Homelessness research and the influence of targeted policy

Is it more effective to target interventions at those most in need? Or should we  'act universally', addressing the basic welfare of all? When it comes to addressing the causes of inequality, there is growing evidence to suggest that universal approaches are more effective. Despite this, policymakers find a continuing appeal in targeted social programs, reserving social services for the most disadvantaged groups in society.  In taking this road, are policymakers failing to address the root causes of disadvantage?  In light of Homeless Persons' WeekDr Pauline McLoughlin (@PJ_McLoughlin), from RMIT University, and Dr Gemma Carey (@gemcarey) from ANU explore how homelessness research in Australia must grapple with a strong current of highly targeted policy. 

The debate over targeted versus universal action is a perennial issue in health and social policy. In recent years, targeted policy approaches have drawn criticism from researchers for being ‘palliative’ rather than preventive, and ultimately less effective. While perhaps cheaper in the short term, the argument is that - by targeting only risk groups and the poorest members of society - we get caught up in the endless task of managing the impacts of disadvantage as they unfold. In the process, we are failing to prevent the deeper structural roots of disadvantage. This arguably costs us more over the longer term.

Still, the intuitive logic of targeting has a strong pull in policy-making. It appeals to the 'new right' ideologies of individual responsibility. It also captures historical leanings towards targeted welfare relief, and our tendency to adopt means testing and workfare policies in Australia.  This preference for targeted approaches is also known as ‘residualisation’. The term evokes the image of a scaled-back welfare state, where welfare itself has become a residual safety net: a ‘last resort’ for the extremely poor and disadvantaged. For homelessness researchers, the crisis-driven focus of homelessness policy is an important but often overlooked example of residual welfare at play. It is also an issue that the research community must understand better, if we are to have a more effective impact on policy processes.

For many years, homelessness policy in Australia has concentrated narrowly on the extremes of rough sleeping and long-term homelessness - primarily through the provision of crisis and transitional accommodation. As highlighted by the 'Which Way Home' Green Paper, this system has not been working, with 'mainstream' services such as schools, health, justice and social security systems providing fragmented and ill-equipped responses. We also know that rough sleepers and single adult homelessness are the proverbial tip of the iceberg.

Since the 1970s, urban growth, increased migration, unemployment, rising house prices, and changes to traditional family structures have all played a role in increasing the vulnerability of ‘new’ segments of the population to becoming homeless. This includes women, children and young people, Aboriginal and Torres Strait Islanders, migrant families, and people leaving state care and prison. There is a growing recognition that these 'new homeless' groups are likely to experience homelessness which is hidden ‘under a roof’ – including couch surfing, displacement through domestic violence, and other forms of insecure and unsafe housing. Homelessness is not limited to rooflessness.

In recent decades, research and inquiries in Australia have brought new recognition of these realities. This has led to a broadening of the range of living conditions and groups that are counted as homeless in policy-relevant statistics and legislation. Australia is one of a minority of countries to have a specific method for counting homeless cohorts in official census data.

Until recently (when census definitions of homelessness were revised), a set of standardized census definitions have enabled us to enumerate homelessness, using researchers Chamberlain and Mackenzie's categories of “primary”, “secondary” and “tertiary” homelessness. The primary end of the homelessness spectrum referred to people lacking “conventional accommodation”. This includes rough sleeping, living in improvised shelters and street dwelling. Secondary homelessness referred to people in various forms of temporary accommodation, including couch surfing, supported accommodation and staying with friends and family. At the tertiary end of this spectrum, were people staying in insecure housing, including medium term boarding and lodging arrangements with no security of tenure.

The intention here has been to improve the capacity of governments to identify and respond to the complexities of homelessness, and the services needed to address and prevent it. Many researchers have also sought to challenge the targeted  approach taken to homelessness in Australia and elsewhere.

To some extent, these research efforts have been effective. The definitions used in the census were central to the former Labor Government’s goal of ‘reducing the risk of homelessness’ as a key priority area under their Social Inclusion Agenda (SIA). This in turn was the impetus for the release of the ‘Road Home’ Homelessness White Paper, which paved the way for an extra $5 billion funding for affordable housing, homelessness services reforms, and research from 2008-2013.

Yet, these broadened understandings of homelessness have had to contend with the powerful pull of targeted policy frameworks in Australia. This is clear even from the widely lauded Homelessness White Paper, and its implementation through measures such as the National Affordable Housing Agreement (NAHA) and the National Partnership Agreement on Homelessness. Beyond the notable investment in social housing provision under the NAHA, the majority of initiatives and programs implemented through the White Paper were highly targeted towards specific groups, identified –through research - as being at significant risk. This included low income families, people exiting state care and prisons, and addressing Indigenous homelessness.

While the White Paper set the target of ‘halving overall homelessness’ by 2020 (a target which will likely not be met), the key focus of the reform measures was to address primary and long-term homelessness. Despite recognition of the structural drivers of homelessness – including unemployment, entrenched disadvantage, poverty traps, gendered inequalities, and a lack of affordable housing - homelessness policy reforms under the White Paper were not substantively linked to universal programs to address these issues at the whole population level.

Categories, such as ‘primary homelessness’, risk groups, or the ‘bottom 5%’ of society, are used extensively in research to frame inquiry into social issues, and in government to develop targeted policy priorities. Here, the population is carved up into groups of greater or lesser ‘need’, based upon measurable shared characteristics, such as income, housing, education and occupation. These categories are divorced from the everyday reality of disadvantage and homelessness, and limited in their ability to guide the design of policies and programs, as they do not provide information on pathways to change.

This leads us further away from universal approaches. Even in the case of groups that are given priority, these categories do not necessarily capture the diversity of needs and experiences of the homeless. At the individual level, targeting of policies and services can promote stigma, leading to service barriers and unmet need.

Rather than inviting more sophisticated efforts to address underlying determinants of homelessness, the research push to create policy-relevant, ‘objective’ understandings of homelessness may risk being sucked in to an existing model targeted at homelessness risk groups. This has arguably drawn attention towards the most extreme forms of homelessness (i.e. sleeping rough), with less energy given to the longer-term goals of prevention and early intervention.

To overcome these entrenched difficulties, researchers need to go beyond ‘evidence into policy’, and reflect on broader tendencies in policy design and implementation. We also need to deal with an important question that is too often obscured beneath the ‘palliative’ approach of targeted intervention: what does a truly preventative policy framework ‘look like’ when it comes to homelessness, anyway? We see a lot of early intervention and targeted responses. But how do we go about investing in a population-wide approach to homelessness?

Recently, the Marmot Review of Health Inequalities called for an approach that uses ‘proportionate universalism’ to effectively address the social determinants of ill health. The idea is that services and programs are delivered universally, but different groups across the ‘social gradient’ receive added support proportionate to an estimated need.

Universal approaches are supported by a body of research, which shows that social democratic countries that address the underlying causes of inequality (through universal social services and programs), perform better on major health and quality of life indicators than liberal welfare state regimes, which use more targeted policies of welfare relief for the most disadvantaged.

This is an important consideration for homelessness researchers. If we are serious about improving the capacity of governments to prevent, rather than simply ‘manage’ and ‘treat’ homelessness, we must be careful that we are not unintentionally creating new openings for policymakers to target efforts at smaller and smaller groups. We must also be careful that the fruits of homelessness research knowledge support policy changes that prevent people from falling through the cracks.

Posted by Pauline McLoughlin