Australian Bureau of Statistics

Rate this page
ABS Home > Statistics > By Release Date
ABS @ Facebook ABS @ Twitter ABS RSS ABS Email notification service
4102.0 - Australian Social Trends, 1998  
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 03/06/1998   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product  
Contents >> Income & Expenditure >> Income Distribution: Poverty: different assumptions, different profiles

Income Distribution: Poverty: different assumptions, different profiles

Using the Henderson Poverty Line as a benchmark, there were over 3.4 million Australians living in poverty in 1995-96. However, other poverty lines can give quite different results.

The term 'poverty' means different things to different people. Some think of poverty as being without a home or food. Others may think that a family is poor if it cannot afford to send the children on school excursions. An elderly couple living in a wealthy suburb may feel poor because they find it difficult to pay high council rates for their home.

The literature on poverty reflects these various views. Briefly, experts recognise three different approaches to defining poverty:

  • absolute poverty, where a family's income does not pay for basic necessities such as shelter and food;
  • relative poverty, where a family's income is low in comparison to the income of other families; and
  • subjective poverty, where a poor family is defined as one that believes its income is inadequate for its needs.

Units of measurement

Income units are made up of one person or a group of related persons within a household, whose income is assumed to be shared. Income sharing is assumed to take place within married (registered or de facto) couples and between parent(s) and dependent children.

Net income is gross income minus direct taxes.

Poverty Lines are threshold income values. If a family's income is below the value applicable for that family, then that family is deemed to be in poverty.

Dependent children are those aged less than 15 years, or 15-24 and both studying full-time and living with parent(s) and do not have a spouse or offspring of their own living with them.

People in relative poverty
In developed countries such as Australia, poverty studies usually adopt the relative approach. All families are ranked according to levels of equivalent income, and a poverty line is drawn at some point on the distribution. If the family's income is below that line, then they are said to be in poverty.

However, the level at which that poverty line is drawn is subjective. As illustrated in the table opposite, poverty lines set at different levels result in different estimates of the number of people in poverty.

The first column presents estimates of people in poverty using the 'All Costs' Henderson Poverty Line (HPL), the line most commonly used in Australian studies. It was adopted by the Commission of Inquiry into Poverty in the early 1970s1 and has been regularly updated since then2. In 1995-96, the HPL for two adults (both working) and with two teenage children was net income of $496 per week.

The second poverty line shown uses an approach commonly used in European countries and in international comparisons of poverty. The line is set at 50% of median equivalent income (MEI) for all income units. The resultant line for the same family unit of two adults and two children was a weekly net income of $353.

The results are clearly different. In 1995-96 there were approximately 8.9 million income units in private dwellings in Australia. Of these, 1.8 million income units (3.4 million people, 1 million of whom were dependent children) were below the HPL.

The alternative poverty measure of 50% of MEI produced a much lower estimate of those in poverty: 904,800 income units or 1.8 million people.

There were also some marked differences in the types of income units below the two poverty lines. In both cases, most of the 'poor' were one-person units. However, with the HPL a higher proportion of those in poverty were elderly units (19%). Using the line of 50% of MEI, very few of the poor were elderly units, and a high proportion (32%) were young one-person units.


Net income less than

HPL(a) $496(c)
50% of MEI(b) $353(c)

Income units

Income units

(a) Henderson Poverty Line.
(b) Median equivalent income for all income units.
(c) Weekly net income for two adults and two children.

Source: Unpublished data, Survey of Income and Housing Costs, 1995-96.

Equivalence scales

When measuring income poverty, it is necessary to adjust family (or income unit) incomes to take account of the different living costs to families of different sizes and composition. Obviously, a family with three children needs more income than a single person to achieve a similar standard of living. These adjustments are made by applying a set of equivalence scales to family income and deriving an equivalent income.

Many different sets of equivalence scales have been devised and there is no general agreement among analysts about the most appropriate set to use. The choice is subjective, and affects the types of families deemed to be in poverty. For example, those that give high costs to providing for children will result in more families with children being deemed to be in poverty.

Two sets of equivalence scales have been used in this review. The first, called the Henderson Simplified Scales (used by the Commission of Inquiry into Poverty) adjusts the net income of families according to: numbers of family members, the differing costs of family members depending on their age and engagement in employment, and, for the family as a whole, their housing costs. These Henderson Scales are applied to income for use with the Henderson Poverty Line (HPL). The second set of equivalence scales is simpler, and is used before applying the poverty line set at 50% of Median Equivalent Income (MEI). These scales, devised by the Organisation for Economic Cooperation and Development (OECD), accommodate differences only in the numbers of adults and children in families.

Sensitivity of poverty measure
The numbers of people (or units) in poverty can be greatly affected by small changes in the level of the line, because of the tendency for incomes to be clustered.

Income clustering is particularly common for elderly people because of the flat rate of pension paid by government. This means that all those on the full rate of age pension receive almost exactly the same amount, with only minor differences due to the added value of rental assistance.

If this pension rate is very close to the particular poverty line chosen for analysis, then it takes only a very small change in either the pension rate or the poverty line to markedly alter the numbers deemed to be in poverty.

For elderly single people in 1995-96, the HPL fell just above the maximum rate of their pension. The result was that, of the 341,000 elderly single units below the HPL, some 226,000 were clustered between 90% and 100% of the poverty line. Therefore, a very slight change in the pension rate or the value of the poverty line may have excluded this group from measured poverty. It would not, however, have reflected any real changes in the purchasing power or living conditions of this group.

For this reason, an additional measure of a range across the poverty line is often used. Poverty may be measured by including those groups that are 10% or 20% above the HPL. Again, this results in different numbers and profiles of the poor.


Below HPL
Below 50% of MEI

Income unit type
    Under 25
Couples only
    Under 65
Couple with children
All units

All units

(a) Women aged 25-59 and men aged 25-64, the upper bounds reflecting ages below eligibility for government Age Pension.
(b) Women aged 60 years and over; men aged 65 years and over.
(c) Reference person aged 65 years and over.

Source: Unpublished data, Survey of Income and Housing Costs, 1995-96.

People at risk
The risk of poverty for different groups of people is measured using poverty rates. This is the proportion of the group who fall below the poverty line.

Again, there were differences between the two approaches. The income units most likely to be below the HPL were elderly one-person units (37% poverty rate) and one-parent units (34%). However, using the 50% of MEI measure, young one-person units (21% poverty rate) and one-parent units (17%) were most likely to be below the poverty line.

The largest variation in poverty rates between the two measures was in the rates for aged one-person units. Approximately 37% of these were below the HPL compared to 6% below the alternative poverty line of 50% of MEI.

This divergence in poverty rates for the same group is due in part to the effects of clustering of aged peoples' incomes and the close relationship that exists between government pension rates and the HPL.


(a) Income units with nil or negative income.

Source: Unpublished data, Survey of Income and Housing Costs, 1995-96.

Statistical units
The discussion in this article has been based on 'income units' as the unit of analysis. This means that it was assumed that sharing of income within families and households takes place only within the restricted nuclear family of couples and parent(s) and dependent children.

The choice of the statistical unit for analysis has a major impact on poverty estimates. Generally, the estimates of poverty rates decrease as the size of the statistical unit increases. That is, there are higher proportions of income units in poverty than of families, and higher proportions of families than households.

The main difference between income units and families is that, for income units, children aged 15-24 who are living with parents but are not full-time students are considered to be financially independent and therefore constitute separate one-person income units.

Yet some of these young people have very low income. In 1995-96, around 104,000 young one-person units had no income. Many of these were still living with parents. This raises the question of whether these young people are actually financially independent of their parents as classified in the surveys.

Because the classification of youth as dependent or non-dependent is so subjective in poverty analysis, it is useful to look not only at their own income but also that of their parents.


Below HPL
Below 50% of MEI
All units
Income unit type

    Under 25
Total one-person
Couples only
    Under 65
Total couples only
Couples with children
All units

(a) Women aged 45-59 and men aged 45-64, reflecting age below eligibility for government Age Pension.
(b) Women aged 60 years and over; men aged 65 years and over.
(c) Reference person aged 65 years and over.

Source: Unpublished data, Survey of Income and Housing Costs, 1995-96.

Young income units living with parents
In 1995-96, 918,000, or two out of every three one-person income units aged 15-24, were living with parents. Of these, 51% were living rent free. Of these young income units, 249,000 had income below the HPL. This gives a poverty rate for this group of 27%. However, when the income of their parents is examined, it is clear that most of them live in larger family units that are not below the poverty line. Of those 249,000 youth who were in poverty when their own income was measured, only 35,000 had parents who were also deemed to be in poverty according to the HPL. This represents only 4% of all 'independent' young income units who are living with parents. This is a very different picture from their poverty rate of 27% when their income alone was taken into account.

Of course, this much lower figure of 4% assumes the other extreme, that income is shared evenly between parents and children. The real situation may fall somewhere in between these two conditions. However, it does serve to illustrate that different measures will produce quite different poverty profiles.


Youth income

Below HPL
At or above HPL
All Youth
Parental income

Below HPL
At or above HPL
All Parents

Source: Unpublished data, Survey of Income and Housing Costs, 1995-96.

Resources other than income
While estimates of poverty based on cash income vary according to the measures used, even more variation is likely when the definition of economic resources is extended. For example, living standards of families are also affected by in-kind income such as fringe benefits and pensioner concessions. Services such as free child care by relatives will also increase the overall resources available to families who would otherwise have to pay for them.

The government also provides substantial indirect benefits through expenditure on education, health, housing and welfare.

Receipt of irregular, lump sum money and benefits from families' holdings of assets may also be important.3

A family's stock of wealth is particularly important for its economic wellbeing, and may affect it in a number of ways. Some forms of wealth provide a return in the form of regular income. This interest, rent or dividends has been included in cash income for poverty analysis.

However, for most Australian families, their major wealth holdings are tied up in their homes. While home ownership may not provide for a cash return in the form of income, it offers other economic benefits to its owners such as cheaper housing. It may also offer greater economic security in that such an asset may be borrowed against or realised through sale.

In recognition of the varying costs of housing, Henderson provided an alternative poverty line for income after housing costs had been deducted. By applying this 'After Housing Costs' HPL to the 1995-96 data, the poverty rate is reduced from 20% of income units to 18%.

The reduction in measured poverty when housing costs have been taken into account was much greater for the elderly as they have very high rates of home ownership. In the case of aged one-person units, the proportions below the 'All Costs' HPL and the 'After Housing Costs' HPL were 37% and 9% respectively. This reflects the fact that about 69% of the elderly single units below the 'All Costs' HPL in 1995-96 owned their homes outright.

Other poverty measures
In the above analysis, poverty is measured indirectly - that is, via the cash income resources available to families to purchase goods and services. However, in other parts of the world, Scandinavian countries in particular, more direct measures are used to examine the actual living standards achieved by families.

In such studies, families are questioned on their ownership of particular goods, how well they manage on their income and whether there are things they have to do without because of lack of income. In Australia there have been a number of living standards projects that have caused a resurgence of interest in these types of measures.4,5,6

In their recent study of a sample of people receiving social security pensions and benefits, Travers and Robertson reported that there was a weak correlation between those who were 'poor' on the income measure and those who were 'poor' on the more direct living standards measure.5

It is fairly safe to assume that the use of this type of measurement technique would change the estimates and profiles of those groups in Australia who are most in need.


1 Commission of Inquiry into Poverty, 1975, First Main Report, Poverty in Australia, AGPS, Canberra.

2 Institute of Applied Economic and Social Research (IAESR), June 1996, Poverty Lines: Australia, IAESR, Melbourne.

3 Australian Bureau of Statistics, 1995, A Provisional Framework for Household Income, Consumption, Saving and Wealth, cat. no. 6549.0, ABS, Canberra.

4 Travers, P., Richardson, S. 1993, Living Decently; Material Well-being in Australia, Oxford University Press, SA.

5 Travers, P., Robertson, F. 1996, Relative Deprivation Among DSS Clients: Results of a Pilot Survey, National Institute of Labour Studies, Monograph Series No 2, NILS, SA.

6 de Vaus, D. 1996, The Australian Living Standards Study in Australian Institute of Family Studies, Family Matters, No 43, Autumn.

Previous PageNext Page

Bookmark and Share. Opens in a new window

Commonwealth of Australia 2016

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.