«DiscoVeriNg HomelessNess Volume 13, Number 1 • 2011 U.S. Department of Housing and Urban Development | Office of Policy Development and Research ...»
The Existence of the Overrepresentation Studies on homelessness after 1980 have consistently found the population to be much more racially diverse than it was before 1980 (Hopper, 2003; Rossi, 1989a; Rossi, 1989b). Before the 1980s, the homeless population was primarily composed of White middle-aged men.3 After 1980, Blacks became overrepresented in the homeless population with respect to their share of the national population (12.8 percent) and their share of the poverty population (28.4 percent of individuals and 26.1 percent of families).4 In one of the most reliable studies of the homeless population, Burt (1992) found 41 percent of the homeless population to be Black and 56 percent of the adult female homeless population to be Black. Shlay and Rossi (1992), in their review of 52 national and local studies of the homeless, found, on average, that 44 percent of the homeless were Black, with percentages ranging from 6 to 90 percent across the studies. According to the Census S-Night5 count, in cities with more than 5 million people, 47.9 per 10,000 Black men and 24.4 per 10,000 Black The NSHAPC is a representative sample of the service-using homeless population. As such, the study tests the broader pull of homeless services, rather than just the pull of homeless shelters.
Kusmer (2002) argues that this finding is biased, because most studies of the homeless population prior to 1980 were of skid row homeless people, who were disproportionately White.
March 1997 Current Population Survey.
As part of the 1990 Decennial Census, the U.S. Census Bureau “conducted a ‘Shelter and Street-Night’ (S-Night) operation to count selected components of the homeless population in preidentified emergency shelters and open locations in the streets and other places not intended for habitation” (Martin 1992: 2).
women were homeless in 1990. These statistics compare with a rate of 14.1 for White males and
5.6 for White females (Hudson, 1998).
Explanations of the Overrepresentation Explanations for the Black overrepresentation in the homeless population have focused on factors thought to push Blacks out of housing or pull Blacks into homelessness at higher rates than Whites.
As stated previously, the major push factors examined include poverty, declines in affordable housing supply, increases in affordable housing demand, housing discrimination, residential segregation, and lack of access to mental health and substance-abuse services; the main pull factor examined is access to shelter space. There is little evidence that access to mental health and substance-abuse services is responsible for the overrepresentation, although the lack of these services may be responsible for increasing homelessness in general (National Academy of Sciences, 1988; HHS, 1989).
Push Factor 1: Poverty Because landlords require rent in exchange for housing, a household’s income could be considered a factor in the risk of losing housing. Thus, if we hold housing prices constant, we can hypothesize that the lowest income groups have the highest risk of being pushed out of housing into homelessness. If Blacks are represented at greater rates than Whites in the poverty population, we can expect their risk of homelessness to be greater. As mentioned previously, Blacks are overrepresented in both the homeless and poverty populations, although they are overrepresented to a greater degree in the homeless population. Among the homeless population, in the late 1980s, Blacks also reported less income from working than did Whites and Hispanics (Burt, 1992; Burt and Cohen, 1990; Burt and Cohen, 1989). This disparity perhaps places Blacks at greater risk of longer homeless spells once they become homeless.
Since they were developed in the early 1960s, official poverty thresholds have not been adjusted to account for area differences in housing costs. Although the original poverty measure may have been a valid indicator at the time it was developed, as rental-housing costs increased during the 1980s and 1990s, the proportion of poverty income spent on housing increased. For this reason, income must be analyzed in relation to its purchasing power in the housing marketplace. This study analyzes affordable housing supply in relation to the size of the population below 50 percent of the poverty threshold in order to take into account both the size of this population and the number of affordable rental units available to them.
Push Factor 2: Declines in Affordable Housing Supply and Increases in Affordable Housing Demand Although explaining homelessness in terms of the availability of affordable housing may seem tautological, housing is but one of many possible explanations, including poverty, mental health problems, drug abuse, and disaffiliation, as to why people become homeless (Hopper, 2003). Even if affordable housing supply is high, individuals may be evicted from their homes into homelessness if their income, mental health problems, or drug abuse make it difficult to make rent payments.
Conversely, if affordable housing supply is low, then income, mental health, and drug abuse problems may play less of a role than housing supply in pushing people into homelessness.
36 Discovering Homelessness From Exclusion to Destitution: Race, Affordable Housing, and Homelessness Many studies of the homeless population have found significant positive associations between the lack of available affordable housing, increasing housing prices, and the size of the homeless population (Burt, 1992; Bohanon, 1991; Eliot and Krivo, 1991; Honig and Filer, 1993). These studies evaluated the relative importance of affordable housing supply on the size of the homeless population using the U.S. Department of Housing and Urban Development’s (HUD’s) 1984 homeless survey (HUD, 1984). Eliot and Krivo (1991) found availability of affordable housing, along with access to mental health care, to be the strongest predictors of lower levels of homelessness.
Areas with higher poverty rates, higher concentrations of Blacks, and more female-headed families had higher rates of homelessness.
Trend studies examining the structural causes of homelessness take it as a given that homelessness increased in the 1980s and use historical trend data to assess the effects of historical factors. Shinn and Gillespie (1994) found that a small surplus of the least expensive units existed in 1970. A gap between the supply of these units and the demand for these units by low-income individuals developed after 1970. In 1985, a gap of 4.54 million units existed between the number of low-income units and the number of low-income households, which became a 5.22-million-unit gap by 1991.
Four million affordable units were lost from the housing market between 1970 and 1990 when the units were upgraded, converted to condos, or demolished. The shortage of affordable units was greatest in central cities. As affordable housing supply declines were followed by income declines in the 1980s and 1990s, rent burdens grew among unsubsidized renters, putting some at greater risk of becoming homeless (Jencks, 1994). Although affordable housing supply studies have made great progress exploring the link between affordable housing and the homeless problem, they tend to assume that all groups have equal access to the affordable housing that is available. These studies do not explain how access to housing may influence Black homelessness differently than White homelessness.
Push Factor 3: Housing Discrimination and Residential Segregation Residential segregation has been associated with negative outcomes for Blacks. Massey and Denton (1988) argued that residential segregation has been the missing factor in explaining the existence of the urban underclass and the concentration of poverty in central cities. Other segregation researchers have focused on negative outcomes for Blacks at the neighborhood level and the individual level. For instance, segregation has been found to lead to lower high school graduation rates, idleness, lower earnings, and single motherhood among Blacks (Cutler and Glaeser, 1997).
Although this research has examined a multitude of negative outcomes, it has not focused on the individual housing outcomes of Blacks.
Some researchers have pointed to residential segregation as a reason for the overrepresentation of Blacks in the homeless population (Baker, 1994; Shinn and Gillespie, 1994; Wright, 1989;
Wright, Rubin, and Devine, 1998). Some have argued that the racial composition of the homeless population is a function of the racial composition of the communities in which homeless people are found. Because homelessness rates are higher in inner-city areas, the homeless population will be Black––if such areas are inhabited primarily by Blacks (Hudson, 1998; Rossi, 1989a; Rossi, 1989b).
Thus, residential segregation may play a key role in Black homelessness. Theories about the role of housing discrimination and residential segregation in the overrepresentation, however, have never been tested.
Research exists on the connection between segregation and people most at risk of becoming homeless: those living in inadequate and overcrowded housing. Housing quality has been identified as a risk factor for homelessness (Ringheim, 1990; Rosenbaum, 1996; Stacey, 1972). Because data linking segregation to racial composition of the homeless population are not available, this study examines the link between segregation and the probability of Blacks living in substandard housing conditions. Mounting evidence indicates that Blacks do not have equal access to goodquality housing (Grigsby, 1994). Previous cross-sectional and trend studies have linked residential segregation to increased rents and decreased housing quality for Blacks (Massey and Denton, 1988;
Rosenbaum, 1996). Rosenbaum (1996) found that living in a highly segregated city (New York) and being Black were positively related to living in inadequate, dilapidated housing.
To date, studies that address segregation in analyzing housing outcomes have focused on individual cities rather than on the national level. Using data from the 1997 AHS, the 1990 Decennial Census, and the 2000 Decennial Census, this study is the first national study to examine the effects of Black headship and residential segregation on two measures of housing quality: housing inadequacy and overcrowding. I contend that residential segregation limits housing opportunities for Blacks by shrinking the market in which they make housing choices. In this sense, residential segregation leads to reduced housing opportunities for Blacks. Because of the high demand among Blacks for housing in neighborhoods with high proportions of Blacks, it should be expected that Blacks will be more crowded than Whites in their housing units and will be more likely to encounter landlords lacking the incentive to maintain properties. These patterns can be expected to increase as Blacks are increasingly separated from the White housing market.
An alternative explanation for the discrepancies in housing quality between Blacks and Whites is provided by Johnston (1982), who argued that increasing nationwide levels of homeownership, supported by Federal policies favorable to homeownership, have had detrimental effects on renters, who are disproportionately Black, poor, and young. He contends that the rents these groups can afford “are insufficient to provide a reasonable return to landlords, let alone cover the rising costs of maintenance” (Johnston, 1982: 184). Thus, as homeownership increases, the quality of rental housing diminishes for Blacks, because there is less incentive for landlords to maintain the rental properties that are available. In this way, it is reasonable to expect increasing levels of homeownership at the city level to also increase housing inadequacy for Blacks.
Pull Factor: Access to Shelter Space Some researchers have argued that homeless shelters perpetuate long-term homelessness and pull people out of inadequate, substandard housing into homelessness (Gounis, 1990; Jencks, 1994).
Although Blacks have less access to high-quality affordable housing, they have greater access to shelter space. Baker (1994) found that shelters were more likely to be placed in communities with high percentages of Blacks, and Lee and Farrell (2004) found that shelters were more likely to be placed in communities with high percentages of minorities. Assuming equal preferences for the use of homeless services, closer proximity to homeless services can be hypothesized to increase the use of those services.
Because Blacks on average are located closer to homeless services, it is logical to hypothesize that precariously housed Blacks will use those services more and will be more likely to become part 38 Discovering Homelessness From Exclusion to Destitution: Race, Affordable Housing, and Homelessness of the service-using homeless population than will precariously housed Whites. Whites who use homeless services will be more likely than Blacks using homeless services to have to move to use services or may end up not using any services because no services are available near where they became homeless. Distance barriers may serve to keep precariously housed White people doubledup in the homes of friends and family or on the streets out of view of surveys of the service-using homeless population.
Data on prior residences, service locations, and socioeconomic characteristics of precariously housed and currently homeless people are needed to determine the causal effect of service location on service utilization. Unfortunately, such data do not exist. Working under the assumption that close proximity to homeless services decreases the need to migrate for such services, this study examines racial differences in migration for homeless services using data from the 1996 NSHAPC.
Data and Methodology This section addresses hypotheses, data, and methods used in Parts I and II of the research. Part I of the study, using data from the 1990 and 2000 Decennial Censuses and the 1997 AHS, analyzes the relationship between residential segregation and two measures of housing quality: housing inadequacy and overcrowding. Part II of the study, using data from the 1996 NSHAPC, analyzes the migration of Black and White homeless clients for homeless services.
Part I: Analysis of the Relation Between Residential Segregation and Blacks at Risk of Homelessness Part I of the study analyzes the relationship between residential segregation and two housing outcomes thought to be risk factors for homelessness: housing inadequacy and overcrowding. It tests