were the averages and variations in the neighborhood poverty rates and percentages of white residents where they lived, by assisted housing status?
2. 2.    What were the distinct temporal trajectories and their incidences for these two conditions of the neighborhoods in which they lived, by assisted housing status??
3. 3.    To what degree may have residence in assisted housing resulted in distinctive trajectories involving lower poverty, higher white composition neighborhood conditions?

Our preliminary results indicate that a spell of residence in public housing is associated with longer durations of residence in poorer, blacker neighborhoods, but there are remarkably few statistically significant differences in the patterns of sequences that our three comparison groups experienced.  These results provide a mixed set of implications for housing policymakers.  On the one hand, it appears that traditional public housing did not distort the residential trajectories of low-income families in as detrimental ways as some have portrayed, though prolonged isolation from white neighbors was one distinctive consequence.  On the other hand, it appears that the second-generation of site-based assistance programs did no better than traditional public housing in promoting residential occupancy sequences that were notably different in exposures to poor or white neighbors.