Appendix 9 - The demography of NTER prescribed areas
A report to the Board of Review of the Northern Territory Emergency Response
by John Taylor Centre for Aboriginal Economic Policy Research The Australian National University, Canberra
Migration and mobility
While Indigenous people in the Northern Territory move often between town and country for a wide variety of reasons, little statistical information is available to quantify the number, frequency, direction and duration of such movements. The five-yearly census provides information on more permanent changes of residence over one and five-year periods prior to the census. In 2006, these data indicated a net migration of Indigenous population from NTER prescribed areas (including town camps) to urban centres of the Northern Territory of 1,037 persons over the five-year period from 2001 to 2006. However, these same data indicated a net gain to prescribed areas of 1,685 migrants from other States. Consequently, the net movement of people actually contributed to population growth in prescribed areas over this period. It should also be noted that it is also the case that the average annual growth of Indigenous population resident in prescribed areas over the past 20 years has not been markedly different from that recorded in the Territory's urban centres (2.1% versus 2.6%). This is despite the greater potential for new census identification to occur in urban counts due to intermarriage.
Despite these statistics, a perception of 'urban drift', meaning a gradual relocation to urban areas of ruralbased Indigenous population, has been widely held in the Northern Territory since the 1980s and both expectations and observations of a rise in movement to urban centres were heightened by the NTER. Over the 12-month period of the NTER roll-out a number of Indigenous organizations, bureaucracies and media reported unusually high movement into towns both in the Northern Territory and in neighbouring jurisdictions. Not surprisingly, a number of submissions to the Review Board highlighted this issue as an unintended and problematic consequence of the NTER.
Accordingly, the Review searched for statistical evidence that might support the idea of heightened movement out of prescribed areas and into urban centres over the period of the NTER. In a report prepared by the Northern Territory Government Treasury (NTG 2008) various NT government and local agency data sources were examined for their tendency to display marked increases in service episodes in urban areas post-NTER. Among the data sets assessed were school enrolments, priority housing applications, community patrols of public places, sobering up shelter admissions and protective custody admissions variously for periods from 2003 through to 2008. All of these data reported aggregates of service episodes and at best are only proxy indicators of movement.
Across all of the measures considered, the results did not reveal a rise in urban service episodes over the past year that was out of the ordinary compared to the trends evident over the past five years.
The review therefore examined an additional source of administrative data in the form of Centrelink customer monthly change of usual residence address on the understanding that this might provide a more direct measure of rural-urban movement, at least that most likely to be associated with longerterm residential shift. These data clearly have shortcomings as they refer only to those Centrelink customers who report such changes and therefore they do not track non-compliant customers or those who are not registered with Centrelink. At the same time, given the extent of Indigenous registration with Centrelink the data refer to an average of around 24,000 Centrelink customers and if the NTER had stimulated a significant shift of population from rural settlements to urban areas then it is likely that this would show up in increased change of address notifications because of breaching rules built into the system that require this.
Monthly change of address notifications for Indigenous residents of communities on prescribed lands were analysed over a five-year period from August 2003 to August 2008. These were classified into changes of address that occurred between NTER prescribed areas and Northern Territory urban centres, as well as between prescribed areas and other States. What they indicate is a good deal of monthly churn between town and the bush but no clear evidence of an overall net shift of population from any one area to another. On average, each month over the five-year period, around 210 Indigenous Centrelink customers notified a change of address from one of the 73 prescribed communities either to an urban centre in the Northern Territory, to an otherwise non-prescribed area of the Northern Territory, or to an interstate destination. On average, over the same period, slightly fewer each month (196) notified a change of address in the reverse direction. This represents an average monthly churn of around 1.8% of elligible numbers.
From Figures 1 and 2 it is clear that these flows experience peaks and troughs such that in some months the numbers moving in either direction can be substantially higher than in others. It is also clear from the trend lines shown in Figures 1 and 2 that the numbers involved in both flows has gradually increased over time and so while there is certainly more movement into urban areas over time, this appears part of a steady trend that has built over the period in question (the last five years) and it is matched over time by a roughly equivalent pattern of movement back again. As for flows interstate, these are fewer and more stable over time with movements out of prescribed areas once again matched by movements in (Figures 3 and 4). In short, the best available source of administrative data with which to consider the question of urban drift does not support the proposition, either before, or subsequent to, the NTER.
This is not to deny the possibility of an increased presence in urban areas, nor one that has been enhanced by the implementation of NTER measures—it is more to suggest that a quantum residential shift of population is not in evidence.
Many submissions to the Review commented on this subject, although in many cases this was based on opinion only. Among those providing hard evidence was the Larrakia Nation Aboriginal Corporation. They documented an unusual influx of Indigenous visitors from remote areas of the Territory to Darwin in July 2007. A conservative estimate revealed an additional 150 Long Grass residents increasing the base-line population by around 40%. Furthermore, they report that the announcement of the NTER was a key contributor to this increase, influencing Aboriginal peoples' decision to leave their home communities and stay in Darwin's Long Grass. Reasons given for such movement were many and interconnected. They included perceived loss of rights, autonomy and purpose; an increase in violence and conflict in communities; fear and confusion over the new laws; a lack of future; and restricted access to alcohol. While reasons for movement into town are always complex, two particular issues stand out from submissions to the Review.
Figure 1: Monthly Indigenous change of address notifications from urban NT to NTER communities: 2003–08
Figure 2: Monthly Indigenous change of address notifications from NTER communities to urban NT: 2003–08
Figure 3: Monthly Indigenous change of address notifications from NTER communities to other states and territories: 2003–08
Figure 4: Monthly Indigenous change of address notifications from other states and territories to NTER communities: 2003–08
First, it is noted that many people have been motivated in their search for alcohol to explore options beyond prescribed areas where alcohol bans are now in place and that this has led them inevitably into less regulated town areas in greater numbers than previously. A submission from the Umoona Community Council Inc. at Coober Pedy provides compelling evidence to this effect. They report, with supporting statistics, a very marked increase in Sovering Up Centre and Mobile Assisted Patrol Service episodes each month since July 2007 compared to equivalent months in the previous year. They also report that this reflects an influx of people from the APY lands who would normally have traveled to Alice Springs for alcohol. The effect of this has been to almost double monthly demand for Patrol Service activity from 860 clients in 2006–07 to 1,520 clients in 200–08.
Second, the imposition of income management appears to have impacted on the increased presence of people in urban centres in two ways—by restricting quarantined purchases to town-based stores and by limiting the discretionary capacity of people to purchase fuel. Thus, in Tennant Creek where people used to 'come for day trip or the weekend and go', now they 'can't spend their money in communities or in roadhouses because they got to come into the main shop here in Foodbarn' and 'got nowhere to stay because that income management brought them into town and they can't leave, they say I got to wait and sort this out first for money side'1 (submission by Barbara Shaw). In other cases when people come into Tennant Creek, especially after the Show 'their money being income managed was in their community store in the remote community, and in town they can't access it. So people were stranded here'.2 (submission by Barbara Shaw).
On the subject of the NTER and urban drift, what these collective data and voices suggest is not so much a structural relocation from the bush into urban areas—the NTER has not brought about wholesale shift of people into town—rather, it has brought about heightened dislocation and inconvenience. Aboriginal people in the Northern Territory have always been mobile to access services, shopping, and alcohol. The NTER introduced a major shock to the existing order and one of the effects appears to have been increased itinerancy in urban centres.
Population estimates and projections
Because of the operational nature of the NTER as a place-based set of initiatives, little reference is to be found within official reporting regarding the numbers and categories of people targeted. This is ironic since the Intervention measures are ultimately about people (especially children), rather than places, yet no overall sense emerges about how many children reside in prescribed areas, what their ages are, and whether their numbers are increasing or in decline. The same can be said for their parents and other adults. In short, even 12 months on, we still have only a cursory and fragmentary sense of the size and composition of the population that has been, or was intended to be, subject to NTER measures. This is more than just an inconvenience, it goes to the heart of determining the adequacy of the scale and nature of program responses, and it is fundamental in determining rates of population coverage for particular measures and how these have changed.
This is not to say that population data were not compiled. One of the initial tasks of the NTER Evaluation Strategy was to compile Community Profiles for each of the prescribed communities. These included population counts from the 2006 Census although no reference was made at the time to the 19% undercount of the Northern Territory Indigenous population as this only became public knowledge in mid-August 2007 (this estimate was later revised by the ABS to 16% in August 2008). Just how systematic and comprehensive this process was remains uncertain. Indeed, a major difficulty for the Review has been the lack of a consistent set of population data in respect of prescribed areas and their constituent settlements. Community Status Reports made available to the Review contain a mix of census figures, population numbers from the Community Housing and Infrastructure Needs Survey (CHINS), and local estimates (of uncertain origin), and it is not always clear whether associated outstation numbers are fully accounted for. In short, if the question was asked at any time over the stabilization phase of the NTER as to the size and composition of the population in prescribed areas, then the answer could only have been approximate at best.
To be fair, there is a sense in which approximation is inevitable. Apart from the problem of census undercount, it is a mistake to assume some definitive distinction between Indigenous people resident in prescribed areas and those in the rest of the Northern Territory as well as in neighbouring parts of other jurisdictions. Given the frequent movement of people between towns, town camps and country areas, and given the density of social networks that connect them, to talk of prescribed and nonprescribed populations is a false binary. Nonetheless, certain aspects of the NTER (notably income management) are triggered by address of usual residence, and NTER measures are clearly focused on prescribed areas, so an important task for the Review Board was to establish best estimates of the numbers of people implicated.
It is possible, using the Australian Geographical Standard Classification, to construct a statistical geography that matches the prescribed areas of the NTER, Using a composite of Census Collection Districts (CCDs) selected to match prescribed areas and town camps, the ABS produced a customized Indigenous population estimate for June 2006 of 44,229. This was based on the final re-based ABS series of Indigenous estimates released in August 2008 (ABS 2008). Because of population growth since June 2006, it is necessary to project the prescribed area population to match the NTER period. Table 1 shows the results of such projection to 2008 for critical sub-groups in the population.
| Target populations | Number | Per cent |
|---|---|---|
| 0–3 (Infant) | 4,166 | 9.1 |
| 4–5 (pre-school) | 2,408 | 5.3 |
| 6–15 (compulsory school age) | 9,811 | 21.5 |
| 0–15 (child health checks) | 16,386 | 35.9 |
| 10–20 (diversionary programs) | 10,558 | 23.1 |
| 15–24 (school to work) | 9,200 | 20.2 |
| 16+ (adults) | 29,268 | 64.1 |
| 25–49 (working age) | 15,998 | 35.0 |
| 50+ (aged) | 5,026 | 11.0 |
| Total prescribed areas | 45,654 | 100.0 |
Based on projection from 2006 ABS customised ERP
In 2008, the Indigenous population of the areas prescribed by the NTER is estimated to be 45,654. This is substantially higher that the figure of 35,929 cited in the NTER Monitoring report of August 2008 (OIPC 2008: 10). Of this number, 36% (16,386) are children aged 0-15 years. These children can be disaggregated into different age groups relevant to the various aims of the NTER-almost 10% of the prescribed area population are estimated to be infants (4,166), a further 5% are of pre-school age (2,408) and over one-fifth are of compulsory school age (9,811). Almost two-thirds of the population are adults (29,268), and just over half of these (15,998) are of prime working-age. In line with continuing high adult mortality, aged persons over 50 years are relatively few.
By 2021, the Indigenous population of prescribed areas is projected to reach 54,766, an increase of 9,112 of 20%. This projection is based on assumptions that current fertility and mortality remain constant and that there is no net migration loss from prescribed areas. Since all of these assumptions may vary, the projected figures shown here are indicative only and designed to stimulate future-thinking supported by some sense of likely outcomes.
Figure 5 shows the age distribution of this possible future population compared to the original population in 2006. It points to sustained growth at younger ages but with the greatest increase in numbers at older ages over 35 years.
Figure 5: Indigenous population distribution by age: prescribed areas of the Northern Territory 2006 and 2021
Table 2 shows what this means in terms of the likely future size of social policy target populations. Comparison with Table 1 shows that the number of children will increase by over 1,500 but that children as an overall share of the population will decline from 36% to 33%. Consequently adults will be more prominent and this is especially the case among aged persons over 50 years of age.
| Target populations | Number | Per cent |
|---|---|---|
| 0–3 (Infant) | 4,562 | 8.3 |
| 4–5 (pre-school) | 2,614 | 4.8 |
| 6–15 (compulsory school age) | 10,767 | 19.7 |
| 0–15 (child health checks) | 17,942 | 32.8 |
| 10–20 (diversionary programs) | 11,829 | 21.6 |
| 15–24 (school to work) | 10,306 | 18.8 |
| 16+ (adults) | 36,824 | 67.2 |
| 25–49 (working age) | 19,697 | 36.0 |
| 50+ (aged) | 7,909 | 14.4 |
| Total prescribed areas | 54,766 | 100.0 |
Based on projection from 2006 ABS customised ERP
Changing settlement pattern
This scenario of growth in the population of prescribed areas is consistent with the experience of the past 20 years. However powerful the perception of urban drift in the Northern Territory, the fact is average annual growth of the Indigenous population resident in the Territory's main urban centres of Darwin, Katherine, Tennant Creek and Alice Springs over the past 20 years has not been markedly different from that recorded in the rest of the Territory (2.6% versus 2.1%). This is despite the greater potential for new census identification to occur in urban counts due to intermarriage.
Of interest to the Review is the degree to which this population growth away from the main urban centres of the Northern Territory is itself leading to urban development in the form of emerging large population clusters, many of which are former mission and government settlements on Aboriginal lands. To examine the evolution of remote area settlement, information on the size of localities in prescribed areas was derived from the 1986 Aboriginal Communities data base of the Northern Territory Government Department of Community Development and compared with results from the ABS 2006 Community Housing and Infrastructure Needs Survey updated in part with information from NTER Government Business Managers. The results are illustrated in Figure 6.
Figure 6: Rank size settlement distribution in prescribed areas: 1986 and 2008
This reveals a substantial expansion of population in medium and large settlements of 100 persons or more. So, a place like Palumpa, which was an outstation of 120 persons in 1986, is now a small town of 430, while Gunbalunya, with a population of 550 in 1986, is now an urban centre town of over 1,000. At the same time, a number of places have declined in population and there has been a proliferation of very small family-based outstations many of which are only intermittently occupied. The overall effect, then, has been steady growth in situ with the emergence of a dispersed network of service centres. Twenty years ago, there were only 3 Aboriginal towns with a population over 1,000 covering barely 12% of the remote area population. According to the latest information from CHINS and Government Business Manager reports there are now 10 such towns covering more than one quarter of the prescribed area population with four more settlements lining up to join them (Angurugu, Numbulwar, Yuendumu and Ramingining). This represents a major shift in the nature of places that people live in and it creates a rapidly changing set of opportunities for development and service delivery.
While the Board recognises that the quality of population data at the level of individual communities is questionable, nonetheless it can be used to gain an overall picture of the living arrangements of Aboriginal people by distributing the various localities within the prescribed areas according to their reported population size. This is done for 2008 in Table 3 and comparison is made with the situation 22 years earlier in 1986 to get a sense of how things have changed.
| >1,000 | 500–999 | 50–499 | 0–493 | |||||
|---|---|---|---|---|---|---|---|---|
| No. | Pop | No. | Pop | No. | Pop | No. | Pop | |
| 1986 | 3 | 3,809 | 11 | 6,909 | 112 | 14,784 | 515 | 6,878 |
| 2008 | 10 | 16,234 | 9 | 6,345 | 122 | 18,772 | 607 | 5,738 |
Note: Populations reported here are estimates only based on DCD field officer assessments in 1986 and a mix of 2006 CHINS and 2008 NTER Community Status Reports
The emergence of towns over the past 20 years in the areas now prescribed for the NTER is striking. Barely 4,000 people lived in such places 20 years ago compared to more than 16,000 today. The medium-sized places (500-999) seem to have remained fairly stable (although a number of these have grown into towns) whereas smaller settled places from 50 persons up 500 have also expanded both in both number and population. This points to substantial momentum for further growth in the settlement hierarchy from a base of 122 localities. Although the final category of very small places appears to have expanded in terms of the number of localities, the fact that the population is now lower compared to 20 years ago indicates that many of these places are effectively abandoned. This partly reflects a rationalisation over that time in the nature of funding for small outstations.
Projection methodology
State and local government planning authorities routinely develop future population scenarios to determine budgetary allocations on the basis of anticipated needs. A key part of this process is the production of small-area population projections or forecasts. While the ABS provides official Indigenous projections of State and Territory populations, these will not be available until mid- 2009. A standard cohort-component methodology is applied for this purpose and this practice is adopted here to project the Indigenous population of the NTER areas to 2021.
The cohort-component method carries forward the 2006 Indigenous ERP for the NTER area to 2021 by successive five-year periods. The projection is based simply on ageing the population by five-year blocs, subjecting each group to age- and sex-specific mortality, fertility and net migration regimes according to the following assumptions using published data that were available at the time of the review:
- Survival rates from the latest Indigenous life tables for the Northern Territory (ABS 2007) are applied and held constant for the projection period. Ideally, a life table that is more customised to reflect the particular mortality profile of the NTER communities would be applied but this is currently not available.
- Age specific fertility rates based on births between 2001 and 2004 to Indigenous women in the Northern Territory Midwives Notification System are applied. These data produce a Total Fertility Rate of 2.5. These ASFRs are held constant for the projection period. Ideally, ASFRs that reflect the particular fertility profile of the NTER communities would be applied but these are currently not available.
- Migration is the most difficult to measure and yet most crucial component of regional population change in the sense that it has the potential to have the greatest demographic impact. One complicating issue for the Indigenous population is the prevalence of short-term circular movement in the overall context of total mobility. While 2006 census data indicate a net loss from NTER areas to non-NTER areas of the Northern Territory of 207 Indigenous persons per annum over the five-year period between 2001 and 2006, they also indicate a net gain to NTER areas from other states of 337 Indigenous persons per annum. If we add to this the evidence from Centrelink change of address data that points to considerable churn, but no net relocation, it is prudent to set the net migration assumption to zero at all ages.
Against these parameters, the projection is conducted separately for males and females in five-year blocs from 2006–2021. Projected births for the 2006–2011 period are added to the existing 2006 population and each cohort is then subjected to respective survival and net migration rates to arrive at an estimate of the population in each age-group in 2011. This process is continued through to 2021.
References
Australian Bureau of Statistics (ABS) 2007. Deaths, Australia 2006, cat no. 3302.0, ABS, Canberra.
Australian Bureau of Statistics (ABS) 2008. Experimental Estimates of Aboriginal and Torres Strait Islander Australians, June 2006, cat no. 3238.0.55.001, ABS, Canberra.
Northern Territory Government (NTG) 2008. Urban Drift Data Sources Review: Darwin Region, Social Analysis Unit, Northern Territory Treasury, Darwin.
- Intervention Rollback Action Group Submission
- Ibid
- Includes locations with some infrastructure. In 1986, only 374 of the 515 listed had a recorded estimate of some resident population. In 2008, only 348 of the 607 listed had a recorded estimate of some resident population. As far as can be established all remaining localities in this category had no resident population.