Abstract

50 years since the first screening of the powerful TV drama ‘Cathy Come Home’, and 30 years after the first homelessness legislation in the UK, this paper reflects on the continued invisibility of women’s homelessness in scholarly and policy debate, It argues that homelessness is inherently gendered, yet rarely recognised as such, and that new conceptualisations of homelessness that take account of gender differentials is urgently needed.

Introduction

In 2007 we contributed a paper to the first volume of People Place and Policy (PPP) online (Casey et al., 2007; see also, Casey et al., 2008) focused on homeless women’s use of public spaces. The paper challenged the dominant narrative that homeless women are neither present, nor do they belong ‘on the streets’ or in other public places. The research on which the 2007 paper was based¹ had sought to draw out the influence of gender on homelessness (Reeve et al., 2006: 2007). Thus, although focused on the geographies of women’s homelessness, the 2007 PPP paper aimed, more broadly, to make women’s homelessness visible and, implicitly, to demonstrate the inherently gendered nature of homelessness.

This is a good time to reflect on current understanding of women’s homelessness. It is a decade since the first volume of PPP, but it is also 50 years since the screening of Cathy Come Home – the powerful TV drama that charted a woman’s descent into homelessness – and 40 years since the first homelessness legislation was introduced in the UK. If the Housing [Homeless Person’s] Act 1977 had been in place in 1967 then ‘Cathy’ – or the women she represented – would (should) have been owed a statutory duty by the local authority and been offered temporary and then permanent housing. She would (should) not have become roofless and, as a result, had her children removed from her care. Whatever criticisms have been levied at the implementation of the homelessness legislation over the years, the priority it gave to families benefited women by virtue of their common role as primary or sole carer of children (Mayock et al., 2016; Fitzpatrick and Pawson, 2016; Pascall, 1991).

Gendered understandings of homelessness

At the time of writing the 2007 paper, there seemed to have been little progress in developing a conceptualisation of homelessness that recognised gender differentials. Seminal work by feminist scholars Sophie Watson and Helen Austerberry in the 1980’s (Watson, 1988; Watson and Austerberry, 1986), followed by broader scholarship drawing attention to gender inequality in the housing market (Neale, 1997; Gilroy and Woods, 1994; Morris and Winn, 1990; Banion and Stubbs, 1984) had not prompted a sea change in homelessness analysis. Unfortunately, that fact still stands a decade later. A group of housing and gender scholars are slowly nudging forward with efforts to develop gendered conceptualisations of homelessness, producing sporadic evidence and calling existing accounts of homeless into question (e.g. Wardhaugh, 1999; Mayock and Bretherton, 2016; Moss and Singh, 2015; Watson, 2011, May et al., 2007; McCarthy, 2017). But this is a small, discrete area of academic homelessness scholarship and has barely touched policy debate or development. Notwithstanding these efforts, there remains little recognition that women’s homelessness is worth scrutiny, and women are rarely the subjects of homelessness research in the UK. As a result, academic and policy knowledge generated about homelessness – knowledge on which policies, legislation and responses to homelessness are based – remains inadvertently dominated by the experiences of men.

Historically, women’s homelessness has been framed by normative assumptions about gender roles, with associated categorisations – for example women as ‘domestic’, as ‘homemakers’, and as ‘deviants’ when failing to adhere to these normative categories – important in the conceptualisations of homelessness that have emerged (O’Sullivan, 2016 Neale, 1997). And the way women are categorised, has profound implications for access to housing, support, and recognition (i.e. literally being ‘counted’). Cultural images of women impact on homelessness policy, and also inform methods of enumeration (Löfstrand and Quilgars, 2016). Pleace demonstrates clearly the ways in which methods of recording and measuring homelessness render women invisible, for example by defining homelessness in ways that inadvertently fail to capture women, or through active decisions not to record gender that are rooted in assumptions about women’s absence from certain spaces or forms of homelessness (Pleace, 2016).

Picking up from Watson and Austerberry we take the position that gender roles and expectations, women’s socio-economic position, patriarchal attitudes embedded in housing production and allocation, cultural images, power structures, and (gendered) forms of disadvantage impact on women’s access to housing and, therefore, on their vulnerability to homelessness.² It is gender-based inequalities, alongside the cultural images that perpetuate these inequalities that both explain women’s homelessness but also produce certain gender specific (or gender-relevant) needs and experiences. We consider some of these specific experiences below. Women’s homelessness can be understood in the context of sexual division of labour that produces and reproduces inequality and disadvantage in the labour market (Tessler et al., 2016; Morris and Winn, 1990) but also a distinctive (if ambiguous) relationship with the home – and, by implication, with non-domestic space also (McCarthy, 2017; Tomas and Dittmar, 1995; Watson and Austerberry, 1986). Women are, therefore, more vulnerable to homelessness, face specific barriers to resolving their housing crisis, the loss of home may have profound implications for identity, and is more likely to be seen as transgressive (i.e. of gender norms) or deviant.

The relationship between domestic violence and homelessness exemplifies some of these points. Domestic violence commonly emerges as a cause of women’s homelessness in research, monitoring and statistics. Rarely, however, is the question posed: ‘why do women become homeless when they leave a violent partner?’ Economic and social constraints can leave women dependent on a male partner /breadwinner for access to housing (private housing, in particular) and therefore homeless when that relationship breaks down or they need to escape due to violence (Rose, 1994; Morris and Winn, 1990). Financially independent women with savings, property, well paid employment, and affordable childcare can avoid or escape homelessness in this situation. It is the fact that women are more likely to be the primary carers, the part-time and low-paid workers, unable to afford their own family home without assistance that is the ’cause’ of their homelessness, although domestic violence may well be the trigger. We have recently interviewed Mandy and Michelle³ for a study that is ongoing at the time of writing.⁴ Both became homelessness when they left violent partners. Their experiences are discussed further below but it is noteworthy that when Mandy explained why she became homeless when her partner changed the lock on their home she said she ‘had no money whatsoever’, while Michelle ‘didn’t have nowhere for myself to go’.

From our 2006 study, we were able to generate understanding of the ’causes of homelessness’ based exclusively on the experiences of women. It is extremely rare for explanations of homelessness to explore for gender differentials, or, indeed, to include enough homeless women in the sample for their experiences to be fully represented in the results (but see Tessler et al., 2001 for a quantitative study in the USA). Our study found that, most commonly, the following experiences set in train a trajectory that led to homelessness:

• sexual abuse
• neglect, abandonment and other family problems
• ‘maternal trauma’ (reproductive health issues, loss of children)
• experience of violence
• bereavement

These experiences are certainly not specific to women, but they are mostly gendered. They are more commonly experienced by women (sexual abuse and violence, for example), or they are experienced differently (maternal trauma, for example), or with potentially differential impact given women’s greater dependence on others for their housing (in relation to bereavement or violence in the home, for example).

Having briefly discussed wider gendered aspects of housing and homelessness we now emphasise this point with reference to three specific aspects of homeless women’s experience: rough sleeping; the relationship between motherhood and homelessness; and survival sex.

Women and rough sleeping

Contrary to all assumptions about rough sleeping, and the underpinning official statistics, homeless women do sleep rough in relatively high numbers. In our 2006 study, 62 per cent of the 144 single homeless women surveyed had slept rough. Rough sleeping was also the most common housing situation the 44 women interviewed for the study entered when they first became homeless. Subsequent qualitative and quantitative research we have conducted has only corroborated this fact. In a qualitative study of female homeless street sex workers in 2009, for example, rough sleeping was the most commonly relied upon accommodation situation amongst the 30 respondents, and nearly half had slept rough in the previous month (Reeve et al, 2009).  In a study in 2011 where both men and women were surveyed, 66 per cent of women (and 79 per cent of men) reported having slept rough (Reeve and Batty, 2011).

But, as we highlighted in the 2007 paper in this journal, women occupy public space (including when sleeping ‘on the streets’) differently from men. They employ strategies of invisibility such as sleeping in sites that are hidden from view, and remaining in plain sight but disguising their homelessness status. It is no surprise, then, that although 66 per cent of women in the 2006 study had slept rough, only 12 per cent had been in contact with a rough sleeper team.⁵

Homeless women who have participated in our research over the past 15 years have described seeking out places that were less visible such as public toilets, garages, bin bays, or spaces located away from busy city centres. As the following two women interviewed for our 2006 study explain:

Flats…have these little sort of huts where they put the bins in and so you get behind the bins and you’re hidden, you know they can’t see you.

There’s places we used to go, car parks where, you know, they were closed in and that, and nobody would see you there, so I wouldn’t feel unsafe there.

Women also make themselves invisible by disguising their homelessness status, making efforts to ‘blend in’ with their surroundings by ‘looking like everyone else’ and appearing to use the space as it was intended, rather than for sleeping. One woman, for example, wore a large woollen poncho when sleeping (upright on a bench) rather than carrying a blanket that would identify her as a homeless person. Another carried a small suitcase so she would look like any other traveller when sleeping in an airport. In this way, women managed to sleep, very publically, in airports, business parks and 24 hour transport without drawing attention to themselves. Safety is a key motivation for women to remain hidden, as the following quotes suggest:

Just thought I wouldn’t be able to cope with living on the streets because I just felt I wouldn’t be safe and there’s no way I’d put myself at threat from men or anything like that….

I got attacked, not beaten, but I got sexually attacked a few nights ago.

The men tended to be in the shop front which I would never contemplate doing…for me it was a safety thing.

Women ‘do’ rough sleeping differently because they are less safe on the streets, and, I would argue, they are less safe on the streets because of patriarchal power structures and associated cultural imagery that renders them vulnerable. For example, women occupy public space differently, and are more vulnerable as a homeless person within it, because gender expectations and representations demand their association with ‘the home’. To be obviously home-less is, therefore, problematic (Wardhaugh, 1999; Radley et al., 2006) when ‘…homeless women’s bodies can be seen to represent a challenge to the feminine body, the mother or wife located in the home (Watson, 1999: 96).

Rough sleeping amongst homeless women is a good example of the way in which a partial (ungendered) understanding of homelessness results in tools of measurement that reinforce existing misconceptions. Rough sleepers are thought to sleep mainly in doorways, city centres, and other visible locations – an assumption based on the rough sleeping strategies of men, not women. The method of enumeration developed for this ‘visible’ population is to count visible rough sleepers. These statistics repeatedly show that the vast majority of rough sleepers are male. In 1990, in response to rising rough sleeping, the government established the Rough Sleepers Initiative (RSI) to tackle this growing problem. In the mid 1990’s, as part of the drive to resolve rough sleeping and in order to monitor the success of the RSI, local authorities were required to conduct a ‘count’ of rough sleepers in their area. This comprised an actual count of people bedding down, conducted on one night of the year. Since 2010 LAs have had discretion about whether to conduct a street count and they can, instead, submit a robust estimate of the number of people sleeping rough on a typical night. Street counts have typically been conducted in city centres and known rough sleeper sites and the most recent guidance offers LAs some advice on how to conduct their count.

With limited resources likely to be available it is better to focus efforts on areas that rough sleepers are known to use (DCLG, 2010: 5).

With efforts focused on areas that (mostly male) rough sleepers are ‘known’ to use, it is no wonder that just 12 per cent of rough sleepers recorded in the 2016 official rough sleeper count were women.

It may well be the case that women are slightly less likely than men to sleep rough, and the results from the 2011 study cited above suggest this is the case. There are a small number of options that are, perhaps, more available to women than to men (see survival sex below). But, for women to sleep rough in significantly fewer numbers than men would require temporary accommodation that is only available to women. Why women would have some kind of magic ‘alternative’ solution unavailable to men, is inexplicable. More likely, they do sleep rough but we do not see them, and as long as rough sleeping is defined and measured as the visible face of homelessness, they will remain invisible.

Homelessness and Motherhood

The final scene of Cathy Come Home, once seen, cannot be forgotten. The sight of social workers wrestling Cathy’s children from her at a bench in a train station is one of the most moving scenes in television history. Cathy’s children are being removed from her care because she is homeless. Ten years later, the introduction of the Housing (Homeless Persons) Act 1977 gave priority to homeless people with dependent children, providing vital protection to parents in Cathy’s situation. The Act placed a duty on local authorities to provide housing for families with a local connection to the area, who were homeless through no fault of their own (i.e. not ‘intentionally homeless’).

However, a Freedom of Information Request revealed that in 2014/15 one in three local authorities took children into care because the family was homelessness (Inside Housing, 18^th November 2016). In our research we have repeatedly encountered women who are separated from their children because of homelessness, either because their children have been removed by the local authority, or because women place their children temporarily in the care of others (grandparents, the child’s father) while they seek suitable housing (see also Mayock et al., 2015).

Our 2006 study was focused specifically on single homeless women⁶ and we recruited all survey and interview respondents through services working with single homeless people. We were surprised, then, to discover that 30 per cent of our survey respondents⁷ reported having children under the age of 16. In around one third of cases, women’s children were permanently in the custody of someone else, whether the children’s father, other relatives, or the local authority (including those who had been adopted). But, in two thirds of cases, women’s children were in the care of others temporarily (50 per cent with family or friends and 17 per cent in temporary LA care). These women generally still regarded themselves as primary carer of their children, albeit amid a hiatus while they stabilised their lives.  Research in 2009 (Reeve et al., 2009) and, more recently, in 2017 revealed a similar picture. Of the 30 women (all homeless street sex workers) interviewed in 2009, all were mothers but none had retained custody or guardianship of their children. ‘Mandy’ was interviewed in 2017 and described the circumstances leading to her separation from her daughter:

I had a partner… that got very violent, he broke my jaw and various other things…so then I had to end up handing my daughter over again which was heart-breaking. I got back from picking my daughter up from school to [ex-partner] changing the locks and I’m stood there with my daughter out in the cold. That night I finally got into the hostel…this was a couple of days before Christmas eve so we’re in this horrible place, I can’t even afford a tree…I didn’t have any of my belongings, I had no money whatsoever…I sent my daughter to my mum’s.

Michelle, like Mandy, became separated from her children when she left a violent relationship with a man who ‘for 17 years…put me through hell, literally’. She has never regained custody of her children:

The day I walked out on him I had to leave four kids behind. I didn’t have nowhere for myself to go so I wasn’t going to drag four kids out. That turned out to be a bigger mistake than ever. Something’s happened, they’ve got taken off him, he’s in jail, my youngest daughter’s been adopted and there’s fuck all I can do about it.

Homelessness was not always the primary reason for women’s separation from their children (this was particularly true in the case of the female sex workers interviewed in 2009, many of whom were also drug dependent) and some of those we have interviewed placed their children in the care of others before they became homeless. But homelessness or precarious housing was sometimes the only or the primary reason, and was always part of the problem.

Whatever the circumstances under which women separated from their children, once this had occurred, they ceased to be parents in the eyes of those who make and enact policy. They were treated as ‘single’ in legislative terms. The protection offered to mothers by the homelessness legislation – and often cited as a reminder that women have benefits positively by public policy – often falls away when their children live elsewhere, even if they are elsewhere only because the mother has nowhere adequate to house them. Elaine, interviewed in 2009, explained what had happened to her when she approached the local authority as homeless a couple of years previously:

She [housing officer] was going like this: ‘right, so what’s your problem?’ I says ‘well I’ve got nowhere to live’. ‘Right, have you got children?’ ‘Yeah, that’s the problem see, I can’t get my children till I’ve got a house’. She’s gone ‘right, so where are the kids now?’ ‘At their dad’s so they’ve got a stable life’. And then she goes ‘so they’re not living with you are they?’ I says ‘well technically they are, technically, yeah, because I’m supposed to be with them, I’m supposed to have them, I raised them but I’ve got nowhere to stay so of course they’re with their dad because I’m not dragging them on the street..’ But she just said ‘so they’ve got a roof over their head then, haven’t they? They’re not actually homeless’. But of course they’re homeless because they should be with me.

Leona, interviewed in 2017, was living with her son in a hostel for women and children. Her son spent part of the week with his father so Leona stayed with a friend during this time, not thinking it appropriate for her to remain in a family hotel without a child. However, by spending three nights away from the hostel she breached the tenancy rules and was evicted. She returned to the homelessness department to find she would no longer be assisted as a mother. She explained:

They said ‘no’ coz they see it as ‘we just gave you a hostel and you’ve failed’…that’s why they recommended I did it [made an application] as a single person cos I failed as a parent in their eyes.

Not treated as a single woman, Leona had no priority and so spent the following few years sleeping rough or staying in hostels for single people where she could not have her son with her. Unable to provide a stable home for her son, and with escalating anger management issues, her ex-partner was able to successfully apply for full custody. Leona’s comment about ‘failing as a parent’ are also worth noting, as an example of the way in which homelessness, and homelessness practices, can interact with gender identity and expectations, to produce specific impacts on women.

Survival sex

Survival sex is the exchange of sex for material support. As a personal strategy – or a form of capital – responding to a structural problem, it is inherently gendered, with women and men having different resources to draw upon to manage their homelessness (Watson, 2011). O’Grady and Gaetz (2004), researching homeless youth in Canada, argue that masculinity and femininity are organised (and reproduced) in particular ways in homelessness spaces. In short, men have more power in this arena because categories of ‘homeless’ and ‘the streets’ are male-defined, while homeless women are alienated from their culturally defined comfort zones of the private and domestic. Specific, gendered subsistence strategies develop, rooted in these power relations. As O’Grady and Gaetz (2004) succinctly put it: Since the ‘streets’ have traditionally been defined as male space, the money-making opportunities that are available to homeless youth are likely to be structured accordingly’ (p.401). They find, for example, that male homeless youth access more lucrative sectors of the informal economy than women and that, unlike men, when homeless women engage in sex work they typically work under the control of a pimp. Watson’s study in Australia also uncovers evidence of a gendered discourse (and accompanying expectation) of willingness amongst homeless women to participate in survival sex (Watson, 2011).

Like Watson, in our research we have found that survival sex is sometimes one of the few resources at women’s disposal to put a roof over their head. In our survey of more than 400 single homeless people in 2011, for example, 20 per cent of women had engaged in sex work to pay for a hotel (compared with three per cent of men), 28 per cent had spent the night with someone specifically to accommodate themselves (compared with 14 per cent of men) and 19 per cent had engaged in sex work because it offered the opportunity to spend the night with a client (compared with three per cent of men). In this research and in our study of homeless women in 2006, survival sex emerged as a subsistence strategy. The ways in which women deploy sex in this context is, however, varied with ‘sex work’ only one dimension of survival sex. In our studies, five distinct forms of survival sex have emerged:

• Actively seeking out someone with whom to spend the night. One women interviewed in 2006 explained her strategy here, also making it clear that her sole motivation for this sexual encounter was accommodation: “a guy walked past who I knew…who I know has a bedsit, so I sort of jumped on him and I was sort of ‘J… alright mate’ and so I went back with him and got into bed with him just so I could have a bed, unfortunately.”
• Developing ongoing sexual relationships with housed men. This was more common as a strategy for avoiding, rather than escaping homelessness amongst women who had just become homeless or knew this was imminent. Some relationships lasted several years but respondents were very clear that their primary motivation for becoming involved with the man in question was to secure housing.
• Temporarily reuniting with ex-partners to secure accommodation for a few nights. In some cases the ex-partners with whom respondents reunited were violent.
• Sex work or ‘clipping’ to raise money for accommodation. This was rare, not least because of the relatively high cost of hotel accommodation, but is a way in which women unable to secure alternative housing deploy sex to obtain accommodation.
• Engaging in commercial sex work in the hope of finding a punter who will allow them to spend the night. These women were usually already sex workers, working to obtain income, but would actively seek out clients they knew sometimes let them stay the night. One women rough sleeper, for example explained that a couple of her clients employed her in their own homes and she made concerted efforts to seek them out when she was desperate for a bed for the night.
• Engaging in sex for items such as food and clothing, or in order to use their facilities (shower, washing machine).

Conclusion

Reflecting on the state of knowledge about homelessness reveals that little has changed in the past decade. In fact, notwithstanding the efforts of a few scholars, little has changed since Waston and Austerberry’s seminal work on women’s homelessness in the 1980’s. Yet, homelessness is inherently gendered. Referring to some specific themes, this paper has attempted to demonstrate how homelessness interacts with, and is informed by gender roles and expectations, institutions, and power structures. Homelessness is not only its visible manifestations and until this is acknowledged, homeless women will remain invisible and our understanding of the nature, character, and extent of homelessness can only be partial at best. This paper has shown, for example how current understanding – for example regarding the visibility of rough sleeping, or common causes of homelessness – begins to crumble once a gendered analysis is applied to subjects about which existing evidence is comfortably consensual.