Indigenous Socioeconomics Indicators, Benefits and Expenditure


Social Policy Group

Indigenous Socioeconomic Indicators

Last updated 7 August 2001

This paper is a companion paper to two other briefs prepared by the Parliamentary Library's Information and Research Services: Indigenous Affairs Expenditure and Indigenous Individual Benefits.

Comparisons between the socioeconomic indicators of Indigenous and non-Indigenous people have become commonplace in public debates. The following paper takes as its starting point the often-used 'snap-shot' type comparisons and then moves on to consider the limited value of such comparisons and the need for tracing trends in indicators if some correlation is to be hazarded between inputs by way of Government expenditure and outcomes by way of improved socioeconomic status. Problems with tracing such trends are examined before finally moving on to attempting to identify trends in a series of socioeconomic indicators.

Snap-Shot Comparisons

There are many sets of 'snap-shot' comparisons between Indigenous and non-Indigenous socioeconomic indicators available. See, for example, Aboriginal and Torres Strait Islander Social Justice: Statistics, a compilation published on the website of the Human Rights and Equal Opportunity Commission, or the chapter Social Conditions of Aboriginal and Torres Strait Islander People in the Australian Bureau of Statistics publication, Australian Social Trends 2000 (Catalogue no. 4102.0). Professor Jon Altman of the Centre for Aboriginal Economic Policy Research summarised some key data in the table reproduced below:

Table 1: A Synoptic View of Socioeconomic Differences Between Indigenous and Non-Indigenous Australians, 1996

Social Indicator

Indigenous (1)

Non-Indigenous (2)

Ratio (1)/(2)

Unemployment rate (%)




Employment rate (%)




Labour force participation (%)




Occupation unskilled (labourers) (%)




Managers, administrators, professionals




Median income, adults (per week)




Median income, families (per week)




Income less than $200 (per week) (%)




Income more than $800 (per week) (%)




Currently renting (%)




Home owner or purchasing (%)




Household size




Did not go to school (%)




Left school aged <15 years (%)




Currently attending tertiary institution aged 15–24 years (%)




Post-school qualification




Male life expectancy at birth (years)




Female life expectancy at birth (years)




Population age over 55 years (%)




Note: Indigenous households are defined as households in which the reference person or the reference person's spouse is Indigenous. It should also be noted that is some circumstances home ownership is not possible for Indigenous people owing to the communal nature of land tenure.

Source: J. C. Altman, The Economic Status of Indigenous Australians, Centre for Aboriginal Economic Policy Research, Discussion Paper, no. 193, 2000, p. 6.

More recently still the Commonwealth Grants Commission inquiry into the distribution of funding for programs that affect Aboriginal and Torres Strait Islander peoples has been attempting to define in more detail outstanding Indigenous needs and to develop measures of relative disadvantage that can be used to target resources more effectively. The Commission's Indigenous Funding Inquiry Final Report is available on the Internet.

Complicating Factors

Though Indigenous/non-Indigenous comparisons are useful for identifying continuing disparity in needs, and though identification of needs may be essential to better targeting resources, neither by themselves tell us a lot about the success or otherwise of Government programs and efforts. Insight into the latter can only be gained by tracing trends in Indigenous socioeconomic indicators. Plotting such trends is not, however, easy. There are at least half a dozen complicating factors.

Data of Variable Quality

Often the necessary comparable longitudinal data is either not available or not available across regions. As Gray and Auld suggested in Towards an Index of Relative Indigenous Socioeconomic Disadvantage (Discussion Paper, no. 196, Centre for Aboriginal Economic Policy Research, 2000, p. 14):

When analysing the changes in ranking according to relative socioeconomic disadvantage it is critical to bear in mind that while changes may be due to real changes in relative socioeconomic disadvantage, they may also be a product of variable data quality, both across regions and between censuses. The sensitivity of the ranking to data quality means that small changes in ranking between 1991 and 1996 should not necessarily be interpreted as a change in the overarching socioeconomic disadvantage.

The Commonwealth Grants Commission Indigenous Funding Inquiry Final Report, has recently made similar observations though they have attempted to arrive at preliminary estimates of relative Indigenous socioeconomic advantage and produced a map based on these figures.

The Trend Context

That which may appear to be a trend in one direction, often proves upon closer examination to be a trend in another (for example, absolute numbers may increase but those same numbers relative to Indigenous population or relative to the relevant number for the total population may be decreasing). The position of Indigenous people may be improving on some scores by some indicators, but, if (as is often the case) it is not improving at the same rate as the wider population then the gap between the socioeconomic status of the two groups may be widening rather than closing. Similarly, the position of Indigenous people may be improving relative to the total population by some indices (e.g. Altman and Hunter calculating in 1998 that relative to the total population, Indigenous poverty has declined from a factor of 2.7 to 3.7 in the early 1970s to a factor of 2.0 in the 1990s), but there may be other considerations to take into account. For example, the status of the total population by those indices may not be improving (this may be the case in the difficult area of poverty. See Geoff Winter, 'Measuring the Numbers of People in Poverty', Research Note, no. 31, Department of the Parliamentary Library, 1999–2000). The Indigenous people may be coming off a particularly low base with respect to that indice; the improvement is largely the result of greater access to welfare payments or work for the dole, rather than engagement in the non-welfare mainstream economy.

Even if some indigenous socioeconomic indicators are found to be improving, they may not be improving as markedly as those same indicators in comparable overseas countries. For example a recent study, (M. Moran, Housing and Health in Indigenous Communities in the USA, Canada and Australia: the significance of economic empowerment, Aboriginal Environments Research Centre, University of Queensland, 2000) has suggested that the socioeconomic status of Indigenous people in Australia compares unfavourably with that of those in Canada (and the US). The report identifies two possible reasons for this. Firstly, greater weight is given in Canada and the US to recurrent housing funding and innovative ways to improve home ownership (e.g. in the US programs, to use local contributions of labour and material as 'sweat equity' and to favour the Indigenous enterprises for construction programs). Secondly, in Canada and the US responsibility for Indigenous Affairs is structured differently to in Australia—with States and Provinces having little role in what is effectively a two-tier system of Federal and Tribal governments.

The Broadening of the Group Identifying as Indigenous

The number of people who identified as Indigenous increased from 265 458 (or 1.6 per cent of the population) in the 1991 Census to 352 970 (or 2.0 per cent of the population) in the 1996 census. This increase of 33 per cent is far in excess of the rate of natural increase of either the Indigenous or total population. The major component of the increase appears to be children of 'mixed-couples', who in terms of their parents' identity may equally well identify as non-Indigenous, identifying or being identified as Indigenous. Indeed, according to recent censuses the proportion of couples (whether married or de facto) composed of one indigenous and one non-indigenous partner, has increased from 46 per cent in 1986 to 51 per cent in 1991 and 64 per cent in 1996. As the rate of 'marrying out' can be identified as higher in the more metropolitan areas it is not surprising that the increase in the indigenous-identifying population since 1991 was highest in the relatively more urban jurisdictions of the ACT and Tasmania (79.7 per cent and 56.3 per cent respectively) and lowest in the less urbanised jurisdictions of the Northern Territory and Western Australia (16.3 per cent and 21.2 per cent respectively). Moreover, as the rate of 'marrying out' is higher for younger age groups, and as there has been a trend for indigenous people to move into metropolitan areas, it can only be concluded that the trend to more intermixed couples will continue. This will result in a further increase in the proportion of Australian children potentially being identified as indigenous.

Whether this 'broadening' of the group identifying as Indigenous is having an upward, downward or neutral effect on Indigenous socioeconomic indicators is, however, difficult to say. It is possible that it may have played a small part in the detected increase in average Indigenous income, tertiary participation, numbers in employment and in the decline in average household size—but if this is the case, then it could only have been playing a minor part as in so many other areas Indigenous socioeconomic indicators have not been improving, and the gap between Indigenous and the wider population has not been narrowing.

The Geographic Distribution of Population

The socioeconomic position of Indigenous people varies significantly between states/regions and within states/regions between urban and rural areas. It can be improving with respect to non-Indigenous in some geographic regions/areas but deteriorating in others. For example, Gray and Auld calculated that the ATSIC regions of Alice Springs and Cairns improved their relative socioeconomic position between the 1991 and 1996 census (in this period the housing situation in Alice Springs improving markedly and the strong economic performance of Cairns appearing to have a positive impact on the economic status of Indigenous people in these regions). By contrast, the general decline in the economic well-being in regional New South Wales appears to have had a flow on effect on the Indigenous people in the Coffs Harbour, Tamworth and Wagga regions, where socioeconomic disadvantage can be calculated as having increased between 1991 and 1996 relative to other ATSIC regions.

Most of the Indigenous population live in major urban and other urban areas and many of these are in mixed households sharing the lifestyle of the surrounding non-Indigenous population. Many Indigenous people do, however, live in rural and remote areas and there they form a larger proportion of the total population than they do in the more urban areas. In the 1996 Census, approximately one in four Indigenous people was counted in non-urban areas compared with only about one in seven in the total population (ABS urban/rural classification criteria including population density, land use and spatial contiguity). Data on Indigenous people will therefore tend to reflect some of the disadvantages associated with living in rural and remote areas, e.g. distance from services and limited employment opportunities. (See Australian Bureau of Statistics, Australian Social Trends 2000, Catalogue no. 4102.0, 2000, p. 22).

Adding to the difficulty of arriving at comparable data is the fact that in rural and remote regions in Northern Australia many Indigenous people with access to their lands maintain key elements of their original subsistence economy and lifestyle—hunting, gathering, living in multi-family households and, though not rejecting introduced goods, not being as interested as some other Australians in the accumulation of personal or household material wealth.

The Age Profile of Population

One implication of the Indigenous Australian's lower life expectancy than other Australians is that the age structure of the Indigenous population is considerably younger than that of the general population. At the time of the last census 68 per cent of the population identifying as Indigenous was under 30 years of age and only 3 per cent was aged 65 and over, compared with 44 per cent and 13 per cent respectively in the general population. This younger age structure has a bearing on some social indicators where age is an associated factor. For example, birth rates may, in part, be higher among the Indigenous population compared to the total population due to the greater proportion of women of child-bearing age within the Indigenous population. (Australian Bureau of Statistics, Australian Social Trends 2000, Catalogue no. 4102.0, 2000, p. 22). The age profile compounds problems associated with high Indigenous unemployment—the ratio of young dependents to employed Indigenous people being about three times the corresponding ratio for all Australians.

The Interconnectedness of Factors

Even when the trend does appear clear, the problem remains of identifying which, if any, Government program, may have been contributing to this trend—and determining if socioeconomic indicators in a particular area are changing because of Government programs and expenditure in this same area or despite it and because of developments in other areas. The well-documented relationships between various kinds of disadvantage mean that it is not always easy to say which programs address which needs. For instance, the experience of arrest, living with someone who has been arrested and inadequate housing are all demonstrably related to poor educational success. Consequently, money spent under the housing or legal headings may help to reduce the need for targeted education assistance. In turn, educational success is closely linked to employment opportunities and improved income, and these to health. Similarly, the correlation between inadequate housing and poor health mean that spending on housing may be the most effective way to address many health problems, as well as the presenting problem of adequate accommodation. Consequently, figures under one heading may need to be interpreted with an eye to their flow-on effects as well as their stated goals. The Royal Commission on Aboriginal People in Canada dubbed this interconnectedness The Circle of Well-Being, and linked the socioeconomic needs with still wider needs. The problem is indeed, not simply common to consideration of indigenous disadvantage in different countries but perhaps common to consideration of the disadvantage of any group in any context.

With the above considerations in mind, we might attempt to trace trends in a few key areas.



Secondary Participation and Retention rates

There was a dramatic improvement in Indigenous secondary school participation rates (the percentage of school age children at school) and retention rates (the percentage of students who continue to year 12 from their respective cohort at the commencement of secondary schooling) in the 1970s and early 1980s. The degree to which these rates may be judged to have continued to improve between into the mid/late 1990s depends, however, on the data used. A comparison of 1986 and 1996 Census data reveals an increase in the percentage of students that left school at ages 17 and over (from 11.5 per cent in 1986 to 20.8 per cent in 1996) and this increase was more accentuated than that found in the general population. Similarly, the report by Chris Robinson and Lionel Bamblett, Making a Difference: The impact of Australia's indigenous education and training policy, National Centre for Vocational Education Research, 1998, calculated the retention rate for male Aboriginal and Torres Strait Islander students as improving from 12.3 per cent in 1989 to 29.2 per cent in 1996. The Robinson and Bamblett figures have since been cited in press releases and reports to support claims of successful policy outcomes, most recently in Katu Kalpa, the 2000 Senate Employment, Workplace Relations, Small Business and Education References Committee report on the effectiveness of education and training programs for Indigenous Australians (p. 3).

A slightly less rosy picture results, however, if in place of Robinson and Bamblett's very low base figure, another figure was used: either the 19.4 per cent which Rosaleen Smyth calculated for the year 1987 or 14.4 which 1995 National Review of Education for Aboriginal and Torres Strait Islander Peoples used as the retention rate for all Aboriginal and Torres Strait Islander students in 1989 (though derived from data from only three states). The picture is slightly less rosy again if we accept R. G. Schwab's 1999 calculation that there was a fall in retention rates between 1994 and 1996 and that a detectable rise between 1996 and 1997 'appears more akin to recovery from a stall and decline than a trend to improvement'. Similarly, in a 1998 paper devoted to the question of 'Have School Retention and Participation Improved for Indigenous Students?', researchers Michael Long and Tracey Frigo analysed a set of measures and found:

The answers to the question vary by State and measure. There are indications of both absolute and relative improvement in New South Wales, Victoria, Western Australia and the Australian Capital Territory…[but] any improvement in retention and participation has not been sufficiently strong to be detected by all the measures used in this report.

and concluded:

Despite the evidence of positive outcomes from individual programs for Indigenous students in terms of engagement in learning and personal development, and despite government funding support for schools and individuals, school retention and participation for Indigenous youth has not improved markedly in recent years in Australia.

Since 1996 the trend has certainly been generally positive. Whilst the retention rates for all Australian Year 12 students have remained steady on 72 per cent the Indigenous rate passed the 1994 figure, increasing to 35 per cent in 1999 and 38 per cent in 2000. (See Australian Bureau of Statistics, Schools 2000, Catalogue no. 4221.0, 2001 and the 2001 Budget paper, Our Path Together, statement by the Hon. Philip Ruddock, 22 May 2001, p. 13). It is also clear that the number of indigenous students in secondary school has gradually increased from 27 810 in 1996 to 33 420 in 1999.

Tertiary Participation Rates

In the area of tertiary participation it is possible to observe that there has been a significant growth over the last decade in the number of Indigenous people in higher education—rising from 3600 in 1990 to 7789 in 1998, and census comparisons suggests the proportion of the Indigenous population attending a tertiary institution almost doubled over the 10 year period from 1986 (2 per cent) to 1996 (4.1 per cent). The fact that the number of ABSTUDY recipients appears to have plateaued over that period (possibly as a result of the advent of means testing in the late 80s) may, however, suggest that the participation rate has been increasing only for those with better means or for those working and studying part-time and thus not eligible for ABSTUDY. It is certainly the case that Indigenous students are much more likely than their non-Indigenous counterparts to be attending a TAFE institution and/or to be doing 'non-award' courses' (20 per cent compared with only 1.4 per cent of non-Indigenous students). Indeed most of the increased Indigenous tertiary participation has been in the non-university sector. Although the Indigenous university participation rates did rise over the period 1986 to 1996, it did not rise any more than did the non-Indigenous rate, while the Indigenous TAFE participation rate over this period rose at a time when the non-Indigenous rate for participation in this sector was falling. (See M. Gray, B. Hunter and R. G. Schwab, A Critical Survey of Indigenous Education Outcomes, 1986-96, Discussion Paper, no. 170, Centre for Aboriginal Policy Research, 1998. Summary available here. See also M. C. Gray, B. Hunter and R. G. Schwab, 'Trends in indigenous educational participation and attainment, 1986–96', Australian Journal of Education, vol. 44, no. 2, 2000, pp. 101–117).

Between 1991 and 1996 there was a significant narrowing of the gap for all educational levels (that the proportion of the Indigenous population having post secondary school qualifications rose from 10.6 per cent in 1991 to 13.6 per cent in 1996 and the proportion with Bachelor degrees rose from 0.6 per cent to 1.5 per cent), but the gap in the ratio between Indigenous and non-Indigenous people holding different degrees closed the least when it came to higher degrees and closed most when it came to lower level qualifications.

Although Indigenous tertiary participation rates rose between 1986 and 1996 it appears that they have subsequently plateaued and perhaps even fallen. The publication Students 2000: Selected Higher Education Statistics on the DEST web-site offers the following tables.

Table 2: Commencing and All Indigenous Students by Gender, 1991 to 2000(a)












% Change on 1999

Commencing Students




































All Students




































(a) Data in this table does not agree with the total reported in 3.12 because of the different definitions used in compiling the Equity data.

Table 3: Commencing and All Non-Overseas Students by Equity Group(a), 1991 to 2000

Equity Group











Commencing Non-Overseas Students
Students from a Non English speaking background

9 707

10 284

11 474

12 072

13 354

13 062

12 418

11 211

10 342

9 643

Students with a disability






4 647

5 761

6 126

6 149

6 414

Women in Non-Traditional Area

32 054

30 940

32 044

34 706

38 428

42 387

44 721

45 283

46 602

46 350


2 551

2 682

2 973

3 581

3 718

3 695

4 186

4 111

4 316

3 655


39 837

37 723

37 811

39 234

41 773

43 939

44 056

43 715

44 085

45 260


4 382

4 530

4 173

4 801

4 906

4 863

4 921

4 880

5 095

5 024

Low Socioeconomic Status

30 805

28 867

29 265

31 474

34 207

35 785

36 150

36 117

36 926

37 060

All Non-Overseas Students
Students from a Non English speaking background

20 769

23 912

26 327

28 554

31 224

32 179

31 448

29 275

26 168

23 674

Students with a disability






11 656

15 019

17 574

18 084

18 926

Women in Non-Traditional Area

80 278

85 470

89 037

93 790

99 609

107 908

117 430

121 312

125 619

125 376


4 790

5 084

5 558

6 375

7 000

7 166

7 741

8 031

8 367

7 682


93 126

97 738

98 878

100 794

102 362

105 694

107 966

108 850

109 642

110 914


9 500

10 373

10 347

10 968

11 234

11 180

11 348

11 191

11 386

11 218

Low Socioeconomic Status

74 231

76 813

77 611

80 359

83 399

86 932

90 155

91 557

92 779

93 011

  1. Equity Data statistics are compiled using the broader definition of enrolment.

Using table 2 it is apparent that the number of indigenous students recorded as commencing higher education courses though generally rising leading up to 1999, fell by 15.2 per cent in 2000. Using table 3 (with a different definition of the equity group) the same trend is apparent in the number of Indigenous students in higher education—a gradual rise to 1999 then a fall in 2000 of 8.1 per cent. Indeed recorded commencements and enrolments in 2000 fell to either below or just above what they were in 1996. By contrast overall non-overseas student commencements for the last 5 years are almost static, as were overall non-overseas student enrolment numbers. Influencing the year 2000 statistics may, however, be changed reporting requirements. These changes may have led to many enrolments which the previous year might have been recorded as ATSI being recorded in 2000 as 'ATSI status unknown' (the number of the later being 5037 more than in 1999). (See Jane Richardson, 'Numbers Disguise Black Students', The Australian, 18 April 2001.)

ABSTUDY's Influence on Participation Rates

The number of people assisted by ABSTUDY, as a program to help Aboriginal people pursue post-secondary studies, rose from 115 in 1969 to 24 095 in 1987 and the number assisted by the ABSEC, the program to ensure that Aboriginal people would have the opportunity to gain the necessary secondary education to take advantage of ABSTUDY, had risen from 19 621 in 1973 to 26 183 in 1987. In 1988, figures such as these led both the then Minister for Aboriginal Affairs, Mr Hand, and the House of Representative Standing Committee on Aboriginal Affairs to claim ABSTUDY and related Commonwealth initiatives had been contributing to improved indigenous education participation rate. Indeed, when Indigenous tertiary enrolments appeared to fall in 2000 (see above) many were quick to see this as due to changes to ABSTUDY (aligning it with Youth Allowance and New Start rates) which took effect from January 2000.

The Report of the 1994 ABSTUDY Evaluation, prepared for the Department Employment Education, Training and Youth Affair by Anne Byrne et. al., November 1997, was the first report since the advent of ABSTUDY which questioned the importance of student financial assistance as a determinant in Indigenous educational participation. The evaluative work found that a number of factors influence the educational decisions of Indigenous students. 'Approximately half of tertiary students and a just over a third of secondary students indicated they would drop out of education if ABSTUDY were unavailable', but 'in general terms, students considered that lack of self-esteem, lack of encouragement, lack of school and family support and racism were more important factors in causing Indigenous students to discontinue or drop out of study'. About two-thirds of those surveyed indicated that ABSTUDY was enough to meet their needs. Secondary students were more likely to say that ABSTUDY was adequate than either TAFE, college or university students. 'Importantly, adequacy was more of a problem for those planning to continue with education than those who were not. This suggests the somewhat secondary importance of financial factors in participation decisions.' It was also observed that ABSTUDY recipients in tertiary education were most likely to be in general studies, basic literacy, arts, education or vocational courses. The fact that ABSTUDY provided incidental allowances to help cover course costs, equipment and book costs suggested these were not major factors in the course preferences being expressed, but that basic literacy and numeracy skills may have been.

Other Influences on Participation Rates

The identification in the Report of the 1994 ABSTUDY Evaluation of lack of self-esteem, lack of encouragement, lack of school and family support and racism as more important than financial difficulties in causing Indigenous students to leave education, was noted in the Government's 1997 ABSTUDY Community Discussion Paper. It suggested that while ABSTUDY may have had the effect of removing financial disincentives to educational participation 'financial factors are only one among a range of factors limiting the educational participation of Indigenous people, and that removing financial disincentives alone is not enough to increase participation to the same levels as the rest of the community'.

The lower level of educational participation and attainment in the Indigenous population has significant flow on effects for employment and, illustrating still further the interconnectedness of factors, the lack of employment opportunities in many communities may also be contributing to poor educational outcomes. As observed in Australian Social Trends 2000 (Australian Bureau of Statistics, Catalogue no. 4102.0, 2000, p. 22).

Aboriginal and Torres Strait Islander people have reduced access to employment opportunities, which may affect their motivation to participate in education beyond the compulsory years of schooling. Educational attainment limitations in turn affect the ability of Indigenous people to secure employment, and can contribute to a cycle of poverty.

Language may also present a barrier to participation in the education system with 5 per cent of Indigenous people reporting in the 1996 Census that they did not speak English at all and 26 per cent that they did not speak it well. (Australian Bureau of Statistics, Australian Social Trends 2000, Catalogue no. 4102.0, 2000, p. 22.)


There are two main data sets for tracing trends in Indigenous Labour force status. Firstly there is the Census data collected every five years from 1971 to 1996. Secondly there is the Australian Bureau of Statistics Labour Force Survey (LFS) conducted on a sample of 63 000 persons from 30 000 households each month). From 1994 a question has been added to the February (and only February) questionnaire which sought to identify the Indigenous status of household members. This Indigenous related data was released for the first time in December 2000 with the caveat that the figures are experimental estimates only. (See Australian Bureau of Statistics, Labour Force Characteristics of Aboriginal and Torres Strait Islander Australians, Catalogue no. 6287.0.) In addition to these two main sets, there have also been figures derived from the 1994 National Aboriginal and Torres Strait Islander Survey (NATSIS) and 1995 National Health Survey (NHS). In 2001 researchers Boyd Hunter and John Taylor represented the main findings of the above four sets of data as follows:

Table 4: Indigenous Labour Force Status, 1971-2000


Employment/ Population Ratio

Unemployment rates

Participation Rates

Census 1971




























NHS 1995




LFS 1994




























a. The 1995 NHS data refers only to the 15 to 64 year-olds and is not strictly comparable to other statistics in this table that refer to the population aged 15 years or more. Notwithstanding, NHS data are broadly consistent with the LFS results.
b. The 1996 Census estimate of participation rate increases to 52.7 per cent if the 'not stated' category is allocated proportionately across all labour force states (rather than assuming respondents who do not reply are outside the labour force). The LFS publications do not report a category for 'not stated' because respondents are always prompted for a response to this question.

Boyd Hunter and John Taylor, 'Reliability of Indigenous Employment Estimates', Agenda, vol. 8, no. 2, 2001, p. 115. ABS (1995); ABS (1998); ABS (2000); and Daly (1995). The NHS estimates are based on unpublished data.

A more detail tabulation of data from the LFS collection might look as follows:

Table 5: Labour Force Status of Indigenous Australians Aged 15 Years and Over









Employed ('000)








Unemployed ('000)








Labour Force ('000)








Not in the labour force ('000)








Total Population ('000)








Labour Force Participation Rate (%)








Unemployment Rate (%)








Employment to Population ratio (%)








Note: The ABS classifies participants in the CDEP scheme as being employed.

Source: Australian Bureau of Statistics, Labour Force Characteristics of Aboriginal and Torres Strait Islander Australians, Catalogue no. 6287.0, Table 3.2.

Although a variety of errors will affect the data in all four of the above collections (flowing from sample size, geographic distribution of sample, sample timing, the cross cultural nature of the respondent/interviewer interaction etc.) where the collections overlap the data reveals remarkable consistency. From 1971 to 1986 the trend in employment/population ratio appears to have been down (and the concurrent unemployment trend up) and then from 1986 these trends reversed. Indeed, the 1996 Census recorded an increase in both the proportion of Indigenous people in work (from 37 per cent in 1991 to 40.7 per cent in 1996) and the level of employment (an increase of 25 000 or 44 per cent over the 1991 level). Although the employment to population rate for the non-Indigenous population was also higher in 1996 than in 1991, the gap between Indigenous and non-Indigenous rates narrowed slightly between these censuses.

The above good news may, however, need to be offset to some extent with two observations.

Firstly, overall Indigenous employment to population rate is still below three-quarters of the level for non-Indigenous adults. (See John Taylor and Boyd Hunter, The Job Still Ahead: Economic costs of continuing indigenous employment disparity, 1998, p. 12.)

Secondly, a large portion of the increase in Indigenous employment can be accounted for by increase in the number of participants in the Community Development Employment Program (CDEP) scheme, who, for the purpose of the LFS, are classed as employed. The number of CDEP participants rose from about 22 178 in February 1994 to 31 050 in February 2001 and CDEP participants have been calculated to account for between 15 and 25 percent of employed Indigenous persons. Although proving to be an attractive minimum-income support program, the CDEP, as a job-creation program, is of debatable value and a very different picture emerges if CDEP participants are classed as unemployed. If the approximately 31 000 CDEP participation in February 2000 are classed not as employed but unemployed then the unemployment rate for indigenous persons in February 2000 would climb dramatically from 17.6 per cent to 40.6 per cent. If they are classed as neither employed nor unemployed (and it is certainly the case that in many remote and regional communities with minimal employment opportunities, in the absence of the CDEP many would not count themselves as looking for work) then the unemployment rate would rise from 17.6 per cent to 22.9 per cent. As Hunter and Taylor conclude:

The key finding of significant decline in unemployment rates since 1998 resonates with an analysis of trends in increasing CDEP scheme employment and with the fact that purely administrative changes to the scheme are likely to have raised overall employment levels over the same period. In other words, it is very unlikely that the recent decline in Indigenous unemployment has formed part of the general labour market trend…

According to Taylor and Bell, the gap between Indigenous people potentially eligible for mainstream work and those actually in mainstream work continues to widen. Gray and Hunter reach a similar conclusion, noting that the ratio between non-CDEP employment and population actually fell between 1991 and 1996 and that the degree to which Indigenous educational attainment had improved relative to non-Indigenous over this same period was insufficient to improve overall employment prospects.


The health status of Indigenous Australians—at every stage of life—is widely understood to be much lower than that of other Australians. (For up-to-date snap-shot comparison see Summary of Indigenous Health Status, Australian Indigenous HealthInfoNet, June 2001). Many factors have been identified as contributing to this discrepancy in health status (poor housing and infrastructure, poor nutrition, alcohol etc.). Many factors have also been identified as contributing to a reduced likelihood of an Indigenous person taking actions to improve their health, e.g. problems of access to doctors, pharmaceuticals and culturally appropriate services and lack of awareness of entitlements. Identifying trends is, however, difficult. Though the need for improved data collection in the area of indigenous health has been widely recognised and although processes have been put in place over the last few years to facilitate this improved collection (e.g. the 1994 National Health Information Development Plan has led, through various working parties, to the development of 50 key health performance indicators) in many jurisdictions the data required to report on these indicators is still not available or of poor quality. Similarly the development of Aboriginal and Torres Strait Islander Health Framework Agreements are also serving as a vehicle for facilitating improved data collection, but a reliable set of cross-jurisdictional post 1996 census data is still yet to be produced.

General Mortality Rates

Between 1984 and 1994 overall male Indigenous mortality rates fell by 1.5 per cent and in the 1990s Indigenous infant and maternal mortality rates were also falling (e.g. Australian Institute of Health and Welfare, Australia's Health 1998, Canberra, 1998). However, non-Indigenous mortality rates in these categories have also been falling and a threefold discrepancy between Indigenous and non-Indigenous mortality rates remains. Life expectancy for Indigenous people in South Australia, Western Australia and the Northern Territory, the three jurisdictions with the most complete coverage of Indigenous deaths, is 60 years for males and 61 for females, fifteen to twenty years less than that of other Australians. Latest research from these jurisdictions reveals for the period of 1995–1997 enormous disparity. Death rates were higher for Aborigines than for others in every age group. The death rate for Indigenous people aged 35–54 was 6 to 7 times higher than for the rest of the community. Of those who died, 53 per cent of Indigenous males and 41 per cent of Indigenous females were less than 50 years old, while only 13 per cent of all male deaths and 7 per cent of all female deaths in Australia occurred among people less than 50 years old.

All-causes mortality rate for Australia's Indigenous population is approximately twice the Maori rate and 2.3 times that of Indigenous population of the United States and while the rate hardly fell in Australia in the two decades between the mid 1970s and mid 1990s, the Indigenous mortality rate declined by 44 per cent in New Zealand and 30 per cent in the United States.

Infant Mortality Rates

The Indigenous infant mortality rate has clearly declined dramatically over the last few decades, from estimated rates of 143 per 1000 in the Northern Territory between 1958 and 1960; to between 75 and 80 per 1000 in the Northern Territory and Queensland in the early 1970s; 30 to 40 per 1000 in these same jurisdictions in the early 1980s; to 29.4, 13.6 and 8.5 per 1000 in the Northern Territory, Western Australia and South Australia respectively in 1997. All these rates, however, have been and still are far greater than the general infant mortality rates which have also been trending down and which in 1996 were 6.5 for males and 5.0 for females. In most States and Territories babies born to Indigenous mothers are twice as likely to be of low birth weight and more than twice as likely to die at birth than are babies of non-Indigenous mothers.

Incidence of Disease

Many comparisons are made between the health status of Indigenous and non-Indigenous Australians. The incidences of diabetes and diseases of the circulatory system among young and middle aged Indigenous adults are more than 10 times higher than in the wider community, and up to 17 times higher for Indigenous females than for non-Indigenous females. Twenty per cent of Indigenous children under 10 years of age have the early stages of the Chlamydia trachomatis infection which gives rise to a chronic conjunctivitis which can lead to opacification of the cornea and blindness. The figure is up to 55 per cent in some areas of the country. (See H. Taylor, Eye Health in Aboriginal and Torres Strait Islander Communities: the report of a review commissioned by the Commonwealth Minister for Health and Family Services, 1997.) The incidence of tuberculosis among Indigenous Australians is seven times the rate of incidence among the general population. (Australian Institute of Health and Welfare, Australia's Health 1996, Canberra, 1996, p. 66.) Rates of hospitalisation for nearly every cause and for every age-group are higher and lengths of stay are longer. Death by disease of the circulatory system are more than two and a half times higher, death by accident and injury are three and a half times higher, and death by disease of the respiratory system are five times higher. Latest research also revealed that among Indigenous people there are 40 per cent more deaths from cancer than would be expected in the population as a whole.

It has been reported that the death rates for diabetes mellitus increased between 1985 and 1994 by almost 10 per cent per year for Indigenous males, and by over 5 per cent per year for Indigenous females (Human Rights and Equal Opportunity Commission, Statistics: Aboriginal and Torres Strait Islander peoples, citing P. Anderson), but in general there are few studies of disease trends over time.

Violence and Neglect

Although there are few consistent cross-jurisdictional longitudinal data collections from which trends might be drawn in the area of violence and neglect, there is an abundance of quantitative information which shows that Indigenous people are at greater risk of both than are non-Indigenous. (For an overview of data sources on types of violence, injury and crime tracking place within the Indigenous population see Paul Memmott et al., Violence in Indigenous Communities, Report to Crime Prevention Branch of the Attorney-General's Department, January 2001.) Indigenous people are more likely than non-Indigenous people to be victims of violence and to suffer intentional injuries which require hospitalisation. Indigenous women are grossly overepresented among women in hospital for intentional injuries (constituting 46 per cent of such hospital separations in 1996–97) and in the case of indigenous homicides, a higher proportion of women are likely to be victims or offenders than is the case with non-indigenous homicide (the former much more likely to occur within the family environment and in the presence of alcohol than the latter). See Jenny Mouzos, Indigenous and Non-Indigenous Homicides in Australia: A Comparative Analysis, Australian Institute of Criminology, Trends and Issues, no.210, June 2001, p. 5. Moreover, although there are differences by State and Territory, Indigenous children are more likely than non-Indigenous children to be the subjects of substantiated cases of abuse and neglect (with rates about 2–8 times higher in most jurisdictions in 1997–98), under care and protection orders (about 4 times higher in 1998) and out-of-home placements (almost 6 times higher in 1998).

Two trends can, however, be identified. Firstly, a recent study of statistics related to that most extreme form of violence, homicide, between 1989–90 and 1999–2000 reveal that indigenous victimisation rates and offender rates fell in the mid 1990s. The former from being about 13 per cent per 100 000 for most of the early 90s to 9 per cent per 100 000 for most of the later 90s (in either case, however, consistently higher than for non-Indigenous Australians whose victimisation rate for the decade fluctuated between 1.3 and 1.8 per cent per 100 000. The later peaked in 1990–91 at 20.9 per 100 000 population, declined in 1993–94, rose a little in 1994–95, then declined to a 1999–2000 rate of 9.9 per 100 000 (the non-Indigenous rate fluctuating between 1.2 and 1.8 per 100 000). (See Jenny Mouzos, Indigenous and Non-Indigenous Homicides in Australia: A Comparative Analysis, Australian Institute of Criminology, Trends and Issues, no. 210, June 2001, p. 3.) Secondly, data on the number of Indigenous children nation-wide on care and protection orders reveals a rise from 13.6 per 1000 at 30 June 1996 to 14.6 at 30 June 1998.


Household Size

According to the Census data the average size of Indigenous households fell from 4.6 in 1991 to 3.7 in 1996 while average non-Indigenous household size rose slightly from 2.6 in 1991 to 2.7 in 1996. It might be possible to say that in both absolute and relative terms Indigenous household size, while still much larger than non-Indigenous household size, is trending down. Indigenous household size in 1986 (4.5) was, however, slightly lower than it was in 1991 so the trend may not have been fully established yet and as household size has been shown to be more likely to be lower in the urban areas, the areas where most of the recent increase in Indigenous-identifying population has occurred, it is unclear how much of this trend is actually attributable to a broadening of the identifying population. In any case it is clear that the gap between Indigenous and non-Indigenous housing status remains significant. According to the 1996 census Indigenous people were more likely than other Australians to live in improvised and/or over-crowded dwellings. Almost one-third of all households living in improvised dwellings were Indigenous and nearly half of all households of 10 or more people were Indigenous. Almost 7 per cent of Indigenous people in Australia live in dwellings with 10 or more residents—50 times greater than the proportion of other Australians living in such conditions.

Home Ownership

The number of houses owned or being purchased by Indigenous people increased by 63 per cent between the 1976 and 1991 Census. However, this was in line with the increase in the number of Indigenous households so the proportion of those owning or buying showed little intercensal variation. Between the 1991 and 1996 census the relationship between the Indigenous and non-Indigenous population in housing tenure remained steady.

Table 6: Indigenous and Non-Indigenous People Owning, Purchasing and Renting Housing

Census data (%)



Indigenous people owning own dwelling



Non-Indigenous people owning own dwelling



Indigenous people purchasing their home



Non-Indigenous people purchasing their home



Indigenous people renting



Non-Indigenous people renting



Community Infrastructure

In the 12 months prior to a survey published in April 2000 of 348 communities with a population of 50 or more, 58 had their water supply fail water testing, 57 had frequent power interruptions and 204 had leakages or overflows in their sewerage system. The same ABS report found some improvement since 1992 when 44 per cent of houses were found to be seriously rundown but found that 29 per cent of dwellings administered by Indigenous housing organisations needed major repair or replacement. In many places, lack of basic facilities such as clean water, sewerage and electricity combined with overcrowding to create a direct link between substandard housing and poor health. In 1996 ATSIC estimated that 'at the current rate of funding…it will take twenty years to clear the backlog of housing and infrastructure needs, currently estimated at $3.1 billion'. See ATSIC's Community Housing and Infrastructure Program Outline.


Census data reveals that between 1976 and 1996 the proportion of indigenous males with no income fell by more than 50 per cent and of indigenous females by 23 per cent. This fall is, however, probably the result of increased coverage of the social security system and represents more a vanishing 'bottom' than a much improved 'middle'. Indeed, although the median weekly income of Indigenous people in Australia rose between the 1991 and 1996 Census by $43 while that for the total population rose by only $20, the rise varied enormously between States and the Indigenous increase was coming off a lower base than the total population increase. In the ACT the 1996 census showed Indigenous median weekly income as less than in 1991, and in all jurisdictions the income was still much less than general medium income. In the Northern Territory, the median income in 1996 of $182, although $40 higher than in 1991, was only half of the median income of the general population ($367). Overall, the mean individual income for those identifying as Indigenous rose from being 63.8 per cent of that of the general population in 1991 census to being 74.1 per cent of that of the general population in 1996 census. It is possible, that some of this rise may be attributable to the higher employment rate, but, as has been discussed above, much of the employment increase is attributable to an increase in low paying CDEP employment, therefore another factor may also have been at work. It is possible some people in families of average or above average incomes who did not identify as Indigenous in the 1991 census, by identifying as Indigenous for the first time in the 1996 census, may have contributed to an apparent relative increase in Indigenous incomes. Indeed, the Census found the higher incomes to be associated with Indigenous people in the more urban areas, and it is in the more urban areas that the number of people identifying as Indigenous has been rising most.


According to statistics compiled by the Australian Institute of Criminology, the number of indigenous prisoners in the years 1994, 1995, 1996, 1997 and 1998 has been 2800, 2985, 3273, 3580 and 3750—representing a steady increase. Indeed it has been observed that there has been an average increase of Indigenous people in custody of 6.9 per cent for each year of the decade 1988–1998 and this is 1.7 times the average annual growth rate of the non-Indigenous prison population. The ratio of the rate of indigenous imprisonment (by population) to rate of non-Indigenous imprisonment has for the whole of the last decade revealed a high rate of Indigenous overrepresentation in the prison system. This overrepresentation ration has, however, declined from 14.2 in 1988 to 11.0 in 1998, possibly partly because the increasing number of people identifying as aboriginal, has slowed the apparent rise in the rate of indigenous imprisonment vis--vis the rate of non-indigenous imprisonment. See C. Carcach, A. Grant and R. Conroy, Australian Correction: The imprisonment of indigenous people, Australian Institute of Criminology, Trends and Issues in Crime and Criminal Justice, no. 137, Canberra, 1999.

Over these same five years from 1994 to 1998 Indigenous deaths in all custodial circumstances have been 14, 21, 18, 15 and 16, or if leaving out deaths in police operations to have number of deaths in Institutional settings only, have been 12, 17, 13, 11 and 14. These numbers by themselves do not reveal either an upward or downward trend, but it has been noted that the number of Indigenous deaths in custody in the decade since the Royal Commission into Aboriginal Deaths in Custody concluded in 1991 has been 147, compared to 99 in the decade before the Royal Commission, and that 17.2 per cent of all prison deaths in the 1990s have been Indigenous people, compared to 12.1 per cent in the 1980s. (Aboriginal and Torres Strait Islander Social Justice Commissioner, Social Justice Report 2000, p. 66).


While snap-shot comparisons of Indigenous and non-Indigenous socioeconomic indicators may reveal the need for continued Government effort in the area of Indigenous education, employment, health, housing, infrastructure and custody rate (that which has been called 'practical reconciliation'), it is only an analysis of trends which can reveal how effective government efforts have been to date. An effort has been made in this paper to trace as many relevant trends as possible, but interpreting these trends has been found to be highly problematic. Often trends which appear highly encouraging, on closer analysis are less so.

Commonwealth of Australia
Research Papers Online are written for Members of Parliament, being located on the Internet they can be read by members of the public, however some linked items are available to Members of Parliament only, due to copyright reasons.

Back to top