Social engineering and indigenous settlement: policy and demography in remote Australia.

Taylor, John

Abstract: In recent years neo-liberals have argued that government support for remote Aboriginal communities contributes to social pathology and that unhindered market engagement involving labour mobility provides the only solution. This has raised questions about the viability of remote Aboriginal settlements. While the extreme view is to withdraw services altogether, at the very least selective migration should be encouraged. Since the analytical tools are available, one test of the integrity of such ideas is to consider their likely demographic consequences. Accordingly, this paper provides empirically based speculation about the possible implications for Aboriginal population distribution and demographic composition in remote areas had the advice of neo-liberal commentators and initial labour market reforms of the Northern Territory Emergency Response been fully implemented. The scenarios presented are heuristic only but they reveal a potential for substantial demographic and social upheaval.

**********

In recent analyses of competing principles in Indigenous affairs policy, political scientist Will Sanders observes a paradigm shift in policy back to ideas of guardianship based on a conviction that governments can, and should, intervene to shape and enhance Indigenous participation and life circumstances (Sanders 2008, forthcoming). The most forceful expressions of this approach are found in the work on capabilities by Pearson (Cape York Institute for Policy and Leadership 2007) and Hughes (2007) on remote settlements. Policy-wise, it is most manifest in the package of measures first announced by the Howard government in relation to its Northern Territory Emergency Response (NTER) (Altman and Hinkson 2007; Sanders 2008; Commonwealth of Australia 2008). Accompanying this shift to what Sanders calls the 'directive right' of politics, there has also been a resurgence of right liberal thinking in which government support for remote Indigenous communities is seen as contributing to a social pathology that only unhindered market engagement involving labour mobility can resolve (Johns 2006a, 2007, 2009). Not surprisingly, this discourse on the right of politics has generated discussion regarding the viability of remote Aboriginal settlements. As a policy issue, this was first raised by former Indigenous affairs minister Vanstone (2005), while the most recent contribution argues for the withdrawal of services from locations that are not economically viable (Johns 2009). A critique is provided by Moran (2009).

All of this has stimulated interest in the issue of rural-urban migration. A good deal is known about the migration of Indigenous peoples. In 2004 a summary of the substantial body of research on Indigenous mobility in North America and Australia and New Zealand (Taylor and Bell 2004) concluded that a political economy framework was essential for analysis of this phenomenon--political, because of the influence of government policy in framing life course options for many Indigenous peoples, and economic because of the relatively limited participation of Indigenous people in mainstream institutions, either deliberately or otherwise. It is significant, then, that both of these influences have become elemental in recent years to Commonwealth government policy formulation around the notion of achieving 'practical reconciliation', now measured as a 'closing of the gap' between Indigenous and other Australians in employment, income, health, housing and education.

Thus, a common signal to Aboriginal people that became crystallised in the NTER was to embrace the institutions of mainstream Australian life with (it is argued here) potential migration-inducing implications. Of particular note here was the Howard government's plan, as part of the NTER, to scrap the Community Development Employment Projects (CDEP) program and lift remote area exemption for unemployed persons (Altman and Johns 2008). Also significant was the questioning of continued public support for remote settlements. However, events move quickly in Indigenous affairs and while much remains uncertain about the consequences of current (post-Howard) policy settings for settlement options, there does now appear to be a firmer commitment to infrastructure support and economic development, at least for 26 of the larger remote Aboriginal towns, as a consequence of the National Partnership on Remote Indigenous Housing. It is also the case that the Rudd government reinstated CDEP in NTER communities, albeit in a modified form. These developments aside, if we return to the original NTER measures, there did seem every possibility that these could result in the most profound alteration in Aboriginal population distribution in remote Australia since the impact of the pastoral industry award decision in the 1960s. Certainly there was a mood that change was imminent (Kilgariff 2007) and this posed key spatial questions--in which direction might policy-induced and market-driven redistribution occur, and to what extent?

To answer these questions, this paper reports some empirically based speculation made about the possible implications for remote area population distribution and demographic composition had the advice of neo-liberal commentators and the labour market reforms of the Howard government been fully implemented. These were first presented at a conference, Which Way? Directions in Indigenous Housing, which was sponsored by the Royal Australian Institute of Architects and held in Alice Springs in October 2007, just weeks into the implementation of the NTER measures. Using data available at that time (mostly from the 2001 Census), two scenarios were modelled based on policy prescriptions that had been clearly enunciated--one involving wholesale depopulation of the bush, the other involving partial depopulation based on labour market reform. Since the spatial and demographic tools were available for such analysis, it was felt that an exploration of likely outcomes for population distribution should form part of the test for integrity of these ideas, as some had already suggested (Phillpot 2007). All too often in Indigenous affairs, debate has tended to be reactive after the fact, and far less proactive in considering the consequences of actions taken before they become entrenched (Ah Kit 2004). What follows, then, is an attempt to demonstrate the possibilities that exist for thinking along the latter lines as an aid to rational policy making. In this context, it is salutary to note that the idea of understanding and anticipating population stocks and spatial flows as a fundamental principle and requirement of remote Indigenous settlement planning was spelled out more than a decade ago out in the pioneering work of Pholeros et al. (1993:22-3, 116) on housing needs, and yet policy development has tended to proceed with little quantification of possible demographic impacts.

The changing political economy of remote Aboriginal settlement

Some 40 years ago, towards the end of the assimilation era, Charles Rowley described the various mission and government settlements that were established across remote Australia for the purposes of welfare administration as having been instrumental in frustrating urbanisation (Rowley 1970). In his view, they functioned as 'holding institutions' serving to prevent the inevitable migration of Aboriginal people to towns and cities. 'A prime purpose of policy', he argued, 'must be to open up the way for Aboriginal re-location, resulting from Aboriginal choice' (Rowley 1970:29).

With the benefit of fairly long hindsight during which time Indigenous people have been free from the institutional and legislative shackles that governed their movements and places of residence--if you like, having now exercised Rowley's choice--the original proposition is seen to have been only partially upheld. While migration from the bush to towns and cities has undoubtedly occurred, the more striking observation concerning Indigenous population distribution of the past 40 years (precisely because it runs counter to expectations such as those expressed by Rowley) is that remote settlements have continued to grow in size, while population has also decentralised to out-stations on Aboriginal lands (Taylor 2006). During this time, the Indigenous share of the non-metropolitan, and especially remote area, population has steadily increased and it is projected to continue to do so (Brown et al. 2008). In effect, there has been considerable continuity of non-urban and remote residence across the country despite rising urbanisation and this has led to a very different distribution of Indigenous population to that observed for the majority population (Figure 1). This situation is best illustrated in the Northern Territory.

[FIGURE 1 OMITTED]

The Northern Territory experience

Since 1971, as the Indigenous population of the Northern Territory has grown, so too has its rural component. Indeed, the most striking feature of Indigenous residential distribution over recent decades has been its fragmentation and widespread dispersal into numerous family-based outstation groups--if anything, people were being held back from country, as much as from town. Having said that, there is no doubt that some movement into town has occurred, and continues to occur. This movement was stimulated initially by forced and spontaneous migration off stations and into burgeoning town camps following implementation of the Pastoral Industry Award from 1968. Since then, a common refrain of Northern Territory public policy discourse has been the notion of Indigenous urban drift.

While differential growth rates of Indigenous population estimates in the Territory's main urban centres over the past ten years would appear to support the notion of drift (3.3 percent per annum in urban areas compared to 1.5 percent per annum elsewhere), it should be noted that there is greater potential for new census identification to occur in urban counts due to higher intermarriage. This aside, the fact remains that the vast majority of Indigenous people in the Territory (65 percent) still choose to reside away from town. This is not to deny that many rural residents spend large amounts of time in town away from their home communities, but that is another story concerned mostly with periodic access to services, alcohol and kin (Commonwealth of Australia 2008: Appendix 9). Ironically, NTER-related health checks and police interventions could enhance this temporary movement to urban centres if they result in higher rates of urban-based hospitalisation and incarceration, since patients and inmates are frequently accompanied by kin from home communities who are intent on staying close to, and caring for, relatives (Coulehan 1995).

The facts concerning this continuity of remote settlement in the Northern Territory have long been known. Successive reporting in the form of community profile databases throughout the 1970s and 1980s by the Commonwealth Department of Aboriginal Affairs and the Aboriginal Development Commission, as well as by the Northern Territory Department of Community Development, then subsequently through the 1990s by the Aboriginal and Torres Strait Islander Commission-sponsored Community Housing and Infrastructure Needs Survey (CHINS)--first in 1992, then in 1999 and 2001--made it quite clear that remote settlements were proliferating in the Northern Territory and that some of these (Wadeye, Nguiu, Maningrida etc.) were growing into sizeable towns. Significantly, these same reporting mechanisms also revealed that levels of government-provided social, economic and environmental health infrastructure that would normally be directed towards a growing population base elsewhere in Australia (and especially to low socio-economic status populations that are heavily reliant on government services) were either absent altogether or were substantially inadequate in relation to expanding needs (Taylor 2007; Taylor and Stanley 2005).

The neo-liberal response

Not surprisingly, given the ideological shift in Indigenous affairs policy outlined by Sanders (2008), by 2005 questions regarding this relationship between population distribution and the structural economic position of Indigenous Australians finally caught the attention of government and various neo-liberal commentators. Their joint stance was, and to some degree remains, unequivocal. Either implicitly, or very explicitly, they required as a matter of policy that Indigenous people in remote communities be encouraged to migrate to mainstream labour markets located, for the most part, in regional urban centres. Select statements in regard to this are as follows:

... we need to think about the large numbers of very small settlements or homelands ... it is unlikely that many of these homelands will grow to become viable towns ... They raise some important issues for the future such as: How viable are they really? ... Perhaps we need to explicitly draw a line on the level of service that can be provided to homeland settlements ... No more cultural museums that might make some people feel good and leave Indigenous Australians without a viable future (Vanstone 2005).

The remote Aborigines are ... loaded with twin economic burdens: they inhabit regions that have no jobs or business opportunities, and the state gives them an income with no effort on their part. The only solution is to stop funding and thus close down all those settlements where unemployment is chronic and where there are no economic prospects, which is most of them (Windschuttle 2006).

If people want the goods and services that a modern economy can provide, they will have to generate an income in order to purchase them. In order to generate an income, they will have to work. If work is not available where they live, they will have to move to find it ...

While some shall be granted Disability Support or other such pensions, those judged to be able to work or train shall have to look for opportunities. In all likelihood there would be an enhanced out-migration from remote areas ...

The challenge for government is to stop funding programs that militate against the migratory solution ... [and] to manage those who are neither eligible for work or school ...

The extent to which Aborigines from remote regions will be more akin to refugees than migrants will be a measure of the difficulty of their adjustment to new circumstances.

Fortunately, Australia has vast experience in catering for both (Johns 2006a:7).

[Policy] changes will spark a move away from remote communities. Those people will create problems at the fringes of towns in Northern Australia and government must be prepared (Johns 2006b).

The underlying logic here seemed to be based on an assessment that life chances would be improved by the resettlement of people into fewer larger locations that have more services and job opportunities. Accordingly, the general intent would be to shift the light grey bars in Figure 1 to the left, with a corresponding flattening of the population share curve. Across northern and central Australia, given the lack of mainstream labour market opportunity out bush and without the dual prop of CDEP and remote area exemptions, and given the nature of longstanding intra-regional migration flows, any such redistribution would inevitably be towards regional centres. This relationship between settlement size and opportunity had been observed before and the question had been asked about the probable impact of seeking to shift people into larger communities--famously described by a former Minister for Indigenous Affairs (Brough) as 'living hell holes' (Altman 2006)--or into larger towns, like Alice Springs, for example, that have their own housing crises. To consider this latter question, the idea of Johns (2006a), that those judged able to work or train should migrate to look for opportunities, is simulated here to construct a series of scenarios of migration to Alice Springs based on different assumptions about workforce participation.

Modelling migration impacts: urban catchments in remote Australia

For many Indigenous people in remote communities, access to banks, hospitals, retail outlets, government offices, employment opportunities, sporting events, education and training institutions involves considerable travel to a regional service centre, though mostly as temporary sojourners rather than as migrants (Foster et al. 2005). The effect of this mobility is to create a functional pool or catchment of population around each service town. Some sense of the spatial pattern of these population catchments is provided by data from the 1999 Australian Bureau of Statistics (ABS) CHINS, which indicated the nearest town that members of each community in remote Australia travel to in order to access higher order services such as banking and retailing. The 1999 CHINS identified a total of 96 service centres across remote Australia servicing 1100 smaller communities with a collective population of 80 000 (Taylor 2002). Figure 2 provides an indication of the spatial pattern of centripetal flows into each centre from locations within their catchment areas.

As shown, these movements are not spatially random, and mobility occurs within well-defined spaces or networks defined mostly by historical administrative connection to particular centres, as well as the location of kin. Mobility fields of this type were first charted across central Australia by Young (1990) and Young and Doohan (1989), and most recently in the Mount Isa region by Memmott et al. (2006). The map highlights the major role played by Alice Springs in servicing vast areas of central Australia. In all, Alice Springs (population 24 000) services some 260 small Indigenous communities encompassing a combined estimated population of 15 000. Moving north, Katherine and Darwin emerge as other major regional centres of attraction, while Cairns stands out in north Queensland. In Western Australia, a string of smaller catchment areas is evident, especially in the Kimberley. In each case, the direction of movement for services is towards a local or regional centre, although not all populations access their nearest service centre. This is partly a function of variable transportation links, but in some instances it reflects patterns of cultural affiliation. It is important to know that the movement of Indigenous people away from remote areas to other parts of Australia is relatively small (Taylor 2006), as this suggests that the patterns shown in Figure 2 cover most of the rural-urban interaction that occurs for remote area dwellers.

[FIGURE 2 OMITTED]

Using these catchments and their (2001) populations as a framework for analysis, two scenarios are constructed.

* First, that there is a wholesale transfer of population from remote communities into respective regional urban centres. A further assumption is that migration to urban centres follows the well-established patterns of movement shown in Figure 2. On this basis, population figures for each remote locality provided by the ABS from the 2001 CHINS are used to derive estimates of the size of individual migration streams. This scenario is developed as a reality check on the sorts of suggestions made above that most, if not all, remote communities are unviable and should be shut down completely in the sense of having their services withdrawn.

* Second (and more plausible), that something akin to the suggestion of Johns (2006a) occurs, whereby what he describes as 'the able-bodied' are encouraged to move out of remote communities in search of opportunity. To operationalise this, we can invoke the original NTER strategy of transferring workers in remote communities off CDEP and into mainstream work in situ. The focus in this scenario is on Alice Springs and its hinterland in the Northern Territory. It is assumed that, as a consequence, the Indigenous mainstream employment rate recorded for the Alice Springs Indigenous Area in 2001 (27 percent) would apply across the surrounding Apatula Indigenous Area. This has the effect of increasing mainstream employment in the bush, though it falls far short of absorbing all unemployed persons. This surplus labour is then assumed to move into Alice Springs in order to comply with job search requirements.

Scenario 1

The population effect of evacuating remote communities and relocating their populations to regional centres that currently service them is shown in Table 1 for a selection of regional service towns in the Northern Territory and the Kimberley region of Western Australia. In each case, the percentage growth of Indigenous population would be substantial, exceeding 300 percent in the case of Alice Springs, with Indigenous numbers there rising from 4600 to more than 20 000. A further impact would be on the overall size of these centres--again, if we focus on Alice Springs, this would almost double in size to a town of over 41 000. An immediate implication is flagged in the last two columns, which show the average Indigenous housing occupancy rate as reported in the 2001 Census compared to what it would be post-migration in the absence of any additional housing provision to accommodate these new arrivals. If we take Halls Creek this time as an example, the relatively high occupancy rate already reported for that town would almost triple to a level seen only in some of the worst instances presently reported for remote communities (Taylor 2004). In fact, if a population transfer on the scale hypothesised in Table 1 was not accompanied by a massive urban housing construction program, then the effect would be to simply transfer into town the overcrowd ed dwelling conditions currently experienced in many discrete communities. Clearly, this would simply represent a relocation, rather than a resolution, to a housing crisis.

Another demographic effect of this population transfer would be to increase the Indigenous share of urban populations. Presently, only two of the selected towns (Halls Creek and Fitzroy Crossing) have majority (over 50 percent) Indigenous resident populations. These would be joined by two more towns (Derby and Tennant Creek), while Alice Springs and Kununurra would both achieve a 50 percent share. Elsewhere, Katherine, followed by Broome, would also be approaching a parity situation. Only Darwin, on account of its much larger total, would remain an overwhelmingly non-Indigenous town, although even here we are looking at an Indigenous population that would be almost one-fifth of the total.

Clearly, such a strategy, if ever enacted, would represent nothing less than a destruction of the base of the Australian settlement hierarchy across central and northern Australia and a radical social, economic and demographic transformation of key regional service centres, with massive policy implications, on a scale hitherto unknown . Perhaps alive to the absurdity of such a proposition, it is interesting to note that some nuance is added to neo-liberal prescriptions for social change. Johns (2006a:6), for example, envisages a situation where 'some shall be granted Disability Support or other such pensions, [and] those judged to be able to work or train shall have to look for opportunities'. Accordingly, he suggests, 'the challenge for government is to stop funding programs that militate against the migratory solution ... [and] to manage those who are neither eligible for work or school' (Johns 2006a:7). What would rural and urban populations look like if policy sought to encourage selective migration in this manner?

[FIGURE 3 OMITTED]

Scenario 2

It is apparent that the current government intends some continuation (even enhancement) of support for remote residence. However, as we have seen, a key part of the original NTER measures was the aim to transfer some 2000 CDEP participants into mainstream jobs and shift any surplus CDEP workers (the vast majority) into job search activities. One way to simulate the first part of this strategy is to suppose that the Indigenous mainstream employment rate in Apatula (6 percent) recorded by the 2001 Census rises to match that of Indigenous people in the urban setting of Alice Springs (27 percent). This latter is applied by age and sex to the Apatula population and any surplus CDEP workers and unemployed are required to migrate, together with a corresponding estimate of their children. The outcomes are shown in Figure 4.

While the level of movement to Alice Springs is reduced compared to the wholesale shift under scenario 1, there is still significant relocation, with the population of Apatula declining by 29 percent and that of Alice Springs increasing by 48 percent. Also of note is a rise in the sex ratio in Alice Springs from 0.99 to 1.07 and a corresponding decline in Apatula from 0.94 to 0.83. The obvious visual effect is to reverse the Indigenous demographic profile of Alice Springs and its hinterland. Whereas Apatula initially has a burgeoning school-age population and sizeable cohorts of youth and young adults, Alice Springs has more of an aged profile, with higher proportions in older age groups. Under the selective migration proposed here, these relativities are essentially reversed, as younger cohorts move at relatively higher rates than older ones, leading to a swelling in the ranks of children, youth, young adults and even older working age groups in Alice Springs.

The actual numeric and proportional changes envisaged for different age cohorts that are typically of interest in social policy are shown in Tables 2 and 3. Thus, the Indigenous population of Apatula declines overall by almost one-third, with losses in all age groups but especially among those of prime working age. Alice Springs, on the other hand, increases in population by almost half and experiences substantial additional numbers of school-age children, young adults and persons of prime working age.

[FIGURE 4 OMITTED]

Aside from the implications for housing construction and associated infrastructure to cater for additional families and young people, which has already been hinted at, there are considerable consequences from this modelling for the full range of social services provision in Alice Springs --for child health services, schooling, training and work programs, job creation, criminal justice and even aged care. Given the difficulties that have been encountered in Alice Springs in meeting demand for new housing (Kilgariff 2007; Tilmouth 2007), the pursuit of a migratory policy could produce demographic outcomes in such towns that place enormous pressure on social cohesion, to say nothing of the effects of diseconomies of scale created by depleted populations in rural communities.

Conclusion

While the aim of government policy is to encourage convergence in socio-demographic trends, the review of the NTER found little evidence of this after 12 months of implementation, at least in regard to migration and population distribution (Commonwealth of Australia 2008). Of course, key elements of the NTER (notably the abolition of CDEP) did not last the course and so the effects modelled here are no doubt diluted. It is equally the case that the sheer depth of human capital deficits and Indigenous preferences for non-urban livelihoods may defy attempts to stimulate migration for some time to come, while the Northern Territory government's recently announced remote service delivery framework (A Working Future, available at www.workingfuture.nt.gov.au/index. html), which will focus infrastructure and service investments on 20 select communities, may also distort the directions of movement and should itself be modelled.

In this context, it should be noted that under-investment and lack of access to services in remote communities have contributed over the years to high rates of rural-urban circular movement; the key difference in the current policy era, with its emphasis on mainstreaming, might be a switch from circulation to permanent migration, with major consequences for urban housing needs and other major services. Ironically, the historic lack of investment in remote communities is one of the reasons why many Indigenous people may have failed to respond to market stimuli for migration in the first place for want of the necessary capabilities to exercise such an option. Encouraging location closer to labour markets will not of itself do anything to overcome this fundamental shortcoming.

The scenarios presented here are heuristic only. They simply reflect the logic of following through on some of the suggestions that have been tabled to address development dilemmas for remote Indigenous communities. As such, they are designed to test the merit of policy prescriptions that have either been proposed for, or enacted in, remote areas, especially in the Northern Territory, and to demonstrate a capacity for debate around the statistical scale and composition of demographic change that they imply. What they indicate gives pause for thought.

First of all, the demographic and social upheaval of following through on ideas that remote communities should be abandoned are so great that such ideas are simply abandoned. A quick calculation of the demographic affect is all that it takes to draw this conclusion. As for other suggestions that policy should act as a stimulus to rural-urban migration, this is more in line with current practice but is less easy to model, given the lack of understanding of policy impacts on human behaviour and since key fundamentals that might impact on movement keep changing. However, there was sufficient intent in the labour market reform policies rolled out in the initial phase of the NTER to provide some guidance as to likely demographic outcomes if such courses of action were to be sustained. Based on assumptions that those not gainfully employed should (i.e. would be required to) seek opportunities elsewhere, the effect on population redistribution is still substantial and raises questions about the scale and nature of consequent infrastructure and service delivery requirements in both rural and urban areas if such an outcome was to eventuate. Of course, there is a very real sense in which these outcomes simply reflect the assumptions made, but the challenge for governments and others in any proposition for policy intervention is to demonstrate that things would necessarily be otherwise than modelled here and, if so, in what way, and to what extent.

REFERENCES

ABS (Australian Bureau of Statistics) 2002 Australian Indigenous Geographical Classification Maps and Census Profiles 2001, ABS, Canberra (Cat no. 4706.0.30.001).

--2008 Experimental Estimates of Aboriginal and Torres Strait Islander Australians, ABS, Canberra (Cat no. 3238.0.55.001).

Ah Kit, Jack 2004 'Why do we always plan for the past? Engaging Aboriginal people in regional development', paper presented at the Sustainable Economic Growth for Regional Australia (SEGRA) 8th National Conference, Alice Springs, 7 September.

Altman, Jon 2006 In Search of an Outstations Policy for Indigenous Australians, Centre for Aboriginal Economic Policy Research, The Australian National University, Canberra (CAEPR Working Paper 34).

--and Melinda Hinkson (eds) 2007 Coercive Reconciliation: Stabilise, normalise, exit Aboriginal Australia, Arena Publications, Melbourne.

--and Melissa Johns 2008 Indigenous Welfare Reform in the Northern Territory and Cape York: A comparative analysis, Centre for Aboriginal Economic Policy Research, The Australian National University, Canberra (CAEPR Working Paper 44).

Brown, Dominic, John Taylor and Martin Bell 2008 'The demography of desert Australia', The Rangeland Journal 30(1): 29-43.

Cape York Institute for Policy and Leadership 2007 From Hand Out to Hand Up: Cape York Welfare Reform Project, Cape York Institute for Policy and Leadership, Cairns.

Commonwealth of Australia 2008 Northern Territory Emergency Response Report of the NTER Review Board, Department of Families, Housing, Community Services and Indigenous Affairs, Canberra.

Coulehan, Kerin 1995 'Keeping company in sickness and in health: Yolgnu from northeast Arnhem Land and medical related transience and migration to Darwin' in G Robinson (ed.), Aboriginal Health: Social and cultural transitions, NTU Press, Darwin, p.7.

Foster, Denise, Julia Mitchell, Jane Ulrik and Raelene Williams 2005 Population and Mobility in the Town Camps of Alice Springs, Desert Knowledge Cooperative Research Centre, Alice Springs.

Hughes, Helen 2007 Lands of Shame: Aboriginal and Torres Strait Islander 'homelands' in transition, The Centre for Independent Studies, St Leonards, NSW.

Johns, Gary 2006a 'Social stability and structural adjustment', paper presented at the Bennelong Society Sixth Annual Conference: Leaving Remote Communities, Sydney, 2 September.

--2006b 'What is to become of Aborigines forced to move', <http://www.bennelong.com.au/articles/ johnsozoct2006.php>. Accessed 30 June 2009.

--2007 'The task ahead', presentation to the Bennelong Society Seventh Annual Conference: The Task Ahead, Melbourne, 2 September.

--2009 No Job No House: An economically strategic approach to remote Aboriginal housing, The Menzies Research Centre, Canberra.

Kilgariff, Fran 2007 'Alice at the tipping point', paper presented at the Bennelong Society Seventh Annual Conference: The Task Ahead, Melbourne, 2 September.

Memmott, Paul, Stephen Long and Linda Thomson 2006 Indigenous Mobility in Rural and Regional Australia, Australian Housing and Urban Research Institute, Brisbane.

Moran, Mark 2009 'What job, which house?: Simple solutions to complex problems in Indigenous affairs', Australian Review of Public Affairs, Digest March.

Phillpot, Stuart 2007 'The future of remote Aboriginal communities: A series of relic settlements of people created by the ebb and flow of contact with non-Aboriginals', paper presented at the Bennelong Society Seventh Annual Conference: The Task Ahead, Melbourne, 2 September.

Pholeros, Paul, Stephen Rainow and Paul Torzillo 1993 Housing for Health: Towards a healthy living environment for Aboriginal Australia, Health Habitat, Newport Beach, NSW.

Rowley, Charles 1970 The Destruction of Aboriginal Society, Penguin Books, Ringwood, Vic.

Sanders, William 2008 'In the name of failure: A generational revolution in Indigenous affairs' in C Aulich and R Wettenhall (eds), The Fourth Howard Government, UNSW Press, Sydney, pp.187-205.

--forthcoming Ideology, Evidence and Competing Principles in Australian Indigenous Affairs, Centre for Aboriginal Economic Policy Research, The Australian National University, Canberra (CAEPR Working Paper).

Taylor, John 2002 The Spatial Context of Indigenous Service Delivery, Centre for Aboriginal Economic Policy Research, The Australian National University, Canberra (CAEPR Working Paper 16).

--2004 Social Indicators for Aboriginal Governance : Insights from the Thamarrurr region, Northern Territory, ANU E Press, Canberra.

--2006 Population and Diversity: Policy implications of emerging Indigenous demographic trends, Centre for Aboriginal Economic Policy Research, The Australian National University, Canberra (CAEPR Discussion Paper 283).

--2007 'Demography is destiny--Except in the Northern Territory' in JC Altman and M Hinkson (eds.), Coercive Reconciliation: Stabilise, normalise, exit Aboriginal Australia, Arena Publications, Melbourne, pp.173-83.

--and Martin Bell 2004 Population Mobility and Indigenous Peoples in Australasia and North America, Routledge, London and New York.

--and Owen Stanley 2005 The Opportunity Costs of the Status Quo in the Thamarrurr Region, Northern Territory, Centre for Aboriginal Economic Policy Research, The Australian National University, Canberra (CAEPR Working Paper 28).

Tilmouth, William 2007 'Saying no to $60 million' in JC Altman and M Hinkson (eds), Coercive Reconciliation: Stabilise, normalise, exit Aboriginal Australia, Arena Publications, Melbourne, pp.231-8.

Vanstone, Amanda 2005 'Beyond conspicuous compassion', address to the Australia and New Zealand School of Government, The Australian National University, Canberra, December.

Windschuttle, Keith 2006 'Manhood whitewashed'. Australian, 23 May.

Young, Elspeth 1990 'Aboriginal population mobility and service provisions: A framework for analysis' in B Meehan and N White (eds), Hunter-Gatherer Demography: Past and present, The University of Sydney (Oceania Monograph 39), pp.186-96.

--and Kim Doohan 1989 Mobility for Survival: A process analysis of Aboriginal population movement in Central Australia, North Australia Research Unit, The Australian National University, Darwin.

John Taylor

Centre for Aboriginal Economic Policy Research, The Australian National

University

John Taylor is a Senior Fellow at the Centre for Aboriginal Economic Policy Research at The Australian National University. He is a social scientist who for 25 years has been engaged in applied research on the measurement and policy implications of demographic and socio-economic change among indigenous Australians at varying scales of analysis from the local to the regional and national.

<j.taylor@anu.edu.au>

Table 1: Indigenous and total populations of select regional service
centres in 2001 and post-Indigenous migration (customised data from
the ABS 2001 CHINS and ABS 2002)

 Initial Indigenous Percent Initial
 Indigenous population increase Total
 population post-migration Population

Darwin 9,691 20,451 111.0 106,476
Katherine 2,230 5,217 133.9 8,571
Tennant 1,176 3,617 212.2 3,182
 Creek
Alice 4,673 20,445 337.5 25,480
 Springs
Kununurra 755 2,987 295.6 3,678
Halls Creek 944 2,774 193.8 1,319
Fitzroy 750 2,399 219.8 1,139
 Crossing
Derby 1,159 3,089 166.5 2,950
Broome 2,514 4,845 92.7 10,487

 Total Indigenous Indigenous
 population housing housing
 post- occupancy occupancy
 migration rate 2001 rate post-
 migration

Darwin 117,236 4.0 8.6
Katherine 11,558 4.5 12.4
Tennant 5,677 6.0 18.7
 Creek
Alice 41,250 4.7 19.4
 Springs
Kununurra 5,910 4.7 17.8
Halls Creek 3,149 7.3 19.7
Fitzroy 2,788 5.3 16.9
 Crossing
Derby 4,880 4.5 11.4
Broome 12,818 4.5 8.0

Table 2: Post-migration change in the size of Aboriginal social
policy age groups: Apatula

Age group Population Change

 Now Post-migration Number Percent

0-4 849 602 -247 -29.1
5-14 1,871 1,328 -543 -29.0
15-24 1,739 1,420 -319 -18.3
25-49 2,667 1,728 -939 -35.2
50+ 968 658 -310 -32.0
Total 8,094 5,736 -2,358 -29.1

Table 3: Post-migration change in the size of Aboriginal social
policy age groups: Alice Springs

Age group Population Change

 Now Post-migration Number Percent

0-4 549 795 246 44.8
5-14 1,101 1,643 542 49.2
15-24 855 1,172 317 37.1
25-49 1,799 2,738 939 52.2
50+ 611 921 310 50.7
Total 4,915 7,269 2354 47.9