«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 ...»
Family characteristics, such as the number of small children and a dummy variable indicating whether any member has a physical or mental disability, are included. The age of the head of household is measured in years and also as a dummy variable separating seniors (aged 55 or older) from the rest. Race/ethnicity is measured as a series of dummy variables for the categories of Asian, African American, Native American, and Hispanic. When these variables are used in the multivariate analysis, the excluded group is White non-Hispanics. Alternatively, another dummy variable is computed to differentiate White non-Hispanics from all other groups. The respondent’s potential support network is measured by a variable indicating the number of close friends and family members living in the same neighborhood. Gender is measured by a dummy variable, taking the value of 1 for a male head of household. The employment status of the head of household at year 1 (coded as a dummy variable) is also included as a control variable.2 Exhibit 7 presents the findings for OLS regressions with changes in neighborhood satisfaction, sense of safety, neighboring behaviors, and economic security as dependent variables. Neighborhood change variables are unrelated to the change in the first three dependent variables—respondents’ neighborhood satisfaction, sense of safety, and neighboring behaviors.
In any case, the relationship is the opposite of what program advocates hope for; that is, relocation to neighborhoods with a smaller proportion of non-Whites may reduce economic self-sufficiency.
A number of individual-level variables are significantly related to the dependent variables. The age of the head of household is important in all the models. Younger heads of households show greater improvement in neighborhood satisfaction, sense of safety, and economic security (at the more marginal p.10 level) than do older heads. Seniors are much more likely to increase their neighboring behaviors after relocation than are younger families. Families with fewer young children also report more positive change in neighborhood satisfaction and sense of safety than those with more young children. Respondents with a high school diploma or more education report more neighboring behaviors and less economic insecurity after relocation than do those who lack a high school education. Asian respondents, most of whom are recent Hmong immigrants to the United States, report a significantly greater reduction in neighborhood satisfaction and sense of safety than do Whites. African-American respondents also report less improvement in neighborhood satisfacTwo additional individual-level variables were left out of the final analysis. First, English-language proficiency was omitted because of its very high correlation with the variable measuring Asian racial status. Second, whether the family made an intermediate move (between the time of relocation and the survey in year 3) was omitted because it was unrelated to any of the dependent variables examined, and its inclusion did not improve the explanatory power of any of the models. In addition, interaction terms between the desire to move and the neighborhood change variables were examined. These terms were statistically insignificant in all cases and did not change the substantive findings from those discussed in this analysis.
tion than Whites. Families in which someone has a physical or mental disability show a significant improvement in economic security after the move compared with other families. This relative improvement seems counterintuitive but may reflect a greater level of income stability for families receiving disability assistance.
The last variable added to the model is the attachment of the family to the Harbor View neighborhood. This variable is extremely important to an increase in both neighborhood satisfaction and sense of safety. Families that wanted to move report significant positive changes in safety and satisfaction compared with those that did not want to move. Those more attached to the Harbor View development reported significantly less change in satisfaction and sense of safety. Neighborhood attachment, however, is unrelated to changes in neighboring behaviors and economic security.
22 HOPE VI Better Neighborhoods, Better Outcomes? Explaining Relocation Outcomes in HOPE VI Summary and Implications These findings are based on a single case study and cannot be generalized to all HOPE VI sites.
Nevertheless, many of the attitudes and outcomes reported by residents in Duluth are consistent with those reported in other locations. Relocation outcomes and neighborhood change among displaced families in Duluth, for example, mirror the national pattern: most families remained in the central city and moved to neighborhoods that exhibited significantly less disadvantage on a range of measures based on census-tract data. Also mirroring national trends, the Duluth families reported very little overall improvement on a range of individual-level outcomes. Thus, the Duluth case offers the potential for understanding why self-reported individual benefits from the HOPE VI Program have been so limited for displaced households.
One explanation for these findings is that, because neighborhood benefits are not a linear phenomenon, relocated people must experience a certain threshold of change before reporting short-term benefits. It might be the case that HOPE VI does not move families to neighborhoods that are good enough to generate benefits. This explanation is suggested by the findings here and elsewhere showing that, although HOPE VI families’ new neighborhoods are better than their original ones, the new areas are nevertheless high in conditions such as poverty, unemployment, and racial segregation compared with local averages. One form of the threshold hypothesis was tested in this study, but it was found that residents who had experienced the greatest degree of change across three different dimensions—racial segregation, poverty, and housing market value—did not differ from others in the extent to which they reported individual-level benefits.
A second explanation for the lack of relationship between objective improvements in neighborhood environment and subjective assessments of individual benefits is that individual attributes play a more central role in determining how and whether families benefit from displacement and relocation. The Duluth case supports this explanation. The age of the head of household and the presence of young children are consistently important predictors of the benefits from relocation that respondents report. Race, health, and the education level of the head of household are also important predictors of the individual benefits examined in this analysis. By contrast, the indicators of neighborhood change as measured by census-tract data are statistically insignificant in virtually all cases.
This analysis provides empirical support for a third explanation of why objective measures of neighborhood change are unrelated to individual benefits among relocated families. The attachment to place as measured by the willingness of the families to move away from the original public housing site was significantly related to improvements in neighborhood satisfaction and perceptions of safety. Respondents who expressed a desire and readiness to move away from the public housing site experienced greater benefits from relocation than those who did not wish to move.
This finding is an extension of previous evidence that those participating in voluntary programs of dispersal report greater benefits than those who are involuntarily displaced (Goetz 2003). The Duluth findings show that even among those who were involuntarily displaced, some families are ready and willing to move, and that these families report the most benefits from relocation.
In Duluth and elsewhere (see, for example, Gibson, 2007; Goetz, 2003; Kleit and Manzo, 2006;
Vale 1997), however, the number of people who do not wish to move is very sizable (more than
one-half of the people who were relocated in most cases in which researchers have collected that information).
The preceding analysis contains two additional findings that are instructive for a more general assessment of the policy of dispersion. First, individual-level outcomes from relocation tend not to be consistent across a range of measures. That is, relocated people who show positive change on one type of outcome do not necessarily show the same magnitude of change on other outcomes.
Equally, negative outcomes in some areas do not imply a negative experience across the entire spectrum of outcomes. This finding suggests a need to refine the logic of the HOPE VI model by, at the very least, differentiating between the presumed individual-level benefits and the processes that are presumed to bring them about.
Second, this analysis of HOPE VI outcomes in Duluth directly relates the degree of neighborhood change to the degree of individual changes reported by residents. The findings confirm evidence from other studies indicating that neighborhood change is largely unrelated to the individual-level benefits. The Duluth case found no relationship between any measures of neighborhood change— even indices that compounded positive changes across three dimensions of change—and several measures of individual outcomes. The one exception was a finding that ran counter to the dispersal hypothesis: a move to neighborhoods inhabited by more Whites was associated with a decline in economic security among relocated people.
As argued previously, these findings invite a systematic deconstruction of theoretical linkages so as to provide a more realistic assessment of changes from involuntary relocation in a HOPE VI project. Among the range of possible outcomes from relocation, for example, previous studies (although not this one) seem to indicate that perceptual changes, such as feelings of greater safety and perceptions of greater social order, can be influenced by a change in neighborhood. On matters related to actual behavioral change, to the achievement of goals such as employment and self-sufficiency, and to improvement in physical conditions such as health, environmental change alone is unlikely to produce consistently positive results. As Levy and Woolley (2007) argue in relation to employment outcomes and as Clampet-Lundquist (2007) argues in the context of social networks, the policy assumptions and program interventions of HOPE VI probably underestimate the complexity of the social and economic changes they aim to induce.
Changes in employment, income, health, and social interactions involve systems that are complex and not fully determined by environment. Perhaps the most obvious is employment and related indicators such as income and economic self-sufficiency. These and other outcomes are likely to be influenced by a mix of systems operating at different scales. Varying factors—such as the availability of appropriate jobs in a metropolitan area, traditions of urban segregation by class and race that vary by region, the willingness of employers to hire, individual attributes such as adequate training and education, and the variable social interactions involved in the job search, the interview, and the hire—play different roles in determining economic outcomes for poor and minority households. Similarly complex systems could be described to help explain the development of social networks and neighboring behaviors that vary by race and income.
Thus, a reevaluation of the dispersal hypothesis requires a more explicit set of theoretical connections between neighborhood change and specific individual-level outcomes. Indeed, one step 24 HOPE VI Better Neighborhoods, Better Outcomes? Explaining Relocation Outcomes in HOPE VI would be to identify those areas in which positive change may reasonably be expected and those that involve larger systems that may be more resistant to such a simplified and problematic stimulus for change such as forced relocation.
Of significance, the degree of neighborhood change was unrelated to feelings of greater safety and neighborhood satisfaction among relocated people in Duluth. Instead, the most prominent factor associated with these outcomes was the desire to move away from the Harbor View site. This finding is important for two reasons. First, it locates the origin of attitudinal change in residents’ evaluation of their original neighborhood. If residents found the existing environment wanting and desired to move away, they were likely to experience the short-term perceptual benefits hypothesized by the program model. At the same time, if this finding is replicated elsewhere, it suggests that the HOPE VI model of involuntary displacement will probably not produce consistent benefits for a substantial number of relocated families. Among the residents in many public housing redevelopment projects, a substantial portion does not wish to move. In Duluth, for example, onehalf of the residents did not want to move; in Portland, two-thirds did not want to leave (Gibson, 2007). Thus, voluntary relocation programs might be a more appropriate approach for achieving outcomes such as a greater sense of safety and a higher level of neighborhood satisfaction.
For families involuntarily displaced from their homes, the questions of safety and neighborhood satisfaction may be more dependent on the families’ networks of social support. To the extent that forced displacement disrupts those informal webs of support, HOPE VI may engender resentment among the displaced and fail to produce the outcomes desired by the program’s architects.
26 HOPE VI Better Neighborhoods, Better Outcomes? Explaining Relocation Outcomes in HOPE VI Acknowledgments The author thanks Rick Bell, Dan Moore, and Susan Jordan of the Duluth Housing Authority for their support of the research and assistance in compiling the data. All interpretations are the author’s.
Author Edward G. Goetz is director of the Center for Urban and Regional Affairs and professor of urban and regional planning in the Humphrey Institute of Public Affairs at the University of Minnesota.