Deconcentration and Social Capital: Contradictions of a Poverty Alleviation Policy

ABSTRACT

Deconcentration is a policy aimed at reducing poverty by relocating residents of distressed public housing complexes into private mixed income neighborhoods. This change is presumed to offer new social opportunities and better public facilities that can facilitate improved economic status. HOPE VI is a federal U.S. Department of Housing and Urban Development (HUD) program, which has effected this policy in a large number of U.S. cities. This paper reports the findings from research in two relocation sites (high and low poverty) in Tampa, Florida, based on interviews with HOPE VI relocatees and their homeowning neighbors. Results indicate that relocation does not enhance social capital for former public housing residents. Social networks are diminished in comparison with prior conditions in public housing. There is very little interaction with homeowners in relocation sites, and considerable resistance by homeowners. Relocatee satisfaction with housing is greater in the low poverty site, but social networks are not different across sites.

KEYWORDS:

• Deconcentration
• HOPE VI
• poverty
• urban neighborhoods

Notes

1. Those figures have now reversed, with a 2005 estimate of 13.3% of families below the poverty line in the U.S. (Berube & Kneebone, 2006). It is important to note that Jargowsky adds cautionary language to his “stunning progress” assessment by observing the apparent problems this movement has caused for older inner ring suburbs. Berube and Kneebone's recent study confirms that worry.

2. In Tampa, a city with a pronounced reduction in high poverty tracts between 1990 and 2000, our analyses indicate that more than half of the effect was due to HOPE VI relocations. These data will be reported in an article currently under preparation.

3. MTO was a quasi-experiment designed to assess the effects of assisted mobility. HUD launched this $70 million project in 1994 in five cities (Baltimore, Boston, Chicago, Los Angeles, and New York). Three groups were established drawing from current residents of public housing—experimental households were relocated with vouchers into low-poverty neighborhoods, a comparison group had vouchers they used without assistance or restriction, and a control group remained in public housing. The program was not refunded in 1995, due primarily to backlash from residents of relocation neighborhoods in Baltimore.

4. Improvement in housing conditions, when compared with public housing where original construction was typically poor and maintenance progressively neglected under the funding formulas of HUD, is not difficult to achieve.

5. Atlantic City (NJ), Chicago, Durham (NC), Richmond (CA), and Washington, DC.

6. The authors do not address potential reactivity problems in these questions, but youth misbehavior is cause for eviction from vouchered apartments. It could be surmised that respondents may be less truthful about such issues.

7. Funded by an award from NSF (BCS-0241178), the project ended on December 31, 2006. Opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

8. The extent of randomness in the HOPE VI recruitment was affected by the age of the contact list (which was two years old when we began), and by the relatively small size of potential interviewees in each neighborhood in relation to desired sample; for about half of the interviewees, we had to rely on other means of identifying eligible families and contacting them one by one in order to achieve our target of 20 in each area.

9. Occupations were: nurse, doctor, nurse's aide, public official, police officer, business owner, store manager, teacher, principal, lawyer, judge, paralegal, bank employee, and realtor. We included nurse's aide, a fairly low status occupation, in our list. We expected that H6 interviewees might be likely to know someone in that occupation, and thus included it to detect potential response sets in the answers. H6 interviewees were just as likely to know a nurse's aide as the HO respondents, but far less likely to be acquainted with a physician (X² [1, n = 79] = 21.1, p < .001).

10. This compares with 66% of H6 respondents who claimed one or more acquaintances with non-H6 residents. However, the questions on this issue were not quite parallel, which may account for the discrepancy. H6 respondents were asked how many non-H6 neighbors they knew. Individuals they identified would include unsubsidized renters in their housing complexes, as well as homeowners. HO were asked if they knew any residents who had been relocated there from public housing. In this regard, the HO responses are more pertinent to the issue of interaction across these two conditions.

11. This is a federal program begun in the early 1990s that combines law enforcement and community development to eliminate lawless elements (weeding) and nurture new social practices and projects (seeding). It is aimed at distressed, high crime areas like Riverbend.