Counting the homeless 2006: Western Australia

1 Definition of homelessness

The Australian Bureau of Statistics (ABS) uses the cultural definition of homelessness to enumerate the homeless population on census night (Chamberlain and MacKenzie 1992). This definition distinguishes between people in primary, secondary and tertiary homelessness.

Primary homelessness describes the situation of all people without conventional accommodation, such as people living on the streets, sleeping in parks, squatting in derelict buildings, living in improvised dwellings (such as sheds, garages or cabins), and using cars or railway carriages for temporary shelter.

Secondary homelessness describes the situation of people who move frequently from one form of temporary shelter to another. On census night, all people staying in emergency or transitional accommodation provided under the Supported Accommodation Assistance Program (SAAP) are considered part of this category. Secondary homelessness also includes people residing temporarily with other households because they have no accommodation of their own, and people staying in boarding houses on a short-term basis, operationally defined as 12 weeks or less.

Tertiary homelessness describes the situation of people who live in boarding houses on a medium to long-term basis, operationally defined as 13 weeks or longer. Residents of private boarding houses are homeless because their accommodation does not have the characteristics identified in the minimum community standard (Chamberlain and MacKenzie 1992): they do not have a separate bedroom and living room; they do not have kitchen and bathroom facilities of their own; their accommodation is not self-contained; and they do not have security of tenure provided by a lease.

2 Overcounting and undercounting

Chapter 2 summarises how the national homeless count enumerated the homeless population using census and other data sets. It contains a discussion of how there can be both overcounting and undercounting of homeless people. Undercounting is most likely in the census category 'improvised homes, tents and sleepers out', and overcounting is more likely in boarding houses because of misclassification.

The problem of establishing reliable figures is compounded by the fact that the homeless population changes over time. There will always be people who are entering and leaving homelessness, as well as people moving between different locations. The challenge is to identify patterns in the population data that might inform the policy process.

3 Accommodation on census night

The homeless population in Western Australia was distributed differently from the national homeless population (Table 1). Nationally, 20 per cent of the homeless were in boarding houses on census night whereas the comparable figure was 12 per cent in Western Australia. Across Australia, 19 per cent of the homeless were in SAAP accommodation, but in Western Australia the figure was 11 per cent. There were significantly more people staying temporarily with other households in Western Australia (59 per cent compared with 45 per cent across Australia), and there were more people in improvised dwellings, tents or sleeping out (18 per cent compared with
16 per cent nationally).

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

4 Age distribution

The age profile of the homeless population in Western Australia was significantly younger than the age profile of the national population (Table 2). Sixty-two per cent of the homeless in Western Australia were aged 34 or younger compared with the national figure of 58 per cent. One-third (32 per cent) of the homeless in Western Australia were teenagers aged 12 to 18 (mainly on their own). Nine per cent of the homeless were children under 12 who were with one or both parents. Another eight per cent were young adults aged 19 to 24, and 13 per cent were adults aged 25 to 34.

Altogether, 38 per cent of the homeless in Western Australia were aged 35 or older, compared with the national figure of 42 per cent.

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

5 Males and Females

In 2006, men outnumbered women in the national homeless population, 56 to 44 per cent (Table 3), and in Western Australia men outnumbered women, 54 to 46 per cent. In Western Australia, there were more females in the 12-to-18 age group (53 to 47 per cent), but more males in the 19-to-24 age group (55 to 45 per cent). From age 25 onwards, men outnumbered women, about 60 to 40 per cent.

Table 3: Percentage of males and females by age group

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

6 Indigenous and Non-Indigenous

In Western Australia, 3.2 per cent of people identified as Indigenous at the 2006 Census. Table 4 shows that Indigenous people made up 3.9 per cent of people staying with other households, 11.8 per cent of persons in boarding houses, 18.8 per cent of those in improvised dwellings and 40.3 per cent of people in SAAP. Indigenous people were overrepresented in all sections of the homeless population in Western Australia.

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

* Figures have been adjusted for missing data on Indigenous status, except in 58 cases where there was inadequate information to make the adjustment.

7 Perth

In Perth, the rate of homelessness was 109 per 10 000 in the Central Metropolitan subdivision, where there were 1368 homeless people (Table 5). Central Metropolitan had nine per cent of Perth's population, but it contained 30 per cent of Perth's boarding house residents, 40 per cent of its SAAP residents, and 35 per cent of persons in the improvised dwellings category. It is common to find a higher rate of homelessness in the inner suburbs of the capital cities. This is the case in Sydney, Melbourne, Brisbane, Adelaide and Hobart.

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

The rate of homelessness was 34 per 10 000 in East Metropolitan where there were 843 homeless people (Table 5). The rates were 38 and 39 per 10 000 in the South East and North subdivisions, where there were 1247 and 1719 homeless people respectively. In South West Metropolitan, the rate was 50 per 10 000 (1543 homeless people).

Altogether, there were 5352 homeless people in suburban Perth compared with 1368 in Central Metropolitan (Table 5). The rate of homelessness was lower in the suburbs, but the number of homeless people was substantial. People often become homeless in the suburbs or in regional communities, but they migrate to the inner city. The provision of services in suburban areas assists people in the early stages of homelessness, including those at risk, and reduces the move to the inner city.

8 Regional and Remote

There are eight statistical divisions covering regional and remote Western Australia. Regional and remote Western Australia has a population of 514 000 people, spread across a large geographical area. Some statistical subdivisions have high rates of homelessness, but relatively few homeless people.

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

The overall picture is summarised in Table 6. There were 6671 homeless people in regional and remote Western Australia, where the rate was 131 per 10 000. For some policy purposes, marginal residents of caravan parks are considered part of the tertiary population. If these residents are included, the rate of homelessness was 50 per 10 000 in Perth and 160 per 10 000 in regional and remote Western Australia (Table 6).

Table 7: People in different sectors of the homeless population, Perth and Regional/Remote Western Australia

Source: Census of Population and Housing 2006; SAAP Client Collection 2006; National Census of Homeless School Students 2006.

Table 7 shows that there were twice as many people in boarding houses in Perth, compared with regional and remote communities (1097 as opposed to 555) and three times as many people in SAAP (1070 compared with 325). In regional and remote Western Australia, more people were staying with other households than in Perth (4166 versus 3786) and twice as many people were in improvised dwellings, tents or sleeping rough (1625 compared with 767).