Number 6: Trends in the incomes and living standards of older people in Australia

This report was published by the former Department of Families, Community Services (FaCS).

Executive Summary

Introduction

In common with other wealthy societies, Australia has a complex system of social protection for retirement and old age. This system includes government income support mainly provided through the Age Pension and Service Pension, mandatory superannuation contributions and tax concessions for superannuation, the public health system, community services such as homes and hostels for the aged and public housing, plus other social institutions including private home ownership.

The most important objectives of government assistance are to alleviate poverty and to assist individuals to maintain adequate living standards. Other important objectives include encouraging self-provision and avoiding undesirable incentive effects.

The Australian public pension system differs from those in most similar societies by giving greater priority to the objective of poverty alleviation, achieved through a general revenue financed system that provides flat-rate and means-tested benefits. The distinctive nature of the Australian system makes concerns about adequacy of payments particularly salient.

In assessing the implications of population ageing for income support and related policies, it is important to have a soundly based analysis of the effectiveness of existing policies in achieving distributional objectives. To assist in such analysis, this paper provides a detailed description of trends in the incomes and living standards of older people in Australia, using a wide range of indicators and alternative approaches to measuring living standards.

Assessing income distribution and pension adequacy - conceptual and measurement issues

This paper argues that it is necessary to have a comprehensive framework for measuring material living standards. This framework should include all components of cash income, and should also take account of government services and subsidies and indirect taxes. It is also important to take account of household wealth.

In addition, in analysing levels of living standards and trends in wellbeing, it is useful to use a number of measures and indicators. A multi-indicator approach may capture a broader range of circumstances, and it is important to be aware of the sensitivity of measured outcomes to the choice of measures of wellbeing.

The Age Pension - Assessing adequacy

Just over 80 per cent of the eligible population receive an Age Pension or Department of Veterans' Affairs (DVA) Service Pension. Just under 60 per cent of age pensioners have incomes below the pension free area, and around one-third have incomes above the free area. Male age pensioners are more likely to have additional income than female pensioners. Around 10 per cent of age pensioners have no additional income, and so are completely 'dependent' on the pension.

Since the mid-1960s, the real value of the basic pension has increased by around 80 per cent for single people and 63 per cent for couples. Over the same period, the standard pension has increased from 22 to 25 per cent of male total average earnings. Compared to low paid workers, the pension has increased from 35 to 45 per cent of a process worker's wage. The pension has fluctuated widely as a percentage of GDP per capita and as a percentage of Household Disposable Income per capita, although in both cases the long-term trend appears to be a reduction. The standard rate of pension has also fluctuated as a percentage of the 'Henderson Poverty Line' for a single older person, but is currently at its lowest level relative to this indicator. Variations in the rate of pension relative to these indicators reflect the differential movements in their real values, the effects of economic cycles, and the differing composition of income measures used.

While the concept of replacement rates is not entirely relevant to a flat-rate, means-tested payment such as the Age Pension, it can be calculated that the current pension goal of 25 per cent of male average earnings provides higher replacement rates after tax is netted out, higher rates for women and for low-paid workers, and higher replacement ratios when housing costs and employee superannuation contributions are taken into account.

Private incomes and assets of age pensioners

The proportion of age pensioners with incomes above the free areas has increased from around 10 to 15 per cent in the 1960s to one-third in the late 1990s. There have been wide fluctuations in the intervening period, reflecting changes in income testing policy. The most common form of income is from savings and investments in banks, building societies or credit unions (90 per cent), and around 10 per cent of age pensioners have incomes from superannuation.

Two-thirds of age pensioners own their own home. Ninety-two per cent of pensioners have additional assets (including personal effects). In 1998, the mean value of these assets was $40,800. Thirty per cent of pensioners have assets (not including the home) of $50,000 or more, but about one-quarter have no assets or assets of less than $5,000, including personal effects. Women have lower asset levels than males, particularly divorced, separated or single female pensioners.

Trends in the cash incomes of older people

This paper reviews trends in the cash incomes of the older population since the early 1980s using Australian Bureau of Statistics' (ABS) Income Surveys. The average total incomes of older people have increased at a faster rate than for the population generally (13 per cent compared to 2 per cent). As a result, the average income of older people has risen from 50 to 55 per cent of the average for the entire population. The main factor associated with this appears to be an increase in the 'other private income' of couples, mainly from superannuation, property and investments. This has resulted in a decline in the proportion of older couples who receive between 50 and 90 per cent of their income from government benefits, but little change in the proportion who rely more substantially on government benefits. The income composition of single older people appears to have changed only to a limited extent.

Home ownership rates have increased for both older singles and couples, although ownership rates remain considerably higher for couples.

Older people are very highly concentrated in the lowest 40 per cent of the income distribution, but somewhat less concentrated when account is taken of their lower family size. The degree of their concentration in the poorest income quintile is highly dependent on the equivalence scale used to adjust for family size.

The reason for this sensitivity is that the older population is extremely concentrated around the level of the pension or slightly above. The degree of this concentration appeared to reduce significantly between 1986 and 1990 - probably reflecting the effects of high interest rates around 1990. Concentration increased again between 1990 and 1996, but not back to the 1986 level.

Comparative analysis suggests that the cash income distribution for older people in Australia is more concentrated (in this sense) than in many European countries and in North America.

This paper presents a further range of data where incomes are 'equivalised' - adjusted for the number of people in the income unit. The main conclusion drawn from this analysis is of apparent relative stability in the income circumstances of older people over the period 1986 to 1995–96, although trends between 1986 and 1990 were partly reversed in the 1990s. The analysis of equivalent incomes shows, for example, that the relative incomes of older people compared to the rest of the population have not changed greatly, and that around half of the older population remain in the bottom 30 per cent of the overall income distribution in each of the three years analysed. Inequality among pensioners appears to have fallen slightly, but this may reflect changes in the composition of the pensioner population rather than changes in incomes within pensioner groups.

Trends in household expenditure levels

This paper uses Household Expenditure Survey data to analyse trends in expenditure levels of older households between 1974–75 and 1993–94. Trends in household incomes and expenditures over this period are significantly affected by changes in household size, which have fallen, but more substantially for younger households than for older households.

Over this period, the per capita real incomes of older households fell by 6.6 per cent, but real expenditure per capita rose by 15.6 per cent. This compares with a real increase of 0.4 per cent in the real incomes per capita of all households and an increase of around 10 per cent in real per capita expenditures. This underlines the propensity of older people to 'dis-save' over time.

As a result, the average incomes per capita of older households appear to have fallen from 83 to 76 per cent of the per capita household income of the population as a whole. On the other hand, the expenditure trend has been in the opposite direction, with per capita expenditures of older households rising from 84 to 89 per cent of the population generally.

The impact of non-cash benefits and indirect taxes

Government non-cash benefits in the form of services and subsidies also have a substantial impact on the living standards of the population generally, and particularly on older people. The ABS has estimated that in 1993–94 the value of government services and subsidies for households with a reference person aged 65 years and over was $145 per week compared to cash benefits of $185 per week. Indirect taxes paid by older households are estimated to be roughly equal, on average, to their income tax liabilities.

Health benefits and other welfare services are most significant for the older population and education benefits are most important for the younger population.

Taking account of these services, subsidies and indirect taxes has a significant impact on the measured living standards of older households. For example, in 1993–94 the private income of older households was only one-quarter of that of all households. Government cash benefits increased this ratio to 48 per cent, and the higher income taxes paid by younger households increased this further to 54 per cent. After taking account of services, subsidies and indirect taxes the ratio rose to 66 per cent. Put another way, the cash disposable incomes of older households are about 73 per cent of their final incomes.

Between 1984 and 1993–94, indirect benefits and taxes have become slightly more 'pro-aged'. This appears to reflect an increase in the relative contribution of public health benefits for older couples and older single person households, and an increase in the relative contribution of other welfare services for older couples.

Trends in relative low incomes

In assessing trends in the wellbeing of the Australian population, a common form of analysis is to estimate how many people have incomes below some measure of poverty or of relative low income. This paper includes a number of estimates of the level of relative low income among the older population and trends over time.

The different measures produce different results. For example, for single older people, the proportion with relative low incomes ranges from 32 per cent using the Henderson Poverty Line to 14 per cent below half median annual income and 5 per cent below half median weekly income. Including the effect of non-cash benefits and indirect taxes reduces the low-income rate to around 2 per cent for older single people.

Part of the variability of these results reflects technical choices made in measurement, the interaction between these choices and the very high degree of concentration in the incomes of older people that were discussed earlier. Because so many older Australians have incomes in a relatively narrow income range of between 40 and 60 per cent of average income, small differences in the level of the low-income line used can have a large impact on rates of low income.

The Henderson Poverty Line shows the largest increase in poverty over the period 1981–82 to 1995–96. A major contributor to this is that the Henderson line has been rising faster than average incomes in the income surveys, because it includes imputed income from owner- occupied housing and the earnings of superannuation funds - neither of which are taken into account in income surveys. There are also doubts about the comparability of the annual income data in the ABS Income Surveys from 1994–95 onwards.

Studies using the Henderson line give a mixed picture of trends in the circumstances of older income units. King (1998) estimates that between 1972–73 and March 1996 the Henderson poverty rate (before housing costs) among single older people rose marginally (but was more than 30 per cent in both periods) and among older couples it fell slightly (from 5 to 3.8 per cent). After housing costs, poverty rates were substantially lower for singles but not for couples, and they fell over this period. In contrast, Saunders (1994) estimated that between 1981–82 and 1989–90 'Henderson poverty' increased from 10 per cent to 28 per cent, while among older couples it increased from 4.3 to 6.7 per cent.

Housing wealth

A significant factor contributing to the living standards of older people is their ownership of homes. Home ownership is widespread among the older population, and the level of home ownership is more equally distributed by income level than most other forms of private income.

The value of dwellings owned by people aged 65 years and over is lower than among most of the younger population, but the level of loans outstanding is much lower than for most groups of younger people. As a result, older people have higher average equity than people under the age of 45 years.

Conclusions

The picture that emerges from this analysis is mixed. The average incomes of older people have increased at a faster rate than the population generally. As a result, their average incomes have risen as a proportion of the average for the community as a whole. The average expenditures per person among older people have also increased. Taking account of government non-cash benefits would further improve the relative position of older people.

At the same time, administrative data suggest that there are sizeable proportions of the age pensioner population who have little or no income apart from their pension, and little or limited assets. However, the extent to which this is the case appears to have decreased over time. Older people are also over-represented in the lower income quintiles of the population.

One of the most striking features of the incomes of the older population is the degree of concentration of incomes around pension levels. This complicates interpretation of trends in incomes and the relative position of this age group, including their vulnerability to low incomes.

In considering future trends in the circumstances of older people, it is necessary to take account of factors impacting on the distribution of incomes of those in the pre-pension age groups. The wellbeing of the older population in future is likely to be enhanced by a wide range of factors, including increasing superannuation coverage, increasing labour force participation among women, higher real wages, and higher average levels of housing wealth. At the same time, there are trends that may tend to offset these, including the long-term decline (until the early 1990s) in the workforce participation of men aged 50 to 64 years, and higher wage inequality among those of working age. In addition, family trends, including the growth in the incidence of sole parent families, may also have adverse effects. Separated, divorced and single older women appear to have lower incomes and assets in retirement than men or couples. The compression of life course events related to women being older at the birth of their first child and increased educational participation among young people may also impact on people's capacity for self-provision in retirement.

In terms of future monitoring of these and related trends, it is desirable to have improved information about the dynamic processes that are associated with these developments. This would be best achieved through an ongoing longitudinal survey. To capture the diversity of outcomes among the older population, it is also necessary to use a broad range of indicators to monitor trends.

1 Introduction

All developed societies have a range of policies to provide for income protection in retirement and old age. The Australian retirement income system includes income support provided through the Age Pension and the Service Pension, plus the mandatory superannuation system and tax concessions for superannuation. The public health and health insurance systems; concessions to defray certain costs (such as for health, public transport or utilities); community services and other services (such as public housing and institutional and community care); and other social arrangements (such as private home ownership) are also important components of social protection for older people. The retired may also benefit from private transfers in cash or in kind from their families. Government activity to promote social protection for older people has a range of objectives, including the alleviation of poverty or the maintenance of pre-retirement living standards (Donald 1984; Foster 1988); the encouragement of self-provision; the avoidance of undesirable incentive effects; or the minimisation of government expenditures. Typically, systems have multiple objectives, with the result that objectives may conflict.

The Australian social security system differs from those in most other countries. One of the most striking features of the Australian system is the extent to which it distinguishes the poverty reduction objective from the income replacement objective of retirement provision, with the public system concentrating on poverty alleviation and the second and third pillars concentrating on income replacement. Consequently, the level of public pension spending is low by the standards of the OECD countries with which Australia is usually compared, reflecting the provision of flat-rate and means-tested benefits. At the same time, coverage of the system is comprehensive. The system is also highly redistributive to groups often poorly served by social insurance systems, such as women, those with long-term disabilities, low wage earners and others with marginal or incomplete attachment to the labour force.

The design features of the Australian system give rise to a number of questions about the effectiveness and efficiency of current arrangements. As noted by Creedy and Disney (1989, p. 357), such a system has a number of built-in tensions, including between the adequacy of benefit levels at the very low end of the income distribution and the high marginal tax rates implied by means-testing. The tension between concerns for adequacy, efficiency and incentives is likely to remain important as the Australian population continues to age over the next half century or more. The Age Pension is the largest income support program currently provided by the Commonwealth Government. Between 1965 and 1998, spending on age and related pensions increased from $2,900 million to $13,100 million (in $1996–97), or from 1.65 to 2.45 per cent of GDP. Concerns about the adequacy of the Age Pension have been significant contributors to this increased spending. On average, real expenditure grew by around 5 per cent per annum between 1965 and 1995, with increases in real rates of payments estimated to result in an average increase of 1.9 per cent per year in Age Pension spending over this period (Whiteford and Morrow, 1998).

Despite the insights arising from the more intense policy investigations of the past decade or so, there remains some ambiguity about the living standards of older people. For example, the 1988 OECD report on Reforming Public Pensions noted that 'old age has always been associated with a fall in economic status, deprivation, destitution and poverty' (1988, p. 44). But the report also points out that 'the income level of the retired population has improved significantly in many OECD countries. In some countries this improvement has brought the disposable per capita income of retirees above the equivalent income of working families with children' (p. 7). Similarly, the World Bank in Averting the Old Age Crisis (1994) claims that it is now a 'myth' that old people are particularly vulnerable to poverty (p. 11). Nevertheless, the OECD report emphasised that 'despite the recent decrease in poverty among the elderly, the incidence remains surprisingly high in most OECD countries' (1988, p. 48).

The OECD's 1996 report on ageing in OECD countries summarised policy findings with the caveat that 'Because there are so many differences among older people, any statement that treats older people as a single group should be treated with caution. Nevertheless, it is generally true that, in many member countries, their economic status has improved over the past two decades relative to workers' (p. 13). Between nations, the variation in outcomes for older people is amply demonstrated in the OECD's 1998 outline of policy challenges in Maintaining Prosperity in an Ageing Society. For Australia and three other of the 12 nations for which estimates were cited, the relative disposable income of individuals in older households fell over the period 1975 to 1994, while their overall share of income rose, contrary to the overall trend of rises in both relative income and income shares for the other countries (p. 57).

International experience and local considerations indicate considerable policy significance in assessing how well off or poorly off the older population is. In 'Demographic Change in Australia - Conference Background Paper' for example, the authors note that among the issues to be considered with the public expenditure implications of ageing are 'the need to scale back pension payments and coverage', the 'inter-generational equity implications of various pension/superannuation scenarios', and the 'implications of the rising incidence of one-parent families within an ageing population for future income support requirements' (PC&MIAESR 1999, pp. 493, 495). In considering these or related issues, it is crucial to have a clear understanding of the effectiveness of current arrangements in securing the living standards of older people. It is also important to understand the distribution of income within the older population, particularly the importance of choosing measures of income and the technical decisions these entail.

The objective of this paper is to provide information for assessing the current distributional impact of Australian public pension arrangements. The main basis for this assessment is an analysis of trends over time in the incomes of persons of age pension age. The paper presents a wide range of alternative indicators of living standards, showing that different indicators can reveal quite different pictures of the position of older people.

The paper is structured as follows. Section 2 discusses conceptual and measurement issues involved in assessing the adequacy of pensions and their impact on the incomes and living standards of the older population. The paper argues that to assess changes over time (and differences in outcomes between countries) it is necessary to have a comprehensive framework for measuring living standards. Without such a framework, comparisons are likely to be misleading, leading to incorrect conclusions about the effectiveness of public policies.

Part 3 provides a range of measures of the adequacy of the basic pension for those completely reliant on social security income. Part 4 provides data on the private income and assets of age pensioners. Part 5 then presents results of an analysis of trends in the disposable incomes of the older population using ABS data from the Income and Housing Surveys. This part includes an analysis of trends in the proportion of the older population with cash disposable incomes below a number of alternative measures of relative low income, including the 'Henderson Poverty Line' and 50 per cent of median and mean incomes for the population as a whole.

This part also explores the sensitivity of outcomes to these different approaches to the measurement of living standards. Part 6 presents a range of alternative measures of living standards drawn from Household Expenditure Surveys, including data on trends in the relative expenditures of older people. Part 7 discusses the impact of government non-cash benefits and indirect taxes on the relative living standards of the older population. Part 8 looks at estimates of relative low income among the older population and Part 9 briefly summarises information on the housing wealth of older people. The paper concludes with a summary of the main findings and a discussion of the information required to monitor the distributional impact of public pensions in the future.

[ Return to Top   Return to Section ]

2 Assessing income distribution and pension adequacy - conceptual and measurement issues

2.1 Measuring living standards

How have the living standards of older people1 in Australia developed over time, and how do they compare with those of older people in similar societies? In attempting to answer these questions, a number of measurement and technical issues must be addressed. These include what is the measure of resources - income, expenditure or consumption - and how is wealth to be taken into account? What is the unit that is assumed to share resources - household, family, benefit unit, or person? How should we treat units of different types or composition -  by using equivalence scales? What is the period of assessment - current, annual or lifetime? What is the low-income standard, and how is it defined?

A different choice in relation to any one of these issues will alter results, perhaps to a significant degree. Indeed, a major objective of this paper is to show that very different conclusions flow from these different methodological choices, in particular the measure of resources used. The choices between different approaches will depend upon research objectives, what is practicable, and researchers' judgements about what is technically more correct. For example, as noted by Atkinson (1989), living standards can be measured either by income or expenditure, and a particular indicator may understate or overstate living standards in different cases. It can also be argued that, for some purposes, resources should be assessed over a very long period, perhaps the lifetime (Creedy 1992, 1994; Piggott 1987). But the requisite data for such an analysis are uncommon, although there have been recent studies simulating lifetime income (Harding 1992).

The comparative literature has usually taken 50 per cent of median income as the measure of relative low income. Such a choice is arbitrary. Again, the precise choice of equivalence scales can have a strong effect on estimates of the extent of relative low income, but there is no universally accepted set of equivalence scales. In these circumstances, it is appropriate to highlight the sensitivity of the results to differing choices of equivalence scales, but it should not be thought that there is one correct answer. In analysing living standards, it is also necessary to use specific measures to compare standards of living. Quinn (1987) notes that measures to assess the adequacy of incomes available to older people include absolute measures, such as how do resources compare with what is needed to achieve a satisfactory life. Relative measures include how the resources of older people as a group compare with the rest of the population, or how individual resources after retirement compare with those available to the same person or family before retirement. In summary, Atkinson (1990) has suggested that it is most useful to present a range of estimates, based on different approaches.

That is the approach adopted here.

2.2 Assessing trends in income distribution

Figure 1 compares two ways of analysing data on the distribution of income. One is employed in most standard income distribution studies. The other is used by the ABS in its series The Effects of Government Benefits and Taxes on Household Income (ABS catalogue number.6537.0).2 In the standard approach, income from wages and salaries, self-employment and property add up to 'factor incomes'. Factor incomes plus superannuation or occupational pensions give 'market incomes'. Public transfers, private transfers, and any other cash income, when added to market income, produce 'gross income'. Gross income minus personal income tax and employees' social security contributions (in other countries) gives 'net cash income'. The degree of redistribution effected either by public transfers or by income tax (and social security contributions) can be assessed in several ways. These include calculating the relative change in income levels for different individuals or by calculating income shares at different stages in the 'process' described above.

Like the standard methodology, the framework of the ABS studies of government benefits and taxes is well known and widely accepted. The ABS also sets out its methodology in the way shown in Figure 1. The concept of final income is a more comprehensive measure of living standards, that includes all impacts covered by the standard disposable or 'net cash income' measure, plus the effects of indirect taxes and other government social spending on subsidies or services to households.

Figure 1: Comparison of different income concepts
Income distribution surveys
Fiscal incidence studies

Sources: Adapted from the Australian Bureau of Statistics The Effects of Government Benefits and Taxes on Household Income (ABS, 1996) and the UK Central Statistical Office Economic Trends (UKCSO, 1991)

Wages and salaries and self-employment income
+
Wages and salaries and self-employment income
+
Investment and property income
=
Superannuation and annuities
+
1. Factor income
+
Investment and property income
+
Occupational pensions and annuities
=
Other income (for example, alimony)
=
2. Market income
+
1. Private income
+
Government cash benefits
+
Government cash benefits
=
Private transfers
+
2. Gross income
-
Other cash income
=
Income tax
=
3. Gross income
-
3. Disposable income
+
Income tax (and employees social security)
=
Benefits in kind (health, education, and so on)
=
4. Net cash income
4. Disposable income plus indirect benefits
-
 
Indirect taxes
=
5. Final income

One obvious point to be made from these comparisons is that a household's resources can be measured in a range of ways, with neither of the two measures shown here being fully comprehensive. The major gap in both frameworks is the failure to take account of household wealth. This may have significant implications for the relative wellbeing of the older population, in part because of the life cycle pattern associated with wealth accumulation. The potential importance of taking account of wealth in the framework for assessing the relative wellbeing of the older population is illustrated by the alternative definitions of pension replacement rates, shown in Figure 2. International comparative studies of retirement income systems commonly use pension replacement rates as the basis for ranking the relative generosity of different pension systems. Replacement rates are usually calculated by comparing the levels of statutory entitlements to some measure of incomes in work, thus showing what percentage of earnings is 'replaced' by benefits.

Figure 2 shows how conventional replacement rate measures could be augmented to provide more appropriate indicators of pension adequacy. Conventional replacement rates are usually calculated by reference to only the top panels in the figure. A more comprehensive approach would take account of the complete range of income sources and costs before and after retirement. Attempting to implement this broader approach to the measurement of living standards is complex. There is no single study that incorporates all of these components of material living standards. Therefore, the discussion that follows looks first at the simplest measures of social security adequacy. This is followed by an analysis of cash disposable incomes, and then the analysis incorporates indirect benefits and taxes. Some of the available information on the relative asset holdings of older people is then discussed.

Figure 2: The definition of the net replacement rate in retirement
Numerator: post-retirement consumption
Denominator: pre-retirement consumption
Positive items:
Cash benefits
Negative items:
Direct taxes
Positive items:
Labour earnings
Negative items:
Direct taxes
Social insurance contributions
Source: Adapted from Wolfson (1987)
Possible refinements to the definition of the net replacement rate
Additional positive items:
Additional positive items:
Occupational and private pensions
Investment income

Investment income
- Interest income
- Interest portion of annuity income
- Imputed rent on owner-occupied housing

- Interest income
- Imputed rent on owner-occupied housing

Government non-cash benefits
- Health
- Housing
- Education
- Transport

Government non-cash benefits
- Health
- Housing
- Education
- Transport

Dis-saving
- Drawing down savings
- Capital portion of annuity income
- Sale of house or reverse annuity mortgage

Additional minus items:
Indirect taxes Work-related expenses

Additional minus items:

Indirect taxes

Saving
- Bank deposits
- House downpayment, capital portion of mortgage payments

Private and occupational pension contributions

[ Return to Top   Return to Section ]

3 The Age Pension - Assessing adequacy

3.1 The Age Pension - Assessing adequacy

Table 1 provides details of the Australian Age Pension system at September 1999. In addition to the basic rates of payment set out in Table 1, pensioners qualify for a pension concession card and may qualify for additional assistance, depending on their circumstances. This includes Rent Assistance, Pharmaceutical Allowance, Telephone Allowance, Remote Area Allowance and pension concession cards. Pension concession cards entitle the cardholder to Commonwealth health concessions, such as low-cost pharmaceuticals. State-based concessions may include reductions in property and water rates; reductions in energy bills; reduced fares on public transport; reductions on motor vehicle registration; and other health, household, educational and recreational concessions. These concessions are provided and funded by State and local governments (the former with Commonwealth assistance) and some private organisations, and the type of concession may vary between States.

Table 1: The Australian Age Pension system, 1999
Feature
Value at September 1999
Standard (single) pension rate
Married pension rate (each)
Supplementary rental assistance
$366.50 per fortnight
$305.90 per fortnight
Up to
$76.00 per fortnight single,
$71.60 per fortnight couple
Source: Adapted from Wolfson (1987)

Free areas (disregards)
Single
Combined married
Cut-out points
Single
Combined married

$102.00 per fortnight
$180.00 per fortnight

$835.00 per fortnight
$1,403.60 per fortnight

Assets test
Single home owners
Single non-home owners
Married home owners
Married non-home owners

Allowable assets - no rate paid above:

$127,750
$219,250
$181,500
$273,000

$251,750
$343,250
$387,500
$470,000

The Age Pension rate is indexed in September and March, in line with movements in the Consumer Price Index (CPI). The Government has also legislated to maintain the single rate of pension at the indexation rates at a minimum of 25 per cent of Male Total Average Weekly Earnings MTAWE) with flow-ons to the married rate of pension. While CPI indexation is intended to protect the real purchasing power of the pension, maintaining the pension at 25 per cent of MTAWE is intended to enable pensioners to share in community living standards.

Table 2 shows the number and characteristics of age pensioners and spending on age pensions since 1965. At 30 June 1998, there were a total of 1.73 million age and wife pensioners. The Age Pension is received by around two-thirds of the population of pensionable age. Veterans' pensions are received by a further 390,000 people, giving a total coverage of 81 per cent of the older population. A small number of people receive other benefits,3 but the bulk of the remainder are excluded from payments by their private incomes or assets.

Table 2: Number and characteristics of age and service pensioners, Australia, 1965 to 1998
  1965 1970 1975 1980 1985 1990 1995 1996 1997 1998

Sources: Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years

Age Pensions 628.1 779.0 1,092.2 1,321.9 1,331.8 1,340.5 1,578.7 1,602.8 1,680.2 1,682.6
Wives 3.5 6.6 21.9 30.8 22.9 23.8 39.6 41.1 36.6 36.2
Total Social Security (DSS)
pensioners and beneficiaries
849.2 1,054.7 1,707.7 2,338.2 2,848.5 2,808.8 3,741.0 3,912.4 3,994.9 n.a. DVA
Pensions 65.2 74.4 121.6 264.7 412.3 440.5 346.8 335.0 389.5 n.a
Total of DSS and DVA cash benefits 914.4 1,129.1 1,829.3 2,602.9 3,260.8 3,249.3 4,087.8 4,247.4 4,384.4 n.a
Characteristics of age pensioners (including wife/carers)
% of total population 5.56 6.21 8.06 9.21 8.51 7.92 9.01 9.03 9.32 9.24
Coverage (%) of pensionable population 53.2 60.3 72.6 76.8 66.5 58.2 64.9 64.7 66.4 67.3
Single rate 61.1 61.1 57.4 55.2 56.0 57.2 49.5 46.9 46.9 46.5
With Rent Assistance (%) 10.8 13.4 14.9 14.1 16.2 18.2 n.a 15.6 15.4 9.6
Reduced rate (%) 13.4 20.6 10.4 33.5 28.7 29.5 32.7 34.6 32.6 32.0
Nil income assessed (%) n.a n.a 19.4 9.5 14.9 11.2 10.1 15.1 8.9 9.6
Spending on Age Pensions 1996–97 $ million 2,918 3,765 7,205 9,339 10,005 9,844 12,552 12,551 13,118 13,142
% of GDP 1.65 1.65 2.60 2.91 2.70 2.22 2.50 2.41 2.45 -

Over the past 33 years, the number of age pensioners has nearly trebled, rising from 5.5 per cent to 9.2 per cent of the total population. Coverage of the pensionable population has fluctuated markedly, reflecting changes in policy towards income and assets testing. In recent years, fewer people reaching Age Pension age qualify for a DVA Service Pension as the cohort who served in the Second World War have now all retired. Consequently, a higher proportion of those reaching retirement age qualify for the Age Pension, a factor reflected in the increasing coverage (from 58 to 67 per cent) since the late 1980s.

Over the period, the characteristics of age pensioners have changed. Until the early 1970s, around 60 per cent were single and 40 per cent married. The proportion who are single has fallen to around 47 per cent. The percentage receiving a reduced rate (because their incomes are over the 'free area') has fluctuated, but there has been an increase in the proportion receiving a reduced rate from 10 to 15 per cent in the 1960s to one-third in the 1990s.

Correspondingly, the proportion that is completely dependent on the Age Pension (nil income assessed) has fallen from more than a quarter in the early 1970s to around one in ten in the 1990s. Also over the same period, the proportion with some income under the free area has risen from one-fifth to more than half.

The 10 per cent of age pensioners who are completely dependent on the pension comprise about 7 per cent of the population of Age Pension age. In addition, there are DVA service pensioners with nil private income, plus a very small number receiving other payments with no private income. There are also a further group of about 5 per cent of age pensioners with extremely low private incomes (under $1 a week). For these groups, it is the level of the Age Pension itself that is the primary determinant of their incomes in retirement.

Figure 3 and Table 3 show trends in the real level of the Age Pension, in 1996–97 terms. Since 1965, the real value of the single rate of pension has increased by 79 per cent, while for couples there has been a 63 per cent real increase. For those receiving Rent Assistance, total real payments have doubled for single people and increased by 75 per cent for couples.

Figure 3: Single pension, constant 1999 dollars.

Figure 3, Single person, constant 1999 dollars

Sources: Australian Bureau of Statistics Consumer Price Index, Australia (ABS, 1999a) and Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

Table 3: Trends in the real value of social security payments for different family types, 1965 to 1999
$ per year ($ 1999)
Year 1965 1970 1975 1980 1985 1990 1995 1996 1997 1998

Sources:  Australian Bureau of Statistics Consumer Price Index, Australia (ABS, 1999) and Department of Social Security annual reports, and DSS Customers: A Statistical Overview, various years

Pension, no Rent Assistance  
Single, no children 5,156 6,014 8,629 8,637 8,600 8,873 9,178 9,193 9,311 9,396
Couple, no children 9,453 10,544 14,330 14,396 14,338 14,793 15,312 15,336 15,534 15,683
Pension, no Rent Assistance  
Single, no children 6,016 6,673 9,675 9,569 9,644 9,903 11,155 11,170 11275 11,367
Couple, no children 10,313 11,203 15,376 15,328 15,328 15,382 19,041 19,068 19,242 19,396

Most of the real increase in pension rates was achieved in the early and middle part of the 1970s. Since 1976, price indexation has generally maintained the real value of the pension, with an upward trend reflecting an 'indexation lag effect',4 plus a number of explicit policy decisions to increase the real value of the payment, including the recent formal linking of pensions and average weekly earnings.

Figures 4 to 7 compare the single rate of pension with a range of alternative indicators of community living standards. This includes GDP per capita and Household Disposable Income per capita (HDIPC); male total average weekly earnings; the process worker's wage (Metal Trades Award C13); and the Henderson Poverty Line for a single person of Age Pension age. These indicators give a more mixed picture of pension trends.

The pension reached its highest point relative to GDP per capita in 1974, and again exceeded 40 per cent of GDP per capita in 1975 and 1978. Over the past 10 years, the pension rate has fluctuated between 30 and 35 per cent of GDP per capita. The position in regard to DIPC is broadly similar in showing substantial fluctuations, but in contrast to the inflation-adjusted series, they show a downward trend. As the Henderson Poverty Line is adjusted by HDIPC, the relative shifts are the same as for the base series. The single rate of pension was above the Henderson line in five years since 1965, but it is now at an historic low relative to this indicator.5 As discussed below, this relationship is very significant in influencing trends in the proportion of the older population with incomes below this low income measure. In contrast, the single rate of pension has increased relative to MTAWE and in relationship to the process worker's wage (for unskilled workers not in labouring jobs).

By definition, the contrast between these trends and the trend in the value of the pension adjusted for inflation reflects variations in the real values of alternative indicators. For example, while the real value of the single pension increased by 79 per cent between 1965 and 1997, GDP per capita and HDIPC increased by even larger amounts. The real value of male total average earnings increased by some 60 per cent, and the real value of the process worker's wage increased by 37 per cent over the period.

One of the main reasons for the disparity between indicators is that the National Accounts include income components not taken into account in the wage indicators. In the case of HDIPC, the two most important components are the earnings of superannuation funds and imputed income from owner-occupied housing. Imputed rent is not relevant to the wage indicators, although it should be noted that a high proportion of age pensioners own their home outright, including among those completely dependent on the pension. In contrast, increasing superannuation coverage among persons of workforce age would mean that the 'average earner' and the low-paid workers would have enjoyed greater real increases in their remuneration than is indicated by earnings alone. However, some comparisons are problematic, in that the apparent 'generosity' of the pension rises when HDIPC and GDP per capita fall during recessions, as occurred in 1991. The same effect is evident in the wage indicators, although not to the same extent.

The difference between the relativities compared to average earnings and process worker's earnings highlights that the replacement rate offered by the pension will vary, depending on the earnings indicator chosen. Since the Australian Age Pension is flat-rate and directed to poverty alleviation not earnings replacement, it is not entirely appropriate to use replacement rates as a measure of adequacy (Johnson 1998; Whiteford 1995). Despite this, it is sometimes noted that the standard rate offering replacement of only 25 per cent of average earnings is far below the replacement rates apparently available in the earnings related social insurance systems of most other OECD countries. However, for a range of reasons, this is not a fully accurate picture of the generosity of the Australian system.

Figure 4: Single pension rate compared to Male Total Average Weekly Earnings (MTAWE benchmark basis) 1965 to 1999

Figure 4, Single pension rate compared to Male Total Average Weekly Earnings MTAWE

Sources:  Australian Bureau of Statistics Average Weekly Earnings, Australia (ABS, 1999b) and Department of Social Security,Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

Figure 5: Single pension rate compared to low wages 1965 to 1999

Figure 5_Single pension rate compared to low wages 1965 to 1999

Sources: Metal Trades Industry Association and Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

Figure 6: Single pension rate compared to GDP per capita 1965 to 1999

Figure 6_Single pension rate compared to GDP per capita 1965 to 1999

Sources: Australian Bureau of Statistics Australian National Accounts (ABS, 1999) and Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

Figure 7: Relative pension rates 1974 to 1999

Figure 7: Relative pension rates 1974 to 1999

Sources: Melbourne Institute of Applied Economic and Social Research, Poverty Lines: Australia, December Quarter 1999 (MIAESR, 1999) and Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

As a starting point, it can be noted that the 25 per cent standard provides greater assistance to those receiving less than average male earnings, which is more than half of the employed male workforce, and a higher proportion of women. Table 4 provides calculations of the effective replacement rates of the Age Pension for a range of different circumstances, which are illustrated in Figure 8. For example, the standard replacement rate for a single person is 25 per cent of gross MTAWE. This is equivalent to 38 per cent of average gross female earnings. The combined pension for a couple is 42 per cent of gross MTAWE. Moreover, as noted above, someone completely reliant on an Age Pension would pay no income tax, while workers do. Thus, the 25 per cent gross replacement rate is equivalent to a replacement rate of 33 per cent of net earnings. Again, for a minimum wage worker, the single pension replacement rate is 58 per cent of net earnings (and higher for a single income couple on the minimum wage).

Table 4: Alternative definitions of pension replacement rates*
Alternative definitions Replacement rate %

Note: *Calculations are at May 1998. **Assumes that pensioner is an outright home-owner and pays 18 per cent of gross pension rate in housing-related expenses; assumes that worker is purchasing a home and is paying 34 per cent of gross income in housing-related expenses and superannuation contributions. These ratios are derived from the 1993–94 ABS Household Expenditure Survey and are the ratios of the average expenses on these items to the average income for the second quintile of purchasers and owners respectively.

% of gross MTAWE (single) 25
% of gross MTAWE (couple) 42
% of gross FTAWE (single) 38
% of net MTAWE (single) 33
% of gross minimum wage (single) 49
% of net minimum wage (single) 58
% of net MTAWE, net of employee superannuation contributions, and housing costs** 50

Finally, the table shows the effects of taking account of employee superannuation contributions and housing costs, which increase the net replacement rate for an average earner to 50 per cent. The reason for taking account of these is that most working people will face these costs, but retirees are unlikely to be making superannuation contributions or paying mortgages.

Figure 8: Selected pension replacement rates June 1998

Figure 8, Selected pension replacement rates June 1998

Source: See Table 4.

In summary, simple measures of pension adequacy should be regarded with caution. The discussion has shown that components of living standards for those in retirement and those in work are much broader than either the pension alone or a single measure of incomes for those of workforce age. More reliable indicators of living standards need to adopt a comprehensive approach.

[ Return to Top   Return to Section ]

4 Private incomes and assets of age pensioners

4.1 Trends in private incomes

While the proportion of age pensioners completely dependent on the pension is fairly low, the largest group is those with incomes under the free area - $100 per fortnight for a single pensioner and $88 per fortnight for each member of a couple at June 1999, at which time around 68 per cent of age pensioners had incomes within the free area range. Unpartnered women (74 per cent) are more likely to have income in this range than couples (70 per cent) or unpartnered men (73 per cent). The most common form of private income is from savings and investments - usually from banks, building societies or credit unions. In 1998, around 90 per cent of age pensioners had incomes from this source. Currently, around 10 per cent of pensioners have incomes from superannuation (7 per cent of females and 14.5 per cent of males), an increase from around 7.5 per cent of pensioners in the late 1980s.

The proportion of age pensioners receiving a reduced rate of payment reflects access to private income and assets among the retired, and changes in income test parameters. Table 2 showed that the proportion of age pensioners with a reduced rate fell from around 20 per cent in 1970 to 10 per cent in 1975. This resulted from the abolition of the income test for pensioners aged 70 years and over. The reintroduction of the income test in stages from 1978 correspondingly resulted in an increase in the proportion paid at the part-rate, as can be seen in Figure 9.

Figure 9: Percentage of age pensioners receiving reduced rates 1955 to 1999.

Figure 9, Percentage of age pensioners receiving reduced rates 1955 to 1999.

Source: Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

Figure 9 also shows the effects of fluctuations of the real value of the free area over time. Generally, peaks in the value of the free area correspond to troughs in the proportion of pensioners receiving a reduced rate (and vice versa) as in 1967, 1973, 1983 and 1988. The free area was equal to the standard pension rate late in 1972. This fell substantially thereafter, because it was not indexed in line with inflation, while from 1976 onwards the pension rate was. In 1982, the free area was 39 per cent of the pension, falling to around 28 per cent by 1990. Since 1991, the free area has also been indexed and has remained around 28 per cent of the standard rate, with the rate for couples being about 30 per cent of their basic payments. Nevertheless, the general increase in the proportion of pensioners with reduced rate payments in part reflects this fall in the real value of the free area.

4.2 Pensioners' assets

As noted by Foster (1988 p. 41), asset ownership confers a number of advantages on some older people. Assets can be invested to produce an income, or in the case of home ownership can reduce the need for income to pay rent. Assets can also be sold to meet consumption needs. The assets test on pensions was introduced in 1985 to better target assistance to those with greater needs, and to ensure the effective operation of the income test. The rate of pension is calculated under both the income and assets tests, with the test that results in the lower rate being the one applied. While the majority of pensioners have payments assessed under the income test, the proportion directly assessed under the assets test has increased from under 2 per cent in the late 1980s to just over 6 per cent in 1999 (see Figure 10 below).

Figure 10: Proportion of age pensioners paid under the assets test 1985 to 1999.

Figure 10, Proportion of age pensioners paid under the assets test 1985 to 1999.

Source: Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

The increase is most likely due to falling nominal rates of return from investment. From 1992, rates of return have dropped so that, for an increasing proportion of pensioners, the assets test reduces their rate of pension by a greater amount than, for instance, the income from their assets reduces their pension under the income test. Hence, they are paid under the assets test.

Administrative data on assets are collected from all pensioners. Tables 5 and 6 provide details of the distribution and average value of assets held by age pensioners at June 1998, as well as the proportion of those with assets who own their own home. At June 1998, around 92 per cent of age pensioners were recorded as having positive assets (not including the family home).

Table 5: Distribution of assets of age pensioners, June 1998

Table 5: Distribution of assets of age pensioners, June 1998
Percentage of age group by asset holdings
Source: Research and Analysis Section, Retirement Programs Branch, Department of Family and Community Services, 1999.
Age

$0.01
to
$1,000

$1,000
to
$5,000

$5,000
to
$10,000

$10,000
to
$20,000

$20,000
to
$50,000

$50,000
to
$100,000

More
Than
$100,000

With assets
as % of
total

60-64 7.3 8.7 8.8 13.8 25.9 23.0 12.4 93.7
65-69 6.8 8.64 8.5 13.7 26.7 22.9 12.9 94.0
70-74 8.0 10.2 9.9 15.8 27.2 18.4 10.5 92.8
75-79 10.3 13.8 12.1 17.9 23.4 13.5 8.9 90.7
80-84 10.6 15.8 12.1 17.9 23.4 13.5 8.9 90.7
85-89 9.6 16.2 13.1 17.2 19.4 13.1 11.4 89.3
90 plus 8.6 15.7 12.2 15.9 18.5 14.6 14.6 88.0
Total 8.3 11.3 10.3 15.6 24.8 18.4 11.2 92.2

 

Table 6: Average assets of age pensioners, June 1998
Source: Research and Analysis Section, Retirement Programs Branch, Department of Family and Community Services, 1999.
Age $0.01
to
$1,000

$1,000
to
$5,000

$5,000
to
$10,000

$10,000
to
$20,000

60-64 78 45,300 42,500 31,100
65-69 78 46,100 43,300 32,100
70-74 75 40,300 37,400 24,600
75-79 69 34,200 31,000 17,800
80-84 61 34,000 30,500 15,300
85-89 50 37,600 33,600 15,800
90 plus 33 44,000 38,700 18,400
Total 67 40,800 37,600 -

The proportion of customers with assets falls from 94 to 86 per cent from age 65 to age 95 and over. Average assets held begin to decline around age 66 and continue to be less for each of the age categories until the 85–89 category where average asset holdings again begin to increase.

Further analysis of the position of women finds that divorced, separated and never married women are less likely to have assets than widows, who are less likely to have assets than married women (16, 17, 11 and 4 per cent respectively of these groups having no recorded assets). Overall, around 45 per cent of divorced or separated age pensioners and 38 per cent of single female age pensioners have no assets or less than $5,000 worth of assets.

Overall, the data suggest around one-quarter of all of those with assets have holdings of between $20,000 and $50,000, with around 45 per cent having assets below this level and 30 per cent having assets of $50,000 or more. However, it can be noted that by including those with no recorded assets, roughly one-quarter of all age pensioners have assets of less than $5,000, including personal effects.

Figure 11: Percentage of age pensioners who are home owners 1958 to 1999.

Figure 11, Percentage of age pensioners who are home owners 1958 to 1999

Source: Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years.

As Figure 11 shows, in 1999 around 68 per cent of age pensioners owned their home, the proportion having dropped slightly from its peak of 70 per cent in the early 1990s. The chart shows the substantial increase in home ownership after the Second World War and the sustained high levels among age pensioners since the mid 1970s.

Among age pensioners, patterns of home ownership vary according to age, sex and other characteristics. For instance, for those with other assets home ownership is above 70 per cent up until ages 75–79 where it is slightly lower. For those aged 80 years or more, home ownership is substantially lower, down to around 50 per cent for those aged 85–89 years, and 33 per cent for those aged 90 years or more. The age groups at which home ownership is lowest coincide with the ages around which average assets held begin to rise.

Recent administrative data on pensioners' assets and living arrangements support the idea that older pensioners tend to sell their home and move to live with family members, in nursing homes or make other arrangements for accommodation under which they are no longer classified as home owners. In particular, when comparing pensioners aged 80 years or more with younger pensioners, there is an increase in the proportion classed as 'non-home owner in government-funded aged care' and a corresponding fall in the proportion who are home owners. However, at this time there is insufficient earlier data from which to determine trends or to further differentiate changes in assets and tenure.

Recent fluctuations in the rates of home ownership among age pensioners are therefore likely to be the result of many factors, including the increased longevity of men, the associated increase in the greater proportion of married couples and fluctuations in returns from investment income. At present, there is no definite trend to increase or decrease rates of home ownership among this group.

[ Return to Top   Return to Section ]

5 Trends in the cash incomes of older people

5.1 The Age Pension - Assessing adequacy

Table 7 summarises trends in the incomes and characteristics of older income units from 1982 to 1997–98. The table is derived from the published results of the ABS Income Surveys. Results refer to income units, or nuclear families, and the income data are gross (before tax) and not adjusted for income unit size (equivalised).6

In 1997–98, older income units made up just over 17 per cent of all income units. Their mean income was around 52 per cent of the total mean income for the population. Around three in four older income units have government pensions and allowances as their main source of income, compared to just under 30 per cent of the total population. Nearly three-quarters of older income units own their home without a mortgage, compared to around 31 per cent overall. Older households are only half as likely to be renters (16 compared to 35 per cent), but are slightly more likely to be renting public housing (7 per cent compared to 4.5 per cent). More than half of all older income units are single people, and very few have dependent children. Older income units have increased from 15 to 17.5 per cent of all income units over this period. Single person units have declined slightly from 58 to 56 per cent of older income units, and women as a proportion of older single people have fallen from 78 to 72 per cent.

The real average income of older couples has increased by 5.7 per cent, while the real average income of older single people has increased by 6.7 per cent, compared to a real increase of 4 per cent for the population as a whole.7 These trends are shown in Figure 12. As a result, the average incomes of older people have increased as a proportion of the average incomes of all income units in the population - slightly more for singles than for couples.

As Figure 13 shows, recent trends appear quite volatile, suggesting that one should be cautious about apparent year to year changes. At the same time, there does not appear to be a particularly strong long-term trend in pensioner incomes, with the average for older people increasing only from 50 to 52 per cent over this 15-year period. It should also be remembered that the very substantial long run increases in real pension levels are not captured in this figure. Most of the real increase in pension rates occurred in the late 1960s and early to mid-1970s. Thus the impression of relative stability in incomes shown in Figure 13 is consistent with the modest increase in real pension rates over the 1980s and 1990s shown in Figure 4.

Nevertheless, there appears to be a significant decline in the proportion of older couples for whom government benefits are the principal source of income, and correspondingly a significant increase in the role of other private income (from property and investments). In contrast, there appears to be very little change for single older income units in the role of different income sources.

Further disaggregation reveals that, for couples, the overall decline in reliance on pensions and allowances is associated with a fall in the proportion who receive between 50 and 90 per cent of their gross income from pensions. The proportion receiving 90 per cent or more of their income from government payments is virtually unchanged over the period, although showing fluctuations in different years.

Inequality of gross incomes measured by the Gini coefficient (a single figure measuring inequality of income distribution)8 has increased for older couples and for older singles, but the overall level of inequality among older people is substantially lower than among the population generally.

Finally, Table 7 summarises trends in housing tenure. The level of outright home ownership among older couples has increased from 80 to 84 per cent, with the proportion with a mortgage or renting from public authorities falling. The proportion of private renters appears to have been broadly stable at under four per cent. Home ownership rates also increased among single older income units, but are substantially lower than for couples. There has been a small increase in the percentage of single older people in public housing. Among the total population, there has been an increase in the proportion owning their homes outright, and a fall in the proportion with a mortgage. The proportion of the total population renting privately also increased over this period.

In summary, this table suggests that the older population has had larger increases in incomes than the overall Australian population since the early 1980s, and as a result their incomes have increased relative to the population generally. This trend has been stronger for couples than for singles. The trend also appears to have been associated with a reduction in 'partial dependence' among older couples, with the proportion of older couples receiving 90 per cent or more of their income from government benefits little changed over this period.

Figure 12: Change in real average income 1982 to 1996–97.

Figure 12, Change in real average income 1982 to 1996–97

Source: Australian Bureau of Statistics, Income Distribution, Australia, various years.

Figure 13: Trends in average incomes of all older people, 1982 to1997–98.

Figure 13, Trends in average incomes of all older people, 1982 to1997–98

Source: Australian Bureau of Statistics, Income Distribution, Australia, various years.
  1982 1986 1990 1994-5 1995-6 1996-7 1997-8 Change %
Note: Subject to very high sampling variability
Source: Australian Bureau of Statistics, Income Distribution Surveys, various years.
No. of older income units
Couples 439.2 521.5 591.4 643.9 684.8 691.6 701.5 +59.7
Singles 600.3 643.3 733.4 840.6 811.4 867.2 896.4 +49.3
All older 1,039.5 1,166.2 1,327.7 1,484.8 1,496.8 1.561.9 1597.9 +53.7
% of all income units
Couples 6.3 7.0 7.4 7.2 7.7 7.6 7.7 +1.4
Singles 8.6 8.6 9.2 9.4 9.1 9.5 9.8 +1.2
All older 15.0 15.6 16.6 16.5 16.8 17.2 17.5 +2.5
Singles as % of:
All older income units 57.7 55.2 55.2 56.6 54.2 55.5 56.1 -1.6
People in older units 40.6 38.1 38.3 39.5 37.2 38.5 39.0 -1.6
Females as % of older singles n.a 78.1 77.1 72.2 73.9 73.0 72.3 -5.8

Mean income ($pw)

Real %
Couples $208 $270 $423 $410 $429 $481 $460 +5.7
Singles $111 $143 $214 $208 $226 $242 $248 +6.7
All older $152 $200 $307 $296 $319 $348 $341 +7.1
Total population $303 $410 $563 $579 $609 $625 $658 +4.0

Mean income (% of total)

Couples 68.6 65.9 75.1 70.8 70.4 77.0 699 +1.3
Singles 36.6 34.9 38.0 35.9 37.1 38.7 37.7 +1.1
All older 50.2 48.8 54.5 51.1 52.4 55.7 51.8 +1.6

Principal source of income of older couples

Wage or salary   3.6 4.3 3.7 4.6 5.0 5.1 +1.5
Own business/partnership n.a 2.2 2.5 2.9 3.0 3.1 3.1 +0.9
Other private income   19.6 23.3 24.7 21.8 26.2 25.8 +6.2
Government pensions and allowances   74.7 69.8 68.3 70.2 64.9 65.4 -9.3

Principa l source of income of older singles

Wage or salary   *0.3 *0.6 *1.4 *1.1 *0.4 *1.0 *0.7
Own business/partnership n.a *0.9 1.1 *0.8 *0.8 *1.5 *1.4 *0.5
Other private income   16.7 16.7 17.4 17.4 15.6 17.3 0.6
Government pensions and allowances   82.1 81.6 80.0 80.0 81.4 9.7 2.4
Pensions and allowances as % of gross income of older couples
50 and less than 90 n.a 31.0 312 18.4 18.6 20.8 21.1 -9.9
90 and over   43.3 37.9 48.9 50.9 43.1 44.1 0.8
Pensions and allowances as % of gross income of older singles
50 and less than 90 n.a 18.4 27.1 12.3 16.0 17.2 12.8 -5.6
90 and over   63.7 54.1 71.3 63.8 63.9 66.2 +2.5
Gini coefficient
Older couples - 0.03 0.34 0.29 0.31 0.33 0.31 +0.02
Older singles - - - 0.22 0.25 0.26 0.27 +0.05
Total population 0.40 0.41 0.42 0.44 0.44 0.44 0.45 +0.01
Tenure of older couples
Outright owner 80.0 77.1 81.2 84.9 85.2 84.1 84.9 +4.9
With mortgage 7.3 10.0 6.5 5.5 4.2 4.9 3.8 -3.5
Public renters 4.3 3.9 3.8 *2.0 3.0 *2.5 3.1 -1.2
Private renters 3.6 2.5 3.3 3.5 3.9 3.2 4.3 0.7
Tenure of older singles
Outright owner 59.8 60.9 64.5 62.9 64.0 67.0 63.9 +4.1
With mortgage 3.5 3.1 3.5 4.0 *1.9 2.6 2.6 -0.9
Public renters 7.2 8.3 7.9 10.4 9.2 9.0 10.1 +2.9
Private renters 6.9 6.9 5.6 6.8 7.3 5.7 5.4 -1.5
Tenure of total population
Outright owner 27.6 29.2 32.5 32.9 32.4 31.3 30.6 +3.0
With mortgage 25.4 24.3 22.5 20.7 21.9 21.4 23.6 -1.8
Public renters 3.9 4.1 4.5 4.2 4.7 4.4 4.5 +0.6
Private renters 15.7 14.2 15.6 17.5 19.9 20.6 19.8 +4.1

5.2 The equivalent incomes of older people

Adjusting for family size can have a substantial impact on the measured living standards of the older population. Table 8 illustrates the effects of differing income adjustments on the position of older people in the overall income distribution. The first panel shows the distribution of older people by overall gross income quintiles. Nearly 60 per cent of older single persons fall in the lowest quintile (20 per cent) of the overall distribution, with a further 30 per cent in the second quintile. Older couples fall into the second and third quintiles. Subtracting income tax to determine disposable income moves just under 10 per cent of older couples from the second to the third quintile, but appears to have virtually no effect on older singles.

The table uses two different equivalence scales to adjust disposable income. The main effect of the Henderson equivalence scale is to increase the proportion of older couples whose incomes fall into the lowest equivalent income quintile, and to move a substantial proportion of single older people from the first and second quintiles to the second and third quintiles. The effects of using the OECD equivalence scales is even more striking. On these equivalences, a higher proportion of couples than singles are in the lowest quintile, and nearly a quarter of older single people are in the third quintile, compared to around 5 per cent when unadjusted incomes are used.

Table 8: Distribution of older people by weekly income quintile, Australia, 1996–97
Percentage of income units by quintile group
  Weekly income quintile All income units
Lowest Second Third Fourth Fifth
Note: *Subject to relative standard error greater than 25 per cent.
Source: Australian Bureau of Statistics, Income Distribution, Australia, 1996–97, catalogue number. 6523.0, Table 26, pp. 41–42.
Gross income
Older couples 5.1 52.4 26.1 9.2 7.1 100.0
Older singles 59.8 31.7 5.6 1.6* 1.2* 100.0
Gross income
Older couples 5.1 43.3 32.3 11.0 7.8 100.0
Older singles 58.3 33.0 5.8 2.0* 0.9* 100.0
Henderson equivalent
Older couples 10.4 44.2 27.8 8.3 9.4 100.0
Older singles 39.0 33.2 16.8 5.7 5.2 100.0
OECD equivalent
Older couples 24.9 35.0 26.4 6.9 6.8 100.0
Older singles 18.3 49.0 23.4 5.3 4.1 100.0

Overall, this suggests that conclusions about the relative position of older people are sensitive to the adjustment for family size and also sensitive to the precise choice of equivalence scale. The reason for this sensitivity is shown in Figures 14 and 15.

Figure 14 shows the distribution of equivalent income of older people, adjusted with the OECD scales, as a percentage of the average equivalent income of the total population in 1995–96, compared to the distribution for the total population. Figure 15 shows the same figure for older people, but for 1986 and 1990, as well as 1995–96. The extreme concentration of older people with between 40 and 60 per cent of average income is clearly evident. This range encompasses all of those completely dependent on the Age Pension or Service Pension, plus those with relatively small amounts of private income.

Figure 14: 1995–96 income distributions - older people and all Australians.

Figure 14, 1995–96 income distributions - older people and all Australians.

Source: Estimated from unit record files, ABS Income Survey 1995–96 (catalogue number 6541.0.15.001).

Figure 15 overleaf shows that the modal value for the equivalent incomes of older people has increased relative to those of the population generally, and that there was a very large shift in this modal value and a decline in the degree of concentration, between 1986 and 1990. Between 1990 and 1995–96, the modal value did not appear to increase relative to the average incomes of the population, but the degree of concentration again increased, although not back to its 1986 level.

The extreme degree of concentration of equivalent cash incomes of the older population has the effect of making many measures of living standards very sensitive to small differences in measurement. As discussed below, estimates of relative low income or 'poverty' vary substantially over time and according to the low-income standard or equivalence scale used.

Figure 15: Income distributions for older Australians - 1986, 1990 and 1995–96.

Figure 15, Income distributions for older Australians - 1986, 1990 and 1995–96

Sources: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001).

It can also be noted that this feature appears to distinguish Australia from other countries. Figure 16 shows the equivalent9 disposable cash income of older people (of pension age) expressed as percentages of the mean equivalent disposable income of the total population for a range of countries. The shape of the distribution of income of older people falls into two groups. Countries that emphasise earnings replacement (France, Germany, Italy, and apparently the United States)10 share a similar, fairly flat income profile, while those countries with substantially flat-rate pension systems compress the incomes of older people into narrower peaks. It is clear that the Australian distribution of disposable income is more compressed than that of any other of these countries, with more than 30 per cent of the older Australian population falling between 40 and 60 per cent of average income, compared to around 20 per cent in other countries in this group, and under 15 per cent in the European welfare states and the United States.11

Figure 16: Comparison of income distributions of older people, around 1985

Figure 16, Comparison of income distributions of older people, around 1985 AFigure 16, Comparison of income distributions of older people, around 1985 B

Source: Estimated from Luxembourg Income Study datafiles by Whiteford and Kennedy 1995.

Table 9 shows average pensioner incomes as a proportion of the average income of non-pensioners, adjusted using OECD equivalence scales.12 Couples tend to have higher equivalent incomes than single people do, although in 1990 single men have about the same equivalent incomes as couples, and in 1995–96, single men are apparently the most well-off group. In 1995–96, those aged 75 years and over appear to be substantially worse off than those under 75 years, but in the two earlier periods this does not appear to be the case (except for single men in 1990).

Table 9: Incomes of pensioners as a proportion of the incomes of non-pensioners by age group and income unit type, 1986 to 1995–96.
  60 to 64 65 to 69 70 to 74 75 plus All pensioners
Note: *Subject to very high sampling variability.
Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001)
1986
Couples 59.7 57.2 56.0 57.3 57.6
Single men* 81.0 6824 55.4 50.0 65.3
Single women 65.4 53.7 52.5 53.2 55.4
All pensioners 65.3 57.2 54.5 54.2 57.8
1990
Couples 61.1 67.0 64.2 64.3 64.4
Single men* 51.9 65.4 74.9 58.5 62.3
Single women 60.7 58.2 59.5 57.1 58.4
All pensioners 59.6 63.6 64.0 59.9 61.7
1995-96
Couples 53.7 60.9 62.4 58.7 59.2
Single men* 53.4 68.3 70.7 62.6 63.9
Single women 51.4 58.0 60.5 49.1 53.4
All pensioners 53.0 61.0 62.7 54.6 57.6

Figure 17 shows how estimates of the average relative incomes of older people are affected by equivalisation. Adjusting for income unit size increases the average relative incomes of older people, by about 10 percentage points in the mid-1980s and five percentage points in the mid-1990s. The decline in this effect is probably due to the increasing share of older people who are couples and the decline in family size among the younger population. While the effect of equivalisation may not appear large, it can be noted that it is actually greater than the trend increase in the relative incomes of older people over this period.

Figure 17: Relative incomes of older people, unequivalised and equivalised.

Figure 17, Relative incomes of older people, unequivalised and equivalised

Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001).

Table 10 shows the average incomes of quintile groups of persons of pension age as percentages of the average income of the non-pensioner population.

Table 10: Incomes of pensioners as a proportion of the incomes of non-pensioners, by pensioner income quintile, 1986 to 1995–96
Average income of quintile group as percentage of average for non-pensioner population
  1986 1990 1995-96
Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001).
Lowest 30 31 29
2nd 37 40 40
3rd 41 46 45
4th 53 58 57
Highest 120 134 117
All pensioners 58 62 58

The relative position of the poorest quintile has been generally stable. On average, the higher relative income of pensioners in 1990 appears to be associated with a substantial increase in the relative position of the richest quintile of pensioners. Correspondingly, the decline in the average incomes of all pensioners appears to be a result of the richest quintile losing this advantage. This probably reflects the high interest rates applying in 1990, and the effects of declines thereafter. In contrast, the second, third and fourth quintiles of have maintained most of their relative improvement. In this context, it is worth noting that the second quintile of pensioners have a higher degree of reliance on government income support than do the first quintile, primarily because the first quintile include persons of pension age who have low incomes from self-employment. It is also notable that the average income of all pensioners is higher than the average income of the fourth quintile of pensioners in each year. This implies that the distributions are highly skewed, presumably reflecting the coexistence of the high concentration of pensioners around the statutory rates of pension and a very long tail of high incomes.

Table 11 shows the distribution of persons of pension age by (OECD) equivalent income quintile. It can be seen that nearly half of all pensioners are in the lowest 30 per cent of the total income distribution. The table suggests that there has been relatively little change in this situation over the past decade, although the proportion in the richest 30 per cent of the population may have increased slightly.

Table 11: Distribution of pensioners by equivalent income decile, 1986 to 1995–96.
  Proportion of pensioners in each income decile
1986 1990 1995-96
Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001)
Lowest 19.1 20.0 19.6
2nd 15.6 15.8 14.7
3rd 14.4 14.3 13.8
4th 13.0 12.8 12.4
5th 11.2 11.1 10.9
6th 9.0 8.9 9.1
7th 7.1 7.1 7.3
8th 5.2 5.2 5.8
9th 3.7 3.4 4.2
Highest 1.9 1.5 2.3

Table 12 shows income inequality among pensioners by age and income unit type, using the ratio of the incomes of the 90th percentile of each group to the 10th percentile of each group. Inequality among single retired men aged 60 to 64 is highly variable, because of the small sample size of the group. Overall, this measure suggests a small decline in inequality, although the trends for different age and income unit types diverge. Generally, the highest degree of inequality is among the 60 to 64 year age group, inequality is usually greater among single men than single women, and inequality is lowest among those aged 75 years and over.

Table 12: Inequality among pensioners - ratios of the 90th to the 10th percentiles of pensioner incomes

  60 to 64 65 to 69 70 to 74 75 plus All pensioners
Note: *Subject to very high sampling variability.
Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001)
1986
Couples 3.0 2.8 2.5 1.7 2.0
Single men* 4.4 3.6 2.7 2.4 3.0
Single women 3.9 2.5 2.6 2.5 2.5
All pensioners 3.6 2.8 2.7 2.6 3.0
1990
Couples 2.2 1.9 1.7 1.7 2.9
Single men* 7.1 2.9 2.4 2.1 2.8
Single women 3.1 2.3 2.6 2.1 2.6
All pensioners 3.4 3.1 3.0 2.4 2.8
1995-96
Couples 2.7 2.9 2.8 3.2 3.1
Single men* 2.9 3.4 2.3 3.1 3.1
Single women 3.1 2.5 3.2 2.0 2.4
All pensioners 2.8 2.7 2.8 2.6 2.6

Table 13 shows trends in pensioners' income sources by equivalent income quintile. The notable patterns here are the continued dominance of government income support up into the fourth quintile of pensioners. Overall, income support provides just over half the total cash income of older people in 1995–96.

Table 13: Pensioners' income composition by quintile, 1986, 1990 and 1995–96.
  Per cent of quintile income by source
Wages Business Income support Investment Super Other
Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001)
1986
1st 1 1 92 6 0 0
2nd 0 0 95 4 0 0
3rd 1 0 85 13 1 0
4th 5 2 60 23 10 1
5th 23 7 12 42 15 1
All 10.9 3.5 51.2 25.5 8.4 0.5
1990
1st 1 0 88 10 1 0
2nd 1 0 90 8 1 0
3rd 1 0 80 15 3 0
4th 6 2 56 23 13 0
5th 20 5 11 48 14 1
All 10.3 2.7 47.7 29.6 9.3 0.4
1995–96
1st 1 92 6 0 0  
2nd 0 0 96 3 1 0
3rd 1 1 86 9 3 0
4th 4 1 63 16 15 0
5th 26 9 13 31 21 1
All 11.6 3.6 53.8 18.1 12.4 0.4

Results for 1990 differ significantly from the other years, particularly in the greater significance of investment income. The role of investment income for the highest quintile group is much lower in 1995–96 than in either 1990 or 1986, apparently reflecting a large increase in the contribution of superannuation and a more modest increase in the role of earnings. Over the whole decade, between 1986 and 1996, the contribution of investment income has fallen from 25 to 18 per cent. Most of this declining share matches an increase in the contribution of superannuation income.

[ Return to Top   Return to Section ]

6 Trends in household expenditure levels

There are strong arguments that measures of consumption are more appropriate than incomes as indicators of household living standards. This is because incomes may reflect temporary variations, which may be smoothed by borrowing or saving or by running down assets. This is particularly important in the case of older people, who typically have lower incomes than the non-retired population, but who have had the opportunity to accumulate wealth. To the extent that such smoothing is possible, it would be expected that consumption and incomes would diverge, with consumption being the better indicator of long-term living standards. However, available data are limited to household expenditures rather than consumption. The most notable problem with available expenditure data is that it does not include the flow of services from ownership of durables, including the family home. The data should be considered as an imperfect indicator of consumption, albeit in the same way that income is an imperfect indicator of economic resources

Table 14 shows trends in the income and expenditure levels of older households between 1974–75 and 1993–94. Over this period, trends in household incomes and expenditures are significantly affected by changes in household size, which have fallen, but more substantially for younger households than for older households. To partially adjust for this, the table also shows trends in income and expenditure per person.

Table 14: Trends in household expenditures and incomes, Australia, 1974–75 to 1993–94
  1974–75 1975–76 1984 1988–89 1993–94 Change %
Source: Calculated from Australian Bureau of Statistics, Household Expenditure Survey, Australia (catalogue number 6537.0) various years.
Households with reference person 65 years and over
Average income $96.59 $115.64 $229.48 $323.01 $348.68 -11.9
Average expenditure $75.22 $88.58 $196.23 $273.44 $335.81 +8.9
Income per capita $55.19 $66.46 $133.42 $187.80 $211.32 -6.6
Expenditure per capita $42.98 $50.91 $114.09 $158.98 $203.52 +15.5
Food share of total expenditure 24.0% 22.6% 22.6% 21.9% 21.3% -2.7
All households
Average income $205.92 $222.35 $453.60 $636.05 $723.23 -14.3
Average expenditure $157.00 $172.35 $361.84 $502.71 $602.11 -6.4
Income per capita $66.86 $71.96 $159.72 $228.79 $274.99 +0.4
Expenditure per capita $50.97 $55.78 $127.41 $180.83 $228.94 +9.6
Food share of total expenditure 20.6% 19.5% 19.7% 19.1% 18.4% -2.2
 
Ratio of average incomes 46.9% 52.0% 50.6% 50.8% 48.2% +1.3
Ratio of per capita incomes 82.5% 92.4% 83.5% 82.0% 76.8% -5.7
Ratio of average expenditures 47.9% 51.4% 54.2% 54.4% 55.8% +7.9
Ratio of per capita expenditures 84.3% 91.3% 89.5% 87.9% 88.9% +4.6

Figure 18 illustrates these trends. Real per capita incomes of older households fell by 6.6 per cent, but real expenditure per capita rose by 15.6 per cent. This compares with an increase for all households of 0.4 per cent in real income per capita and 10 per cent in real per capita expenditures. As a result, the average income per capita of older households has fallen from 83 to 76 per cent of the per capita household income of the population as a whole. On the other hand, the per capita expenditures of older households rose from 84 to 89 per cent of the population generally.

Figure 18: Trends in household expenditures and incomes, Australia, 1974–75 to 1993–94.

Figure 18, Trends in household expenditures and incomes, Australia, 1974–75 to 1993–94

Source: See Table 14.

The differences in the income trends shown here and those found in the earlier tables are likely to reflect a number of factors. The periods covered differ, particularly because these results go back to the middle of the 1970s, whereas earlier tables showed trends from the early 1980s. In addition, these results refer to household incomes and earlier results to income unit income.

[ Return to Top   Return to Section ]

7 The impact of non-cash benefits and indirect taxes

Government non-cash benefits in the form of services and subsidies have a substantial impact on the living standards of the population generally, and particularly on those of older people. The ABS (catalogue number 6537.0) has estimated that in 1993–94 the value of government services and subsidies for households with a reference person aged 65 years and over was $145 per week, compared to cash benefits of $185 per week. On average, indirect taxes paid by older households are estimated to be roughly equal to their income tax liabilities. Health benefits and other welfare services are most significant for the older population and education benefits are most important for the younger population. The average value of direct government cash benefits is greater than average private income for older households, and is particularly significant for older single-person households, where direct benefits are more than twice as valuable as average private income.

These estimates can be used as broader indicators of household living standards, incorporating the impact of a more comprehensive selection of government policies. However, it should be emphasised that these estimates are the result of many assumptions. They do not show the redistributive impact of the welfare state in an economic sense (Piggott 1987). Nevertheless, they are useful for illustrating that government impacts on living standards encompass much more than cash benefits.

Table 15 compares income components for older household groups with the average for the population generally. For example, the average private income of older households is only 26 per cent of that of the total population. After including cash income support, this ratio rises to 48 per cent, and after taking account of income taxes it increases to 54 per cent. The addition of indirect government benefits and the subtraction of indirect taxes increases the ratio further to 66 per cent. Figure 19 illustrates these effects, and also shows the significance of adjusting for household size.

Figure 19: Effects of benefits and taxes on household income, 1993–94.

Figure 19, Effects of benefits and taxes on household income, 1993–94

Sources: Calculated from ABS, Household Expenditure Survey, Australia, various years and The Effects of Government Benefits and Taxes on Household Income, ABS catalogue number. 6537.0.

Table 16 shows the income components as a percentage of the 'final income' for each household type. Thus it can be seen that, in 1993–94 private income was around 38 per cent of the final income of older households, but 95 per cent of the final income for the population as a whole. Cash benefits were more valuable for older households than their private income, raising it to 80 per cent of final income. Income tax reduces this, so that the cash disposable incomes of older households are about 73 per cent of their final incomes. Indirect benefits net of indirect taxes then contribute the 'remaining' 27 per cent of final income.

Overall, between 1984 and 1993–94 the net effect of indirect benefits and taxes has become slightly more 'pro-aged'. This can be seen in Table 16. While the relative contribution of indirect benefits has remained stable for all households (20.3 to 20.4 per cent of final income), they have risen for older households from 26 to 34 per cent of final income. This appears to reflect an increase in the relative contribution of health benefits for older couples and older single person households, and an increase in the relative contribution of other welfare services for older couples.

Table 15: The effects of government benefits and taxes on household income compared to all households, 1984 and 1993–94
1984
Sources: Calculated from ABS Household Expenditure Survey, Australia, various years, and The Effects of Government Benefits and Taxes on Household Income, ABS catalogue number. 6537.0
Income, benefits and taxes Couple only, reference person 65 and over Single person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Private income 28.0 12.4 80.4 30.0 100.0
Total direct benefits 239.5 159.5 255.8 208.9 100.0
Gross income 52.3 29.4 100.5 50.6 100.0
Direct tax 23.2 12.2 76.6 27.2 100.0
Disposable Income 59.7 33.8 106.6 56.5 100.0
Indirect Benefits
Education       8.0 100.0
Total health benefits 141.2 89.2 159.4 122.5 100.0
Other welfare 212.2 222.9 245.8 221.9 100.0
Total indirect 81.1 61.1 107.6 77.5 100.0
Indirect taxes 52.4 24.2 89.1 46.7 100.0
Final income 64.7 40.2 108.4 61.7 100.0

1993–94

Income, benefits and taxes Couple only, reference person 65 and over Single person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Private income 29.7 10.4 66.1 26.1 100.0
Total direct benefits 210.9 153.0 265.0 191.4 100.0
Gross income 53.9 29.5 92.7 48.2 100.0
Direct tax 22.4 10.1 60.3 22.1 100.0
Disposable income 61.3 34.0 100.3 54.3 100.0
Indirect benefits
Education 27.7 4.9 100.0
Total health benefits 195.3 115.5 198.5 159.5 100.0
Other welfare 267.5 168.9 306.2 227.1 100.0
Total indirect 127.1 79.0 144.4 107.7 100.0
Indirect taxes 62.3 25.7 83.9 49.1 100.0
Final income 74.6 43.9 110.8 65.7 100.0

 

Table 16: The effects of government benefits and taxes on household income by household type, 1984 and 1993–94
1984
Note: * Subject to high sampling variability.
Sources: Calculated from ABS Househol Expenditure Survey, Australia, various years, and The Effects of Government Benefits and Taxes on Household Income, ABS catalogue number. 6537.0.
Income, benefits and taxes Couple only, reference person 65 and over Single person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Private income 42.7 30.5 73.1 48.0 98.6

Direct benefits

Age Pension 34.5 42.8 20.1 32.5 4.5
DVA pension 11.3 6.7 4.8 8.2 1.8
Total direct benefits 47.4 51.0 30.2 43.4 12.8
Gross income 90.0 81.5 103.3 91.4 111.4
Direct tax –8.1 –6.8 –15.9 –9.9 –22.5
Disposable income 82.0 74.7 87.4 81.5 88.9

Indirect Benefits

Education * * 3.0 1.2 9.3
Total health benefits 18.9 19.2 12.7 17.2 8.7
Housing benefits 0.5 1.7 0.5 0.8 0.6
Other welfare 5.8 9.8 4.0 6.4 1.8
Total indirect 25.5 30.9 20.2 25.6 20.3
Indirect Taxes –7.5 –5.6 –7.6 –7.0 –9.3
Final Income 100.0 100.0 100.0 100.0 100.0

1993–94

Income, benefits and taxes Couple only, reference person
65 and over
Single person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Private income 37.7 22.4 56.5 37.7 94.8
Direct benefits
Age Pension 26.7 39.3 22.5 29.4 4.5
DVA pension 13.1 11.1 *5.2 10.6 1.6
Total direct benefits 41.4 51.0 35.9 42.7 14.6
Gross income 79.1 51.0 91.6 80.3 109.4
Direct tax –6.2 –4.8 –11.3 –7.0 –20.7
Disposable income 72.9 68.6 80.3 73.3 88.7

Indirect benefits

Education * * 2.0 0.6 7.9
Total health benefits 23.6 23.6 16.1 21.8 9.0
Housing benefits * 1.7 * 0.9 0.6
Other welfare 10.5 11.3 8.1 10.1 2.9
Total indirect 34.8 36.7 26.6 33.5 20.4
Indirect taxes –7.7 –5.3 –6.9 –6.8 –9.1
Final income 100.0 100.0 100.0 100.0 100.0

[ Return to Top   Return to Section ]

8 Trends in relative low incomes

In assessing trends in the wellbeing of the Australian population, a common form of analysis is to estimate how many people have incomes below the Henderson Poverty Line or other measures of relative low income.13 This is to be expected in a system that gives priority to assisting those most in need and emphasises poverty alleviation. There is considerable controversy about the nature of poverty in wealthy societies such as Australia. Much of the controversy is concerned with whether poverty is purely relative, whether it has an irreducible absolutist component, or whether these terms are at all useful. To review the literature on this topic is outside this paper's scope. We emphasise that our analysis simply refers to relative low income, and does not provide direct evidence on the extent of hardship or deprivation among low-income groups. When discussing the new results, we do not use the term poverty, but refer to relative low income. However, other researchers using the same data and similar methods have described their results as showing estimates of poverty, so when discussing their research, their term is adopted.

Studies using the Henderson line give a mixed picture of trends in the circumstances of older income units. King (1998) estimates that between 1972–73 and March 1996, the Henderson poverty rate (before housing costs) among single older people rose marginally (but was more than 30 per cent in both periods) and among older couples it fell slightly (from 5 to 3.8 per cent). After housing costs, poverty rates were substantially lower for singles but not couples, and they fell over this period. In contrast, Saunders (1994) estimated that between 1981–82 and 1989–90 'Henderson poverty' increased from 10 per cent to 28 per cent, while among older couples it increased from 4.3 to 6.7 per cent. Part of the explanation for these differences is the different time periods used. However, to be consistent this would imply a reduction in poverty among older people between 1972–73 and 1981–82, an increase in the 1980s and a fall for couples in the 1990s.

The variability of these results also reflects technical choices made in measurement, and the interaction between these choices and the very high degree of concentration in the incomes of older people discussed earlier. Because so many older Australian have incomes in a relatively narrow income range of between 40 and 60 per cent of average income, small differences in the level of the low-income line used can have a large impact on rates of low income.

The sensitivity of poverty and low-income estimates to these technical choices is well illustrated in Tables 17 and 18, which give a wide range of estimates of the level of relative low income among the older population and trends over time. All the results in Table 17 refer to incomes over the relevant financial years. Table 17 shows trends over time using the Henderson line, plus half-median income adjusted by different equivalence scales, and a half average income measure. This last measure uses household incomes and is consistent as far as possible with the Households Below Average Income Statistics produced by the United Kingdom Department of Social Security (referred to as the HBAI measure).

The Henderson Poverty Line shows the largest increase in poverty over the period 1981–82 to 1995–96. The low-income rate for older couples rises from 5 per cent to 21.4 per cent over the period, for singles from 11 to 32 per cent, and for the total population from 13 to
21 per cent. As is well known, a major contributor to this is that the Henderson line has been rising faster than average incomes in the income surveys. When the Henderson measure is adjusted only to reflect price changes - as is the case in the second panel of results - then the increase in the overall low-income rate is from 13 to 14.9 per cent, and the increase is much lower for older income units, particularly older singles.

The most consistent result is that low-income rates for older income units are always above those for the non-aged population, although the extent of this difference varies widely. In addition, all the results - except those using half-median income and the 'DSS equivalence scales'14  - show increases in low-income rates over this period. However, the extent of this increase varies enormously. The results using the standard Henderson measure show an increase of eight percentage points for the population as a whole, while the half-median line with the OECD equivalence scales shows an increase that is only 0.8 percentage points.

Table 17: Alternative estimates of trends in the extent of low income, Australia, 1981–82 to 1995–96
Percentage of various groups with low income by low-income measure
Source: Estimates prepared by the Social Policy Research Centre, University of New South Wales, using ABS Income Surveys, unit record files, various years.
  1981–82 1985–86 1989–90 1994–95 1995–96

Henderson detailed

Older couples 5.0 5.6 6.9 16.7 21.4
Older singles 10.8 24.5 27.9 31.1 31.7
All non-older 13.6 15.3 16.1 19.1 20.3
Total population 13.0 15.1 16.1 19.6 21.0

Henderson detailed (CPI-adjusted)

Older couples 5.0 4.8 5.3 14.1 16.9
Older singles 10.8 14.2 13.7 20.6 17.1
All non-older 13.6 14.0 13.2 15.6 14.6
Total population 13.0 13.4 12.7 15.7 14.9

Half-median, Henderson equivalence

Older couples 3.5 3.8 4.1 12.9 14.9
Older singles 4.5 4.6 6.8 16.9 13.9
All non-older 9.4 9.4 9.6 12.1 10.5
Total population 8.8 8.8 9.1 12.4 11.0

Half-median, McClements equivalence

Older couples 5.3 4.6 6.2 14.9 16.8
Older singles 4.9 5.9 9.1 17.8 14.9
All non-older 11.2 10.7 10.8 13.0 11.4
Total population 10.6 10.0 10.4 13.4 12.0

Half median, OECD equivalence

Older couples 5.0 3.8 5.2 13.9 15.8
Older singles 3.9 3.9 6.6 16.8 13.8
All non-older 11.1 10.8 10.3 12.7 10.6
Total Population 10.4 10.0 9.8 13.0 11.2

Half median, DSS equivalence

Older couples 50.5 7.3 10.0 17.7 21.3
Older singles 66.4 15.7 14.2 18.1 17.6
All non-older 10.2 10.1 10.1 12.2 9.5
Total Population 15.4 10.2 10.3 12.9 10.8

Half mean, households, HBAI

Older couples 8.9 17.7 21.3 20.0 24.5
Older singles 25.5 40.3 36.4 26.6 28.4
Total population 13.2 14.5 15.4 15.6 15.1
Table 18: Alternative estimates of low-income rates, Australia, mid-1990s.
Percentage of the older population with low income
Note: The first number in each series is the low-income rate for older couples, and the second number is for older single people. For ‘final income’ the third number in each set is the estimate for the total Australian population.
Source: Estimates prepared by the Social Policy Research Centre, University of New South Wales, using ABS Income Surveys, unit record files, various years.
 

Annual income

Current income

Final income

Income units Households Income units Households Households
Henderson detailed 21.4

-

10.8

-

-

31.7 34.8
Henderson, CPI 16.9

-

-

-

-

17.1
Half-median, Henderson 14.9

-

-

-

-

13.9
Half-median, McClements 16.8 17.3 7.9 8.6

-

14.9 17.9 5.7 7.3
Half-median, OECD 15.8 16.5 7.3 8.4

-

13.8 16.3 5.3 7.2
Half-median, DSS 21.3 21.8

-

-

-

17.6 19.7
Half-mean, McClements 24.0 24.5 12.0 12.9

-

22.2 28.4 9.7 13.5
Half-mean, OECD 23.1 23.4 11.2 11.7

-

17.7 20.7 7.7 9.8
Half-median, 1993, disposable

-

-

-

-

5.7
3.2
8.2
Half-median, 1993, disposable plus social wage

-

-

-

-

2.6
2.6
4.2
Half-median, 1993, disposable plus social wage per capita

-

-

-

-

2.8
2.3
4.9

Between 1990 and 1994–95, all measures except the HBAI results show an extremely large jump in low-income rates for older couples and most also show a jump for older couples. As noted by Harding and Szukalska (1999), there are doubts about the comparability of the annual income data in the ABS Income Surveys from 1994–95 onwards due to a change in the ABS treatment of those who altered family or labour market status during the year.

Table 18 and Figure 20 show there are also substantial differences between estimates of low-income rates at the same point in time, using a wider range of methodological variations. The first column shows results for older couples and older singles, respectively, which are the same as for the corresponding results in Table 17. Table 18 then shows results using households rather than income units, and then using current weekly income rather than annual income. Two general conclusions can be drawn. The use of households rather than income units gives slightly higher low-income rates for all other technical choices. Using current rather than annual income gives very much lower low-income rates, except for single older people using the standard Henderson methodology.

Figure 20: Alternative estimates of low-income rates for single older people, mid-1990s.

Figure 20, Alternative estimates of low-income rates for single older people, mid-1990s

Source: See Table 18

A final set of percentages in the last column of Table 18 shows estimates of relative low income after taking account of non-cash services and subsidies and indirect taxes. Here the relevant income concept is 'final income' as used in the preceding section of this paper. The first set of estimates is simply of the level of low income using half-median equivalent disposable cash income, with subsequent estimates adding the value of non-cash benefits per household and per capita, respectively. These low-income rates are lower for older households than for the population generally.

In summary, these results show that estimates of the size of the low-income population are sensitive to the precise choice of methodological approach made in measuring 'poverty'. Again, this reflects the concentration of older people in a relatively narrow income range around the statutory pension rates. However, a number of conclusions can be drawn from these technical choices. On the basis of cash incomes, low-income rates among older people are higher when households are used as the unit of analysis rather than income units. Similarly, using cash incomes, older people are more likely to experience relative low income than is the non-aged population. Finally, using current weekly income rather than annual income appears to produce lower estimates of relative low income

[ Return to Top   Return to Section ]

9 Housing wealth

The most important form of household wealth is home ownership, which is estimated to have accounted for 49.5 per cent of household assets in 1993 (Baekgard 1998). Home ownership is a significant factor contributing to the living standards of older people. Home ownership is widespread among the older population. Table 19 shows the level of home ownership, with and without mortgages, by life cycle groups in 1996–97.

Table 19: Dwelling tenure type by selected life cycle groups, Australia, 1996–97
Percentage of income units by type of ownership
  Owner without mortgage Owner with mortgage
Source: Australian Bureau of Statistics, Income Distribution, Australia, 1996–97.
One person, under 35 years 1.7 5.6
Couple without dependent children, reference person under 35 years 5.2 46.4
Couple with dependent children by age of oldest child
Under 5 21.2 44.8
5–14 24.1 53.5
15–24 42.4 43.4
One-parent families 11.8 18.7
Couples without dependent children
Reference person 55–64 72.8 15.5
Reference person 65 years and over 84.0 4.9
One person aged 65 and over 67.0 2.6
All units with reference person 65 and over 74.5 3.7
All income units 31.3 21.4

Among the older population, the level of home ownership is more equally distributed by income level than most other forms of private income. Table 20 shows levels of home ownership by equivalent income quintile among the pensioner population in 1986, 1990 and 1995–96. While home ownership increases with income, the extent to which this occurs is relatively slight.

Table 20: Pensioners' housing tenure by income quintile
Percentage of ownership
Source: Estimated from unit record files, ABS Income Surveys 1980 and 1990 (catalogue number 6543.0) and 1995–96 (catalogue number 6541.0.15.001)
  Owned Other

1986

 
1st 72 28
2nd 73 27
3rd 78 22
4th 79 21
5th 81 19

1990

 
1st 75 25
2nd 75 25
3rd 79 21
4th 77 23
5th 88 12

1995–96

 
1st 78 22
2nd 76 24
3rd 77 23
4th 76 24
5th 87 13

Table 21 shows ABS estimates of dwelling values and equity by age group in 1995–96. The value of dwellings owned by people aged 65 years and over is lower than among most of the younger population, but the level of loans outstanding is much lower than for most groups of younger people. As a result, older people have higher average equity than people under the age of 45 years.

Table 21: Dwelling value and equity in the home for owner-occupiers, 1995–96
Age group Mean dwelling value $000 Mean loan outstanding $000 Mean equity $000 Owner-occupier households $000
Source: ABS, Australian Social Trends 1998, catalogue number 4102.0, p. 155.
Under 35 147.6 62.3 85.3 787.3
35–44 179.0 46.7 132.3 1,082.1
45–54 188.8 22.7 166.1 1,063.5
55–64 179.2 6.7 172.5 750.2
65 and over 156.2 1.1 155.1. 1,106.2
Total 170.8 27.1 143.7 4,789.3

By modelling imputed income from owner-occupied housing, the benefits of home ownership can be taken into account in the income distribution. The most notable Australian study to do so is Yates (1991). Whiteford and Kennedy (1995) used Yates' estimates of imputed income and applied them to the 1985–86 Income Survey (in the Luxembourg Income Study). For older people, imputed income from owner-occupied housing (plus the relatively small imputed rental subsidy for those in public housing) was equivalent to 26.7 per cent of cash disposable income, compared to the corresponding value of 8.9 per cent for the population as a whole. The inclusion of imputed income plus non-cash government benefits raised the average income of older people from 73 to 86 per cent of the population mean.

[ Return to Top   Return to Section ]

10 Conclusions

A mixed picture emerges from this analysis. The average incomes of older people increased at a faster rate than for the population generally. As a result, their average incomes have risen as a proportion of the community average. Among older people, average expenditures per
person have also increased. Taking account of government non-cash benefits further improves the relative position of older people, as does imputed income from owner-occupied housing. At the same time, administrative data suggest that there are sizeable proportions of the age pensioner population who have little or no income apart from their pension, and little or limited assets. However, the extent to which this is the case appears to have decreased over time. Older people are also over-represented in the lower income quintiles of the population. The most striking feature of the incomes of the older population is the degree of concentration of incomes around pension levels. This complicates interpretation of trends in incomes and the relative position of this age group, including their vulnerability to low incomes.

In considering likely future trends in the relative position of older people, it is necessary to take account of a wide range of factors impacting on the distribution of incomes of those in the pre-pension age groups. In future, the wellbeing of the older population is likely to be enhanced by a wide range of factors, including increasing superannuation coverage, increasing labour force participation among women, higher real wages, and higher average levels of housing wealth. At the same time, there are trends that may tend to offset these, including the long-term decline and then flattening of the labour force participation of men aged 50 to 64 years (Ingles 1998), and higher wage inequality among those of working age. In addition, family trends, including the growth in the incidence of sole parent families, may also have adverse effects on wellbeing in retirement. Separated, divorced and single older women appear to have lower incomes and assets in retirement than men or couples. The trend for women to defer childbirth until later in life and the consequent compression of their prime working years, along with increased educational participation among young people, may also impact on capacity for self-provision in retirement (Jackson 1998).

In terms of future monitoring of these and related trends, it is desirable to have improved information about the dynamic processes that are associated with these developments. This would be best achieved through an ongoing longitudinal survey. To capture the diversity of outcomes among the older population, it is also necessary to use a broad range of indicators to monitor trends. Finally, the main message of this paper is that the concept of economic resources used in analysing trends in living standards is of fundamental importance. Future analysis should pay particular attention to modelling and measuring comprehensive income measures.

[ Return to Top   Return to Section ]

Appendix A: Pension levels and means tests 1965 to 1999

Table A1: Trends in the level of the standard pension
Year Single pensions as % of:
GDP per capita HDIPC MTAWE* Process worker’s wage Henderson Poverty Line,
single pensioner
Note: *This is the value of the standard rate of pension at September each year compared to the relevant MTAWE benchmark as legislated in November 1997.
Sources: Australian Bureau of Statistics Consumer Price Index, Australia, Average Weekly Earnings, Australia, and Australian National Accounts , Metal Trades Industry Association, Melbourne Institute of Applied Economic and Social Research, and Department of Social Security, Ten Yearly Statistical Summary, annual reports, and DSS Customers: A Statistical Overview, various years
1965 34.5 n.a 22.7 35.5 n.a
1966 33.5 n.a 23.7 35.0 n.a
1967 33.5 n.a 21.9 35.4 n.a
1968 31.9 n.a 22.3 33.1 n.a
1969 31.1 n.a 22.1 34.4 n.a
1970 30.7 n.a 20.9 35.8 n.a
1971 31.0 n.a 20.6 36.0 n.a
1972 32.1 n.a 23.7 35.9 n.a
1973 33.7 n.a 22.7 35.5 n.a
1974 41.3 57.9 25.9 39.0 94.9
1975 40.2 55.7 26.6 39.3 109.1
1976 39.3 55.4 26.0 39.7 110.7
1977 39.7 57.4 26.6 40.4 110.1
1978 40.3 57.1 26.4 41.4 107.7
1979 37.0 52.7 26.7 38.4 100.1
1980 37.8 54.7 26.7 40.0 103.0
1981 36.8 53.6 25.4 39.0 99.3
1982 36.9 53.7 24.0 35.2 102.0
1983 38.2 56.4 25.0 39.1 108.1
1984 37.0 54.1 23.9 39.1 104.3
1985 35.7 52.8 24.6 38.8 108.8
1986 35.4 53.5 24.0 42.0 109.0
1987 35.9 54.7 24.9 43.3 110.3
1988 34.6 53.6 24.9 43.6 107.1
1989 33.3 51.6 25.7 44.0 101.8
1990 33.9 52.5 26.2 43.4 108.6
1991 35.8 57.0 26.5 45.5 120.5
1992 36.0 54.8 25.6 44.7 116.5
1993 35.3 53.4 25.8 45.6 117.2
1994 34.3 52.3 25.7 45.4 112.7
1995 33.2 50.0 25.7 44.9 110.4
1996 33.6 50.3 25.8 47.8 111.5
1997 32.5 49.7 25.3 46.2 109.6
1998 30.5 49.3 25.0 45.4 108.6
1999 30.5 49.3 25.0 45.4 106.4

 

Table A2: Major policy developments - Age Pensions and related payments
Year Policy Development
Source: Department of Social Security, annual reports, various years1996
1965 Guardian's Allowance introduced for widowed and unmarried age or invalid pensioners with care of children; Wife's Allowance extended to wives of pensioners with care of children.
1966 All restrictions on payments applying to Aboriginal Australians were removed; Pensioner Medical Service introduced; extension of period for retrospective pension payable for persons discharged from mental hospitals; pension rate increased.
1967 Sheltered Employment Allowance introduced.
1969 Pension means test taper introduced at 50 per cent rather than 100 per cent.
1970 Standard rate of pension extended to cover married pensioners permanently separated due to illness or infirmity; supplementary (rental) assistance extended to cover these circumstances.
1972 Pension means test including free area liberalised; Wife's Pension introduced to replace Wife's Allowance and extended to all non-pensioner wives of age pensioners; portability provisions for pensioners going overseas to countries with reciprocal agreements, and subject to specified residence qualifications.
1973 Portability provisions available wherever pensioner chooses to live.
1974 Residence requirement removed for Invalid Pension where permanent incapacity or blindness occurred in Australia; Incentive Allowance paid free of means test in lieu of Supplementary Allowance to persons in sheltered workshops; character and not deserving provisions removed from Social Services Act.
1976 Twice-yearly indexation of basic pension rates introduced; pensions income test replaces means test.
1978 Indexation provision altered to annual rather than twice-yearly increases.
1979 Twice-yearly indexation reintroduced.
1980 Pensions became payable to patients in mental institutions.
1982 Supplementary (rental) allowance no longer payable to claimants living in public housing.
1983 Rehabilitation Allowance introduced at Invalid Pension rates and conditions for those in rehabilitation program and up to 6 months after; introduction of Mobility Allowance for persons with severe disability in employment or vocational training and not able to use public transport without assistance; indexation of pensioner fringe benefit limits; introduction of Spouse/Carers Pension.
1984 Introduction of Remote Area Allowance.
1985 Assets test introduced for all income-tested pensioners; Carer's Pension replaced Spouse/Carer's Pension with eligibility extended to other relatives.
1986 Portability provisions altered to provide for proportional payment to pensioners overseas; indexation delayed six weeks, and changed to December/June rather than November/May.
1987 New assessment procedures for Invalid Pension, primary eligibility now medical; Invalid Pension, Rent Assisatance and Sheltered Employment Allowance became subject to preclusion periods for lump sum compensation received, with direct deduction of periodic payments from pension; separate income test on Rent Assistance removed; introduction of pensions earnings credit; Pensioner Education Supplement introduced.
1988 Holders of Pensioner Health Benefits Card allowed to retain for three months after private income exceeded cut-outs by up to 25 per cent; standardised assessment of market-linked and capital-stable investments and trust.
1989 Indexation lag reduced from 24 to 12 weeks, with indexation applied to fringe benefit limits and assets test free area; introduction of pre-departure certificates for pensioners leaving Australia.
1990 Introduction of new bereavement arrangements with lump-sum payments; pensions income test adjusted to align treatment of annuities and funded superannuation pensions with tax treatment; introduction of Pharmaceutical Allowance to compensate those affected by co-payments.
1991 Disability reform package - Disability Support Pension replaced Invalid Pension, with eligibility based on level of impairment sufficient to prevent 30 hours work a week for at least two years, youth rates introduced; disability panels established; Incentive allowance and Sheltered Employment Allowance abolished and Rent Assistance closed to new claims; portability of Disability Support Pension restricted to 12 months for some customers; Rent Assistance subject to six-monthly indexation; introduction of deeming provisions; Pharmaceutical Allowances extended to all pensioners; pensions income test free area subject to indexation; Employment Entry Payment introduced.
1992 Allowed farm assets and liabilities totalled (previously not offset); indexation of pensioner earnings credit; telephone rental voucher system replaced by Telephone Allowance.
1993 Ad hoc pension increase in January to meet 25 per cent MTAWE benchmark; separate income test on pensioner concession cards abolished; Education Entry Payment extended to Disability Support Pension recipients; independent rate paid to those under 18 after 18 weeks.
1994 Disability Wage Supplement introduced at Disability Support Pension conditions for those working under special disability award rates; requirement to claim overseas pensions if eligible; Commonwealth Senior's Health Card for those of age pension age not entitled to pension; calculation of compensation preclusion period altered to increase waiting periods.
1995 Age pension age for women raised to 60.5 years; rRelaxation of deprivation provisions for intergenerational transfer of farms.
1996 Extended deeming introduced; lump-sum advances of pensions introduced; Pensioner Loans Scheme extended.
1997 Legislation to maintain single pension at 25 per cent of MTAWE with flow ons; Earnings Credit Scheme abolished; removal of deeming exemption for first $2,000 ($4,000 for couples); maximum rate of Rent Assistance reduced for those in shared accommodation, but not disabiltiy support pensioners and carers; removal of means test exemption for superannuation assets of those aged 55 but under age pension age after 39 weeks on income support.
1998 Disability Wage Supplement subsumed into Disabiltiy Support Pension; treatment of income streams simplified; date of effect of all income support payments standardised.

Appendix B: Incomes and selected characteristics of income units, Australia, 1982 to 1997–98

Table B1: Incomes and selected characteristics of income units, Australia, 1982

Characteristics

Couples with reference 65 years and over

Single person income units 65 years and over

All with reference person 65 years and over

All income units

Source: Australian Bureau of Statistics, Income and Housing Survey, Income of Income Units, Australia, 1981–82, catalogue number. 6523.0, Table 8, page 16.
Total number of income units (000) 439.2 600.3 1,039.5 6,953.1
Mean gross income ($ per week) $208 $111 $152 $303
Median gross income ($ per week)
Principal source of income (% of income units)
Wage or salary
Own business or partnership
Other private income
Government pensions and Allowances
Per cent contribution of government pensions and allowances to gross income
Nil and less  than 1

{ 72.9

1 and less  than 20
20 and less  than 50 3.9
50 and less  than 90 7.5
90 and over 15.7
Total 100.0
Tenure type
Owner without mortgage 80.0 59.8 68.3 27.6
Owner with mortgage 7.3 3.5 5.1 25.4

Renter

Public 4.3 7.2 6.0 3.9
Private 3.6 6.9 5.5 15.7

Resident relative

Other 1.4 9.8 6.2 17.3
Total renter 9.2 23.9 17.7 36.9
Other 3.5 12.8 8.8 10.0
Total 100.0 100.0 100.0 100.0

T ype of income unit

Couple without children 21.7
Couple with children 29.5
All couples 51.1
One-parent 4.1
One-person 44.7
Total 100.0

 

Table B2: Incomes and selected characteristics of income units, Australia, 1986
Characteristics Couple income units with reference person 65 years and over Single person income units with reference person 65 years and over Income units with reference person 65 years and over All income units
Note: *Subject to high sampling variability.
Source: ABS, Income Distribution, Australia, 1986, catalogue number. 6523.0, Table 4, p. 15; Table 11, p. 23; Table 21, p. 33.
Total number of income units (000) 521.5 643.3 1,166.2 7,464.1
Mean gross income ($ per week) $270 $143 $200 $410
Median gross income ($ per week) $206 $112 $170 $328

Principal source of income (% of income units)

Wage or salary 3.6 *0.3 1.8 58.7
Own business or partnership 2.2 *0.9 1.4 6.9
Other private income 19.6 16.7 18.0 6.8
Government pensions and Allowances 74.7 82.1 78.8 27.7
Total 100.0 100.0 100.0 100.0
Nil and less than 1 10.9 9.4 10.1 48.3
1 and less than 20 6.8 2.4 4.4 21.7
20 and less than 50 8.0 6.2 7.0 2.4
50 and less than 90 31.0 18.4 24.0 6.8
90 and over 43.3 63.7 54.6 20.8
Total 100.0 100.0 100.0 100.0

Tenure type

Owner without mortgage 77.1 60.9 68.1 29.2
Owner with mortgage 10.0 3.1 6.2 24.3
Renter  
Public 3.9 8.3 6.4 4.1
Private 2.5 6.9 4.9 14.2
Resident relative  
Other  
Total renter 6.4 15.2 11.3 18.3
Other 5.7 18.2 12.6 26.3
Total 100.0 100.0 100.0 100.0

Type of income unit

Couple without children 97.5 43.6 24.0
Couple with children 2.5 1.1 26.4
All couples 100.0 44.7 50.4
One-parent *0.1 4.2
One-person 100.0 55.2 4.5

 

Table B3: Incomes and selected characteristics of income units, Australia, October–December 1990
Characteristics Couple income units with reference person 65 years and over Single person income units with reference person 65 years and over Income units with reference person 65 years and over All income units
Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Income Distribution, Australia, 1995–96, catalogue number. 6523.0, Table 4, p. 15; Table 11, p. 23; Table 21, p. 33.
Total number of income units (000) 591.4 733.4 1,327.7 7,986.8
Mean gross income ($ per week) $423 $214 $307 $563
Median gross income ($ per week) $307 $167 $239 $434

Principal source of income (% of income units)

Wage or salary 4.3 *0.6 2.2 58.3
Own business or partnership 2.5 1.1 1.7 6.6
Other private income 23.3 16.7 19.7 8.5
Government pensions and allowances 69.8 81.6 76.4 26.6
Total 100.0 100.0 100.0 100.0

Per cent contribution of government pensions and allowances to gross income

Nil and less than 1 17.9 11.8 14.5 51.2
1 and less than 20 4.9 1.9 3.2 19.2
20 and less than 50 8.1 5.0 6.4 3.2
50 and less than 90 31.2 27.1 28.9 7.9
90 and over 37.9 54.1 46.9 18.6
Total 100.0 100.0 100.0 100.0

Tenure type

Owner without mortgage 81.2 64.5 71.9 32.5
Owner with mortgage 6.5 3.5 4.8 22.5
Renter  
Public 3.8 7.9 6.1 4.5
Private 3.3 5.6 4.6 15.6
Resident relative *0.3 3.5 2.1 9.7
Other *0.7 2.6 1.8 4.0
Total renter 8.2 19.6 14.6 33.8
Other 3.4 11.0 7.6 9.9
Total 100.0 100.0 100.0 100.0

Type of income unit

Couple without children 98.6 43.9 25.1
Couple with children 1.4 0.6 25.8
All couples 100.0 44.5 50.9
One-parent *0.2 4.8
One-person 55.2 44.3
Total 100.0 100.0 100.0 100.0

 

Table B4: Incomes and selected characteristics of income units, Australia, 1994–95
Characteristics Couple income units with reference person 65 years and over Single person income units with reference person 65 years and over Income units with reference person 65 years and over All income units

Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Income Distribution, Australia, 1995–96, catalogue. number. 6523.0, Table 4, p. 15; Table 11, p. 23; Table 21, p. 33.

Total number of income units (000) 643.9 840.6 1,484.8 8,984.3
Mean gross income ($ per week) $410 $208 $296 $579
Median gross income ($ per week) $321 $173 $243 $434

Principal source of income (% of income units)

Wage or salary 3.7 *1.4 2.4 56.2
Own business or partnership 2.9 *0.8 1.7 5.7
Other private income 24.7 13.2 18.2 6.8
Government pensions and allowances 68.3 83.7 77.0 28.8
Total 100.0 100.0 100.0 100.0

Per cent contribution of government pensions and allowances to gross income

Nil and less than 1 16.0 8.7 11.9 50.1
1 and less than 20 5.8 *1.1 3.1 13.3
20 and less than 50 10.4 5.6 7.7 4.9
50 and less than 90 18.4 12.3 14.9 5.2
90 and over 48.9 71.3 61.6 23.4
Total 100.0 100.0 100.0 100.0

Tenure type

Owner without mortgage 84.9 62.9 72.5 32.9
Owner with mortgage 5.5 4.0 4.7 20.7
Renter  
Public *2.0 10.4 6.8 4.2
Private 3.5 6.8 5.4 17.5
Resident relative   4.2 2.6 8.2
Other *1.1 2.2 1.6 3.2
Total renter 6.7 23.7 16.3 33.0
Other *2.3 9.1 6.1 12.6
Total 100.0 100.0 100.0 100.0

Type of income unit

Couple without children 99.0 - 42.9 25.4
Couple with children *1.0 - *.04 21.3
All couples 100.0 - 43.4 46.8
One-parent - - - 4.3
One-person - 100.0 56.6 48.9
Total 100.0 100.0 100.0 100.0

 

Table B5: Incomes and selected characteristics of income units, Australia, 1995–96
Characteristics Couple income units with reference person 65 years and over Single person income units with reference person 65 years and over Income units with reference person 65 years and over All income units

Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Income Distribution, Australia, 1995–96, catalogue. number. 6523.0, Table 4, p. 15; Table 11, p. 23; Table 21, p. 33.

Total number of income units (000) 684.8 811.4 1,496.8 8,888.3
Mean gross income ($ per week) $429 $226 $319 $609
Median gross income ($ per week) $320 $184 $259 $457

Principal source of income (% of income units)

Wage or salary 4.6 *1.1 2.7 55.5
Own business or partnership 3.0 *0.8 1.8 6.5
Other private income 21.8 17.4 19.4 7.0
Government pensions and allowances 70.2 80.0 75.5 29.0
Total 100.0 100.0 100.0 100.0

Per cent contribution of government pensions and allowances to gross income

Nil and less than 1 15.6 12.2 13.7 50.2
1 and less than 20 6.4 *1.3 3.6 13.5
20 and less than 50 8.1 6.0 6.9 5.1
50 and less than 90 18.6 16.0 17.2 6.2
90 and over 50.9 63.8 57.9 22.6
Total 100.0 100.0 100.0 100.0

Tenure type

Owner without mortgage 85.2 64.0 73.6 32.4
Owner with mortgage 4.2 *1.9 3.0 21.9
Renter  
Public 3.0 9.2 6.4 4.7
Private 3.9 7.3 5.8 19.9
Resident relative - 4.2 2.3 7.8
Other *0.3 3.5 1.9 2.8
Total renter 7.3 24.2 16.4 35.2
Other 3.4 9.9 6.9 10.5
Total 100.0 100.0 100.0 100.0

Type of income unit

Couple without children 98.6 - 45.1 24.6
Couple with children *1.4 - 0.6 22.7
All couples 100.0 - 45.8 47.3
One-parent - - - 4.9
One-person - 100.0 54.2 47.8
Total 100.0 100.0 100.0 100.0

 

Table B6: Incomes and selected characteristics of income units, Australia, 1996–97
Characteristics Couple income units with reference person 65 years and over Single person income units with reference person 65 years and over Income units with reference person 65 years and over All income units

Note: *Subject to high sampling variability.
Source:Australian Bureau of Statistics, Income Distribution, Australia, 1996–97, catalogue. number. 6523.0, Table 4, p. 17; Table 11, p. 26; Table 21, p. 36.

Total number of income units (000) 691.6 867.2 1,561.9  
Mean gross income ($ per week) $481 $242 $348 $625
Median gross income ($ per week) $350 $194 $276 $477

Principal source of income (% of income units)

Wage or salary 5.0 *0.4 2.5 54.8
Own business or partnership 3.1 *1.5 2.2 5.8
Other private income 26.2 15.6 20.3 7.3
Government pensions and allowances 64.9 81.4 74.1 30.0
Total 100.0 100.0 100.0 100.0

Per cent contribution of government pensions and allowances to gross income

Nil and less than 1 17.2 9.8 13.1 49.8
1 and less than 20 7.2 2.9 4.8 13.4
20 and less than 50 11.0 5.1 7.7 4.6
50 and less than 90 20.8 17.2 18.7 6.8
90 and over 43.1 63.9 54.7 23.0
Total 100.0 100.0 100.0 100.0

Tenure type

Owner without mortgage 84.1 67.0 74.5 31.3
Owner with mortgage 4.9 2.6 3.7 21.4
Renter  
Public *2.5 9.0 6.2 4.4
Private 3.2 5.7 4.6 20.6
Resident relative   3.1 2.1 7.5
Other *1.3 2.7 1.7 2.5
Total renter 7.1 20.6 14.6 34.9
Other 3.9 9.9 7.2 12.3
Total 100.0 100.0 100.0 100.0

Type of income unit

Couple without children 99.1 - 43.9 24.2
Couple with children *0.9 - *0.4 22.3
All couples 100.0 - 44.3 46.5
One-parent -- - - 5.2
One-person   100.0 55.5 48.3
Total 100.0 100.0 100.0 100.0

 

Table B7: Incomes and selected characteristics of income units, Australia, 1997–98
Characteristics Couple income units with reference person 65 years and over Single person income units with reference person 65 years and over Income units with reference person 65 years and over All income units

Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Income Distribution, Australia, 1997–98, catalogue. number. 6523.0, Table 4, p. 15; Table 11, p. 23; Table 21, p. 33.

Total number of income units (000) 701.6 895.9 1,597.9 9,129.4
Mean gross income ($ per week) $460 $248 $341 $658
Median gross income ($ per week) $358 $191 $268 $499

Principal source of income (% of income units)

Wage or salary 5.1 *1.0 2.8 54.9
Own business or partnership 3.1 *1.4 2.1 5.8
Other private income 25.8 17.3 21.0 7.3
Government pensions and allowances 65.4 79.7 73.4 29.6
Total 100.0 100.0 100.0 100.0

Per cent contribution of government pensions and allowances to gross income

Nil and less than 1 15.1 12.5 13.6 50.3
1 and less than 20 7.5 *1.4 4.0 12.5
20 and less than 50 11.7 6.5 8.8 5.3
50 and less than 90 21.1 12.8 16.4 6.5
90 and over 44.1 66.2 56.5 23.0
Total 100.0 100.0 100.0 100.0

Tenure type

Owner without mortgage 84.9 63.9 73.1 30.6
Owner with mortgage 3.8 2.6 3.1 23.6
Renter  
Public 3.1 10.1 7.0 4.5
Private 4.3 5.4 4.9 19.8
Resident relative   4.5 2.7 8.3
Other *0.9 *2.2 1.4 2.3
Total renter 8.3 22.2 16.1 34.8
Other 3.1 11.3 7.7 11.0
Total 100.0 100.0 100.0 100.0

Type of income unit

Couple without children 98.6 - 43.3 24.3
Couple with children *1.4 - *0.6 22.6
All couples 100.0 - 43.9 46.9
One-parent - - - 5.7
One-person - 100.0 56.1 47.4
Total 100.0 100.0 100.0 100.0

[ Return to Top   Return to Section ]

Appendix C: The effects of government benefits and taxes on household income, Australia, 1984 to 1993–94

Table C1: The effects of government benefits and taxes on household income, Australia, 1984
Income, Benefits and Taxes Couple, only, reference person 65 and over Single  person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Household Expenditure Survey, Australia, 1984: The Effects of Government Benefits and Taxes on Household Income, ABS catalogue number. 6537.0, Tables 3.2, 3.9, 3.40 and 3.47.
Private  income 112.38 49.94 322.59 120.48 401.43
Direct Penefits
Age Pension 90.94 70.08 88.50 81.63 18.16
DVA Pension 29.69 11.01 21.27 20.69 7.21
Total direct benefits 124.98 83.45 133.49 109.00 52.18
Gross income 237.36 133.38 456.08 229.48 453.60
Direct tax 21.22 11.15 70.14 24.95 91.60
Disposable income 216.15 122.23 385.94 204.53 362.01

Indirect benefits

Education 0.71 * 13.21 3.02 37.74
Hospital  care

{ 49.86

{ 31.41

{ 56.13

29.37 23.08
Other health beneifts 13.79 12.14
Housing benefits 1.22 2.75 2.00 1.98 2.59
Other welfare 15.32 16.09 17.75 16.02 7.22
Total indirect 67.11 50.61 89.09 64.18 82.78
Indirect taxes 19.77 9.15 33.64 17.63 37.74
Final income 263.49 163.68 441.39 251.09 407.05
Average persons per household 2.0 1.0 2.8 1.7 2.8
Average income units per household 1.0 1.0 2.2 1.2 1.3

Households in

Population (000) 395.6 399.2 154.7 955.2 5039.2
Persons (000) 791.2 399.2 433.9 1,643.0 14,290.8
Government renters (%) 3.7 7.6 5.6 5.6 5.7
Principal income source from government benefits (%) 75.9 81.7 37.1 72.0 26.2

 

Table C2: The effects of government benefits and taxes on household income, Australia, 1988–89
Income, Benefits and Taxes Couple, only, reference person 65 and over Single  person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Household ExpenditureSurvey, Australia, 1988–89: The Effects of Government Benefits andTaxes on Household Income, ABS catalogue number. 6537.0, Tables 2, 9, 25 and 28.
Private  income 191.80 64.53 439.70 177.62 572.02
Direct Penefits
Age Pension 100.82 83.25 100.60 93.20 21.13
DVA Pension 54.02 19.66 44.28 38.43 9.88
Total direct benefits 165.13 107.19 194.54 145.39 64.02
Gross income 356.93 171.72 634.24 323.01 636.04
Direct tax 33.58 8.28 93.92 32.41 129.45
Disposable income 323.35 163.43 540.32 290.60 506.59

Indirect benefits

Education 1.10 * 18.48 3.65 48.95
Hospital  care       57.71 26.59
Other health beneifts 103.10 60.26 98.71 26.94 20.44
Housing benefits 1.50 3.73 1.42 2.41 2.42
Other welfare 31.13 18.33 33.32 26.13 11.84
Total indirect 136.83 82.56 151.93 116.83 110.23
Indirect taxes 35.19 14.68 59.48 30.58 64.57
Final income 424.99 231.33 632.74 376.85 552.26
Average persons per household 2.0 1.0 2.8 1.7 2.8
Average income units per household 1.0 1.0 2.2 1.2 1.3

Households in

Population (000) 431.0 417.5 156.1 1,008.9 5,240.4
Persons (000) 862.1 417.5 441.6 1,733.2 15,072.9
Government renters (%) 5.0 10.7 *5.4 7.4 6.3
Principal income source from government benefits (%) 71.2 80.5 34.7 69.2 24.4

 

Table C3: The effects of government benefits and taxes on household income, Australia, 1993–94
Income, Benefits and Taxes Couple, only, reference person 65 and over Single  person 65 and over Multiple income units with reference person 65 and over All households, with reference person 65 years and over All households
Note: *Subject to high sampling variability.
Source: Australian Bureau of Statistics, Household ExpenditureSurvey, Australia, 1993–94: The Effects of Government Benefits and Taxes on Household Income, ABS catalogue number. 6537.0, Tables 2, 7, and 29.
Private  income 185.80 65.02 413.86 163.44 626.43
Direct Penefits
Age Pension 131.66 114.27 164.45 127.81 29.97
DVA Pension 64.39 32.23 *38.41 46.05 10.30
Total direct benefits 204.17 148.10 256.47 185.24 96.79
Gross income 389.98 213.12 670.34 348.68 723.23
Direct tax 30.73 13.82 82.58 30.33 136.99
Disposable income 359.24 199.30 587.76 318.35 586.23

Indirect benefits

Education 0.69 * 14.45 2.58 52.19
Hospital  care 77.21 47.68 { 118.12 63.44 29.80
Other health beneifts 39.00 21.05 31.45 29.71
Housing benefits 2.75 5.03 2.90 3.79 3.87
Other welfare 51.84 32.74 59.34 44.02 19.38
Total indirect 171.49 106.58 194.82 145.29 134.96
Indirect taxes 37.97 15.47 50.56 29.55 60.28
Final income 492.76 290.41 732.01 434.09 660.91
Average persons per household 2.0 1.0 2.7 1.6 2.6
Average income units per household 1.0 1.0 2.2 1.2 1.3

Households in

Population (000) 523.1 582.4 173.8 1,285.3 6,616.8
Persons (000) 1,046.1 582.4 475.0 2.118.3 17,394.6
Government renters (%) 4.9 11.4 *6.4 8.0 7.1
Principal income source from government benefits (%) 72.5 82.9 43.9 73.1 29.5

[ Return to Top   Return to Section ]

Endnotes

  1. Older people are persons of Age Pension age or over - 60 years for women and 65 for men for most of period of time for which data were collected, although the pension age for women is now rising gradually. However, much of the published data refer to persons aged 65 years and over, or persons in households where the reference person is aged 65 years and over.
  2. This approach is in turn derived from the United Kingdom Central Statistical Office (CSO) Fiscal Incidence Studies (CSO, 1990).
  3. At June 1998, there were 138,000 persons aged 65 years and over receiving other payments and 34,400 women aged 60 to 64 years receiving other payments.
  4. During periods of falling inflation, the lag between the period used as the base for an indexation increase and inflation in the period of measurement means that the real value of the payment will rise to a small extent.
  5. No Australian government has ever endorsed the Henderson poverty line as a measure of adequacy. As discussed below there are significant conceptual problems with this measure.
  6. It should also be noted that the ABS Income Surveys cover people in private and special dwellings. They exclude people in institutions such as hospitals, nursing homes and hostels and retirement villages.
  7. This is in marked contrast to the trend in the United Kingdom, for example, where between 1979 and 1996–97 pensioners' incomes increased by 60 per cent in real terms before housing costs. However, the difference appears to be explained by the extremely low incomes of pensioners in the United Kingdom at the start of the period. See The Pensioners Incomes Series 1996-97, www.dss.gov.uk/hg/press/press1298/298.htm.
  8. The Gini coefficient 'is a measure of the expected difference between the incomes of any two units in the population and has been scaled to lie between zero and one. It has the value zero when income is distributed equally and the value one when one unit receives all the income.' (ABS 1998, 1996-97 Income Distribution Australia, catalogue number 6523.0, p.61).
  9. Equivalised using the McClements equivalence scale. The methodology used conforms as closely as possible to the United Kingdom series of Households Below Average Income statistics. See Whiteford and Kennedy (1995) for details.
  10. It may be that the United States does not emphasise earnings replacement, but that its pension system does not substantially alter the distribution of income.
  11. This also means that estimates of poverty among the older population in Australia are more sensitive to the choice of poverty line than in other countries.
  12. Pensioners are defined as all those aged 65 years and over, plus persons aged 60 to 64 years who are not participating in the labour market. The methodology follows that of Johnson (1998).
  13. Neither the Henderson Poverty Line nor any other measure has been endorsed as an official benchmark of relative low income by an Australian Government.
  14. These are the equivalence scales implicit in the current structure of income support payments.

[ Return to Top   Return to Section ]

References

Atkinson, A.B. 1989, 'Poverty' in The New Palgrave Social Economics, eds J. Eatwell, M. Milgate & P. Newman, Macmillan, London.

Atkinson, A.B. 1990, 'Comparing Poverty Rates Internationally: Lessons from Recent Studies in OECD Countries', Discussion Paper no.53, Welfare State Programme, London School of Economics.

Australian Bureau of Statistics (ABS) 1980, 1990 and 1995–96, Income Distribution Australia, catalogue number 6523.0, ABS, Canberra.

Australian Bureau of Statistics (ABS) 1984, 1988-89 and 1993-94, Household Expenditur eSurvey, Australia, catalogue number 6537.0, ABS, Canberra.

Australian Bureau of Statistics (ABS) 1987, 1992, 1996, The Effects of Government Benefits and Taxes and Household Income, catalogue number. 6537.0, ABS, Canberra.

Australian Bureau of Statistics (ABS) 1998, Australian Social Trends, catalogue number 4102.0, ABS, Canberra.

Australian Bureau of Statistics (ABS) 1999a, Consumer Price Index, Australia, catalogue number 6401.0, ABS, Canberra.

Australian Bureau of Statistics (ABS) 1999b, Average Weekly Earnings, Australia, catalogue number 6350.0, ABS, Canberra.

Australian Bureau of Statistics (ABS)1999c, Australian National Accounts, catalogue number 5206.0, ABS, Canberra.

Baekgaard, H. 1998, 'Simulating the Distribution of Household Wealth in Australia: New Estimates for 1986 and 1993', Technical Paper no. 14, NATSEM, University of Canberra.

Centrelink 1999, A Guide to Commonwealth Government Payments, 20 March to 30 June 1999, Canberra

Creedy, J. 1992 'Increasing the Time Period in the Measurement of Inequality' in D.J. Slottje (ed.), Research on Economic Inequality, Vol. 2, JAI Press, Greenwich.

Creedy, J. 1994, 'Statics and Dynamics of Income Distribution: An Introductory Survey', The Australian Economic Review, 4th Quarter, pp. 51–71.

Creedy, J. & Disney, R. 1989, 'The Australian Pension System: Some Basic Analytics', Economic Record, 65, 191, December, pp. 357–368.

Department of Social Security, various years, DSS Customers: A Statistical Overview, DSS Canberra.

Department of Social Security, various years, Ten Year Statistical Summary, DSS Canberra.

Donald, O. 1984, 'Government Support of Retirement Incomes in Australia', Research Paper no. 24, Canberra: Department of Social Security, April.

Falkingham, J. & Harding, A. 1996, 'Poverty Alleviation versus Social Insurance: a comparison of lifetime redistribution', in A. Harding (ed.), Microsimulation and Public Policy, North- Holland, Amsterdam.

Foster, C. 1988, 'Towards a National Retirement Incomes Policy', Social Security Review Issues Paper no. 6, Australian Government Publishing Service, Canberra.

Harding, A. 1992, 'Income Distribution and Redistribution Across the Lifecycle: Evidence from Australia', Discussion Paper no. 70, Welfare State Programme, STICERD, London School of Economics

Harding, A. and Szukalska, A. 1999, 'Trends in Child Poverty in Australia: 1982 to 1995-96', Discussion Paper no. 42, NATSEM, University of Canberra.

Ingles, D., 1998, Older Workers, Disability and Early Retirement in Australia, paper for a conference on Income Support, Labour Markets and Behaviour: A research Agenda, Department of Family and Community Services, Canberra.

Jackson, N., 1989, 'Understanding population ageing: a background', in Australian Social Policy, 1999/1, Department of Family and Community Services, Canberra. Johnson, P. 1998, Older Getting Wiser, Institute of Chartered Accountants, Sydney.

King, A. 1998, 'Income Poverty since the Early 1970s' in Australia n Poverty: Then and Now, eds R. Fincher and J. Nieuwenhuysen, Melbourne University Press, Melbourne.

Melbourne Institute of Applied Economic and Social Research (MIAESR), Poverty Lines: Australia, December Quarter 1999, University of Melbourne.

Organisation for Economic Co-operation and Development (OECD) 1998, Maintaining Prosperity in an Ageing Society, OECD, Paris

Organisation for Economic Co-operation and Development (OECD) 1988a, Reforming Public Pensions, Social Policy Studies number. 5, OECD, Paris.

Organisation for Economic Co-operation and Development (OECD) 1988b, The Future of Social Protection, Social Policy Studies number. 6, OECD, Paris.

Organisation for Economic Co-operation and Development (OECD) 1988c, Ageing Populations: The Social Policy Implications, OECD, Paris.

Organisation for Economic Co-operation and Development (OECD) 1996, Ageing in OECD Countries: A Critical Policy Challenge, Social Policy Studies number. 20, OECD, Paris.

Piggott, J. 1987, 'Statistical Incidence Studies: An Economic Perspective', in Redistribution and the W elfare State: Estimating the Effects of Government Benefits and Taxes on Household Income, ed P. Saunders, Social Policy Research Centre Reports and Proceedings number. 67, University of New South Wales.

Productivity Commission and Melbourne Institute of Applied Economic and Social Research (PC&MIAESR) 1999, 'Conference background paper', in PC&MIAESR, Policy Implications of the Ageing of Australia's Population, Conference Proceedings, AusInfo, Canberra.

Quinn, J. 1987, 'The economic status of the elderly: beware of the mean', Review of Income and Wealth, 33(1), pp. 63–82.

Saunders, P. 1994, W elfare and Inequality, Cambridge University Press, Melbourne. United Kingdom, Department of Social Security 1994, Households Below Average Income: A Statistical Analysis 1979–1991/92, HMSO, London.

United Kingdom Central Statistical Office (UKCSO) 1991, 'The Effects of Taxes and Benefits on Household Incomes, 1988', Economic T r ends, 449, March, pp. 107–149, HMSO, London

Whiteford, P. 1995, 'The Use of Replacement Rates in International Comparisons of Benefit Systems', International Social Security Review, 48, 2/95, pp. 3–30.

Whiteford, P. & Kennedy, S. 1995, Incomes and Living Standards of Older People: A Comparative Analysis, UK Department of Social Security, Research Report number. 34, HMSO, London.

Whiteford, P. & Morrow, I. 1998, The Growth of Social Security Spending in Australia, 1965 to 1998, paper given at the Conference of Economists, Sydney.

Wolfson, M. 1987, 'Lifetime Coverage: The Adequacy of Canada's Retirement Income System', in Distribution of Household Wealth, ed E. Wolff, Clarendon Press, Oxford.

World Bank 1994, Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press, Oxford.

Yates, J. 1991, Australia's Owner-Occupied Housing Wealth and Its Impact on Income Distribution, Social Policy Research Centre Reports and Proceedings no. 92, University of New South Wales.

[ Return to Top   Return to Section ]

Content Updated: 24 April 2014