«A Journal of Policy Development and Research HoPe VI Volume 12, Number 1 • 2010 U.S. Department of Housing and Urban Development Office of Policy ...»
Economic insecurity can be one reason for the residential instability of people displaced from public housing (Barrett, Geisel, and Johnston, 2006; Gibson, 2007; Reed, 2006). Nationwide, for example, three out of five people who had been relocated and given vouchers by HOPE VI reported difficulties paying rent or utilities within the previous year (Popkin, 2006). Among displaced public housing residents in Fort Worth, one-half reported that they feared eviction because of their economic insecurity (Barrett, Geisel, and Johnston, 2006). In Portland, one-third of those displaced by HOPE VI reported hardship making their rent payments, 60 percent reported difficulties paying for utilities, and 17 percent were deemed by the local housing authority to be
at risk because of these financial difficulties (Gibson, 2007). Such outcomes are troublesome in and of themselves, but they are especially worrisome because the HOPE VI Program is intended to improve the economic self-sufficiency of public housing families.
Health and Behavior Although data on displaced people’s health are limited, Popkin’s (2006) analysis of interview data from the Urban Institute’s HOPE VI Panel Study found that participants faced serious health problems before relocation. More than one-third of the adults reported having a chronic illness or health condition. More than 20 percent of the adults reported having asthma, and the rate of children’s asthma was more than three times the national average (Popkin, 2006). Mental health problems—including depression, stress, fear, and anxiety—were also common and occurred at a rate nearly 50 percent higher than the national average (Popkin, 2006). Manjarrez, Popkin, and Guernsey (2007) found that these conditions have not improved for people relocated by HOPE VI, despite the passage of time. Three-fourths of the panel study respondents reported no change or a decline in their health. The number of respondents indicating health conditions that required regular, ongoing care increased significantly (Manjarrez, Popkin, and Guernsey, 2007). The mortality rate for African-American women in the HOPE VI Panel Study is higher than for African-American women in general and in the MTO control group that did not move (Manjarrez, Popkin, and Guernsey, 2007).
The HOPE VI Panel Study data also show no improvements in overall health for relocated children (Gallagher and Bajaj, 2007). Boston (2005), on the other hand, reported greater mortality among a comparison group of public housing residents living in projects that were not renovated than among those displaced by HOPE VI-like redevelopments in Atlanta.
Safety Studies consistently show that families that move out of neighborhoods of concentrated poverty report an increased sense of safety (see Buron et al., 2002; Goetz, 2003; Popkin and Cove, 2007).
Residents report a significant decline in drug-related activity, a greater personal sense of safety, and improvements in safety for their children. Among people displaced by HOPE VI, those who moved into other public housing reported fewer benefits (Popkin and Cove, 2007). Interestingly, no statistical correlations exist between these findings and any secondary benefits, such as improved mental or physical health (Popkin and Cove, 2007). Gibson (2007) reported that 30 percent of displaced households in Portland’s Columbia Villa HOPE VI project thought their new neighborhoods were safer, but 18 percent thought they were less safe in their new neighborhoods.
Housing and Neighborhood Characteristics Most (63 percent) of the people relocated by HOPE VI who participated in the panel study reported their new housing to be in “good” or “excellent” condition (Popkin, 2006: 216). Brooks et al. (2005) found a similar outcome for those relocated in Atlanta, although families using vouchers reported much higher levels of satisfaction. Goetz (2003) found that former residents of public housing in Minneapolis were more satisfied with the quality of their new housing than were comparison groups, whether they had moved voluntarily or involuntarily. Satisfaction was greater, 10 HOPE VI Better Neighborhoods, Better Outcomes? Explaining Relocation Outcomes in HOPE VI however, among families who had chosen to move, suggesting that people who were involuntarily displaced—particularly those who were hard to house—may have found it challenging to obtain good-quality housing in the private market.
The HOPE VI Resident Tracking Study found that displaced households reported less crime, better housing conditions, and improved neighborhoods when compared with their former residences (Buron et al., 2002; Popkin et al., 2004). Displaced households do not always view their new neighborhoods in favorable terms, however. People displaced by HOPE VI tend to evaluate their new homes more favorably than the housing projects they left and tend to report fewer neighborhood problems (Comey, 2007). But this is not always the case: in interviews and focus groups with former residents of a Seattle HOPE VI development, Manzo, Kleit, and Couch (2005) found that 60 percent of the people interviewed thought their former public housing residence was a better place to live than their new neighborhoods.
Social Networks Research has shown that HOPE VI redevelopments have done little to help displaced families with social integration into their new neighborhoods. In interviews with 41 displaced Philadelphia families conducted 2 years after relocation, Clampet-Lundquist (2004) found that very few households built social ties in their new neighborhoods, regardless of neighborhood poverty levels.
Young people in these families were more likely to build friendship networks than the adults were;
however, young people were unlikely to view their new neighbors as role models or to interact with adults in their new neighborhoods (Clampet-Lundquist, 2007). Barrett, Geisel, and Johnston (2006) reported fewer neighboring behaviors (such as talking to a neighbor for more than 10 minutes or watching a neighbor’s child) and fewer supportive relationships after a HOPE VI displacement in Fort Worth. Curley (2006) reported similar findings for families in Boston who were displaced by HOPE VI. Such data suggest that the process of forming social networks is complex and may depend on several factors, including attitudes and perceptions of neighbors, whether relocation was voluntary or involuntary, and characteristics of households being displaced.
Kleit and Manzo (2006) found that HOPE VI relocations result in objective improvements in neighborhood conditions but may also include “some negative social outcomes” (Kleit and Manzo, 2006: 271). Place attachment, according to these authors, is important in determining whether residents want to leave the site (see also Vale, 1997). Those who have lived in public housing the
longest are the least willing to move because they regard their particular development as home:
they had put down roots and become attached to the community (Kleit and Manzo, 2006; Vale, 1997). Gibson (2007) found that two-thirds of the residents of the Columbia Villa project in Portland did not want to leave. Even after being forced to move, many residents reminisced about the community and mourned the loss of their neighbors, the open space in the project, and the level of comfort they had felt there. Most did not believe that their new neighborhoods provided as much sense of community as Columbia Villa had; only one-third felt a greater sense of community in their new neighborhood. Goetz (2003) found that relocated people experienced only limited interactions with their neighbors. Families who moved voluntarily—and whose immediate neighbors lived in subsidized, rather than market-rate, housing—reported higher levels of interaction with their neighbors than did families who moved involuntarily. Finally, the Urban Institute’s HOPE VI Panel
Study data showed a greater level of social isolation for children of displaced families, although the authors suggested that the isolation might be good for the children, because it would protect them from the negative influences of their surroundings (Gallagher and Bajaj, 2007).
Summary The findings related to individual benefits from HOPE VI are mixed. No evidence suggests that the program is producing benefits such as increased economic self-sufficiency and access to enhanced social capital. The best outcomes are related to perceptions of safety and of neighborhood quality (as expressed through measures of satisfaction). The next section of the article presents evidence from a single HOPE VI case in Duluth, Minnesota, used to investigate the relationship between neighborhood conditions and individual outcomes.
Harbor View HOPE VI Duluth is a small city (population 87,000) on the edge of the Iron Range region northeast of Lake Superior. The city’s population has shrunk during the past several decades, reflecting a general decline in iron mining. Duluth’s major employers are now hospitals and the University of Minnesota. The city’s Harbor View public housing project, built in 1951, provided 200 townhouse units in a series of two-story, barracks-style buildings on a 20-acre site north of downtown. In 2003, the Duluth Housing Authority (DHA) received a HOPE VI grant for $20 million to completely demolish the project and build a new, mixed-income development on site. Families were relocated during the demolition and redevelopment.
Data The research combines information on where families moved with survey data from two points in time, capturing how residents felt about their lives and living conditions before being displaced and then again up to 30 months after displacement. Measures of neighborhood change are generated by comparing the characteristics of new neighborhoods to the characteristics of the original public housing site. Combining the measures of neighborhood change with personal evaluations of the families allows an examination of the relationship between changes in neighborhood conditions (as measured by census-tract data) and changes in residents’ perceptions.
In 2003, as part of the intake process to facilitate relocation counseling and to establish needs for community and social services, the DHA interviewed residents in 216 households. The intake interviews included a number of questions related to health, income, employment, neighborhood satisfaction, and social integration. In late 2005, the author mailed surveys to 192 addresses, with 111 questionnaires returned (a 58-percent response rate).1 The mail survey asked many of the same questions as the interviews, providing data points for before and after the move. Exhibit 1 At the time of the mailing, the DHA did not have address or contact information for 9 of the 216 households that responded to the initial intake survey, leaving 207 potential respondents to the mail survey. Of the 207 surveys mailed to former residents, 192 turned out to be valid addresses. The response rate for the relocation survey, therefore, is based on a denominator of 192.
12 HOPE VI Better Neighborhoods, Better Outcomes? Explaining Relocation Outcomes in HOPE VI compares the survey respondents with nonrespondents, based on data collected at the initial interviews in 2003. The groups differ significantly on just one attribute: education. Survey respondents are significantly more likely to have earned a high school diploma than those who did not respond.
Survey data show no statistically significant differences in race, gender of the head of household, first language spoken, presence of young children, or whether a family member has a physical or mental disability. In addition, data show no differences in the employment rate in year 1 of the study, in the reported level of satisfaction with the Harbor View neighborhood, or in the reported sense of safety in that original neighborhood.
The neighborhoods to which respondents and nonrespondents moved are largely identical from a statistical standpoint. Exhibit 2 shows census-tract data for nine different social and housing characteristics. The new neighborhoods for respondents and nonrespondents are statistically the same for eight of the nine characteristics. The one difference is that nonrespondents tend to live in neighborhoods with a higher percentage of non-White residents (17.8 and 15.5 percent, respectively). But for a range of other characteristics—employment, median income, poverty, and homeownership—the data show no statistically significant differences across groups.
The following analysis, which focuses on the change in responses before and after the respondents moved, attempts to determine whether and how these changes are related to changes in neighborhood conditions. Neighborhood conditions were measured for two points in time: premove conditions are defined by the characteristics of the census tract in which all families lived while at the Harbor View site and postmove conditions are defined by the characteristics of the census tracts to which families moved as a result of relocation. For residents who moved more than once since the initial relocation, the analysis uses the characteristics of their neighborhoods at the time of the survey. As with the neighborhood data, change variables (the difference between the respondents’ assessment in year 3 of the study and their assessment at intake) were created for all items analyzed.
The following five outcome measures, which come from the survey instruments, are computed as change variables. Appendix A provides details about question wording and the construction of indices. Coding was done so that higher values indicate a positive change in the variable.
1. Neighborhood satisfaction. A summary question about the respondent’s overall satisfaction with the neighborhood provided five answer categories. The computed change variable ranges from -4 to +3.
2. Sense of safety. A summary question about how safe the respondent feels in the neighborhood provided five answer categories. The computed change variable ranges from -4 to +4.