Abstract

Research indicates that homelessness is dynamic and that a significant number of homeless people have experienced multiple episodes of homelessness. This episodic nature of homelessness implies that the issue for such people is not only getting housing and exiting homelessness but staying housed or when there is a need to move, making a transition to another housing without returning to homelessness. This article describes a multidimensional conceptual model that was developed by synthesizing the theoretical and research literature on homelessness. Essentially, this model identifies four dimensions in society within which multi-layered factors that impact on housing outcomes are located. The multidimensional model underscores the fact that exits and returns to homelessness are determined by a complex interaction of individual and structural factors necessitating a multilevel and integrated response to episodic homelessness.

Introduction

“The Committee is gravely concerned that such a wealthy country as Canada has allowed the problem of homelessness and inadequate housing to grow to such proportions that the mayors of Canada’s top ten largest cities have now declared homelessness a national disaster”.
Homelessness has become a major social and political problem in North America over the past two decades. The extent of this problem in Canada can be seen in the declaration by mayors of Canada’s ten largest cities that homelessness is a national disaster worthy of the type of response traditionally reserved for natural disasters like earthquakes and hurricanes. Homelessness is of particular interest to the social work profession because it is frequently connected to other issues that we struggle with in our professional work. For example, research indicates that housing is a factor in admissions of children to care by family service workers of Children’s Aids Society (Chau et al., 2001; Cohen-Schlanger, et al., 1992). It is not an overstatement that when people have adequate housing that meet their needs, they are more able to work on other issues in their life.

Despite a categorical and taxonomic view of homelessness, Tremblay and Ward (1998) describe homelessness as a continuum of the actual current housing situation of the individual. This continuum, they suggest is from sleeping rough, to day and night drop-ins, to shelters, to rooming housing, to unstable housing, to stable housing. Tremblay and Ward point out that the goal of “any homelessness strategy is to move people from the left side to the right side of the spectrum, from sleeping rough to stable housing” (p.7). The second major task, the authors point out is developing strategies that will ensure people do not lose their housing.

Tremblay and Ward’s point is a salient one, particularly when considered against the backdrop of research findings that indicate that a significant number of homeless people have experienced multiple episodes of homelessness (Piliavin et. al., 1993; Wong & Piliavin, 1997; Sosin, Piliavin & Westerfelt, 1990). That is, they move from the left side of the homelessness continuum to the right side and back to the left side in a pattern that reoccurs frequently. This episodic nature of homelessness is well described by Peressini & McDonald (2000) who point out that “homelessness is not a finite or static process, but a fluid and dynamic one, characterized by multiple transitions, role exits and role entries" (p. 526). Thus, for episodically homeless people, the issue is not only getting housing and exiting homelessness, but staying housed or, when there is a need to move, making a transition to housing without returning to homelessness. The ‘homeless career’ of an episodically homeless person is frequently made up of several exits and returns to homelessness interspersed with periods of housing (Wong and Piliavin, 1997; Sosin, Piliavin and Westerfelt, 1990).

For many people who have experienced several exits and returns to homelessness, contributing causes include not just economics, but vulnerabilities such as physical or mental health disabilities and/or drug/alcohol use, a situation that causes community housing workers to consider them as ‘hard to house’. Anderson (1998) describes ‘hard to house’ people as including people who live on the street (e.g., park benches, ravines, under bridges), those who move from shelter to shelter with periods of time on the street, as well as people who have been barred by shelters or released from institutions (e.g., consumers/survivors of the mental health system). Anderson points out that such ‘hard to house’ people have been severely debilitated by the conditions of homelessness.

It is pertinent to point out that as people move from the right side of the continuum i.e. from stable housing to sleeping rough, they are more likely to move from ‘adequate housing’ to ‘inadequate housing’, from housing to homelessness and from inhabiting ‘private space’ to inhabiting ‘public space’. However, these continua are contentious, and the boundaries are often gray and not clear-cut. Particularly contentious is the issue of whose values and agenda should inform criteria that are used in determining what adequate housing is, when a homeless person can rightly be said to have been ‘housed’ and is therefore no longer homeless, and what is private as opposed to public space.

Despite the controversies and gray boundaries on the above continua, the fact that the majority of homeless people fall into the episodic homeless group has important policy, practice, and research implications that have not been fully capitalized on. As Wright, Rubin and Devine (1998) rightly point out, because episodically homeless persons find themselves acceptably housed from time to time, an important goal of policy should be to extend these periods of housing. This paper describes a multidimensional conceptual framework that identifies factors associated with episodic homelessness. Essentially, this model delineates four dimensions in society within which multi-layered factors that impact on housing outcomes are located. These factors have been synthesized from an extensive review of both the theoretical and research literature on homelessness. The multidimensional nature of this model not only underscores the fact that multiple aspects of disadvantage determines vulnerability to homelessness but also draws our attention to the possibility of using these factors as leverage points in the fight against episodic homelessness.

Episodic Homelessness – What are the Causes?

Although there is a marked absence of theoretical models that provide an explanation of why homeless persons who exit homelessness return to homelessness again, theoretical frameworks that address initial vulnerability to homelessness do offer some useful perspectives because factors associated with the first episode of homelessness may also be associated with subsequent episodes. However, in borrowing these theoretical explanations, it is necessary to frame them within a stance of ‘uncertainty’ because of two potential problems identified by Piliavin, Wright, Mare and Westerfelt (1996). The first problem is these theoretical explanations of initial entry into homelessness have not been adequately tested despite the extensive research that has been conducted on homelessness worldwide. The second reason is the possibility that the factors associated with initial homelessness might not necessarily be related to subsequent homeless spells and exits.

In light of such limitations, the theoretical explanations reviewed below are those that have, in various forms and versions, been part of a vigorous discourse within the homelessness and housing literature. These explanations fall into two large groups – those that explain homelessness based on individual vs. structural deficits and those that explain homelessness based on problematic relational issues, such as disaffiliation or social exclusion.

Homelessness: Individual or Structural Deficits? Theoretical explanations that frame homelessness as an individual deficit see it as arising from the personal circumstances or the ‘fault’ of those who are homeless, as in the case of mental illness and addiction which make such individuals unwilling to work. The high prevalence of psychiatric disorders among the homeless generally reinforce such a position. Marcuse (1987) points out that this approach to homelessness and housing is “specialism” because it assumes that homelessness and other housing problems are “the special problems of particular groups within a generally well-functioning housing system” (p. 233). Marcuse notes that the research focus of this position is on the personal characteristics of the ill-housed, for example, the elderly, the poor, large families, ethnic minorities, single-parent households and women. Critiquing such a misguided position, Marcuse notes that it blames the victim and portrays systemic housing issues as separate and individual problems.

Research studies that are framed by such an approach have concentrated on detailing demographic, and diagnostic categories of homeless persons who have mental illness. This approach clearly has its limitations. Snow, Anderson and Koegal (1994) point out that most of the research and policies on homelessness is framed by “a language of disability” which formulates homelessness as a social phenomenon that is caused by individual deficiencies. There has been criticism that such research has diverted attention from tests of the structural conditions underlying homelessness while few studies have empirically examined the effect of multiple factors on homelessness (Elliot and Krivo, 1991). The exceptions would be an emerging, small collection of studies on homeless exits and returns which go beyond demographic characteristics and examine the structural factors that predict exits and returns to homelessness (Dworsky & Piliavin, 2000; Wong, Culhane, Kuhn, 1997; Wong and Piliavin, 1997).

Bogard et al. (1999) note that the assumption that the personal problems of the homeless caused their homelessness is implicit in the goals and character of “service-intensive” shelters which primarily attempt to ameliorate the psychological problems of the homeless. These individual factors range from mental illness to personality “defects”. Wright, Rubin and Devine (1998) rightly argue that studies and analyses that focus on the individual deficits of homeless persons mistake the characteristics of homeless people for the causes of homelessness. Such studies have been critiqued for mistaking homeless peoples’ need for mental health and substance abuse treatment as the reason they are homeless rather than the real reasons which are usually related to poverty, housing, and other structural conditions.

However, some homelessness researchers disagree that individual deficits are to blame for high homelessness rates arguing instead that structural factors, the primary one being inadequate income, have created conditions that literally destine many people to be homeless. From this perspective, homelessness is viewed as a reflection of the organization and distribution of society’s resources. Common structural factors that are often cited include lack of low-cost housing, high poverty rates, poor economic conditions and lack of community mental health care facilities. While some structural analysts emphasize the housing aspects in creating homelessness, others emphasize the income aspects or the reduction of unskilled, entry level employment. Marcuse calls the approach of analysts who only see the income aspects of homelessness as “economism”. Their approach explains the lack of adequate housing as a simple function of the distribution of income, in short, “if everyone had enough money, the “housing problem” would be solved” (p. 233).

Jahiel (1992) argues that homelessness does not occur in a social vacuum but is a side effect of socially condoned activities of certain individuals in society. Jahiel suggests broadening the homelessness discourse into the areas of housing, income production, healthcare, and family life where people that initiate and control the events that make people homeless are situated. The author dismisses the suggestion that individual explanations are involved in homeless-making processes pointing to numerous research evidence that support structural explanations.

The division between individual and structural explanations has been rightly criticized as overly simplistic (Neale, 1997). While structural explanations do not satisfactorily explain the large numbers of people with mental illness and addictions within the ranks of the homeless that research studies show, individual explanations, on the other hand, ignore the ‘well-developed body of scholarship” that suggests that there is a relationship between economic, societal conditions and homelessness (Koegal, Burnam, Baumohol, 1996). As a number of theorists point out, both perspectives are needed to understand contemporary homelessness. Structural factors determine why pervasive homelessness exists now while individual factors explain who is least able to compete for scarce housing (Koegal, Malamid and Burnham 1995).

Fitzpatrick, Kemp & Klinker (2000) note that although the key factors contributing to homelessness have been identified as adverse housing and labour market trends, cuts in social and security benefits, rising levels of poverty and family restructuring, not everyone who are affected by these factors become homeless. They go on to point out that the selection of who becomes homeless is not a random process but individual problems and circumstances significantly increase people’s risk of becoming homeless. Their views have also been echoed by other homelessness researchers (Burrows, 1998; Smith et al., 1998). This interconnectedness between individual and structural factors in homelessness is well articulated by Koegal, Burnam & Baumohol (1996) who point out that homelessness is shaped as much by structural factors as individual ones. They point out that “in a permissive environment full of cheap flops and undemanding work, even outcasts largely remain housed” p. 26.

Homelessness – Disaffiliation or Social Exclusion? Another explanatory framework for understanding homelessness is the disaffiliation model, which is based on the concept of institutional disaffiliation originally proposed by Bahr and Caplow (1973). The disaffiliation model argues that homelessness is largely due to a process of increasingly loosening of an individual’s ties from mainstream society. Wolch, Dear, & Atkita (1998) point out that events that are immediate precipitators of homelessness (i.e. eviction, discharge from an institution, loss of a job, divorce or domestic violence and removal of welfare support) not only mean a loss of housing but also loosen connections to social ties. Grigsby, Baumann, Gregorich & Roberts-Gray (1990) point out that disaffiliation can have deleterious effects on health and well-being as does the loss of food and shelter. Social isolation is associated with stress-related illness such alcoholism and psychopathology. The homeless are not only without housing but also without many markers of citizenship that flow from connections to formal institutions.

In contrast to the disaffiliation/affiliation explanation of homelessness that conceptualizes a homeless person’s lack of social ties as an individual deficit, thus indirectly holding the homeless responsible for choosing to ‘disaffiliate or not affiliate’, social exclusion, offers a structural explanation of how disadvantaged groups are ‘shut out’ of formal structures and institutions of the economy, society and state. Marsh & Mullins (1998) note that the ‘idea’ of social exclusion has emerged over a short period to assume centre stage in political and popular debates about social disadvantage and trace the beginnings of the discourse on social exclusion in earlier debates in France, Britain and the US on exclusion, poverty and ‘the underclass’.

While these earlier debates in both Britain and France were framed by a structural analysis, the authors point out that the American debate on the underclass took a different form that emphasized the negative qualities of disadvantaged groups that were responsible for its’ predicament. This analysis of the differences between North American vs. European conceptualizations of social exclusion are consistent with other criticisms that a majority of North American research studies on homelessness have attracted for focusing on individual disabilities and relational difficulties of the homeless. Wagner (1993) points out that much of the homelessness research is focused on the disempowerment of the homeless and the poor, their marginality, their illnesses (both mental and physical), their isolation, and their vulnerability.

Although both social exclusion and disaffiliation focus on the relational capacity of individuals and the processes that hinder their accessing institutional resources, disaffiliation examines this notion from the individual deficit perspective while social exclusion takes the structural perspective making it easier to link social exclusion to political, economic, and social rights of citizenship. Conceptualizing homeless people as citizens with all the rights therefore has the potential for an empowering praxis.

Sommerville (1998) noting the relationship between housing and homelessness and social exclusion, points out that because housing is a set of relations (including characteristic networks and patterns of activity), housing processes can be looked at as types of processes that either promote social inclusion, or contribute to social exclusion. Sommerville provides an example of the relationship of social exclusion to housing planning as when there is a failure to design and build housing that is accessible to people with disabilities, therefore ensuring their isolation and dependence on others in basic everyday activities. Another example is ‘nimbyism’ that is used as a strategy by powerful residents to exclude housing for lower-income people from certain areas, thereby segregating and isolating them only to other areas.

A Summary of the Research Literature: There is some research support for the above theoretical explanations of homelessness. In a synthesis of research findings on factors associated with homelessness, Hartman (2000) identified the primary cause of homelessness as a lack of affordable housing due in part to high rental prices and a decline in construction, some cases, abandonment of government subsidized housing. The author also identified unemployment and low wages, mental illness and/or substance abuse and domestic abuse as additional causes of homelessness. However, the higher rates of mental illness and/or addictions among homeless people compared to the general population that Hartman noted and several studies have also reported, have been critiqued primarily because the prevalence rate varies from study to study raising questions about their accuracy. Secondarily, these studies do not clearly delineate whether homelessness leads to a higher prevalence of mental illness, or having a mental illness predisposes people to homelessness (Cohen & Thompson, 1992).

One unequivocal factor that research indicates is important is subsidized housing. Research on exits and returns to homelessness overwhelming show that receipt of subsidized housing is the best predictor of who exits homelessness and stays housed. For example, in a longitudinal study that followed 564 homeless families over a five-year period, Shinn and Weitzman (1998) found that 80 percent of homeless families that were given subsidized apartments were able to remain stable despite their personal problems. The authors point out that if given housing first, homeless families will more likely become stable and remain housed.

In addition, research findings indicate that support services help ‘hard to house’ people exit homelessness. The role of appropriate support services is particularly important for homeless people with mental illness with research findings indicating that the illness may be secondary in predicting the likelihood of their maintaining housing tenure in the community. Support services that appropriately target their needs are also important. Research findings also indicate that the type of housing situations that homeless persons exit to significantly affects the likelihood of their becoming homeless again. Homeless people that exit to independent apartments have the lowest return rate.

Dworsky and Piliavin (2000) speculate that there is something about living in one’s own private residence that reduces the likelihood of experiencing a subsequent homeless spell. In a recent Canadian study by Anucha and Hulchanski (2003), study participants, ‘hard to house’ tenants that lived in two shared housing programs, were quite clear and definite that the constant interpersonal conflicts with their roommates increased their risk of eviction. The findings indicate that a major shortcoming of the shared housing model is the lack of privacy that forced participants to conduct private everyday business in public view. The authors’ conclude that privacy is an important aspect of adequate housing particularly for vulnerable populations like ‘hard to house’ tenants.

A Multidimensional Conceptual Model of Factors Associated with Episodic Homelessness

The review of theoretical and research literature suggests that exits and returns to homelessness are determined by a complex interaction of multiple actors. This paper proposes a multidimensional model, (Figure 1) which synthesizes these factors and their impact on the homeless careers of episodically homeless persons into a framework. Essentially, this model identifies four dimensions within which multi-layered factors are located. The four dimensions are the market, the state, civil society and household or individual characteristics.

The private market refers to private enterprises that are in business to make a profit. The state includes all levels of government and agencies that carry out activities for the government. Civil society is separate from the state and from the market economy and is made up of social, economic and cultural activities that result from non-profit and volunteer arrangements between individuals and groups. Civil society consists of all the organized activities and all the associations and groups that are not directly part of the state and that do not have making profit as their main objectives. The fourth dimension is household or individual characteristics, preferences and resources.

Figure 1: A Multidimensional Model of Episodic Homelessness

The first dimension – the Private Market Sector, includes housing and job market realities. Factors layered within this dimension include vacancy rates, cost of rental units, trends within the economy, availability of skilled and unskilled jobs, etc. When opportunities within the rental market is low, it is much harder for the homeless to access housing and exit a homeless spell. Similarly, when unemployment rate is high, and there are few unskilled jobs, it is harder for the homeless to find a job to earn the income needed for rent. Elliot and Krivo (1991) found that more widely available low rent housing is associated with lower rates of homelessness.

Figure 2: The 1 ^st Dimension – The Private Market Sector

The second dimension – the State, includes systemic factors within the social, political and economic realms. This dimension rightly recognizes the differential impact the policies of countries or even regions have on homelessness. Examples of factors layered within this dimension are social welfare and housing policies (which impact on availability of subsidized housing), amount and eligibility criteria of income maintenance programs, health care system and availability of support programs. When income maintenance programs are generous and supports are in place in the community, people with disabilities (i.e. psychiatric consumers/survivors) are more likely to be able to access and maintain a home. Frequently, in the absence of these supports, homelessness becomes a ‘recurring waystation’ with many of them falling into chronic homelessness. Elliot and Krivo (1991) found that more widely available mental health care services of all types prevent people from becoming homeless, while Shinn and Weitzman (1998) provide evidence on the importance of affordable housing in helping people exit homelessness. Findings from their longitudinal study of 564 homeless families that stretched over a five-year period showed that 80 percent of homeless families that were given subsidized apartments were able to remain stable despite their problems. The authors point out that if given housing first, homeless families will become stable and remain housed.

Figure 3: The 2^nd Dimension – The State

The third dimension – Civil Society, includes the social economy, non-profit sector, non-governmental organizations, and social service agencies. Layered within this dimension are factors such as community participation, resources in the community, social inclusion, social capital, etc. The stigma that certain housing programs face when located within communities that have a high degree of NIMBY (Not In My Backyard) syndrome makes it difficult for ‘hard to house’ people to feel connected and have a sense of belonging in their communities. Shuldiner (2000) recommends a broader change from building housing for low-income people (or other special interest housing) to building communities that include low-income people. The author warns that: “What starts out separated is all too soon isolated” (P. 3).


Figure 4: The 3^rd Dimension – Civil Society

The fourth dimension – household or individual characteristics, preferences and resources includes factors such as socio-economic status, disabilities, ethnicity, race, migrant status, age, etc.
Figure 5: The 4^th Dimension – Household or Individual Characteristics
The interaction of these multi-layered factors located within the four dimensions could impact on the homeless careers of episodically homeless people in three possible ways. First, these factors could interact favourably causing an ideal housing outcome where a homeless person is able to access housing, exit homelessness, keep that housing and not return to homelessness again. An example of such a situation is the case of a person with severe mental illness who has the right supports in a city that has adequate and affordable housing for people with mental illness. Such a person will very likely get and keep housing because of having appropriate housing and the right supports.

Secondly, these factors could interact less favourably causing a less than ideal housing outcome where a homeless person does access housing and exits homelessness but quickly loses the housing and returns to homelessness again. An example of such a scenario can be seen in cities that have programs that are heavily weighed towards helping homeless people find housing but have little or no eviction prevention programs such as rent banks, mediation services between landlords and tenants (or tenants and tenants in the case of shared housing programs) and payment plans for rent arrears. Research studies indicate that such prevention programs are effective in preventing homelessness. Lindblom (1991) points out that helping homeless people without preventing new entries or reentries is like bailing a boat without fixing the leaks – it might stop things from becoming worse but the problem will not be solved.

Thirdly, these factors could interact unfavourably causing dire housing outcomes where a homeless person falls into chronic homelessness and doesn’t have the opportunity to access housing and exit homelessness. Such a situation can be seen when people with ‘biographic vulnerabilities’ (Snow, Anderson and Koegal, 1994), such as mental illness and/or addictions, little formal education or employability skills, live in a city with little or no subsidized housing, little spending on mental health services, low vacancy rates within the private rental market, high rents compared to welfare rates and very low demand for unskilled labour. In this type of situation, the individual and structural factors located within the four dimensions interweave and converge in such a way that episodically homeless people experience extreme disadvantage in the housing market and spiral into chronic homelessness.

The multidimensional nature of this model underscores the fact that exits and returns to homelessness are not determined by factors within one sector of society or dimension but is frequently a complex interaction of factors within the four dimensions identified. The model draws our attention to possible leverage points in preventing the reoccurrence of homelessness.

Implications of the Multidimensional Conceptual Model for Policy and Practice

The multidimensional model clearly indicates the need for a multi-dimensional approach to the provision of housing and related supports, rejecting arguments that only more housing is needed (Shinn and Weitzman, 1998) or that just more supports and programs are needed (Baum and Burnes, 1993). Underscoring the need for such holistic and integrated approach, Jahiel (1992) correctly points out: “Preventing homelessness means three things: keeping it from occurring in the first place; providing homeless people with stable homes and incomes, along with needed services, and keeping homelessness from recurring” p.315. Such a responsive and preventative approach will need policies and practices at both micro and macro levels.

Social work has traditionally had difficulties integrating micro and macro interventions, as many of its theories have been located in dichotomizations of individual and society as noted by Rossiter (1996) who points out: “we inevitably end up with social work theories that either focus on the individual, and hold the oppressed responsible for their victimization, or focus on the social, giving social work practitioners no space to think about complex, agentic people”. Also reinforcing the limitations of dividing micro and macro practice, Ife (1997) argues that the separation of social work into macro – micro practice “must be seen as reflecting a political reality, or an ideological imperative, which is about separating the personal and the political, rather than reflecting the reality of good practice” (p.197). The multidimensional model implies that effective policies and practices on episodic homelessness will not be either ’people changing’ or ‘system changing’ but a combination of both.

The cornerstone of effective macro policies and practices that can prevent recurrent homelessness is the development of more affordable self-contained housing units anchored on an understanding that adequate housing is necessary for full participation of people in the community. There needs to be a wholehearted condemnation from social workers and other allied service providers that “warehousing” ‘hard to house’ people or just “maintaining” them in a shelter is not acceptable practice for a civilized society. In a study by Anucha & Hulchanski (2003) in Toronto with participants who had experienced several episodes of homelessness, a participant eloquently articulated the importance of affordable housing and the political context of the homelessness crisis:

“We just need more money, more affordable housing, bottom line, everybody says this. I mean, they can come up with money for the Olympic bid, which we lost, so all that money was wasted. So that money could have been put towards housing. If they want something politicians find the money. So why can’t they come up with money for affordable housing? North America is the richest in the world, this (referring to homelessness) should have never happened. How did it get like this? Politics. We aren’t able to discuss mismanagement, politicians not doing their job properly, not caring” (p. 39).

The building of more subsidized housing units must be integrated with the creation of more job opportunities, increased income supports and large-scale efforts to improve health, education and employability.

Micro level interventions that are needed include eviction prevention programs that work with both landlord and tenants in addressing the issues that threaten housing stability. Housing Again (2002), an internet bulletin published by community based groups in Toronto that are engaged in advocacy for homeless and under housed people, point out that Canada is a signatory to the United Nations Covenant on Economic and Socio-Cultural Rights of which General Comment 7 of the agreement says that signatories are to ensure that forced evictions must not result in homelessness. Canada, the advocates argue, has an obligation to ensure that any party that conducts evictions consults with the people who are affected and ensure that a re-settlement plan is in place. Policies are needed that implement such re-settlement plans as an essential part of any eviction process particularly for no-profit landlords.

People who are homeless should also have access to programs and services that support them in addressing the issues that contributed to their past tenuous housing histories. A participant in the study by Anucha & Hulchanski (2003) pointed this out when asked what would improve the housing stability of people who have experienced several episodes of homelessness:

“Supports, community health supports for drug addiction, counseling, case management, training …on the basis of this housing program it would have to be case management, check and balance with the staff, like when you have staff who are not used to doing that it creates animosity …People need housing and support. They need to deal with the root issues of things. Providing (only) social housing is just providing a band-aid” (p. 41).

Coordinated discharge planning for people leaving institutions such as jails and hospitals is also essential to prevent recurrent homelessness. Many ‘hard to house’ leave jails and hospitals and end up on the streets because of inadequate discharge planning. A participant in the study by Anucha & Hulchanski (2003) offered this advice based on personal experience with addiction programmes:

“Change the program [the rehab program] to help you get housing, it is very important to everybody. Nobody wants to walk out of here and sleep on a park bench, which I have never done in my life, thank God and I don’t plan on starting now. A lot of these people do and if they don’t have housing they will leave feeling really great, sleep on the park bench, meet their old friends they used to party with before and be right back where they started” (p. 65).

As a matter of urgency, social workers, who are usually the professionals that bear the burden of coordinating discharge planning in many institutional settings, must recognize that discharging a client to NFA (no fixed address) is unethical and not good social work practice.

Social workers and other human services professionals need to cease defining their roles exclusively in terms of treatment while looking at housing as a social welfare problem of housing agencies. Clinical roles and models need to be expanded to encompass a recognition and understanding that adequate housing is a major determinant of health. As Prilleltensky, Rossiter, and Walsh-Bowers (1996) rightly point out: “failing to oppose or change oppressive conditions that ruin the mental health of our clients is a moral choice, one that supports the societal status” (p.294). The authors argue that challenging or supporting exploitative social structures that are deleterious to the mental health of clients, is a moral choice that is comparable to ethical standards like client confidentiality and informed consent. The authors insist that not advocating for social and organizational change is to ignore powerful social forces that impinge on the therapeutic relationship.

Conclusion

The conceptual model described in this article clearly indicates the need for a multi-dimensional approach to the provision of housing and supports for people who have experienced several episodes of homelessness. The multidimensional nature of this model underscores the fact that exits and returns to homelessness are not determined by factors within one sector of society or dimension, but is frequently a complex interaction of factors within the four dimensions identified. Lindblom (1991) points out that helping homeless people without preventing new entries or reentries is like bailing a boat without fixing the leaks – it might stop things from becoming worse but the problem will not be solved. The current lop-sided efforts that are weighted heavily towards alleviating literal homelessness have some similarities towards political actions that are only a fight against injustice (limiting hell) but not a contest for justice (attaining heaven). As Simon (1994) elaborates:

“It makes a difference whether we describe our political actions as part of a fight against injustice (limiting hell), against other people’s suffering, or as a contest for justice (attaining heaven). The two labels do not constitute different ways of talking about the same thing. Justice and injustice frameworks create different forms of politics. Justice beckons us to create the positive in the future whereas injustice frantically yells at us to eradicate the negative in the present” (p.1).

References

Anderson, R. (1997). Street as metaphor in housing for the homeless. Journal
of Social Distress and the Homeless, 6(1), 1- 12

Anucha, U. & Hulchanski, J. D. (2003). Where Do They Come From? Why Do They Leave? Where Do They Go? A Study of Tenant Exits from Housing for Homeless People. Ottawa: Canada Mortgage and Housing Corporation.

Bahr, H. and Caplow, T. (1973). Old men: Drunk and Sober (pp. 53 – 64). New York: New York University Press.

Baum, A. & Burnes, D. (1993). A Nation in Denial: The Truth about Homelessness. Westview Press.

Bogard, C. J., Gerstel, N., and Schwartz, M. (1999). Homeless mothers and depression: Misdirected Policy. Journal of Health and Social Behavior, 49, 46-62.

Breakey, W. R. & Fischer, P. J. (1995). Mental illness and the continuum of
residential stability. Social Psychiatry & Psychiatric Epidemiology, 30(4), 147-151.

Chau, S., Fitzpatrick, A., Hulchanski, J.D., Leslie, B. & Schatia, D. (2001). One in five…housing as a factor in the admission of children to care. Centre for Urban Studies Research Bulletin, 3.

Cohen, C .I. and Thompson, K.S. (1992). Homeless mentally ill or mentally ill homeless? American Journal of Psychiatry, 149, 816-822.

Cohen-Schlanger, M., Fitzpatrick, A. Hulchanski, J.D. & Rapheal, D. (1995). Housing as a factor in admissions of children to temporary care: A survey. Child Welfare, 74(3).

Dworsky, A. L. & Piliavin, I. (2000). Homeless spell exits and returns: Substantive and methodological elaboration on recent studies. The Social Service Review, 74, 193 – 213.

Elliot, M. and Krivo, L. (1991). Structural Determinants of Homelessness in the United States. Social Problems, 38(1), 113 – 131.

Grigsby, C., Baumann, D., Gregorich, S. and Roberts-Gray, C. (1990). Disaffiliation to entrenchment: A model for understanding homelessness. Journal of Social Issues, 46, 141 - 156.

Hartman, D.W. (2000). “Policy implications from the study of the Homeless”. A Journal of Clinical and Applied Sociology, Vol. 2 (2): 57-76.

HOUSING AGAIN (2002). Bulletin Number 55, http://www.housingagain.web.net .

Ife, J. (1997). Rethinking Social Work: Towards Critical Practice. Australia: Addison Welsley Longman Australia Pry Limited.

Jahiel, R. (1992). (Ed) Homelessness: A Prevention-Oriented Approach. Baltimore & London: The John Hopkins University Press.

Koegel, P., Burnam, M.A., and Baumohl, J. (1996). “The Causes of Homelessness” in Homelessness in America, ed. J. Baumohl. Phoenix: Oryx Press 24-33

Koegel, P., Melamid, E., & Burnam, M.A. (1995). Childhood risk factors for homelessness among homeless adults. American Journal of Public Health, 85, 1649-1649.

Lindblom, E. N. (1991). ‘Toward a comprehensive homelessness – prevention strategy’. Housing Policy Debate, 2 (3) pp. 957 – 1025.

Marcuse, P. (1987). The Other Side of Housing: Oppression and Liberation. In B. Turner(Ed.), Between State and Market: Housing in the Post-Industrial Era, pp. 232-270.

Marsh, A. and Mullins, D. (1998). The social exclusion perspective and housing studies: Origins, applications and limitations. Housing Studies, 13, 749.

Neale, J. (1997). ‘Theorising homeless: contemporary sociological and feminist perspectives’, in R. Burrows, N. Pleace and D. Quilgars (eds.) Homelessness and Social Policy. London: Routledge. Pp. 35 – 49.

Peressini, T. and McDonald, L. (2000). Urban homelessness in Canada. In T. Bunting and P. Filion (Eds.) Canadian Cities in Transition. Oxford University Press.

Piliavin, I., Sosin., Westerfelt, A. & Matsueda. (1993). The duration of Homeless Careers: An Exploratory Study. Social Service Review. 67: 576 – 98.

Piliavin, I., Wright B., Mare, R. & Westerfelt, A. (1996). Exits from and returns to homelessness. The Social Service Review, 70, 33 – 7.

Prilleltensky, I., Rossiter, A., & Walsh-Bowers, R. (1996). Preventing harm and promoting ethical discourse in the helping professions: conceptual, research, analytical, and action frameworks. Ethics and Behavior, 6(4), 307-320.

Rossiter, A, B. (1996). A perspective on critical social work. Journal of Progressive Human Services, 7(2), 23-40.

Shinn, M. and Weitzman, B. (1998). Housing needs of homeless families. American Journal of Public Health.

Shuldiner, J. (2000). Public housing at the crossroads. Journal of Housing and
Community Development, 57 (4), 19 – 28.

Simon, T. W. (1994). Democracy and Social Injustice: Law, Politics and Philosophy. Rowman and Little Field Publishers, Inc.

Snow, D. A., Anderson, L., and Koegal, P. (1994). Distorting tendencies in research on homeless. American Behavioral Scientist. 37:461-475.

Sosin, M., Piliavin, I., Westerfelt, A. (1990). Towards a longitudinal analysis of homelessness. Journal of Social Issues, 46, 157 – 174.

Sommerville, P. (1998). Empowerment through Residence, Housing Studies, 13, (2) 233 – 257.

Tremblay, J. & Ward, J. (1998). Homeless Voices: Follow-Up to the Homeless, not Helpless Report. Toronto: Toronto Health City Office.

Wagner, D. (1993). Checkerboard square: Culture and resistance in a homeless American. Boulder, Colorado: Westview Press.

Wolch, J., Dear, M., & Atkita, A. (1998). Explaining Homelessness. Journal of the American Planning Association, 54, 443 – 453.

Wong, Y., Culhane, D. P. & Kuhn, R. (1997). Predictors of exit and reentry among family shelter users in New York City. Social Service Review, 442 - 461.

Wong, Y. & Piliavin, I. (1997). A dynamic analysis of homeless-domiciled-transitions. Social Problems, 44, 408 – 24.

Wright, James D., Rubin, B.A. & Devine, J. A. (1998). Beside the Golden Door: Policy, Politics and the Homeless. NY: Aldine de Gruyter.

Copyright 2005 Critical Social Work