«Mixed Messages on Mixed incoMes Volume 15, Number 2 • 2013 U.S. Department of Housing and Urban Development | Office of Policy Development and ...»
Kleit, 2010; Laakso, 2013) and may not feel as comfortable in mixed-income communities (Chaskin and Joseph, 2011, 2010). Public housing residents may feel under increased scrutiny by their
middle-income neighbors, believing that different social norms may lead to disapproval of particular behaviors and habits (Chaskin and Joseph, 2011; Joseph and Chaskin, 2010). Duke (2009) noted that strict rules and expectations in mixed-income communities appear to emphasize the rights of more affluent residents, leading lower income residents to view themselves as not truly part of the neighborhood. The potential costs of mixed-income developments for residents of former low-income communities include loss of support networks, increased stigma, increased isolation, and feelings of relative deprivation (Curley, 2009; Joseph, 2006).
Social networks tend toward homogeneity; neighborhoods usually are homogeneous with regard to socioeconomic status (Kleit and Carnegie 2011; Putnam, 2007), and neighboring relationships are more frequent within homogenous networks than between them (Kleit, 2005; Putnam, 2007).
It is not surprising that artificially created mixed-income communities face a number of challenges, including tensions about youth activities, race, parenting, and differences in the ways that parents restrict their children’s activities (Chaskin and Joseph, 2011) and tensions among different groups of people, including between renters and owners and between parents and nonparents (Joseph and Chaskin, 2010).
Findings regarding ethnic diversity are mixed. In general, in ethnically diverse neighborhoods, trust, altruism, and community cooperation are less common (Putnam, 2007). In their comparison of two mixed-income communities, Chaskin and Joseph (2011) found more contentious social interactions in the more ethnically diverse community. Similarly, in a Seattle study, Whites viewed ethnically homogeneous neighborhoods as being more harmonious than heterogeneous neighborhoods. In mixed neighborhoods, Whites reported more noise and trouble and less trusting relationships (Guest, Kubrin, and Cover, 2008). By contrast, Manzo, Kleit, and Couch (2008) found that most residents of an ethnically and racially diverse community viewed diversity as an asset.
The variety of housing types, central to mixed-income HOPE VI developments, can also present challenges. Renters tend to have fewer social relationships in their community than market-rate owners and relocated public housing residents (Chaskin and Joseph, 2011). Homeowners have noted the lack of integration between homes and rental units (Kleit, 2005), expressing concerns about interactions between owners and renters and feelings that renters have less commitment to the development (Joseph, 2008). Any correlations between housing type and race or ethnicity can add to the challenges of building community.
Joseph (2008) found that former public housing residents expressed general dissatisfaction with the sense of community in the new development. Barriers to social interaction included minimal shared public space, physical and qualitative distinctions between subsidized and market-rate units, stigma, self-isolation by former public housing residents, segregated residents association structures, and perceived assumptions of property management staff about residents.
Development of Neighboring Relationships Questions arise regarding how to promote neighboring relationships across income levels, housing types, and ethnic groups. “Proximity is very important in the creation of neighboring relationships” (Kleit, 2005: 1435). More than proximity is needed to foster community, however. Physical
Cityscape 31Keller, Laakso, Stevens, and Tashiro
integration can foster interaction, but Kleit found that public housing residents in a Seattle mixedincome development were still isolated from nonpoor neighbors more than would be expected, and residents knew more fellow residents who were of similar income and educational level. Ethnicity and native language are important variables in looking at neighboring relationships, and children also can promote ties across housing type and income (Kleit, 2005). Homogeneity in terms of stage of life, homeownership, lifestyle, and values are more important than proximity (Kleit and Carnegie, 2011).
The expectations of the community also may affect neighboring relationships. Joseph (2008) found that market-rate owners expected little personal benefit from living in the mixed-income community, apart from perhaps meeting some interesting people. Public housing residents valued both the demographic makeup and the more idealized environment that they viewed as less chaotic and potentially providing opportunities for their children. According to Joseph, most did not specifically expect benefits from having new neighbors, however, and planned to keep to themselves.
Some public housing residents thought that the potential benefits would include social mobility and also thought that more affluent neighbors would develop more realistic and positive attitudes toward them.
Families in Mixed-Income Communities Families typically interact with those of similar age and stage of development. Whereas children often provide a connection among parents, Kleit (2005) found that in NewHolly, a housing development in Seattle, fewer homeowners had children, limiting opportunities for mixing across incomes and ethnicities. Thus, another challenge of mixed-income housing is attracting a critical mass of families with children (Varady et al., 2005). Middle-income families may not be attracted to mixed-income developments, and those who are may not have much in common with lower income residents (Popkin et al., 2000). Varady et al. (2005) suggested that, to attract families, communities must have strong public schools, work collaboratively with the schools, and actively market to families with children. Middle-class families with children are absent from many developments because of perceived safety issues in the community and the poor reputation of neighborhood schools. In a comparison of three public housing communities, Varady et al. (2005) found that one Louisville development was attracting families with children, although they speculate that it was because housing location did not determine school attendance. A second site appears to have promoted income mixing but not racial integration. Varady et al. concluded that attracting middle-class families with children was not a prominent goal of any of the developments they studied and highlighted the difficulty of maintaining an income mix that will lead to meaningful social interaction across social class lines.
In Chicago, Joseph (2008) found that middle-income families made housing decisions that met their basic interests and needs, looking specifically at the variables of location and affordability but not necessarily at the mixed-income makeup or ethnic diversity of the community. Similarly, Kleit and Manzo (2006) found that place dependence is important in shaping moving preferences, but final relocation choices may be determined more by family factors such as the size of the family, housing options, and employment opportunities. In addition, income differences exist in how
residents’ needs are met in the community. Middle-class residents meet many of their needs outside the geographic community and are less place bound (Joseph and Chaskin, 2010). By contrast, low-income neighboring networks tend to be more place based and homogeneously low income, with more overlapping relationships (Kleit, 2005). Particularly for people who are mobility challenged, poor, or elderly, the neighborhood is still the place where many relationships are formed (Curley, 2010). These differences between low- and middle-income groups can challenge cohesion within the community.
A Northwest Housing Community Salishan, the community addressed in this article, is named for the coastal Salish First Nation peoples, and the name has been loosely translated as “people of many colors coming together.” Thus, this multiethnic ideal is part of the historical fabric and lore of the community. The neighborhood has always been more ethnically diverse than Tacoma, Washington, the city in which it is located.
“Salishan was one of the area’s first residential neighborhoods that was racially integrated on purpose. Diversity by race, language, ethnicity, national origin, and age has remained a signature and appealing aspect of Salishan to the present day, including the redevelopment of New Salishan” (THA, 2009a: par. 4).
This housing development, along with many others in the Pacific Northwest (see Gibson, 2007;
Kleit and Galvez, 2011; Manzo, Kleit, and Couch, 2008), presents a different demographic than many HOPE VI sites in other parts of the country. At the beginning of Salishan’s HOPE VI reconstruction in 2003, nearly 60 percent of the residents were immigrants and refugees; roughly 25 percent were Cambodian, 25 percent were Vietnamese, and 10 percent were from countries in the former Soviet Union (NICF, 2007). Many types of families were represented in the development, including two-parent, multigenerational, and single-parent families; grandparents raising grandchildren; and individuals living alone. This diversity is by contrast to many public housing developments, which consist largely of female-headed, African-American families, many of whom have lived in public housing for their entire lives (Holin et al., 2003; Joseph, 2008).
Another difference is the Tacoma Housing Authority’s (THA’s) goal of eventually increasing housing density, from 855 to 1,278 housing units, although final projections are for 290 public housing units, 471 other subsidized rentals, and 100 homeownership units reserved for low-income residents.
Of those 100 homeownership units, 28 are sweat-equity homes for those whose incomes are less than 40 percent of Area Median Income (AMI) and 72 are homeownership units for those whose incomes are less than 60 percent of AMI (THA, 2009b). Instead of the highrise apartment model, this community initially consisted primarily of single-family homes, interspersed with fewer duplexes and triplexes, all of one story. The new community has a combination of one- and two-story singlefamily homes but many more duplexes, triplexes, fourplexes, and apartment housing for seniors.
Although the Old Salishan (before HOPE VI) community faced significant issues, they were not to the level of the nation’s “most severely distressed” housing, originally the target of HOPE VI. The housing quality was poor, with poor wiring and insulation, no showers, and mold and mildew.
Community challenges included crime, drugs, gangs, and poverty, issues that were targeted by the THA and residents. The census tract comprising this housing development had the highest poverty
Cityscape 33Keller, Laakso, Stevens, and Tashiro
rate in the Tacoma area, with 56 percent of residents falling below the poverty line in the 2000 census. More than 80 percent of the students in the elementary school serving this community qualified for free or reduced-price meals in 2010. In many ways, the residents of this community were socially and geographically marginalized.
Although the poor-quality low-income housing has been replaced with mixed-income housing, market-rate rentals (18 units) are relatively few compared with primarily public housing and sitebased housing choice voucher (Section 8) rentals. Individual family homes consist of market-rate homes (257, primarily on the periphery of the development), fewer (28) sweat-equity (Habitat for Humanity) homes, and some (72) subsidized homes (THA, 2009b). It may seem ironic that, although the goal was a mixed-income development, some displaced residents stated that they were unable to return because they earned too much income to qualify for the rentals but not enough to secure a mortgage.
Methodology Data were gathered across nearly 4 years. The midpoint evaluation, in 2006, consisted of semistructured interviews with 52 current and former residents (20 nonmovers and 32 movers) of the community 3 years into redevelopment. In 2009, 26 followup interviews, 7 focus groups, and interviews with eight community stakeholders were conducted. Initial interviewees were heads of household, randomly selected from THA occupancy lists, which were divided by housing situation. Additional recruitment strategies were employed by caseworkers and members of the various ethnic communities to obtain representation from the predominant ethnic groups residing in Salishan and from each type of housing. Followup interviews were conducted with all initial interviewees who agreed to a second interview and could be located and scheduled. Bicultural translators and interpreters interviewed residents from the three major non-English speaking language groups at Salishan: Khmer, Russian, and Vietnamese. All resident interviews were fully transcribed into English from audio recordings.
The focus groups included former and current Salishan residents; some had taken part in the initial and followup interviews and others were found through snowball and convenience sampling, including youth and young adults, who were not part of the interviews. The focus groups included (1) Russian homeowners, (2) Russian teenagers, (3) Cambodian young adults, (4) Cambodian elders, (5) late-adolescent Cambodian and Vietnamese youth, (6) Vietnamese elders, and (7) long-term female residents (four White and one African American). Focus groups also were conducted in the primary language of the interviewees and transcribed into English. Stakeholders were recommended by community members and THA staff and included representatives of local government, clergy, and social service providers.
The transcripts were reviewed, looking specifically at issues relevant to sense of community and valuing of diversity, including questions about trust, participation in neighborhood activities, and views toward neighbors. The data analysis included descriptive coding to organize data and look for patterns in segments of interviews and common threads in respondents’ accounts of life in Salishan, using cross-case analysis. To verify the original coding, the data were continuously reviewed for discrepancies or errors. The themes and conclusions were compared with those in the literature.
Respondents first moved to Salishan an average of 12.6 years before the beginning of interviews in