Australian Bureau of Statistics

Rate the ABS website
ABS Home > Statistics > By Catalogue Number
4922.0 - Information Paper - A Statistical Definition of Homelessness, 2012  
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 04/09/2012  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product  
Contents >> Process for developing definition of homelessness >> Where did the views of Definition subgroup members merge and where did they diverge?


The subgroup was in agreement on a number of elements (and of the components of those elements), but had divergent views on some components of the definition, as well as on the process of developing advice on the definition.

Areas of agreement

There was large agreement within the subgroup on relevant issues, including that:

  • Homelessness is broader than rooflessness or houselessness;
  • Cultural views of homelessness evolve over time;
  • Lack of security of tenure is a feature of homelessness; and
  • Inadequate dwelling is a feature of homelessness.

Areas of divergent thinking

Scope of the subgroup

Divergent views on the scope of the definition subgroup's were expressed by some members, particularly around the process for progressing the definition work in an Aboriginal and Torres Strait Islander context, and in the review of the cultural definition.

Aboriginal and Torres Strait Islander homelessness definition work

Some members of the HSRG and the Definition Subgroup noted that the definition work should include analysis of issues pertinent to Indigenous populations. They stated that this work should not be undertaken separately as the ABS Homelessness definition should reflect and encompass Indigenous understanding of experiences of homelessness.

Others, including ABS, noted that the specified timelines for completion of the first phase of the definition work would not allow for the necessary consultation and review of the relevant literature to ensure that the Indigenous context had been adequately represented. Work has commenced on considering homelessness in an Aboriginal and Torres Strait Islander context, and a subgroup will be formed to progress this work.

The cultural definition of homelessness

A minority of members were of the view that the discussion of an official ABS statistical definition of homelessness should be confined to reviewing, refining and updating the cultural definition developed by Chamberlain and MacKenzie. These members argued that widening the scope of the discussion was unnecessary and unproductive given the widespread and long-standing use and acceptance of this definition in Australian homelessness policy and in the service sector, and its demonstrated capacity to be operationalised in the Census estimates of homelessness. They noted, for example, that the cultural definition could be refined to encompass severe overcrowding as a state that falls below cultural understandings of a minimum level of accommodation adequacy, as could 'choice' in terms of what the community accepts as not only minimum but 'normal' in modern society in a variety of situations that the sector and other people would not regard as homelessness. Violence and safety more generally, as well as dwelling adequacy in the context of new forms of accommodation, could also be explored.

The majority including the ABS, were of the view, as per the terms of reference of the group (see Appendix 2), that key international and national definitions must be considered to demonstrate that a rigorous, evidence informed, process had been undertaken to develop the conceptual framework /definition with core elements underpinning it. They noted that the purpose of the group was to advise on the conceptual elements of a definition that could be operationalised across a range of ABS datasets, not just using the Census of Population and Housing. It was also noted by ABS and some members that there has been no empirical validation of the Chamberlain and MacKenzie cultural definition in terms of its assumptions about a minimum shared community standard. Moreover views have evolved over the past 20 years or so, suggesting that this standard is historically contingent.

Some members saw merit in the 'community standard' concept being tested in a nationally representative way, either through extensive focus group work or through survey activity (perhaps as part of a deprivation survey or similar measurement exercise).

The ABS agreed that the cultural definition could be modified to address the conceptual elements of a broader definition of homelessness. However, it was not clear to the ABS that refreshing the cultural definition added anything to the conceptual underpinnings of homelessness measurement that is not delivered directly by the development of a conceptual definition alone. A revamped cultural definition would require constant revisiting as the world changed, because without the conceptual underpinnings being established on which to assess the homeless state, new and emerging living situations could not be readily classified.

Consensus was not reached on this issue. However, the subgroup agreed to proceed according to its terms of reference by identifying the elements in the key national and international definitions, and examine the basis for homelessness identification and categorisation in a large list of living situations.

Accommodation alternatives (including choice)

The issue of access to accommodation alternatives (including choice) was the subject of extensive debate and some divergent opinion among members of the group.

The notion of accommodation alternatives was included in the ABS proposal for a definition to recognise that some people have the experience of living somewhere which they would not call 'home' for a period of time, as part of the many life transitions that people both pursue and experience. If those people could have accessed alternative suitable accommodation, but they choose not to in order to pursue their own life goals, they are unlikely to be considered by most people as being homeless. Some submissions on the definition, including from a service peak body, supported particular examples where a living situation might have been classified as homeless under some interpretations of a cultural definition of homelessness, but which the submission proposed should be excluded from the ABS definition of homelessness.

An example of people who might first appear homeless under some criteria but actually have accommodation alternatives, are young Australians travelling in their gap year with no usual address while staying in a variety of short term accommodation circumstances. Other examples include the large numbers of Australians that move for a variety of purposes (jobs, education, tree / sea change) and may not have a 'usual address' established in a new location at a particular point in time (i.e. still looking around before buying or signing a long term lease). Owner builders may have temporary accommodation arrangements that may appear to be 'homeless' for a time under some aspects of the elements of the ABS definition and other definitions. Incorrectly attributing homelessness to these groups could skew both policy and resource allocations and put at risk public concern about the social issue of homelessness and its implication for social exclusion.

The group noted that the idea of lack of accommodation alternatives and choice is not new in homelessness definitions. It is reflected in the SAAP definition, and in the Statistics New Zealand definition (where it is described as 'options to acquire safe and secure private accommodation').

Most of the subgroup members agreed in principle that the idea of accommodation alternatives or choice is an essential consideration in defining someone as homeless. While most members agreed it was easy to conceptualise the issue of choice, many members noted that it is difficult to measure and operationalise the concept of accommodation alternatives or choice in Censuses and, to a less extent, in surveys. It can be difficult to create an operational boundary between having access to financial, physical, psychological and personal resources and not having access to these resources. Some members, for example, advised that an exact definition of accommodation alternatives would be required in the definition and that the definition would need to incorporate a specific and detailed standard (for example having lack of assets or income, physical health or psychological capacity that would prevent someone from accessing safe and secure living arrangements). Others cautioned against a too prescriptive application of this aspect of access to alternatives.

Most subgroup members agreed on many of the aspects of accommodation alternatives (or choice) that should be excluded from a measure of homelessness. Some members considered that this aspect of the definition should be addressed through an explicit list of exclusions rather than through the overarching consideration proposed in the ABS definition.

Some discussion centred on the complete absence of accommodation alternatives in particular locations regardless of the personal resources of individuals in that location (e.g. mining towns or roadside construction). According to the ABS definition, if people chose to move from a location where accommodation choices were available to a location for, say, significant personal benefit, the person would not be classified as homeless. On the other hand, people with no accommodation alternatives in one location who move to another and remain without accommodation alternatives are homeless if so assessed against the key elements of the definition applied to their current circumstance.

The ABS view is that the concept of accommodation alternatives should be addressed in the conceptual definition, and that in the operationalisation of the concept that the risks of error (in both directions) in particular measurement contexts should be clear. Some members put the view that an alternative approach is to exclude the 'accommodation alternatives/choice' element from the concept but in operationalisation to provide sufficient detail to allow users to exclude components of 'homelessness' that are not relevant to the particular purpose for which the statistics are being used.

All subgroup members advised the ABS to produce estimates of the magnitude of under and over estimation which may occur in regard to its operationalisation, and a list compiled of the inclusions in, and exclusions from, homelessness measures that arises from the application of this overarching dimension in particular measurement contexts.

Personal safety

The ETHOS typology of homelessness and housing exclusion includes an operational category for people living under the threat of domestic violence. Subgroup member views differed on whether personal safety should be conceptualised as homelessness or at risk of homelessness. All agreed that domestic violence is a significant cause of homelessness and that personal safety is a concern for people who are subject to, or fleeing domestic and family violence.

Some members considered that a person living under threat of domestic and family violence does not have control of space, privacy, safety, or the capacity to pursue social relations and they are socially excluded. As such they are housed but not “homed” and without accommodation alternatives. Some suggested that there is in fact a potential double standard in relation to this group especially when compared with those living in boarding houses. If people living in boarding houses are classified as homeless, in part, for reasons of safety, lack of privacy and personal discomfort and fear then so too should people living in violent households. Some also asked why people living in unsafe dwellings that are structurally unsound are considered homeless because they are in a living arrangement that is not safe or secure, however those living in unsafe dwellings because of the threat and reality of violence are not homeless although their living arrangement is also not safe nor secure?

Others argued that people in these circumstances who are living in an adequate dwelling with security of tenure are better understood as "at risk of homelessness". In effect these members gave greater weight to the accommodation (security of tenure, adequacy) rather than the social amenity components of the ABS definition (some of these members even favoured a definition without the social amenity component).

Some members were hesitant to include personal safety in a definition of homelessness because domestic and family violence is a broader social problem, and double labelling it as homelessness might narrow the issue to just a housing issue and dilute its profile as a social issue in its own right. However members generally agreed that homelessness was broader than a housing issue (i.e. it is not just rooflessness) and double badging the plight of victims of domestic and family violence as both a safety issue and either a homelessness issue or a precarious housing issue should not dilute the focus for policy (and other) action. Similar issues apply with mental health.

All members agreed that when a person escaped domestic violence, a return to that unsafe dwelling did not constitute an accommodation alternative.

Other views on the definition

Other views on the definition included severe crowding and the security of tenure element.

Severe crowding

The homelessness definition that ABS proposed to the subgroup and HSRG categorised severe overcrowding as an example of the third element - a lack of control of, and access to space for social relations, and being without accommodation alternatives. In extremely overcrowded dwellings inhabitants are generally unable to pursue social relations, or to have personal (i.e. family or small group) living space, or to maintain privacy and/or personal safety (safe from violence including domestic violence), nor do different family / groups within the dwelling have exclusive access to kitchen facilities and a bathroom. The ETHOS typology characterises 'extreme' over-crowding as homelessness within the category of inadequate housing.

Subgroup members agreed with the characterisation of severe or extreme crowding as both a lack of social amenity and inadequate housing that potentially also places tenure at risk. Discussion focused on the operationalisation of the concept i.e. where to draw the line between crowding (and potentially "at risk of homelessness") and the more severe end defined as homelessness. Some members wanted further information about the ABS analysis of 2006 Census data in relation to 3 or more extra bedrooms needed and 4 or more extra bedrooms needed (as determined by the standardised measure of overcrowding, the Canadian National Occupancy Standard). Members sought to analyse the information to determine at what level of the standard crowding measure the line should be drawn at extreme or severe crowding that warranted classification as homelessness.

The definition paper presented by ABS struck the somewhat conservative boundary at 4 or more extra bedrooms needed when using Census data. The Information Paper - Methodology for Estimating Homelessness from the Census of Population and Housing (cat. no. 2049.0.55.001) provides further information about the boundary issues in overcrowding and its implications for homelessness measurement when using Census data.

Security of tenure element

All members agreed that no tenure, and lack of security of tenure, constitutes homelessness when coupled with no accommodation alternatives.

Subgroup discussion of the security of tenure element focused on the homelessness status of specific living situations, particularly transitional housing, and community and private boarding houses. The subgroup noted that tenure arrangements in transitional housing vary across, and sometimes within, states and territories. Some transitional housing models offer, in effect, security of tenure for a fixed, often medium term period (12 months or more), whereas others provide 3 month tenancies or less. Some members of the subgroup generally concluded that transitional housing that is stable, and is the pathway to long term housing, does not represent homelessness. While a legal requirement may exist for the issuing of 60 day notices to quit when a tenant first occupies the dwelling, when the intention and the practice is for much longer tenancy (usually until formal longer term arrangements are in place) then both security and stability of tenure are achieved.

A similar conclusion was reached on tenure in community run boarding houses.

However, while in some jurisdictions the minimum tenure in private boarding houses is 60 days, and the period for a notice to quit is 120 days, the subgroup did not consider this routinely delivered stability of tenure in all cases. Other members noted that, in some jurisdictions, promoting boarding house reform as a component of the solution to homelessness might be hindered by a definition that classified a type of dwelling as representing homelessness.

While the conceptual elements of the homelessness definition should be applied regardless of the particular terminology that might be used to describe a person's accommodation situation, the way in which the definition is operationalised in regard to tenure, may overstate or understate the extent of a person's lack of security, depending on the information that has actually been collected. The more information that is collected regarding tenure, the more likely a person would be classified correctly.

Other issues

ABS intends to publish, along side its estimates of homelessness, estimates for other aspects of marginal housing that fall just outside the definition of homelessness. For example, while the ABS definition of homelessness can be applied to people living in caravan parks - the conceptual elements would form the basis of classification - there is interest in marginal housing that is just outside the definition of homelessness, including for people living long term in caravan parks.

Previous PageNext Page

Bookmark and Share. Opens in a new window

Commonwealth of Australia 2014

Unless otherwise noted, content on this website is licensed under a Creative Commons Attribution 2.5 Australia Licence together with any terms, conditions and exclusions as set out in the website Copyright notice. For permission to do anything beyond the scope of this licence and copyright terms contact us.