Australia has one of the highest home ownership rates among OECD countries. In line with this, 67% of all Western Australian households in 2001 lived in a dwelling that was either fully owned or mortgaged. Historically, a key factor in this privatisation of housing stock has been a community-wide perception of home ownership as a primary means of enhancing economic status through the provision of secure and, over the longer-term, affordable housing. According to observations made by the ABS, equity accumulated in the family home represents the major part of household wealth for many people (ABS 1998: 154). As well as providing financial security for retirement and unemployment, this equity also yields other economic benefits such as collateral for loans. For all these reasons, home ownership has been, and continues to be, encouraged and promoted by governments.

Given the vital role played by home ownership in the financial developmental cycle of most Australian families, and the attempts by ATSIC over the years to raise the level of Indigenous home ownership, it is striking to note that only 2% of Aboriginal households in the Northern East Kimberley lived in a fully owned dwelling in 2001, while just 3% were resident in a mortgaged dwelling (Table 6.3). By contrast, 41% of non-Aboriginal dwellings in the region were either owned outright or mortgaged. Thus, Aboriginal households in the region remain overwhelmingly dependent on rented accommodation. While, on the one hand, this limits access to the property market for Aboriginal people as a means of improving their financial security, on the other hand it is symptomatic of their relatively low economic status, as well as indicating a cultural focus on communal forms of tenure.

It should also be noted that the non-Aboriginal rate of home ownership is also relatively low by Western Australian (and national) standards. The net effect of this is that the private housing market in the region is a fairly small one. No doubt a key factor in this is the temporary nature of much of the workforce—ironically many individuals who migrate to the East Kimberley for employment probably invest their earnings in property elsewhere, such as Perth. Whatever the case, the result is that only 599 dwellings in the region are privately owned, either outright or under mortgage, with barely 7% of these classified as Aboriginal dwellings. According to the 2001 Census, only 24 Aboriginal dwellings in the region are under mortgage—a remarkably low figure in the context of access to regional wealth accumulation.

According to 2001 Census data, a total of 1,182 dwellings in the region are rented, and Table 6.4 shows the distribution of these by landlord type. Clearly, in towns, Aboriginal people depend much more on State-provided rental housing than do non-Aboriginal people. More than half (55%) of Aboriginal dwellings in Kununurra, Wyndham and Halls Creek are rented from the WA Department of Housing and Works (DHW). This compares to only 21% of non-Aboriginal rental dwellings. The reason for this contrast is that non-Aboriginal residents have a wider range of rental options in town, the most common being to rent from an employer who often provides housing as part of a contract employment package in order to attract workers to the region. The other main source is the private rental market from which Aboriginal households are notably absent. Away from the towns, Aboriginal controlled housing organisations provide almost the entire stock of housing for Aboriginal people, whilst non-Aboriginal renters tend to acquire housing from an employer.

Given the more complex rental options evident in urban settings, it is interesting to compare these census data on rental accommodation in the three East Kimberley towns with tenancy data for June 2001 made available by the DHW. The comparison is provided in Table 6.5. According to the 2001 Census, a total of 544 dwellings in these three towns were rented from the State housing department or from a government employer. In effect, this amounts to the same thing since the DHW manages both sets of housing stock. However, in June 2001, DHW tenancy records revealed a total of 757 dwellings under these forms of tenure—a figure 28% higher than provided by the census. That the DHW figure is correct is undeniable as it is based on actual tenancies on the public record. Thus, something appears amiss with the census data. Two possible explanations suggest themselves.

The first possibility is that the census failed to count all such dwellings. While this could be expected to occur to a limited degree, it seems unlikely that it could happen on the scale suggested here, and so it is discounted. The second possibility is that confusion existed either among census interviewers (involved with the Indigenous Enumeration Strategy), census respondents, and census coders, or among all of these, in regard to the proper classification of rental (and possibly of any) tenure. For example, it might be that some households in DHW dwellings either classified themselves, were classified by collectors, or in data processing, as resident in other forms of tenure, with community housing, private rental, and home ownership being the only other options. If the problem is one of misclassification, then the true distribution of dwellings by tenure type is unknown. For this reason, DHW data are preferred for the analysis of urban rental dwellings.

The DHW funding arrangements provide for three types of urban rental outcomes—mainstream public rental, Aboriginal specific rental, and rental dwellings made available to State (or Commonwealth) public servants. Currently, a total of 493 mainstream and Aboriginal-specific public rental houses are available in the three East Kimberley towns—271 in Kununurra, 149 in Halls Creek and 73 in Wyndham. An additional 264 government employee houses are available. The distribution of mainstream and Aboriginal-specific housing in each town is shown in Table 6.6. Overall, 37% of the State rental housing stock is set aside for exclusive Aboriginal occupancy, although this proportion is much higher in Halls Creek than in the other two towns. However, this does not represent the full complement of Aboriginal public housing tenancies since Aboriginal households are also eligible to apply for and occupy mainstream rental housing. Unfortunately, the precise number that do is unknown since Aboriginal identification in administrative records is incomplete. From the June 2001 tenancy records, at least 54 mainstream rental dwellings (16% of the total) were occupied by self-identified Aboriginal tenants, but DHW estimates that the overall figure could be closer to 50%—indeed, if the Aboriginal share (83%) of applications for mainstream rental accommodation in June 2001 is anything to go by, it could be higher still. Thus, it is possible, though not certain, that as much as two-thirds of State public rental properties (325) are occupied by Aboriginal households.

Turning to the actual DHW tenants data, in February 2003 a total of 488 of the 493 rental dwellings were actually occupied and almost half of these (236, or 48%) were identified in DHW administrative records as Aboriginal dwellings (made up of the Aboriginal-specific rental properties and 54 of the mainstream rental properties). This proportion varied between the three towns, with 41% of the 271 State housing dwellings in Kununurra occupied by Aboriginal households, 59% of the 149 dwellings in Halls Creek, and 57% of those in Wyndham. This overall Aboriginal share of the State housing rental market is noticeably lower than the equivalent census-derived proportion of 63%. However, if the estimate mentioned above is accepted—that 50% of mainstream rental dwellings are occupied by Aboriginal households—then the resulting Aboriginal share of the public rental housing market (68%) is actually very close to the census figure and may provide some confirmation that at least the census proportional share in the form of tenure might be valid.

Clearly, there is imprecision in all this, yet some facts seem assured. First, Aboriginal families have fewer urban housing options than others. Second, their access to urban housing is dependent on continued expansion of State public housing stock, and a guaranteed major share of this. However, if the true size of the regional housing stock by tenure type remains unknown, especially in urban areas, then the estimation of future supply and demand for housing is rendered more difficult and less precise.