Australian Bureau of Statistics

Rate the ABS website
ABS Home > Statistics > By Release Date
ABS @ Facebook ABS @ Twitter ABS RSS ABS Email notification service
1301.0 - Year Book Australia, 1996  
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 01/01/1996   
   Page tools: Print Print Page RSS Feed RSS Bookmark and Share Search this Product

POVERTY AND DEPRIVATION IN AUSTRALIA


IINTRODUCTION

This article has been contributed by Dr Peter Saunders, Director, Social Policy Research Centre, University of New South Wales.)

As 1996 has been designated International Year for the Eradication of Poverty, it provides an opportunity to review evidence on the extent of poverty in Australia and to reflect on the meaning and measurement of poverty.

The measurement of poverty cannot be undertaken independently of national customs, values and standards of living. This means that in a relatively affluent country like Australia, the meaning of poverty is quite different from the absolute deprivation or subsistence poverty which exists in many developing countries. The World Bank has defined poverty as an inability to attain a minimal standard of living (World Bank, 1990, p.26). This is a general definition which can be applied to different national situations according to the specific meaning given to the concept of a standard of living and how the minimal level is determined.

Interest in the issue of poverty has grown in recent years, both internationally and within Australia. In the international context, the increasingly integrated world economy has highlighted the global nature of poverty and focused attention on its consequences. The United Nations Development Programme (UNDP) has emphasised this point, noting that: 'Poverty is the greatest threat to political stability, social cohesion and the environmental health of the planet' (UNDP 1994, p.20).

Reflecting these concerns, the reduction and elimination of widespread poverty was one of the major themes of the UN-sponsored World Summit for Social Development which took place in Copenhagen in March 1995. The eradication of poverty throughout the world formed the basis of one of the Commitments contained in the Summit Declaration, which recommended that each country set a target date for eradicating absolute poverty. The Summit also recommended that the General Assembly of the United Nations endorse its proposal that 1997 should represent the commencement of a UN decade for the eradication of poverty.

Within Australia, as in many other industrial countries, poverty has re-emerged as a policy issue in response to the high levels of unemployment and increasing levels of inequality which have been experienced over the 1980s (ABS, 1994; Saunders, 1994). These concerns have reinforced a longer-term interest in the study of poverty in a country where poverty alleviation has been a major goal of wage and social security policies. These concerns were highlighted during the 1987 federal election campaign when Prime Minister Hawke made his pledge that by 1990 no child will need to live in poverty.

Assessing the governments success in achieving that goal has rekindled interest in the extent of poverty among Australian families. Research on the topic has revealed that fundamental differences exist concerning the definition of poverty and how it should be measured. Some academic writers (e.g. Travers and Richardson, 1993) have questioned the usefulness of using a single income-based poverty line to estimate the extent of poverty, a view which has been endorsed by the current Minister for Social Security (Baldwin, 1995).

These debates raise many issues in poverty research which cannot all be dealt with in the space of a single article. Some of them will be mentioned, but the poverty estimates which are presented below have been derived from one particular framework. Because of this, it is important to emphasise that estimates of poverty are estimates, and like all other estimates, are only as reliable as the methods used to derive them.

Poverty and inequality

In its 1990 World Development Report, the World Bank used a poverty line for each country set at the equivalent (in 1985) of US $370 per person per year. On this basis - equivalent in Australia to around $10 per person per week - there is virtually no poverty in Australia. However, the World Banks' focus was on absolute or subsistence poverty - not having enough income to be able to secure minimum levels of food, clothing and shelter.

In rich countries like Australia, poverty is conceived in relative rather than absolute terms. This implies that poverty is defined not in terms of a lack of sufficient resources to meet basic needs, but rather as lacking the resources required to be able to participate in the lifestyle and consumption patterns enjoyed by other Australians. To be relatively poor is thus to be forced to live on the margins of society, to be excluded from the normal spheres of consumption and activity which together define social participation and national identity.

The acceptance of poverty as a relative concept means that poverty and inequality are related. Relative poverty can only exist where there is inequality, but this does not mean that poverty and inequality are the same. Not all forms of inequality will imply relative poverty, only those in which some individuals or groups fall well below the average. The eradication of relative poverty will require some reduction in inequality, but how much and at what cost will depend upon the nature and severity of the poverty problem.

Direct and indirect measures

Most poverty research focuses on whether disposable income (income measured after the receipt of cash benefits from government and after the payment of personal income taxes to government) is sufficient to meet needs. This is justified on the grounds that income is a good measure of living standards in market economies like Australia. However, income is an indirect measure of poverty because it measures the resources on which living standards depend, but does not directly measure the living standards actually experienced.

Where people have access to resources other than income (e.g. financial assets or housing wealth) income may be an imperfect indicator of actual living standards. In Australia, where home ownership is high by international standards, this can have an important bearing on the measurement of poverty, particularly among groups like the aged where home ownership is most prevalent.

These problems might be circumvented by estimating poverty on the basis of consumption expenditure rather than income - as has been done by Eurostat, the Statistical Office of the European Communities to compare relative poverty among EC countries. However, the use of expenditure data is not without its own problems. The most notable of these is that it is generally expenditure on consumption rather than consumption as such which is measured. Furthermore, it may be possible to sustain a standard of living during periods of low income by running down past savings or incurring debts, but this situation will not be sustainable. Low consumption expenditure may thus conceal rather than reveal the existence of poverty.

A better way of overcoming these difficulties is to try and identify poverty directly rather than indirectly. This approach, based on the Level of Living Surveys pioneered by researchers in Sweden, has been applied in Australia by Travers and Richardson (1993) as well as in the Australian Living Standards Study undertaken by the Australian Institute of Family Studies (McDonald and Brownlee, 1994). It involves observing the many dimensions of living conditions, including consumption of private goods, housing, health, education, employment and participation in social and political aspects of community life.

Given the multi-dimensional nature of actual living standards, it is difficult to derive a single poverty measure using the direct approach. The focus in this approach tends to be on quantifying inequality in particular aspects of living conditions and exploring what these imply for policies in specific areas. It is, however, also possible to derive cumulative measures of disadvantage in several dimensions which are indicative of the existence of poverty.

Relative deprivation

The concept of relative deprivation is associated with the work of British social researcher Peter Townsend (Townsend, 1979). It builds on direct observation of various dimensions of the lifestyles actually experienced by people and defines relative deprivation as a situation where people lack the resources to have the living conditions and amenities and participate in activities which are widely encouraged and approved in the society in which they live. It is claimed that below a certain level of income the index of deprivation begins to increase substantially and that this income threshold can be thought of as a poverty line. However, it is the level of deprivation itself, rather than the income threshold attached to it, on which attention is focused.

The concept of relative deprivation leaves unresolved the question of how the customary or approved living conditions and activities are identified. One approach is to leave such judgements to the experts. Another, which attempts to integrate the concepts of poverty and relative deprivation involves identifying a list of necessities which are regarded as such by a majority of the population and then defining poverty as those who experience an enforced lack of several of these socially perceived necessities. These approaches to the measurement of poverty are under-developed in Australia, where relatively little is known about how people conceive of poverty and what they think constitutes being poor.

The role of values

None of the conceptual approaches to poverty described above avoid the need for values to enter into the way poverty is defined. These values need to be made explicit and their validity needs to be debated and endorsed if poverty research is to promote action to alleviate poverty. The fact that the measurement of poverty is not value-free does not mean that poverty research cannot embody scientific principles associated with data quality and methodological technique (Saunders, 1995). What it does imply is that estimates of poverty will always be challenged - and should always be challenged. This should be kept in mind when assessing the estimates of poverty presented.

Indices of poverty

Most of the poverty measures discussed later estimate the percentage of income units whose incomes fall below a poverty line expressed in monetary terms. This measure is referred to as the headcount index of poverty, for obvious reasons. The headcount ratio can be expressed as the percentage of income or family units below the poverty line, or as the percentage of individuals living in poor units or families.

One limitation of the headcount poverty measure is that it takes no account of the severity of poverty - how far below the poverty line the poor actually are. Poverty is an all-or-nothing state. Furthermore, changes in income which cause people to move from just below to just above the poverty line may cause the headcount poverty ratio to decline markedly, but reflect only minor changes in living standards. In addition, government policies which raise the incomes of the poor without pushing them above the poverty line will have no effect on the headcount poverty ratio.

To counter these criticisms, a range of more complex indices of poverty has been developed. These reflect both the severity of poverty as well as the numbers below the poverty line. The most well-known of these measures is the poverty gap, which estimates the total gap between actual income and the poverty line for all those who are in poverty. The poverty gap thus measures the total cost of raising all of the poor to the poverty line but no further. It indicates the severity of the poverty problem and the (minimum) cost of addressing it.

But even the poverty gap can be criticised because it assigns an equal weight to all income shortfalls below the poverty line. If, instead, it is thought desirable to assign more weight to those who are in deepest poverty, then even more sophisticated (and complex) poverty measures are required. The problem here is that such increased sophistication may produce a poverty measure which is less transparent and understandable in the community.

Needs and equivalence scales

Thus far, the discussion has proceeded as if there is a single poverty line. In fact, the poverty line varies according to needs which in turn depend upon the characteristics of each income unit, the most obvious being the number of income unit members. The income unit itself is defined to approximate a basic economic unit comprised of related individuals who are living together. There are four basic types of income unit; single people; married couples (de jure or de facto) without children; couples with children; and sole parents.

For each income unit type, the poverty line most commonly used in Australia varies according to the age, gender and workforce status of each individual member, and according to the total number of individuals living in the income units household. These variations are embodied in the equivalence scale on which the measurement of poverty is based. The scale reflects the relative needs of different family types and is designed to allow the standards of living of different families to be compared on a single scale.

The equivalence scale expresses the needs of each income unit relative to those of a benchmark type - a single person unit, for example. The benchmark or reference unit is assigned an equivalence value of 1.0 and the equivalence scale then assigns values to other units relative to this. Thus, a married couple may have an equivalence value of 1.7, implying that the needs of a couple are 70% higher than the needs of a single person. In this case, the value is less than 2 because of the existence of cost economies in household living arrangements.

The above example implies that a couple needs 70% more income than a single person in order to attain the same standard of living. This in turn implies that the poverty line for the couple should also exceed that of a single person by 70%. This highlights the fact that poverty is concerned not with low income as such, but with incomes that are low relative to needs. It also points to the important role of the equivalence scale itself in determining both the overall extent of poverty and its composition among different income unit types.

Another key assumption concerns the fact that, within income units, there is assumed to be equal sharing of resources. This means that the standard of living, and hence the poverty status, of each individual unit member are the same. Relatively little is known about the extent of income pooling within Australian families. Given its importance, the assumption of equal sharing, like other assumptions, needs to be tested.

The Henderson poverty line

The first systematic attempt to estimate the extent of poverty was undertaken in the mid-1960s by a group of researchers at the Institute for Applied Economic and Social Research (IAESR) at Melbourne University (Henderson, Harcourt and Harper, 1970). The research team, led by IAESR Director Professor Ronald F. Henderson, estimated poverty in Melbourne using a poverty line for a reference two-adult, two-child family set at an income equal to the value of the basic wage plus child endowment payments (later called family allowances).

This poverty benchmark was chosen because of its relevance to Australian concepts of minimum income, but also because as a proportion of average earnings and average incomes it was similar to poverty lines used in other countries. It was regarded by the research team as an austere measure, likely to produce lower-bound estimates of poverty. The equivalence scale used to derive poverty lines for other family types was, in the absence of any reliable Australian estimates of relative needs, derived from a survey of household budgets and costs undertaken in New York in the 1950s. This aspect of the poverty line has been criticised, although the cost relativities implied by the New York scale were regarded at the time by the IAESR researchers as quite plausible for Melbourne.

The poverty line described above was subsequently used by the Commission of Inquiry into Poverty to produce the first authoritative national estimates of poverty in Australia using data from an ABS income survey undertaken in August 1973 (Commission of Inquiry into Poverty, 1975). Following a recommendation in the IAESR study, the poverty line was updated to 1973 in line with the movement in average weekly earnings. The updated line became known as the Henderson poverty line (HPL) after Professor Henderson, the Chairman of the Poverty Inquiry.

Updated estimates of the HPL have been published regularly since 1979 by the IAESR and reported by welfare organisations such as the Australian Council of Social Service (ACOSS). Updating the poverty line by average earnings was replaced by the use of household disposable income per capita in 1981, the latter measure being more comprehensive in scope but also preferable because it incorporates changes in personal tax payments.

Updates of the HPL have been used to estimate poverty as more recent ABS household income data have become available, and also as a benchmark against which to assess the adequacy of government benefits. In light of the significance of both exercises, it is no surprise that the poverty line estimates have been subject to extensive criticism. To date, however, no viable alternative has been proposed and the HPL continues to be widely used to measure poverty.

Table S2.1 presents the poverty lines for income units in the workforce in the March quarter 1995. The methods and assumptions used to derive them are described in detail in Johnson (1987). The dollar value of the poverty line for each income unit type is shown, as well as the relativities for different income unit types - the equivalence scale. In dollar terms, the poverty line for a single person was $215 a week, for a couple it was $288 a week, increasing thereafter by around 20% for each child. The poverty line for a sole parent with one child was about 4% below that of a childless couple with the same workforce status. (Poverty lines for those not in the workforce were around $41 below those shown in table S2.1 for each income unit type). Housing costs vary with family size, but range from around $70 a week for a single person up to $106 a week for a couple with four children.

S2.1 POVERTY LINES FOR AUSTRALIA, MARCH QUARTER 1995(a)

Including housing]
Other than housing


$ per
Relativities
$ per
Relativities
    Income unit type
week
(COUPLE =100)
week
(COUPLE =100)

    Couple
288.4
100.0
210.9
0.7
    Couple, 1 child
346.6
120.2
262.2
0.9
    Couple, 2 children
404.9
140.4
313.4
1.1
    Couple, 3 children
463.2
160.6
364.7
1.3
    Couple, 4 children
521.4
180.8
415.4
1.4
    Single person
215.6
0.8
145.1
0.5
    Single parent, 1 child
276.8
1.0
199.3
0.7
    Single parent, 2 children
335.0
1.2
250.5
0.9
    Single parent, 3 children
393.2
1.4
301.8
1.1
    Single parent, 4 children
451.5
1.6
353.0
1.2

            (a) The estimates assume that the head of the income unit is in the workforce and are based on an estimate of seasonally adjusted household disposable income per head per week for the March quarter 1995 of $313.29.
            Source: Institute of Applied Economic and Social Research, 1995, Table 1.


Poverty in 1990

Estimates of poverty in 1989-90 using the HPL framework have been derived by the Australian Institute of Health and Welfare (AIHW, 1993) and Saunders (1994). The former estimates are shown in table S2.2. They reveal that over 840,000 income units had incomes below the poverty line, leading to an overall poverty rate of 13.8%. Poverty was most prevalent among sole parent income units, whose incidence of poverty was more than three times the national rate.

Poverty after housing costs is estimated by comparing disposable income net of actual housing expenditures with the after-housing costs poverty line (see table S2.1). On this basis, poverty is lower overall and markedly lower for the aged, many of whom have relatively low housing costs. Offsetting this is the tendency for after-housing poverty to be higher amongst younger groups, particularly single people and two-parent families. This partly reflects the modest allowance for housing costs in the HPL (table S2.1).

Table S2.2 indicates that income unit type and housing status both affect the risk of poverty. In general, the risk of poverty is much less in income units with two adult members than in those containing only a single adult. Research indicates that the risk of poverty is also influenced by labour force status, age, the gender of the head of household and whether or not the household head is a recent migrant to Australia (Paul and Podder, 1992).


S2.2 INCOME UNITS IN POVERTY, BEFORE AND AFTER HOUSING COSTS, BY INCOME UNIT TYPE, 1989-90(a)

Total number
Poverty rate before housing costs
Poverty rate after housing costs
    Income unit
('000)
(%)
(%)

    Single person
    aged 15-24
258.1
16.0
24.6
    aged 25-64
1,125.8
17.2
15.4
    aged 65 & over
714.5
19.2
5.8
    Couples (no children)
    head aged under 65
1,266.0
6.5
6.9
    head aged over 65
565.4
6.0
4.2
    Two parent families
1,795.9
10.5
13.8
    One parent families
358.8
46.1
43.8
    All income units
6,084.5
13.8
13.0
    Number of income units in poverty (000)
-
842.1
794.2

            (a) Excludes residents of non-private dwellings, self-employed income units and independent children of the household head. Figures refer to weighted population estimates.

            Source: Australian Institute of Health and Welfare, 1993, Table 3.1.


Poverty and labour force status

The Poverty Inquiry identified the workforce status of the income unit head as the dominant factor which determined the poverty status of the income unit as a whole. That basic relationship remains in force today, although with two-earner couples now outnumbering single-earner couples, the labour force status of any single individual is less important in determining whether the unit as a whole is poor. Furthermore, the widening of earnings inequality and the growth in part-time and casual employment has made employment less of a guarantee against poverty than was the case in the past.

Quantifying the relationship between labour force status and poverty is complicated by the fact that poverty is determined on the basis of annual income while labour force status can vary over the course of the year. Where this occurs, the classification of labour force status is to some extent arbitrary. In constructing graph S2.3, the unemployed have been defined (following the Poverty Inquiry) as those income units where the head is unemployed for eight weeks or more during the course of the year. As a consequence, the residual category of those in the labour force comprises part-time workers, part-year workers and those who have been unemployed for less than eight weeks.



Despite this, graph S2.3 indicates that labour force attachment reduces the risk of poverty, while access to full-time employment virtually removes it entirely. To the extent that the opportunity to achieve the benefits associated with employment are greater amongst couples than income units with only a single adult, the differences in graph S2.3 partly explain the patterns of poverty by income unit type shown in table S2.2.

Poverty among immigrants

The extent of poverty among all immigrants has been estimated systematically by Johnson (1991). His results, using data from the ABS household income surveys conducted in 1982 and 1986 are summarised in table S2.4. The value of these comparisons is affected by two factors: first, the limited degree of detail on place of birth available on a consistent basis for both years; second, the numbers in some birthplace categories are quite small.

Despite these limitations, table S2.4 indicates that although poverty among immigrants is somewhat higher than amongst those income units where the head is Australian-born, this is not a consistent pattern across all immigrant groups. In some cases (particularly for immigrants from the UK and Ireland) poverty is actually lower than amongst those born in Australia.

Without further analysis, it is not possible to attribute the differences shown in table S2.4 to birthplace alone, as there may be other differences which can explain the results. The immigrant groups differ in regard to such factors as length of residence, age, English-speaking ability and educational qualifications and these may account for the observed differences. However, research shows that when age and educational qualifications are held constant, poverty in 1981-82 was higher among immigrants from non-English-speaking backgrounds, as well as among those immigrants who arrived in Australia most recently (Johnson, 1991).
S2.4 POVERTY BY PLACE OF BIRTH IN 1981-82 AND 1985-86

      Place of birth of head of Income Unit
Poverty rate in 1981-82
Poverty rate in 1985-86

      Australia
10.2
11.7
      UK & Ireland
8.1
7.3
      Italy
16.9
15.8
      Other Europe
11.5
12.6
      Asia
16.9
29.1
      Americas
11.4
30.7
      Africa
12.9
10.7
      Oceania
11.3
24.9
      All immigrants
11.4
14.6

      All income units

10.5

12.5

            Source: Johnson, 1991, Table 3.5.


Poverty among Indigenous Australians

It is now well established that the overall socioeconomic status of Aboriginal and Torres Strait Islander people is well below that other Australians. Not surprisingly, that difference extends to a comparison of the poverty status of Indigenous and non-Indigenous Australians. What is more surprising is the lack of reliable data on which to substantiate such a claim - at least as it applies to poverty assessed on the basis of inadequate income.

The Poverty Inquiry was unable to document the poverty status of Aboriginal people on a national basis because of the deficiencies of the data collected (by white interviewers) on the incomes of Aboriginal households (Commission of Inquiry into Poverty, 1975, p.260). The Inquiry was, however, able to produce estimates of Aboriginal poverty in Brisbane and Adelaide in 1973. The estimated poverty rates for Aboriginal people in these two cities (before housing costs) were 47% and 22% respectively - well above the national poverty rate, the poverty rate in all metropolitan areas, and the poverty rates in Queensland and South Australia, respectively.

S2.5 POVERTY, SEVERE POVERTY AND NEAR POVERTY AMONG ABORIGINAL AND NON-ABORIGINAL INCOME UNITS WITH CHILDREN, 1986

Severe poverty
(Income below
80% of the poverty line)
Poverty
(Income below
100% of the poverty line)
Near poverty
(Income below
120% of the poverty line)



    Income unit
Aboriginal people
Non-
Aboriginal people
Aboriginal people
Non-
Aboriginal people
Aboriginal people
Non-
Aboriginal people

    Couples with
    1 child
6.8
2.4
12.2
3.6
33.8
13.6
    2 children
5.2
1.9
27.3
8.0
44.2
12.8
    3 children
19.6
2.4
50.0
14.2
67.4
33.8
    4 or more children
30.8
16.7
48.7
25.1
71.8
47.6
    Sole parents with
    1 child
34.3
14.2
46.3
25.8
77.6
58.1
    2 children
15.9
13.5
77.3
51.0
95.5
73.5
    3 or more children
34.6
40.8
92.3
82.1
96.2
86.2
    All families with children
18.5
5.8
43.2
15.0
63.5
28.2
    Percentage of children
20.4
7.0
49.9
18.0
67.9
31.5

            Source: Ross and Whiteford, 1992, Table 5.


In a recent study, Ross and Whiteford (1992) used information collected in the 1986 Census of Population and Housing and in the Income Distribution Survey conducted in the same year to derive estimates of poverty amongst Aboriginal families with children. Table S2.5 compares their estimates of poverty among Aboriginal and non-Aboriginal families.

The overall poverty rate among Aboriginal families is almost three times that among non-Aboriginal families. Half of all Aboriginal children were living in poverty in 1986, more than two-thirds were in near poverty and a fifth were in severe poverty. In all cases, these figures are well above those estimated for children in non-Aboriginal families. The general pattern of poverty among Aboriginal and non-Aboriginal families is similar, with the risk of poverty for both groups increasing with the number of children and being substantially higher for sole parent units than for couple units. There is also a tendency for the differential between the two groups to narrow as the number of children increases. A major factor identified as underlying these differences is the increased prevalence of joblessness among Aboriginal families.

Trends over time

One advantage of using the HPL framework to estimate poverty is that consistent estimates of the trend in poverty over time can be derived. Table S2.6 brings together the results from the studies by Gallagher (1985) and Saunders (1994) which present poverty estimates for each of the years since 1972-73 in which the ABS has undertaken an income survey. Although the two studies apply the same broad methodology, there are some differences which cause each to produce slightly different estimates for the one year (1981-82) which is included in both studies.
S2.6 ESTIMATES OF INCOME UNITS IN POVERTY, 1972-73 TO 1989-90(a)

1972-73
1973-74
1978-79
1981-82
1985-86
1989-90






No.
No.
No.
No.
No.
No.
Income unit type
('000)
%
('000)
%
('000)
%
('000)
%
('000)
%
('000)
%

Single person
230.1
18.3
319.5
18.3
208.5
10.7
296.0
12.2
418.7
17.0
563.5
20.4
Sole parent with
1 child
n.a.
n.a.
25.8
26.7
27.8
26.4
44.5
31.7
56.8
44.1
95.0
52.9
2 or more children
n.a.
n.a.
38.9
49.1
48.5
46.3
66.6
60.5
72.7
66.4
95.5
64.8
All sole parents
54.1
33.8
64.7
36.8
76.3
36.3
111.1
43.5
129.4
53.7
190.5
58.0
Couples with
0 children
14.0
1.9
53.5
4.1
66.8
5.1
46.6
4.1
49.7
3.8
98.0
6.4
1 child
26.7
2.9
23.3
4.1
24.1
5.1
21.0
5.0
19.0
4.5
38.7
8.6
2 children
26.7
2.9
25.6
3.9
40.4
6.5
33.0
5.7
37.5
6.6
48.1
8.4
3 or more children
30.7
7.2
74.0
12.6
47.1
11.5
42.8
16.8
60.2
20.3
54.0
19.9
All couples with children
57.4
5.2
122.9
9.2
111.6
8.2
96.8
7.2
116.6
9.0
140.9
10.5
All income units
399.4
10.2
560.6
11.1
463.1
9.3
550.2
10.7
714.5
13.4
992.9
16.7

            (a) Estimates exclude all self-employed income units.
            Source: For 1972-73 to 1978-79; Gallagher, 1985, Table 1. For 1981-82 to 1989-90; Saunders, 1994, Table 9.2.

Table S2.6 shows that the overall poverty rate was broadly constant at around 10% in the decade up to 1981-82. Over this period, poverty declined markedly among single people but increased among couples, with and without children. After 1981-82, the overall poverty rate rose steadily, reaching 16.7% by 1989-90 according to Saunders (1994). This period also saw a substantial reversal of the earlier decline in poverty among single people, whose poverty rate almost doubled from 10.7% to 20.4%.

One of the factors underlying the increase in poverty after 1981-82 is the fact that the poverty line itself increased in real terms by 1.2% a year between then and 1989-90 - corresponding to a total real increase of just over 10%. However, this is to be expected of a relative poverty line in a period when real community incomes are rising. The real annual increase in the poverty line in the 1980s is actually lower than between 1972-73 and 1979-80, when the poverty line was adjusted in line with average weekly earnings.

International comparisons

Comparing poverty rates across different countries is a difficult exercise. In the first place, it requires the availability of data which conform to a common definitional and measurement framework with regard to key concepts such as income and the family unit. Second, a method for establishing a poverty line is required which permits meaningful cross-national comparisons to be made whilst at the same time recognising that the poverty line should reflect national conditions and circumstances.

Considerable progress in overcoming the first obstacle (data comparability) has been made since the advent in 1983 of the Luxembourg Income Study (LIS), an international project with the goal of rendering household income survey data comparable across countries by ensuring that they conform to a common definitional framework. To overcome the second obstacle, a value judgement is required which can receive broad support, both within and across countries.

To this end, researchers have adopted a poverty line for each country which is set at a fixed percentage of median equivalent family income. The measure has the advantage that it reflects differences in the standard of living in each country (as incorporated in the level of median income) whilst at the same time aiding comparisons between countries by fixing the poverty threshold at the same relative level in each country. The exact percentage of median income adopted as the poverty threshold can be varied in order to assess the sensitivity of results. Because this poverty line bears no relation to national poverty benchmarks, it is perhaps more accurately described as a low income, rather than a poverty, threshold.



Graphs S2.7 and S2.8 summarise what the existing research using the LIS database indicates about the extent of poverty across a range of countries (including Australia) belonging to the Organisation for Economic Cooperation and Development (OECD). Graph S2.7 shows the percentage of persons in each country who are living in families with incomes below a poverty line set at 50% of median income. The estimated poverty rate varies from around 4% in Belgium and the Netherlands to almost 19% in the United States, with an average value across all thirteen countries of just below 9%. Australia's poverty rate, at 12.9%, is second highest after the United States, similar in magnitude to that existing in Canada, Ireland and Italy.

Graph S2.8 compares the poverty rate among non-elderly families only, defined as those headed by a person aged less than 60 (Forster, 1994, table 1). Again, the cross-country variation in poverty is considerable, ranging from around 5% in Belgium, Luxembourg and the Netherlands to over 15% in Australia, Canada and Ireland, with the United States again having the highest poverty rate (18.7%). The average rate for all countries shown is 10.3%. The fact that this is slightly above the average rate shown in graph S2.7 reflects the use of a different unit of analysis in the two graphs and the differential poverty rates amongst non-elderly and elderly families in each country.

Finally, it is worth noting that Australia's poverty ranking as shown in Figures 3 and 4 does not change markedly if the low-income benchmark is changed from 50% of median income to either 40% or 60%.

Deprivation

As explained earlier, in order to estimate the extent of deprivation it is necessary to investigate peoples lifestyles and living conditions directly and quantify those dimensions in which they are experiencing deprivation. Studies of Australian living standards conducted by Travers and Richardson (1993) and the Australian Institute of Family Studies (McDonald and Brownlee, 1994), though not addressing the issue of deprivation directly, have produced results which are of relevance. This new research complements earlier qualitative studies (e.g. Trethewey, 1989) of the lives and experiences of those on low-incomes which document what poverty and deprivation mean in everyday terms.

The work of Travers and Richardson, based on a national survey of Australian living standards conducted in 1987, reveals that comprehensive measures of full income (defined to include estimates of the value of leisure and ownership of assets) are only weakly correlated with the extent of various forms of social participation. Furthermore, the correlations between equivalent money income and other non-monetary dimensions of full income are also rather low, which suggests that inequalities in the different spheres of material well-being tend to offset each other. These findings imply that those who are defined as poor according to the level of their money incomes will not necessarily be experiencing the lowest levels of deprivation, nor are they always excluded from participating in normal social activities.

The Australian Living Standards Survey undertaken by a research team at AIFS has investigated living standards across fourteen spheres of life for families with children. In addition to economic resources, the aspects considered include health, employment, housing, transport, leisure and recreation, access to local and community services, safety and security, social and political participation and family functioning. Analysis of data on almost 1800 households living in four Melbourne localities reveals that thresholds of living standards exist which mark the lowest fifth of households (ranked in terms of equivalent household income) distinct from the top 30% of households, with the 50% of households in the middle forming a third grouping.

A number of indicators of relative deprivation for those at the bottom are discovered in the data, including a lack of assets, the incidence of financial difficulties, housing problems, below-average health and the perception that children in low-income households are worse off than other children (McDonald and Brownlee, 1994, Figures 1 to 9).
S2.9 INDICATORS OF HOUSING STRESS BY INCOME UNIT TYPE, 1990 (%)

Income unit type

    Indicator
Singel person
Couple (no children)
Two parent family
One parent family
All income units

    High expenditure on housing(a)
7.4
3.8
5.3
21.2
6.6
    No internal bathroom
0.3
0.5
0.3
0.3
0.4
    No internal toilet
0.7
0.4
0.3
0.2
0.5
    No internal bathroom or toilet
0.4
0.1
0.1
0.2
0.2

            (a) High expenditure is defined as equivalent to 30% or more of disposable income. These figures relate only to those income units in the lowest 40% of the distribution of equivalent income (defined using the Henderson equivalence scale).
            Source: Australian Institute of Health and Welfare, 1993, Tables 3.2 and 3.3.


Further information on high housing costs and inadequate housing conditions as indicators of deprivation has been presented by the Australian Institute of Health and Welfare (AIHW). The measures indicate that high housing costs are a problem for many low-income Australians, particularly for sole parent families (table S2.8). In relation to inadequate housing conditions, however, there are few Australians without an internal bathroom and toilet and where these do not exist, there is no consistent pattern of differences across income unit types.

Saunders (1994) utilises data from the 1988-89 Household Expenditure Survey to derive several indicators of deprivation in relation to inadequate housing conditions, lack of assets such as a car, lack of a telephone and no insurance protection of such assets. These indicators present a more pessimistic picture than that shown in table S2.8, although what is most significant about the results is that they again reveal clear differences across family types, with deprivation most prevalent amongst single aged people, larger two-parent families and sole parent families generally. These results, combined with those in tables S2.2 and S2.6, suggest that deprivation and poverty seem quite closely related in Australia.

Concluding comments

The International Year for the Eradication of Poverty provides an opportunity not only to review the research on poverty, but also to promote policies which can eradicate poverty and abolish deprivation. While there is little evidence from the available statistics that absolute poverty is widespread in Australia, groups such as the homeless and Indigenous Australians who tend not to appear in the conventional statistics are poor even on a subsistence basis. In addition, relative poverty is a reality for many more Australians, including many families with children. High levels of unemployment, particularly long-term unemployment, and a continuation of the increase in inequality will only exacerbate the number of Australians at risk of poverty.

Addressing areas of absolute and relative disadvantage will require government action in many different spheres and at many different levels. Government income transfers are essential to provide a comprehensive and adequate social safety net. Improved access to market incomes is also important for overcoming the marginalisation and social exclusion which often accompany enforced joblessness. The provision of adequate housing and community services which contribute directly to living standards is also important, as is the development of an environment in which individuals can achieve their own aspirations for material well-being and personal fulfilment.

Poverty in rich countries like Australia can be eradicated. We have the resources. We need to agree that the eradication of poverty is a worthwhile objective and, having agreed, to pursue that goal with a sense of purpose and commitment.

Bibliography

ABS Publications

Australian Bureau of Statistics (1994), Australian Social Trends 1994 (Catalogue No. 4102.0), ABS, Canberra.

Other Publications

Australian Institute of Health and Welfare (1993), Australia's Welfare 1993. Services and Assistance, AGPS, Canberra.
Baldwin, P. (1995), Beyond the Safety Net. The Future of Social Security, Office of the Minister for Social Security, Canberra.
Commission of Inquiry into Poverty (1975), First Main Report. Poverty in Australia, AGPS, Canberra.

Förster, M. (1993), Comparing Poverty in 13 OECD Countries - Traditional and Synthetic Approaches', Working Paper No. 100, The Luxembourg Income Study, Luxembourg.
Förster, M.F. (1994), The Effects of Net Transfers on Low Incomes Among Non-Elderly Families, OECD Economic Studies, No. 22, Spring, pp.181-221.
Gallagher, P. (1985), Targeting welfare expenditures on the poor: Work in progress on poverty in Australia, 1981-82', Social Security Journal, December, pp. 19-34.
Henderson, R. F., Harcourt, A. and Harper, R. J. A. (1970), People in Poverty. A Melbourne Survey, Cheshire for the Institute of Applied Economic and Social Research, Melbourne.

Institute of Applied Economic and Social Research (1995), Poverty Lines: Australia. March Quarter 1995, IAESR, University of Melbourne.
Johnson, D. (1987), The Calculation and Use of Poverty Lines in Australia', Australian Economic Review, 4th Quarter, pp. 45-55.
Johnson, D. (1991), The Measurement and Extent of Poverty Among Immigrants, Bureau of Immigration Research, AGPS, Canberra.
McDonald, P. and Brownlee, H. (1994), Australian Living Standards: The Next Decade, in J. Disney and L. Briggs (eds.), Social Security Policy: Issues and Options, AGPS, Canberra, pp.49-62.

Paul, S. and Podder, N. (1992), Measuring the Risk of Poverty in Australia', Economic Papers, Vol. 11, No. 3, pp. 23-31.
Ross, R. and Whiteford, P. (1992), Poverty in 1986: Aboriginal Families with Children, Australian Journal of Social Issues, Vol. 27, No. 2, pp.91-111.
Saunders, P. (1994), Welfare and Inequality. National and International Perspectives on the Australian Welfare State, Cambridge University Press, Melbourne.
Saunders, P. (1995), In Defence of a Poverty Line, Just Policy, No. 4, September, pp. 9-16.

Townsend, P. (1979), Poverty in the United Kingdom. A Study of Household Resources and Standards of Living, Penguin Books, Harmondsworth.
Travers, P. and Richardson, S. (1993), Living Decently. Material Well-Being in Australia, Oxford University Press, Melbourne.
Trethewey, J. (1989), The Everyday Life of Children in Poverty', in D. Edgar, D. Kean and P. McDonald (eds.) Child Poverty, Allen and Unwin, Sydney, pp. 50-58.
United Nations Development Programme (1994), Human Development Report 1994, Oxford University Press, New York.

World Bank (1990), World Development Report 1990. Poverty, Oxford University Press, New York.

Source: Year Book Australia, 1996 (ABS Catalogue No. 1301.0)


Bookmark and Share. Opens in a new window

© Commonwealth of Australia 2014

Unless otherwise noted, content on this website is licensed under a Creative Commons Attribution 2.5 Australia Licence together with any terms, conditions and exclusions as set out in the website Copyright notice. For permission to do anything beyond the scope of this licence and copyright terms contact us.