Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 04 June 2020

The Politics of Homelessness in the United States

Abstract and Keywords

As the United States grapples with increasing economic inequality and significant poverty, homelessness represents a thorny political and policy issue. This article explores the debates that contrast two primary responses to homelessness, Housing First and the linear, or treatment first, model. Both are employed to address homelessness, though Housing First has been directed almost exclusively to chronic homeless people who are on the streets for long periods of time and contend with mental illness and substance abuse problems. By reviewing the strengths and weaknesses of each type of policy, this article analyzes the philosophy underlying each approach and its impact on homeless people.

Keywords: homelessness, United States, housing, homeless policy, Housing First


Homelessness represents both a policy problem and a political quandary in the United States. In the wealthiest economy in the world, the fact that individuals and families lack housing and must live on the streets, in their cars, or in congregate shelters calls into question the basic functioning of the social safety net and suggests that something is deeply wrong with the political and economic priorities of the country. Yet the dominant discourse in the United States proposes that at least some percentage of homeless people are at fault for their situations; their dysfunctional behavior, aberrant choices, and lack of a work ethic explain their homelessness more than economic inequalities or policy priorities. Within this framework of contrasting views on homelessness, debates rage over the significance of the homeless problem, the reasons people lose their housing, how best to assist them, and who should be classified as homeless. These questions often are linked to one another; why homelessness occurs and which demographic variables are stressed in describing homeless populations relate directly to whether and how policy is structured to help people become housed.

A relatively invisible political issue affecting a small portion of the population until the early 1980s, homelessness became increasingly problematic throughout the late twentieth and early twenty-first centuries in the United States. Although the economy rose and dipped during these decades, as did the percentage of people living in poverty, homelessness appeared to be a relatively intractable problem. The numbers of people newly homeless generally have not abated, even during periods of economic growth. Still more concerning, the homeless population contains sizable numbers of children and young people, both accompanied by parents and unaccompanied by adults. Most estimates indicate that families with children comprise 30% to 36% of the homeless population (Bassuk et al. 2014; Weinreb, Rog, and Henderson 2010). And the number of homeless children is increasing. Approximately 1.6 million school-aged children were homeless during the 2011–2012 school year: “These numbers represent a 10% increase over the previous school year—an historic high” (Bassuk et al. 2014, 457; National Center on Family Homelessness 2011). As the analysis of policy approaches below suggests, political and policy choices may explain, at least in part, the growing numbers of children who are homeless.

Research indicates that the reasons for homelessness are complicated and multilayered. First and foremost, homelessness is a product of poverty:

[A] variety of complex social system dislocations—an increasing rate of poverty, a deteriorating social “safety net,” the steady loss of low-skill employment and low-income housing, and others—have created a situation … where some people are essentially destined to become homeless. In so many words, we now have more poor and otherwise marginalized people than we have affordable housing in which to accommodate them.

(Wright et al. 1998, 6)

While the official US poverty rate has hovered around 15% since 2010, estimates suggest that “a third of all people were near poor and poor” in the United States (Iceland 2013, 44). In addition, approximately 6.6% of households have an income below 50% of the poverty line (Iceland 2013, 44), or roughly $12,000 annually for a family of four, suggesting that the depth of poverty presents a tremendous barrier to housing stability for a considerable portion of the population. As Edin and Shaefer discovered in their research on impoverished families in the United States, approximately 1.5 million households lived on cash incomes of at most $2 per day per person in 2011, a calculation that includes cash welfare payments but does not include in-kind assistance like food programs (Edin and Shaefer 2015). When poverty is that profound, people clearly struggle to afford basic necessities such as housing, food, clothing, and utilities and are at considerable risk of becoming homeless.

Low income housing is in short supply, and housing subsidies are not widely available to those whose incomes qualify them for assistance. With almost twelve million extremely low income renters (those who earn less than 30% of area median income), there are just over four million available units that are affordable for this group (Harvard Joint Center for Housing Studies 2013). More than 70% of households earning less than $15,000 annually pay more than 50% of their income for rent each month (Harvard Joint Center for Housing Studies, 2013). In addition, among those who qualify for government-funded low income housing vouchers, only a fourth get access to a voucher, and low income households can expect long waiting lists for federal rental assistance (Center on Budget and Policy Priorities 2009). Thus, housing supply has not kept pace with the numbers of people whose incomes require low cost housing if they are to remain stably housed, putting them at severe risk for homelessness. New data on eviction rates demonstrate the impact of the low cost housing shortage; evictions of low income renters occur because of their inability to pay the full amount of rent consistently (Desmond 2016). When Desmond surveyed low income renters in Milwaukee, he found that approximately 12% had “experienced one forced move—formal or informal eviction, landlord foreclosure, or building condemnation—in the two years prior” (Desmond 2016, 330). Among low income renters, women with children were particularly likely to be evicted. Some of those who are evicted double up with family or friends, some find another rental in the private market, and others become homeless.

Struggles with poverty and low income housing shortages interact with several other convoluted causes of homelessness, often a combination of what are termed “structural” and “individual” issues, such as low wages and the declining value of the minimum wage, family violence, lack of access to welfare supports, mental illness, and drug and alcohol use (Williams 2016; Williams 2008). To further complicate matters, in the United States a number of homeless subpopulations exist that may have differing service needs as well as reasons for becoming homeless; for example, families with children, unaccompanied youth, and the mentally ill may diverge significantly from one another. The differences between the “chronic” versus first time or “crisis” homeless populations have become a crucial area of study and debate in the United States. Chronically homeless people are unaccompanied adults who have been homeless for a year or more or who have had at least four episodes of homelessness over the past three years and are either mentally or physically disabled. Culhane and Metraux found that among those who stay in shelters, the chronically homeless had substantially more substance use and mental illness than the rest of the shelter-using homeless population (Culhane and Metraux 2008). Studies that track shelter use over one- or two-year periods indicate that a relatively small portion of the homeless population is “chronic” or long term, but that they use a disproportionate amount of resources and have service needs that differ from those who are homeless for short periods.

Layered upon and connected to the various causes of homelessness, arguments weighing the best policy responses to homelessness present central political questions in the United States. As the primary federal legislation responding to homelessness, the McKinney-Vento Act is the funding vehicle for many local shelters and homeless programs. The Stewart B. McKinney Homeless Assistance Act was passed in 1987 and renamed the McKinney-Vento Homeless Assistance Act in 2000. The federal legislation provides funding for a broad range of homeless programs and services, including emergency and transitional shelters, permanent housing, healthcare, education, and job training. The act provides a definition of homelessness used by most federal and state programs, as well as most scholars engaged in research on homelessness (Pub. L. No. 100-77). Officially, a homeless person is defined by the McKinney-Vento Homeless Assistance Act as

an individual who lacks a fixed, regular, and adequate nighttime residence; and an individual who has a primary nighttime residence that is—a supervised publicly or privately operated shelter designed to provide temporary living accommodations (including welfare hotels, congregate shelters, and transitional housing for the mentally ill); an institution that provides a temporary residence for individuals intended to be institutionalized; or a public or private place not designed for, or ordinarily used as, a regular sleeping accommodation for human beings.

(Pub. L. No. 100-77)

This relatively narrow definition of homelessness was expanded in 2010 with the passage of the Homeless Emergency Assistance and Rapid Transition to Housing (HEARTH) Act (Pub. L. No. 111-22). By adding people who are doubled up, facing eviction, residing in motels, and precariously housed, the HEARTH Act amends McKinney-Vento to include those who are homeless but do not live in shelters or literally on the street. It also explicitly mandates that women fleeing domestic violence be considered homeless and fully eligible for homeless services.

Among the discussions about the purposes and chief concerns of McKinney-Vento is the examination of its priorities, particularly the balance between funding for emergency shelters and permanent housing. The most recent of these debates contrasts the “linear” method to the “Housing First” approach to addressing homelessness. At the core of the linear model, the primary policy for the past thirty years has been to rely on “emergency” shelters to assist homeless people. These shelters proliferate in US cities and are often the only type of homeless service available in a metropolitan area. With limits on stays of thirty to ninety days, this approach in theory provides immediate, short-term shelter for the homeless. Homeless people must meet certain milestones in order to move through a series of steps in an emergency shelter and sometimes transitional housing, toward permanent housing; ergo, the reliance on emergency shelters is considered the first step in the linear approach. “Housing First,” on the other hand, seeks to provide permanent housing immediately and follow up with social supports, addiction therapy, and the like after housing has been secured. Though Housing First is a more recent strategy in the homeless policy arsenal, it has gained stalwart advocates that include some service providers and the United States Interagency Council on Homelessness.

Both the linear and Housing First models, as currently structured and implemented, have been supported and criticized by researchers and advocates, for different reasons. Some argue, for example, that the linear model utilizes social control as the foundation for its programs, while Housing First is perceived to avoid social control and introduce consumer choice. In addition, although the linear model has been the primary response to homelessness for decades, significant numbers of people remain without housing, including thousands of people who are newly homeless every month. On the other hand, Housing First has been conceived and implemented largely to benefit chronically homeless populations. Some advocates argue that the federal government’s narrow interest in Housing First—as a response to chronic homelessness exclusively—has led to a general disregard of homeless families both in policy circles and by the public in recent years. Questions about whether and how the Housing First approach could be executed to benefit other subpopulations in addition to the chronically homeless remain unanswered. Though the criticisms of the two models are mostly different, the models share the inability to address the core reason that people are homeless: a severe shortage of low cost housing in the United States means that the housing stock simply is too small to meet the needs of low income people, thus producing housing instability and homelessness.1 Combined with declining real wages, the dearth of government housing assistance, and waning welfare supports, any homeless policy that does not significantly expand low cost housing options will not address the structural reasons that people become homeless and thus will not prevent or respond effectively to homelessness.

This article explores the political and policy debates that contrast the Housing First and linear models, both used to respond to homelessness in the United States. By reviewing the strengths and weaknesses of each type of policy, this article analyzes the philosophy underlying each policy approach and its impact on homeless people. While the primary function for Housing First to date has been to help long-term or chronically homeless people who struggle with mental illness and substance abuse, this article explores its applicability to other subpopulations, including families, and argues that the need for low cost, permanent housing is paramount for homeless interventions and policies to succeed.

The Linear Approach

In place in some form since the 1980s, the linear model still stands as the dominant type of homeless service provision in the United States; more shelters and programs rely on this version of homeless assistance than the Housing First model. Characterized as “earning your way to housing” (Willse 2015, 143), linear programs are built on the notion that homeless people must move through various stages of social services, for example emergency shelter, transitional shelter, and finally permanent housing. Various other terms have been used to refer to the linear approach, including “mainstream model” (Padgett, Henwood, and Tsemberis 2016), “continuum of care” model (Collins, Malone, and Clifasefi 2013), “staircase model” (Padgett, Henwood, and Tsemberis 2016), “residential treatment first” (Tsai et al. 2010), or simply “treatment first” (West et al. 2014). All of these terms refer to an approach that “anticipates that homeless persons will enter and graduate from a sequence of programs … with progress based on recovery toward self sufficiency” (Kertesz et al. 2009, 496); they are based on the pretext that the individual must be reformed in order to gain and maintain housing (Willse 2015; Williams 2016).

The linear model depends on acceptance into a short-term homeless shelter as the basic first step in the progression toward housing. Shelters range from small, private shelters serving fewer than twenty people to large, armory-style shelters that house hundreds of people (Dordick 1997). Most family homeless shelters provide a bedroom in a larger building or a bedroom and kitchenette in a renovated motel to homeless families (Williams 2016), but also may have shared bedrooms, bathrooms, and public spaces (Fisher et al. 2014). Though they vary, most emergency shelters employ basic regulations that serve as surveillance and control mechanisms. Specifically, clients are expected to reveal to the staff their personal histories, current goals, and daily activities. Shared living space and caseworker surveillance make it difficult for homeless residents to find time to themselves and to keep many aspects of their lives private. Indeed, because of shelter rules and invasive staff practices, many homeless people exhaust other options before turning to a shelter as their last resort. Yet because so many people do have to choose between the street and a shelter, and because there are too few shelters for the number of homeless people, the shelters are almost always full. They constantly have to turn people away for lack of space (Foscarinis 2008; Williams 2016).

In the linear approach, permanent housing is perceived as a resource that could easily be wasted on those who are not “housing ready.” For homeless people to be successful in maintaining housing over a long period of time, they must conquer a series of steps to stabilize their mental health, achieve abstinence from drugs and alcohol, improve their financial status, and regulate their behavior: “Most linear interventions assume that a return to long-term stable housing, in either the private market or a subsidized setting, requires the restoration of behavioral self-regulation and the capacity to interact in a constructive social environment and also that an individual’s tangible resource needs must be addressed” (Kertesz et al. 2009, 500; Padgett, Henwood, and Tsemberis 2016; Willse 2015). Each step in the sequence comes with increasing demands on the homeless person (Padgett, Henwood, and Tsemberis 2016), often including treatment for drugs or alcohol that results in abstinence, or agreeing to take medication to prove psychiatric stability (Greenwood, Stefancic, and Tsemberis 2013; Tsemberis and Eisenberg 2000). For those who are perceived as noncompliant or who may choose another path to housing, their lack of cooperation is presumed to be proof that they cling to the very dysfunctional behaviors and choices that supposedly created their homelessness (Williams 2016). In other words, they are perceived not to be “housing ready” (Padgett, Henwood, and Tsemberis 2016).

Analysis of the Linear Approach

Many of the criticisms of the linear approach begin with its reliance on emergency shelters as the first step in a process that must be completed in order to achieve permanent housing. Though theoretically emergency shelters provide immediate assistance and a place to sleep for all who need it, the reality is that there are not enough shelter beds to meet demand, and emergency shelters often have waiting lists and turn people away (Foscarinis 2008; Williams 2016). In addition, shelters may refuse assistance to people for a multitude of reasons, including that they lack enough space for larger families; do not wish to work with people with active drug, alcohol, or mental health issues, which are perceived as problematic in a communal setting; have only English-speaking staff; will not house families headed by women with teenage sons (boys above age twelve are not allowed to stay in some shelters for homeless battered women); or do not wish to work with women who are homeless due to domestic violence (Williams 2008). The notion that a shelter provides “emergency” housing, then, is only true in some circumstances for particular people; homeless people must meet a fairly rigid set of requirements to gain entrance to most shelters.

Once accepted into an emergency shelter, homeless people may discover that the program is built on the assumption that they are too dysfunctional to make appropriate choices about housing or about what they need in order to become housed. As Willse argues:

[P]opular conceptions of poverty in the United States have maintained that individuals living in poverty produce their impoverished conditions, not social or governmental institutions. Such discourse of personal responsibility has been accompanied by intensive networks of social welfare technologies that seek to regulate the poor by intervening in individual behavior…. [P]ersons living without shelter have been understood as being especially incapable of self-management and in need of invasive social assistance.

(Willse 2015, 140–141).

Invasive social assistance is one of the hallmarks of family emergency shelters, and much of the research literature focuses on the tendency for shelters to utilize social control as a key aspect of their programs (DeWard and Moe 2010; Hartnett and Postmus 2010; Williams 2016, 2008; Willse 2015). Research describes how shelter programs are designed to control, shape, and change perceived deviant behaviors as much as to help people locate housing. Policies such as curfews, specified morning wake-up times, smoking bans, required chores, breathalyzers, and urinalysis tests are paired with requirements to participate in parenting classes, substance abuse counseling, and budgeting classes. The extensive and complicated rules and mandatory requirements create an environment that suggests that families are homeless largely because of their behaviors and choices. Willse maintains that the linear model’s reliance on “invasive social assistance” in making “‘shelter resistant’ individuals ‘housing ready’" (Willse 2015, 140) both reflects and serves to enforce a view of homeless people as undeserving of assistance.

When provided the opportunity to critique the shelter system, homeless families, for example, articulate their desire for permanent subsidized housing as opposed to emergency, transitional, or other nonpermanent shelter solutions. In a study of housing decisions made by homeless families, homeless adults with children articulated anxiety and dissatisfaction with many of the programs currently available to them. While subsidized housing was the most desired assistance, project-based transitional housing was perceived as the worst option because “facilities [were] in undesirable locations, rules that excluded men (including fathers of children), and various environmental stressors” (Fisher et al. 2014, 382), including concerns about crime. Transitional housing, though usually available for a year and thus providing help for longer than an emergency shelter, has many of the same rules and restrictions as an emergency shelter. Transitional housing may be perceived as a necessary step in the linear model in order eventually to achieve permanent housing.

Despite criticisms, it is doubtful that the linear approach—and the shelters and transitional programs on which it depends—will disappear. With an extensive network of shelters, too much infrastructure exists that would be difficult and expensive to transition to other purposes. There is, moreover, a significant investment in the linear approach and its treatment objectives on the part of social service workers. The idea that programs “waste” housing by putting people into apartments who are not “housing ready” persists among some social service programs (Padgett, Henwood, and Tsemberis 2016). As Kertesz and others point out, however, more than twenty years of funding for the linear approach through the federal government’s Continuum of Care process has not abolished or even truly decreased homelessness (Kertesz et al. 2009). Thus without evidence to demonstrate the effectiveness of the linear model, conditions were ripe for a new approach, Housing First, to gain interest and adherents.

Housing First

The roots of Housing First can be traced to programs developed by Pathways to Housing in the early 1990s in New York City (Greenwood, Stefancic, and Tsemberis 2013; Tsemberis, Gulcur, and Nakae 2004). A nonprofit agency serving the chronic or long-term homeless population who also battle mental illness and/or substance abuse, Pathways created a program with the central premise that chronically homeless people could be housed immediately. Supportive services such as primary healthcare, employment assistance, substance abuse counseling, and mental healthcare were offered after housing was secured (Greenberg et al. 2013; Tsemberis and Eisenberg 2000). Pathways founders argue that homeless people have a right to housing, even if it is not a legally recognized right in the United States (Greenwood, Stefancic, and Tsemberis 2013). In addition, their organizing principle of consumer choice creates a level of respect for their clients’ viewpoints about the quality, style, and location of housing, as well as the services utilized once they are housed (Padgett, Henwood, and Tsemberis 2016; (Greenwood, Stefancic, and Tsemberis 2013). The argument that people have to prove they are housing ready, so central to the linear model, was rejected by Pathways and other programs pursuing a Housing First model.

With assistance to locate housing and extensive support services offered but not mandated once a person has been housed, Housing First “offers the independence and privacy that most consumers desire” (Tsemberis and Eisenberg 2000, 491), alongside intensive but flexible and nonmandatory services to help people maintain their housing and attain greater self-sufficiency (Greenberg et al. 2013). Staff members track down and do outreach to invite homeless people who are eligible for housing to apply to the program, and once they have entered the program, help them locate, choose, rent, and furnish an apartment. Programs may lodge clients in a variety of settings: “Supportive housing may consist of individual residential units in market housing; set-asides of some number of units inside larger residential developments; congregate apartment or single room occupancy (SRO) buildings that exclusively house the target population; or mixed-tenancy, congregate buildings that provide affordable housing for people with and without disabilities or histories of homelessness” (Levitt et al. 2012, 413; Collins, Malone, and Clifasefi 2013). For Pathways to Housing programs that serve as a model for many other Housing First programs, “major goals are to … meet basic needs, enhance quality of life, increase social skills and social roles, and increase employment opportunities” (Tsemberis and Eisenberg 2000, 489). Neither psychiatric treatment nor sobriety must be attained by homeless people in order to qualify for housing, and they need not agree to participate in any particular service or form of care to remain housed; rather, programs employ a harm reduction approach with the goal of reducing risks to participants associated with substance use or mental illness (Padgett, Henwood, and Tsemberis 2016). Housing First is expensive because of the intensive supportive services that accompany housing for the chronically homeless; Moulton estimates the cost of housing a chronically homeless person at $55,600 for the first year (2013).

The gradual spread of Housing First approaches set in motion a new view of chronically homeless people; they began to be referred to as “vulnerable populations” (West et al. 2014, 232; Tsemberis and Eisenberg 2000), with an emphasis on their high mortality rates (Willse 2015). While a focus on their presumed dysfunctional choices and personal responsibility for their circumstances has not been eradicated from political discourse or from social service responses to homelessness, Housing First did draw attention to ways that the chronically homeless could be considered a weak and defenseless population who deserved assistance. Chronically homeless people are appreciably more likely than the general population to suffer from cirrhosis of the liver, end-stage renal disease, hypothermia, and HIV/AIDS, among other issues; as a result, they visit emergency rooms and are hospitalized often (Willse 2015). Though vulnerability as an organizing principle has not completely edged out the long-held view of poor and homeless people as responsible for their own circumstances, it was embraced by the US government under the George W. Bush administration when it adopted Housing First for chronic homeless initiatives.

Contrasting with advocates’ and social programs’ calls for housing as a “right,” the US government began to support using a Housing First approach for chronic homeless populations principally as a cost-saving measure (Willse 2015). The federal government largely depoliticized Housing First by deemphasizing both the claims that homeless people have a right to housing and the importance of client input and consumer choice in housing. Foscarinis points out that by the 1990s, advocates for the homeless had little hope that a constitutional right to housing would be recognized at the federal level: “[A]s a practical and political matter, prospects for establishing a constitutional right to shelter or housing were rightly seen as extremely limited, and such litigation was not actively pursued” (2008, 116). Thus, homeless advocates accepted that promoting Housing First initiatives in terms of their potential to save the state money would be politically more viable than an assertion of the right to housing.

In 2003 the Department of Housing and Urban Development (HUD) called for an end to chronic homelessness in ten years and a drop by 50% in five years (United States Department of Housing and Urban Development 2002). HUD matched its behest with funding increases for programs providing permanent housing and supportive services for mentally ill and substance-dependent, long-term homeless people (Moulton 2013). The US government pledged $35 million to fund permanent housing and supportive services for chronically homeless people (Greenwood, Stefancic, and Tsemberis 2013). Concerned over the cost to cities and states of managing the service needs of this group, HUD and others cited data suggesting that chronically homeless people created a significant financial burden for state and local governments as a result of their heavy use of shelters, jails, detoxification facilities, hospitals, and other services (Moulton 2013). Studies showed, for example, that chronically homeless people “represented a relatively small proportion of the homeless (approximately 10 percent), but utilized approximately half of shelter services as measured in days” (Spence-Almaguer, Petrovich, and Paige 2008, 3; Kuhn and Culhane 1998). The average cost per year in services paid by local, state, and federal governments for each chronically homeless person was estimated at $44,733 in 2006 dollars (Moulton 2013, 601).

Analysis of Housing First

A number of studies have found that Housing First for chronically homeless people can be successful in maintaining housing and saving the state money. Early studies done by Pathways to Housing in New York City indicated better long-term housing retention for chronically homeless people using a Housing First rather than a linear model (Tsemberis, Gulcur, and Nakae 2004; Tsemberis and Eisenberg 2000) and lower rates of substance use (Padgett, et al., 2011). Evidence from San Mateo, California, showed that both criminal justice and medical costs decreased considerably for sixteen chronically homeless people assisted by a Housing First program (Greenberg et al. 2013; see also Martinez and Burt 2006). In another study, 111 chronically homeless individuals with “severe” alcohol addiction were housed in a single-site setting in Seattle. “[O]nly 23% (26 of 111) of participants returned to homelessness during the 2-year period, and afterward, 24% (7 of 26) of these individuals returned to the same housing project” (Collins, Malone, and Clifasefi 2013, S272). Some of the participants indicated that they would be willing to accept abstinence-based housing, but did not believe they would be successful in remaining housed in such a program (Collins, Malone, and Clifasefi 2013, S272). Groton reviewed five studies that explored results of Housing First interventions across the United States; all showed better results for housing retention when a Housing First method was used as opposed to the linear model (Groton 2013). Neither the linear nor Housing First approaches, however, demonstrated a significant decrease in substance use or mental illness among participants (Groton 2013).

Despite some promising data on the outcomes of Housing First approaches, a serious problem with Housing First policies is their reliance on access to low income housing to operate. The low income housing stock in the United States is shrinking. Without access to low cost housing, Housing First programs cannot place formerly homeless people in apartments or houses, the first and crucial step in the Housing First approach. Although the federal government has voiced support for Housing First interventions, it has not provided sufficient funding for housing vouchers or succeeded in expanding the low income housing supply. As a result of its tepid attempts to address the lack of low cost housing, some academics have suggested that the government’s defense of Housing First “has been largely rhetorical in practice” (Foscarinis 2008, 123). Without a real and sustained financial investment in low cost housing, then, the government champions Housing First more in the abstract than in practice.

Although Housing First represents a departure from past homeless policy requiring people to prove they are “housing ready” or “deserving” of help before they get assistance, not all portions of the homeless population share in the advantages presented by the new programs. In essence, with its emphasis on chronically homeless individuals, Housing First may be siphoning both funding and the focus of policymakers away from families and other homeless subpopulations:

In recent years, homeless families with children have not received the same federal attention and level of fiscal support as other homeless subgroups. Specifically, the federal government and local agencies have responded to the unique needs of chronically homeless persons with an influx of targeted federal funding…. Even as their numbers increase and more are taking longer to transition to self-sufficiency, homeless families remain largely invisible to the public and absent from the national policy debate.

(ICPH 2011, 1)

Thus the attention to the economic costs associated with allowing chronically homeless people to remain on the street has enabled Housing First advocates to gain funding for programs, obtain housing vouchers, and the like, but specifically for chronically homeless individuals. Families, who are relatively invisible to the public as compared to chronically homeless people and cost the state little comparatively, have not gained the same traction with policymakers (Foscarinis 2008). Data suggest that while chronic homelessness has decreased in the past ten years, there has been an increase in family homelessness (Weinreb, Rog, and Henderson 2010). Low income housing shortages particularly affect single-parent families, in which one income must stretch to support multiple people; single-parent families are more likely to experience “severe” problems affording housing (Harvard Joint Center for Housing Studies 2008).

Analyses of Housing First programs and demonstration projects generally do not address how the choice to assist the chronically homeless population is made at the expense of other subpopulations among the homeless or those in poverty who are precariously housed, but funding, personnel, and most important housing vouchers are a limited resource. This is clearly demonstrated with a program in California designed to assist the fifty most vulnerable chronically homeless in one California county. County social service workers received fifty vouchers from the housing authority to pursue the program; these are vouchers that were then no longer available to other low income or homeless people and families. Judgments regarding “vulnerability” were based on the number of visits to the emergency room, presence of cirrhosis of the liver, advanced age, and consumption of police services, including how many civil penalties were owed for infractions like sleeping on the sidewalk and other laws that criminalize the functions of living while homeless (Cornejo 2014). The program served chronically homeless, single individuals but not families. The method used by this program to identify people eligible for a Housing First intervention has been repeated in many other cities and counties as the norm for operating Housing First programs (Padgett, Henwood, and Tsemberis 2016). Because the concept of vulnerability in this case is tied to the costs incurred by the state for healthcare and criminal justice systems, cities and counties often select the fifty or one hundred people who use the most services and present the clearest financial incentive for the state to remove them from homelessness.

Research on successful exit from homelessness has followed funding priorities; much of the academic research on solutions to homelessness published in the past ten years has monitored chronically homeless individuals. Thus, while the research exploring chronically homeless people’s ability to exit homelessness, especially in the context of a Housing First program, is plentiful, there is less research exploring homeless families’ exits from homelessness (Bassuk et al. 2014). Research suggests that homeless families who exit a shelter without a permanent housing option “frequently double up with others, or move from place to place, with stays in regular housing sometimes interspersed with returns to shelter” (Fisher et al. 2014, 367). Other research demonstrates that homeless families with alcohol or drug issues mirror chronically homeless people to some degree and likewise would benefit from a Housing First approach. For example, a study of homeless families in Massachusetts found that those testing positive for alcohol or drugs stayed in shelters longer than those who did not (Weinreb, Rog, and Henderson 2010). The same study, however, noted that “for families, personal circumstances may have less bearing on the timing of shelter exits than policy and program factors” (598). Specifically, some families remained in the shelter in order to obtain a housing subsidy, though the wait was long and resulted in further instability for families who had to live in a congregate setting, not to mention the high financial cost of shelter stays relative to housing subsidies.

Another issue that may affect the success of Housing First is the impact of chronically homeless people on other residents in housing approximate to program units. One study compared chronically homeless people to long-term homeless people who did not meet the chronic definition, but had stayed in shelters at least 730 nights over a four-year period. Both were tenants in a program in which they received housing and supportive services. Property managers associated with the program assessed chronically homeless people as needing more time and effort to adjust to living in an apartment, as less hygienic than the long-term shelter residents, and as requiring more support from case managers to make the transition to permanent housing (Levitt et al. 2012). In this case, property managers were all associated with the program, so they were trained to work with chronically homeless people and willing to provide the extra care and attention required by residents. Circumstances may differ, however, in cases where property managers are private employees who are not apt to want to expend more energy on homeless clients; evidence suggests that low income tenants who are perceived to cause trouble may be evicted from low cost housing (Desmond 2016). Since Housing First agencies prefer scatter-site housing(placing clients in apartments or houses “scattered” among numerous complexes) as an alternative to housing all clients together in one apartment building, it is likely that property managers would not be agency employees, at least in some cases. Given chronically homeless people’s struggle to transition to behavioral norms associated with long-term housing, the potential for complaints from neighbors and problems with landlords or property managers presents a hurdle for the Housing First model, though the question has not been addressed in a sustained manner in the literature.

Although most studies suggest substantial savings associated with housing chronically homeless people as compared to their costs to the state when not housed, some researchers have argued that these claims are overgeneralized (Kertesz et al. 2009). Gilmer et al. found that clients with serious mental illness who were enrolled in a program utilizing a Housing First philosophy increased their use of outpatient mental health visits in comparison to a control group of similarly mentally ill people who did not have access to supportive housing (Gilmer et al. 2010). The large-scale study evaluated over twenty thousand clients and thus compared costs associated with a larger and broader group of homeless people than many of the smaller scale studies of individual programs to date. Importantly, annual service costs were $12,056 higher for homeless people in supportive housing than in the control group. These “findings contrast with previous studies that have found that the costs for more intensive services and subsidized housing are mostly or entirely offset by reductions in inpatient, emergency, and justice system costs” (Gilmer et al. 2010, 1123). The authors suggest that when larger programs serve clients beyond the fifty to one hundred most expensive homeless people recruited by some programs, savings to the state diminish. Because many of the twenty thousand homeless people surveyed in the Gilmer study did not have significant numbers of hospitalizations, jail stays, or use of detoxification services, their costs to the state were minimal before they entered the program, and thus monetary savings to the state could not be used as a justification to pay for Housing First programs.


Perhaps the most pressing question regarding the future of both the linear and Housing First interventions is whether and to what degree Housing First will expand from its roots in programs for chronically homeless people to aid other groups. Questions abound about whether to extend the model to homeless people who are not mentally ill or substance users, to families, or to those who are homeless for short periods. Folded into the Housing First definition is the concept of “harm reduction,” so it has been conceptualized with a substance-dependent population in mind (Greenwood, Stefancic, and Tsemberis 2013). Since homeless families exhibit less drug and alcohol use and mental illness than the chronically homeless, and point to low wages, underemployment, and low income housing shortages as primary reasons for homelessness, their principal needs will not be addressed unless Housing First’s supportive services are reoriented to provide assistance for locating employment and child care and improving employment options. It is doubtful, then, that simply applying the model to families or to those who do not have substance abuse or mental illness issues will be fruitful. Given the level of diversity in the homeless population, a one-size-fits-all model does not make sense.

If the rationale for public funding of Housing First programs lies in the argument that the state saves money relative to the linear model or to simply not assisting people, then there will be little political will to assist families or other portions of the population who do not utilize criminal justice and public health systems to a great degree and thus do not produce state expenditures at the same level as the chronically homeless. Families also tend to have relatively short periods of homelessness, so generally do not stay in emergency shelters for long periods of time (Fisher et al. 2014). In this sense, programs that rely on showcasing net reductions in state costs when chronically homeless people are provided permanent housing have employed a limited strategy. In addition, families and others who do not meet the chronic definition represent a much larger group than the people currently targeted for Housing First interventions, because there are many more people who are homeless for short periods of time than are homeless for a year or more (Poulin, Metraux, and Culhane 2008). Studies that track shelter use over one- or two-year periods indicate, for example, that approximately 10% are chronically homeless (Burt 2001). Consequently, there will necessarily be significant expenditures associated with extending the Housing First model to the rest of the homeless population. Chief among these costs would be creating more low cost housing alternatives. Given the dearth of affordable housing now plaguing the system, this would require a wholesale rethinking and reinvestment in the American social safety net. Perhaps because this seems almost inconceivable given the political climate in the United States, the unavailability of housing units has not received sustained attention in the research on the linear and Housing First policies.

In the face of these challenges, however, Housing First has the potential to resolve many of the problems with emergency shelters. Although not yet widespread, there have been attempts on the part of battered women’s shelters in particular to establish Housing First programs that are not designed to serve a chronically homeless population and to address concerns about the social control orientation of emergency shelters. This seems like a promising start, and such organizations may provide a model to create programs for a population with severe housing problems but less intensive social service needs (Fisher et al. 2014). An Oregon program for homeless battered women that incorporates a Housing First model is an illustrative case in point (Williams 2008). In 2003 it closed its domestic violence shelter in order to use the resources previously expended on running the shelter to focus more directly on battered women’s housing needs. It continues to provide emergency help to battered women through the use of motel vouchers; it now aids five times as many women as had been housed in the shelter. Under the prior program, many women were not served because the shelter was full. The Oregon agency now pours the money saved by not operating a shelter into “mobile advocacy,” working to house women and to offer them multiple support services to address simultaneously their histories of domestic violence, economic marginalization, and housing requirements. The program helps some women to get into other shelters and some to become housed immediately (Williams 2008).

Another important barrier to expanding Housing First models to families is the dearth of well-designed, methodologically sound research on families exiting homelessness. Evidence-based studies, similar to the research used to assess Housing First for chronically homeless people, are necessary to explore what kinds of assistance homeless families desire to become stably housed and whether and what type of supportive services are required alongside permanent housing. In a review of six studies of family homelessness, Bassuk points out that homeless families do not fare well in a variety of interventions currently used, including emergency and transitional shelters: “Compared with their homelessness at enrollment, the housing circumstances of families generally improved at exit from these programs; that is, families were no longer literally homeless, but many were not residentially stable. Furthermore, their work status was slightly improved, but most families were not earning a livable wage” (Bassuk et al. 2014, 470). The larger the family, the more complicated it becomes to locate affordable housing. Apartments for families are more expensive and scarcer than studio apartments that an individual could rent (Padgett, Henwood, and Tsemberis 2016), and strict rules instituted by the Department of Housing and Urban Development govern the number of bedrooms in which a family with children must live. But these hurdles do not alone explain the reduced attention to family homelessness. Rather, the tendency to ignore this group relative to their chronic counterparts has much to do with the government’s reliance on cost saving as a motivating rationale for assisting homeless people. It can also be explained by the public’s propensity not to envision families when they consider homelessness, largely because of the stereotypical view of homeless people as “street” people who are dirty, mentally ill, eating out of trash cans, and sleeping on sidewalks. Unlike chronically homeless people, families may not “look” homeless in the context of popular stereotypes, so they become relatively invisible, with less political and public support for solving their homelessness.

Given that a focus on cost savings associated with housing the chronically homeless has proven to be a limited strategy, and in fact has caused homeless families to be ignored by policymakers, perhaps it is time for homeless advocates and Housing First supporters to reemphasize the concept of a right to housing. The contention that all people have a right to housing, and that it is the government’s responsibility to assist people in meeting a minimum standard of living that includes stable housing, is not widely embraced in the United States and calls into question the neoliberal model that has been the foundation for the dominant political discourse on homelessness as well as policy prescriptions in the United States since the 1980s. It is, however, a political argument that would enable homeless advocates to maintain that all portions of the homeless population deserve housing, not just those who are visible or chronically homeless or use state services heavily. Since the severity of income and wealth inequality has captured the attention of both politicians and the public in the United States, the argument that all people deserve a roof over their heads may be politically more acceptable now than it has been for the past one hundred years.


The author wishes to thank the Circle of Giving and the College of Liberal Arts at California Polytechnic State University for awarding grants to support this project. Many thanks also to Caitlyn Morrison, Stacy Okoro, Sydney Tanimoto, and Nestor Veloz-Passalacqua for their assistance with research for this project.


Bassuk, E., C. DeCandia, A. Tsertsvadze, and M. Richard. 2014. “The Effectiveness of Housing Interventions and Housing and Service Interventions on Ending Family Homelessness: A Systematic Review.” American Journal of Orthopsychiatry 84 (5): 457–474.Find this resource:

Burt, M. 2001. Helping America’s Homeless: Emergency Shelter or Affordable Housing? Washington, DC: Urban Institute Press.Find this resource:

Center on Budget and Policy Priorities. 2009. “How Many Families Use Federal Rental Assistance?”

Collins, S. E., D. K. Malone, and S. L. Clifasefi. 2013. “Housing Retention in Single-Site Housing First for Chronically Homeless Individuals with Severe Alcohol Problems.” American Journal of Public Health 103 (2): 269–274.Find this resource:

Cornejo, A. 2014. “New Program Aims to House 50 of SLO County’s Chronically Homeless Residents” San Luis Obispo Tribune, August 26. this resource:

Culhane, D., and S. Metraux. 2008. “Rearranging the Deck Chairs or Reallocating the Lifeboats? Homeless Assistance and Its Alternatives.” Journal of the American Planning Association 74 (1): 111–121.Find this resource:

Desmond, M. 2016. Evicted: Poverty and Profit in the American City. New York: Crown Publishers.Find this resource:

DeWard, S., and A. Moe. 2010. “‘Like a Prison!’: Homeless Women’s Narratives of Surviving Shelter.” Journal of Sociology & Social Welfare 37 (1): 115–135.Find this resource:

Dordick, G. 1997. Something Left to Lose: Personal Relations and Survival among New York’s Homeless. Philadelphia, PA: Temple University Press.Find this resource:

Edin, K., and H. L. Shaefer. 2015. $2.00 a Day: Living on Almost Nothing in America. Boston: Houghton Mifflin Harcourt.Find this resource:

Fisher, B., L. Mayberry, M. Shinn, and J. Khadduri. 2014. “Leaving Homelessness Behind: Housing Decisions among Families Exiting Shelter.” Housing Policy Debate 24 (2): 364–386.Find this resource:

Foscarinis, M. 2008. “The Evolution of Homelessness: Trends and Future Directions.” In Robert Hartmann McNamara (Ed.), Homelessness in America, volume 3 (pp. 113–132). NY: Praeger Publishers.Find this resource:

Gilmer, T. E., A. Stefancic, S. Ettner, W. G. Manning, and S. Tsemberis. 2010. “Effect of Full Service Partnerships on Homelessness, Use and Costs of Mental Health Services, and Quality of Life among Adults with Serious Mental Illness.” Archives of General Psychiatry 67 (6): 645–652.Find this resource:

Greenberg, B., S. Korb, K. Cronon, and R. Anderson. 2013. “Supportive Housing Best Practices in a Mid-sized US Urban Community.” Housing, Care & Support 16 (1): 6–15.Find this resource:

Greenwood, R. M., A. Stefancic, and S. Tsemberis. 2013. “Pathways Housing First for Homeless Persons with Psychiatric Disabilities: Program Innovation, Research, and Advocacy.” Journal of Social Issues 69 (4): 645–663.Find this resource:

Groton, D. 2013. “Are Housing First Programs Effective? A Research Note.” Journal of Sociology & Social Welfare 40 (1): 51–63.Find this resource:

Hartnett, H., and J. Postmus. 2010. “The Function of Shelters for Women: Assistance or Social Control?” Journal of Human Behavior in the Social Environment 20 (2): 289–302.Find this resource:

Harvard Joint Center for Housing Studies. 2008. The State of the Nation’s Housing 2008. Boston: Harvard University.Find this resource:

Harvard Joint Center for Housing Studies. 2013. The State of the Nation’s Housing 2013. Boston: Harvard University.Find this resource:

Homeless Emergency Assistance and Rapid Transition to Housing (HEARTH) Act. 2009. Helping Families Save Their Homes Act of 2009, Division B. Pub. L. No. 111–122.Find this resource:

Iceland, J. 2013. Poverty in America: A Handbook. Berkeley: University of California Press.Find this resource:

Institute for Children, Poverty & Homelessness (ICPH). 2011. “Missing the Mark: Federal Funding Off Target for Homeless Families.”

Kertesz, S. G., K. Crouch, J. B. Milby, R. E. Cusimano, and J. E. Schumacher. 2009. “Housing First for Homeless Persons with Active Addiction: Are We Overreaching?” Milbank Quarterly 87 (2): 495–534.Find this resource:

Kuhn, R., and D. Culhane. 1998. “Applying Cluster Analysis to Test a Typology of Homelessness by Pattern of Shelter Utilization: Results from the Analysis of Administrative Data.” American Journal of Community Psychology 26 (2): 207–232.Find this resource:

Levitt, A. J., J. J. Jost, K. A. Mergel, A. Hannigan, J. DeGenova, and S. Y. Chung. 2012. “Impact of Chronically Street Homeless Tenants in Congregate Supportive Housing.”  American Journal of Orthopsychiatry 82 (3): 413–420.Find this resource:

Martinez, T. E., and M. R. Burt. 2006. “Impact of Permanent Supportive Housing on the Use of Acute Care Health Services by Homeless Adults.” Psychiatric Services 57 (7): 992–999.Find this resource:

Moulton, S. 2013. “Does Increased Funding for Homeless Programs Reduce Chronic Homelessness?” Southern Economic Journal 79 (3): 600–620.Find this resource:

National Center on Family Homelessness. 2011. “The Characteristics and Needs of Families Experiencing Homelessness.”

Padgett, D. K., B. F. Henwood, and S. Tsemberis. 2016. Housing First: Ending Homelessness, Transforming Systems, and Changing Lives. New York: Oxford University Press.Find this resource:

Padgett, D. K., V. Stanhope, B. Hewood, and A. Stefanic. 2011. “Substance Use Outcomes among Homeless Clients with Serious Mental Illness: Comparing Housing First with Treatment First Programs.” Community Mental Health Journal 47 (1): 227–232.Find this resource:

Poulin, S. R., S. Metraux, and D. P. Culhane. 2008. “The History and Future of Homeless Management Information Systems.” In Homelessness in America, ed. Robert Hartmann McNamara, 3:171–179. Westport, CT: Praeger Publishers.Find this resource:

Spence-Almaguer, E., J. Petrovich, and J. Paige. 2008. “Chronic Homelessness: Portraits on the Streets.” In Homelessness in America, ed. Robert Hartmann, 2:1–30. Westport, CT: Praeger Publishers.Find this resource:

Stewart B. McKinney Homeless Assistance Act. 1987. 42 U.S.C. 11301.Find this resource:

Tsai, J., A. S. Mares, and R. A. Rosenheck. 2010. “A Multisite Comparison of Supported Housing for Chronically Homeless Adults: “Housing First” versus “Residential Treatment First.” Psychological Services 7 (4): 219–232.Find this resource:

Tsemberis, S., L. Gulcur, and M. Nakae. 2004. “Housing First, Consumer Choice, and Harm Reduction for Homeless Individuals with a Dual Diagnosis.” American Journal of Public Health 94 (4): 651–656.Find this resource:

Tsemberis, S., and R. F. Eisenberg. 2000. “Pathways to Housing: Supported Housing for Street-Dwelling Homeless Individuals with Psychiatric Disabilities.” Psychiatric Services 51 (4): 487–493.Find this resource:

United States Department of Housing and Urban Development. 2002. Final FY 2003 Annual Performance Plan. Washington, DC: United States Department of Housing and Urban Development.Find this resource:

Weinreb, L., D. Rog, and K. Henderson. 2010. “Exiting Shelter: An Epidemiological Analysis of Barriers and Facilitators for Families.” Social Service Review 84 (4): 597–614.Find this resource:

West, S., D. Patterson, A. Mastronardi, K. Brown, and R. Sturm. 2014. “Individual Predictors of Community Cost before and after Housing First.” Journal of Poverty 18 (3): 231–253.Find this resource:

Williams, J. C. 2016. ““A Roof Over My Head”: Homeless Women and the Shelter Industry.” Second Edition. Boulder: University Press of Colorado.Find this resource:

Williams, J. C. 2008. “Homeless Women: Domestic Violence and Housing Instability.” In Robert Hartmann McNamara (Ed.), Homelessness in America, volume 2 (pp. 57–74). NY: Praeger Publishers.Find this resource:

Willse, C. 2015. The Value of Homelessness: Managing Surplus Life in the United States. Minneapolis: University of Minnesota Press.Find this resource:

Wright, J. D., et al. 1998. Beside the Golden Door: Policy, Politics, and the Homeless. New York: Aldine de Gruyter.Find this resource:


(1) Thanks to the anonymous reviewer and the editor for Oxford Handbooks Online for articulating this point as an important critique of both models.