Exploring gender and homelessness

Conceptualisation of lone adult homelessness has changed over time (Pleace, Reference Pleace2016). Earlier explanations that emphasised behavioural drivers, homelessness as a ‘deviant’ choice, were replaced by the idea of a ‘deserving’ homeless population, who were mentally ill (Zlornick et al., Reference Zlornick, Robertson and Lahiff1999). Cuts to public welfare and health programmes, falls in social and affordable housing supply, alongside shifts in labour markets towards more relatively low paid, part-time work, were also seen, by some, as creating conditions that triggered increasing levels of homelessness (Marcuse, Reference Marcuse1988). Debates about the nature of lone adult homelessness have been described as ‘sin-talk’ (the individual pathology of homelessness as deviance), ‘sick-talk’ (homelessness triggered by illness) and ‘system-talk’ (homelessness triggered by changes in welfare systems and economies) (Gowan, Reference Gowan2010). A ‘new orthodoxy’ had arisen by 1990 arguing homelessness was a result of the interplay between individual and structural variables. The argument was that retrenchment from progressive social and housing policy, linked to economic and social change, created a higher overall risk of homelessness for particular groups of people, whose decisions, characteristics, experiences and needs could be the trigger factors for homelessness and also cause repeated and sustained homelessness (Vincent et al., Reference Vincent, Deacon and Walker1995; McNaughton-Nicolls, Reference McNaughton-Nicholls2009; Lee et al., Reference Lee, Tyler and Wright2010; Pleace, Reference Pleace2016).

These theories have been partially modified over time. There is growing evidence that differences in welfare systems may be associated with variations in the nature and extent of homelessness. There appears to be more homelessness associated with extremes of poverty in countries with less extensive social protection (Kuhn and Culhane, Reference Kuhn and Culhane1998; Metraux et al., Reference Metraux, Culhane, Raphael, White, Pearson, Hirsch, Ferrell, Rice, Ritter and Cleghorn2001; Aubry et al., Reference Aubry, Klodawsky and Coulombe2012; Chamberlain and Johnson, Reference Chamberlain and Johnson2013; Metraux et al., Reference Metraux, Manjelievskaia, Treglia, Hoffman, Culhane and Ku2016), alongside evidence of homelessness tending to be experienced by small groups with high and complex needs, who appear to have fallen through the more extensive safety nets of highly developed welfare systems (Benjaminsen and Andrade, Reference Benjaminsen and Andrade2015; Benjaminsen, Reference Benjaminsen2016; Pleace et al., Reference Pleace, Knutagård, Culhane, Granfelt, Nichols and Doberstein2016; Dyb, Reference Dyb2017).

These ideas and arguments about the nature of homelessness are all associated with an evidence base that reports that lone adult homelessness is a highly gendered phenomenon throughout the economically developed world, an extreme of poverty and socioeconomic marginalisation that is experienced mainly by lone, adult men. Women are present in these homeless populations, but they are never a majority, nor do they ever really get close to equalling the numbers of men (Anderson et al., Reference Anderson, Kemp and Quilgars1993; Jones and Pleace, Reference Jones and Pleace2010; Busch-Geertsema et al., Reference Busch-Geertsema, Edgar, O’Sullivan and Pleace2010).

The standard explanation has been that welfare and social work systems tend to protect women from potential homelessness at a higher rate than men, particularly when a woman has dependent children with her (Baptista, Reference Baptista, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010). There is also obfuscation stemming from distinct systems dealing with domestic abuse and violence, refuges and related services, which record activity around helping the overwhelmingly (although not exclusively) female population who experience domestic abuse and who are made homeless by that abuse, but who are often recorded in administrative systems as victims of abuse, not as homeless people (Busch-Geertsema et al., Reference Busch-Geertsema, Benjaminsen, Filipovič Hrast and Pleace2014). Just as lone adult homelessness is supposedly a gendered experience in that most of the people who experience it are men, family homelessness, across the UK, Europe and North America, appears equally gendered, the bulk of homeless families are lone women parents (Toohey et al., Reference Toohey, Shinn and Weitzman2004; Baptista et al., Reference Baptista, Benjaminsen, Busch-Geertsema and Pleace2017).

Working from this assumption, it is logical to argue that study of lone adult women’s homelessness, while it is still a visible social phenomenon, can never be more than the study of a small subset of the homeless population. This is because many women, who become, or who are at risk of homelessness, get assisted by domestic violence and social/family services, and in the case of the UK, are also protected by family-orientated statutory homelessness systems (Fitzpatrick and Pleace, Reference Fitzpatrick and Pleace2012).

Feminist scholars have argued that women’s homelessness is interpreted and responded to within the culturally expected ‘norms’ of women fulfilling specific roles, as partner/wife, mother or carer (Watson with Austerberry, Reference Watson and Austerberry1986; Watson, Reference Watson2000; see also Edgar and Doherty, Reference Edgar and Doherty2001). Within a patriarchy, women are not expected to exist outside the father/male partner controlled ‘home’, which means lone female homelessness is, from a patriarchal perspective, seen as an extreme and rare event (Löfstrand and Thörn, Reference Löfstrand and Thörn2004; Baptista, Reference Baptista, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010; Hansen Löfstrand and Quilgars, Reference Hansen Löfstrand, Quilgars, Mayock and Bretherton2016). These arguments centre on the idea that society does not expect women to be homeless, because women are viewed, essentially, as being the core of family structures, the core of what society understands as ‘home’. The idea that lone women may be experiencing homelessness at any sort of scale collides with dominant social and cultural paradigms about who women are, or how they live their lives.

While it is possible to argue about the relative importance of these different potential influences on the extent of social scientific research on women’s homelessness, it is clearly the case that the experience of homeless women has received only sporadic attention. There has been research focusing on women, but the bulk of the evidence base on lone adult homelessness has been characterised by recording what proportion of a studied population were women and noting any broad variations between genders, rather than by detailed examination of women’s experiences (Baptista, Reference Baptista, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010; Mayock and Bretherton, Reference Mayock and Bretherton2016).

Explanations of homelessness that effectively disempower homeless people, i.e. they are homeless because of ‘sickness’ or ‘sin’ or ‘systems’ (Gowan, Reference Gowan2010), have been subject to sustained criticism (Somerville, Reference Somerville1992; Neale, Reference Neale1997; Somerville, Reference Somerville2013; Pleace, Reference Pleace2016), a key aspect of which is the lack of consideration of the impact of the agency of homeless people (McNaughton-Nicolls, Reference McNaughton-Nicholls2009). Likewise, earlier explanations of women’s homelessness that centred on patriarchy have been criticised as both depriving women of agency, i.e. they are portrayed as being swept into homelessness in ways they cannot control by patriarchy, and as failing to take account of other social and economic factors (Drake, Reference Drake1987; Neale, Reference Neale1997; Bretherton, Reference Bretherton2017).

Some researchers are now arguing that there is evidence of women choosing particular routes through homelessness at higher rates than men. There is older evidence of homeless lone women parents ‘exhausting’ informal options before seeking formal assistance in the USA and UK (Shinn et al., Reference Shinn, Weitzman, Stojanovic, Knickman, Jimenez, Duchon, James and Krantz1998; Pleace et al., Reference Pleace, Fitzpatrick, Johnsen, Quilgars and Sanderson2008) and also some more recent studies indicating that women are more likely to seek informal solutions, relying on friends, family and acquaintances. Further, when women do approach services, it is quite often only at the point at which they have exhausted these other, informal, options (Bretherton, Reference Bretherton2017). In essence, women may be experiencing ‘hidden’ homelessness at higher rates than men, sleeping on floors, on sofas and in spare rooms, with no legal rights to the space they are occupying, possibly limited control over that space and limited privacy, and while not on the streets or in a shelter, definitely without a home (Baptista, Reference Baptista, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010; Bowpitt et al., Reference Bowpitt, Dwyer, Sundin and Weinstein2011; Mayock and Sheridan, Reference Mayock and Sheridan2012; Mayock et al., Reference Mayock, Sheridan and Parker2015; Bretherton, Reference Bretherton2017; Johnson et al., Reference Johnson, Robar and Zhu2017).

Defining ‘hidden’ homelessness in exact terms is difficult. If the focus is on the physical situation in which someone is living, defining who should be seen as ‘legitimately’ homeless can become mired in political, ideological and cultural arguments about what constitutes actual homelessness (Phillips, Reference Phillips2000). However, the work around ETHOS (the European typology of homelessness) and the MPHASIS project on enumerating homelessness has been useful (Busch-Geertsema, Reference Busch-Geertsema, O’Sullivan, Busch-Geertsema, Pleace and Quilgars2010; Pleace and Bretherton, Reference Pleace and Bretherton2013; Busch-Geertsema et al., Reference Busch-Geertsema, Benjaminsen, Filipovič Hrast and Pleace2014). Under these definitions, homelessness is the absence of any legal rights and secure personal space, i.e. homelessness includes situations where a woman is sleeping on someone’s floor, on a sofa, or in a spare room, without anything that defines that space as her own. Homelessness is the absence of the legal, physical and cultural constructions that are used to create a home (Busch-Geertsema, Reference Busch-Geertsema, O’Sullivan, Busch-Geertsema, Pleace and Quilgars2010). There is a consensus across many European and OECD countries that homelessness is the absence of a recognisable ‘home’, rather than just the absence of a roof (Busch-Geertsema et al., Reference Busch-Geertsema, Benjaminsen, Filipovič Hrast and Pleace2014).

Women experiencing lone adult homelessness also appear to have different characteristics from men. Domestic violence and abuse are causes of female homelessness in a way that does not apply to lone men to the same extent (Jones, Reference Jones1999; Reeve et al., Reference Reeve, Goudie and Casey2007; Mayock et al., Reference Mayock, Bretherton, Baptista, Mayock and Bretherton2016). Lone homeless men can be fathers who are separated from their children, but the experience of being a parent separated from children appears to be much more widespread among lone homeless women (Jones, Reference Jones1999; Reeve et al., Reference Reeve, Goudie and Casey2007; Mayock et al., Reference Mayock, Sheridan and Parker2015). These routes into homelessness and differences in life experiences may, because they are clearly distinct from those of many lone adult homeless men, also influence the ways in which women experience and respond to homelessness.

A comparative absence of data may also be leading to assumptions that are not backed by robust data. One of the key tropes around women’s homelessness is the idea that it is closely linked to sex work, essentially for the purposes of survival and for the purposes of maintaining shelter, i.e. sex is exchanged for a roof over one’s head. While there is evidence that these kinds of association can exist (Harding and Hamilton, Reference Harding and Hamilton2009), some research has raised questions about how widespread it actually is and highlighted the dangers of negative, false, stereotyping of homeless women associated with patriarchal imagery that – again – processes women’s homelessness as ‘deviance’ from the expected family/partner centred roles, being outside the ‘home’ being conflated with deviance from ‘expected’ female sexual norms (Löfstrand and Thörn, Reference Löfstrand and Thörn2004; Hansen Löfstrand and Quilgars, Reference Hansen Löfstrand, Quilgars, Mayock and Bretherton2016).

This article has two goals. The first is to add to the evidence base specifically focused on the experience of lone adult women who become homeless. The second is to draw on the results of a large, longitudinal, qualitative study of an economic integration programme for homeless people which created an opportunity to look at women’s homelessness over time. The article looks at women’s trajectories through homelessness and begins to explore the possibility that those trajectories may have patterns associated with gender.

The women

The forty-seven women were mainly in their thirties and forties. Two-thirds of the women were of White European ethnic origin, and one third were of Black British/African Caribbean ethnic origin or from other ethnic backgrounds.

Four of the forty-seven women self-reported a criminal record. Poor health, including limiting illness and/or a disability were self-reported by nearly half the women (twenty-two). Seven women self-reported problematic use of drugs and/or alcohol and nearly two thirds self-reported that they had a history of mental health problems.

Nineteen were homeless at first contact with the programme and nineteen had a recent history of homelessness. The others had been assessed by the programme as being at imminent risk of homelessness. Experience of living rough was quite widespread but appeared to be mainly on a short-term basis, although some women reported repeated experiences.

Thirteen women had not completed formal education, seventeen had finished school and another seventeen had reached further and/or higher education (post-sixteen education, including University). All were unemployed at first contact with the researchers.

Becoming homeless

The main, self-reported, trigger events for homelessness given by the women were mental health problems, and violence, both domestic violence and abuse, and threats, abuse and violence from neighbouring households.

Differing experiences of the associations between mental health and homelessness were reported by the women participating in the research.

Violence and abuse in their former homes were reported as trigger events for homelessness. However, domestic abuse, while clearly present, was not as widely reported among this group of homeless women as some earlier studies have suggested (Jones, Reference Jones1999; Mayock et al., Reference Mayock, Bretherton, Baptista, Mayock and Bretherton2016).

However, violence and abuse as a trigger for homelessness existed in a wider form that some earlier research has indicated. Women reported losing their homes not because there were risks of abuse and violence from within their own household, but from neighbouring households, anti-social, violent and abusive behaviour had driven some of the women out of their homes.

The ‘trigger’ events that women identified, like mental health problems or domestic abuse had led to homelessness because the women lacked personal, practical and financial resources, did not have, or had exhausted, informal support from family or friends, had either avoided services, or been turned away when seeking assistance. Some women emphasised an experience of poverty as the immediate cause of homelessness and as the main obstacle they faced to exiting homelessness.

Pathways through homelessness

The women’s experiences of homelessness were diverse, but there was evidence supporting the conclusions of some of the more recent work on gender and homelessness (Mayock and Bretherton, Reference Mayock and Bretherton2016; Reeve, Reference Reeve2018). Women reported contact with formalised homelessness services but were also often relying on informal support from family, friends and acquaintances when they experienced homelessness. There was evidence that in order to keep some sort of roof over their heads, many of these women had relied on informal support that they had organised for themselves and that those who were making use of formal services had often relied on informal support, on family, friends or acquaintances, at some point during their homelessness. Six of the women described some of these experiences.

If the definition of homelessness and therefore ‘hidden’ homelessness discussed earlier in this article is accepted (Busch-Geertsema, Reference Busch-Geertsema, O’Sullivan, Busch-Geertsema, Pleace and Quilgars2010; Busch-Geertsema et al., Reference Busch-Geertsema, Benjaminsen, Filipovič Hrast and Pleace2014), the findings of this research support the assertion of some recent studies that at least some women’s homelessness may not be visible, because women use informal responses, choosing to rely on family, friends and acquaintances (Mayock et al., Reference Mayock, Sheridan and Parker2015; Mayock and Bretherton, Reference Mayock and Bretherton2016). However, there was also evidence that access to formal services could be limited, in that while women were drawing on their own resources to find informal solutions to homelessness, this was in part a reaction to not being able to secure help elsewhere.

Barriers to formal services were reported. The different homelessness laws that operate in the four main administrative regions of the UK (England, Wales, Scotland and Northern Ireland) are increasingly focused on prevention, but the original legislation was designed to protect people at risk from violence and homeless children (Wilson and Barton, Reference Wilson and Barton2018). The women had for the most part not received accommodation from the statutory homelessness system, or significant support from associated preventative services. The study predated the most recent legislative reforms in Wales and England that are designed to enhance preventative services. There is longstanding evidence that lone adults, including lone women, can face multiple barriers to assistance under the homelessness laws (Niner, Reference Niner1989; Anderson et al., Reference Anderson, Kemp and Quilgars1993; Pleace et al., Reference Pleace, Fitzpatrick, Johnsen, Quilgars and Sanderson2008; Bretherton et al., Reference Bretherton, Hunter and Johnsen2013; Dobie et al., Reference Dobie, Sanders and Teixeira2014; National Audit Office, 2017) and this appeared to be the case for this group of women. Resources have become concentrated on homeless families containing dependent children over the last twenty years (Carr and Hunter, Reference Carr and Hunter2008; Bretherton et al., Reference Bretherton, Hunter and Johnsen2013; National Audit Office, 2017). When the women had sought help from local authorities their experience had generally not been positive:

Contact with emergency accommodation and temporary, congregate, supported housing for homeless people, generally known as hostels, was reported only by a minority of the women. When women did experience these services, it could involve sometimes long stays and the experiences were sometimes reported as being stressful.

It has been hypothesised that women may avoid services dominated by homeless men (Mayock et al., Reference Mayock, Sheridan and Parker2015). There was not clear evidence that this had been a major factor in women’s engagement with homelessness services, although of course this was a group of women using an employment, education and training programme that was also used by a largely male population. The programme offered some women-only activities, but most service provision, other than one-to-one support with workers, often involved being in a mixed gender setting (Bretherton and Pleace, Reference Bretherton and Pleace2016). The experience of women in this instance tended towards services not offering the right range of support, or as inaccessible, partly perhaps of the presence of men, both influencing service design and limiting availability, but the presence of men per se, did not seem to explain why use of formal services was not greater. One additional finding was potentially important here, which was that some women also reported that information on service provision in different areas could be limited.

The women could be grouped into one of four pathways through homelessness during the course of two years of data collection, allowing that not all forty-seven women completed all four interviews. The largest group, of twenty-five women, had pathways that were characterised by a broadly ‘usual’ life, in the sense of living an existence in which they were stably housed, had often been in paid work or full time further or higher education and had the conventional family and personal relationships. For this group, homelessness or the risk of homelessness had been associated with a sudden shift, which in many cases was a deterioration in health, particularly mental health, which had disrupted their normal life and the stability of their housing. Over the course of the analysis, this group tended to self-exit from homelessness, with the support of the programme and other services, alongside informal social supports, i.e. their homelessness was a transitional state that had been associated with illness or another ‘shock’, which in a few cases had included domestic abuse, that had been overcome.

Three smaller groups had differing trajectories. There were eight women who had experienced social and economic marginalisation for prolonged periods, including sustained and recurrent homelessness. Contact with the programme had beneficial effects for this group, which meant that they began to change direction and move out of homelessness, as both the programme itself and other services, sometimes accessed via the programme, provided the correct mix of support that had hitherto been absent. There was no sense in which this was a reorientation of the individual, i.e. a ‘choice’ not to be homeless, instead the evidence suggested that being offered a route out of homelessness, by a service model that emphasised and respected individual choice, had been the key factor in altering pathways (Bretherton and Pleace, Reference Bretherton and Pleace2016).

For two other groups, the pathways through homelessness over the course of 2014–2015 were not positive. One small group of five women had made steps towards exiting homelessness, including moves towards further or higher education and towards paid work, but deteriorations in mental and physical health had reversed these gains, even where access to treatment was being facilitated, movement away from homelessness had not occurred.

Finally, there was a group of nine women, characterised by recurrent and sustained homelessness and higher support needs, whose situation remained a constant, they received some benefits from participating in the programme and were often being supported and accommodated by other homelessness services, but the right combination of support and access to housing that would have enabled an exit from homelessness had not been found. The study slightly predated the increasing use of ‘Housing First’ models as a response to homelessness among people with complex needs, which has shown some significant successes with long-term and repeatedly homeless women (Quilgars and Pleace, Reference Quilgars and Pleace2018).

Exiting homelessness

At the point of last contact, thirty-three of the forty-seven women described themselves as moving away from homelessness and towards mainstream social and economic life. Eight had secured work and the other twenty-five were in education, training, working and/or in a position where they could actively seek work. These women were in the largest group who had experienced transitional homelessness as a result of a ‘sudden shock’ and those who, after more sustained experience of homelessness and socioeconomic marginalisation had been assisted away from homelessness by the employment, education and training programme that was being evaluated (Pleace and Bretherton, Reference Pleace and Bretherton2017).

Twenty-four of these thirty-three women reported they were housed. Of these, the largest group, of sixteen women, were in social rented housing, another six were in the private rented sector, one was in owner occupation and another in accommodation tied to their work. Of the remaining nine, two were housed, but reported they were going to have to move because their private rented sector tenancies were time limited. Four women were in supported temporary accommodation and three were still experiencing hidden homelessness but regarded themselves - and were established - as being clearly en route to their own housing. For the women in the other two groups, the five whose progress away from homelessness had been curtailed by a deterioration in mental or physical health and the nine whose situations had not been changed by contact with the programme, there was not a clear trajectory to settled housing.

There are dangers in associating routes out of homelessness (and routes into homelessness) with individual action or inaction and seeing exits from homelessness in terms of ‘determination’ rather than luck (O’Flaherty, Reference O’Flaherty, Ellen and O’Flaherty2010), having access to the right supports and actually being in a position in which it is possible to exercise meaningful choices (Parsell and Parsell, Reference Parsell and Parsell2012). For some of the women themselves, the idea of there being a straightforward ‘exit’ from homelessness was problematic. These women questioned the underlying logic by which some homelessness services worked; the idea homelessness could be exited on a sustained basis through individual action.

A key finding was that many of the women led lives in which precarity, in terms of income, housing and, to a lesser extent, relationships was a near constant. Those who had ‘exited’ homelessness had not always done so in a way that they viewed as necessarily being sustainable. If they had work, the amount they earned per week could vary, jobs were often short-term and their housing, which was generally in the private rented sector, was on short-term tenancies. Rent levels, relative to what the women were able to earn and/or secure through the benefits system, were high, meaning that the women who had exited homelessness were often in financially challenging situations. Some of the women who had ‘exited’ homelessness did not view themselves as being secure from becoming homeless again, instead they were sometimes returning to situations that they saw as uncertain.

Evidence of populations that exist in a near constant state of housing precarity is not new (Meert and Bourgeois, Reference Meert and Bourgeois2005; Mayock et al., Reference Mayock, Sheridan and Parker2015). However, this group of women were quite often describing shifts between one precarious housing, or accommodation, situation after another, which could be punctuated with actual homelessness, rather than a very simple pathway of housing stability, then homelessness and then a return to stability.

From this perspective, homelessness was an extreme point in the lives of women that were characterised by physical and legal housing precarity. Homelessness existed within a pattern, rather than being a break from a ‘normal’ trajectory through housing, which raises questions about what we should regard as a ‘successful’ exit for a woman from lone adult homelessness (Meert and Bourgeois, Reference Meert and Bourgeois2005). The view that ending homelessness is somehow a sufficient or desirable ‘outcome’ in itself, has been criticised in other contexts (Johnson et al., Reference Johnson, Parkinson and Parsell2012; Quilgars and Pleace, Reference Quilgars and Pleace2016). This research raises some questions around how we need to define what constitutes a true ‘exit’ from homelessness in general and, in particular, what it means for a woman who has experienced homelessness to feel that the risk she will ever experience it again has been minimised.

While there have been breakthroughs in service design, such as Housing First, the results of this study showed that there are still groups of homeless people, in this instance groups of homeless women, whose mix of needs is not being properly recognised and responded to. The evidence of a mutually reinforcing interplay of addiction, crime, mental illness, poor physical health and long-term and recurrent homelessness is longstanding (Kemp et al., Reference Kemp, Neale and Robertson2006) and this research, alongside other studies, shows that women are within this group of homeless people and need the right mix of services, which may need to be gender specific, to exit homelessness (Bretherton, Reference Bretherton2017; Quilgars and Pleace, Reference Quilgars and Pleace2018).