Australian Bureau of Statistics
6523.0 - Household Income and Income Distribution, Australia, 2007-08
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 20/08/2009
|Page tools: Print Page Print All RSS Search this Product|
SUMMARY OF FINDINGS
METHODOLOGICAL CHANGES AND INCOME MEASURES
Appendix 4 describes the most recent changes, made with this issue, to household income measures published by the ABS. These changes are the last in a succession of changes made over recent cycles to implement new international statistical standards for household income measurement and analysis.
The changes improve coverage of income by:
Excluding the effects of the improved coverage in 2007-08 reduces the Gini coefficient from 0.331 to 0.319. The 0.319 measure for 2007-08 is up 5.6% on the 1994-95 measure of 0.302. While other methodological changes introduced with the 2003-04 and 2005-06 survey results have contributed to this difference, the residual movement (after methodological changes) in this summary indicator is very likely to be statistically significant. Some other indicators of income distribution show a similar pattern. As the table below shows, the income share going to first four income quintiles fell, while the share for the fifth quintile rose. While one of the three ratios of the higher percentile income levels to the lower percentiles rose significantly (the P90/P10 ratio was up by about 9%) the other two fell slightly, and there was little change in the P20/P50 ratio.
A major contributor to some of the changes in the income distribution measures in 2007-08, when compared with 2005-06, was the strong rise in wage and salary incomes (up 28%). After adjusting for the change in scope of the measures, a 22% increase remains, and is associated with the numbers of employees rising 8% (while the number of households rose only 2%), and with rising average wages and salaries.
In 2007-08, average (mean) equivalised disposable household income for all persons living in private dwellings (i.e. the income that a single person household would require to maintain the same standard of living as the average person living in all private dwellings in Australia) was $811 per week (Table 1). There were approximately 20.6 million people living in private dwellings (Table 2).
In real terms, average equivalised disposable household income in 2007-08 ($811) was 16% higher than in 2005-06 ($699) reflecting in part the break in series due to the improvements in measuring income introduced in this cycle (see Appendix 4 for more information). Adjusting for this break in series the net increase was 13% between 2005-06 and 2007-08 and 50% between 1994-95 and 2007-08.
For low income people (i.e those people with household income in the second and third deciles) average equivalised disposable household income grew by 12% ($44 per week) from 2005-06 to 2007-08 or 10% when adjusting for the break in series. For middle income people the rise was 14% (11% when adjusted for the break in series) and 20% for high income people (16% when adjusted for the break in series).
Households with different characteristics tend to have different income levels, as shown in table 6, and summarised in the following table. Wages and salaries were the principal source of income (PSI) for households with middle and high income levels in 2007-08, while government pensions and allowances dominated for low income households. However, low income households had the highest incidence of full ownership of their home, reflecting the high proportion of older people in the low income category.
Middle income households contained more people on average than high income households (2.9 compared to 2.5) but contained fewer employed persons (1.6 compared to 1.9). In part, this reflects the different age profiles of the two groups. Table 6 shows that middle income households had an average of 0.8 persons under the age of 18 and 0.2 aged 65 and over, compared to 0.4 and 0.1 respectively for high income households. Low income households had an average of 0.7 employed persons, and housed an average of 2.6 persons. Of these, 0.8 were under 18 years, 1.2 were 18 to 64 years, and 0.6 were aged 65 years and over (Table 6).
The characteristics of Australian households are changing over time. Table 3 shows that the average number of persons per household declined from 2.69 to 2.56, or about 5%, between 1994-95 and 2007-08. The proportion of couple only households increased from 23.7% to 26.5%, a higher increase than with any other family composition type. Each principal source of income retained its relative importance between 1994-95 and 2007-08, with 61.5% of households primarily dependent on wages and salaries in 2007-08. The proportion of households reliant on government pensions and allowances was 23.2% in 2007-08, down from 26.1% in 2005-06 and 28.5% in 1994-95. Over the last decade, home ownership remained relatively stable at around 70%.
Life cycle stages
Income levels across the population partly reflect the different life cycle stages that people have reached. A typical life cycle includes childhood, early adulthood, and the forming and maturing of families, as illustrated in table 12. Other family situations and household compositions are shown in table 11. The following table compares households in different life cycle stages.
Younger couples without children had the highest mean equivalised disposable household income of $1,155 (Table 12) per week, with an average of 1.8 employed persons in the household. For couples with dependent children only, and with the eldest child being under five, mean equivalised disposable household income was $871 per week (25% lower than for the young couples without children) (Table 12). This lower income principally reflects the lower average number of employed persons in these households (1.5) and the larger average number of persons in these households (3.4) over which incomes are shared.
Average incomes were higher for households with non-dependent children, reflecting higher proportions of employed persons in these households, but were lower for households comprising older couples and lone persons, where the numbers of employed persons were substantially lower.
People living in households where the reference person was aged 65 and over had the lowest mean incomes, with lone persons' incomes at $434 (Table 12) per week. This was lower than for older couple only households where the reference person was aged 65 and over and mean incomes were $558 per week. Older lone persons were more likely than older couples to have government pensions and allowances as their principal source of income (76% compared to 65%), while older couples were more likely to fully own their home (86% compared to 69%) (Table 12).
Households comprising one parent with dependent children had a mean income of $520 per week (Table 11), similar to that of older couples ($558 per week) (Table 12), but only 8% fully owned their home and therefore a substantially greater proportion were making mortgage or rental payments from their income. Of these households, 45% had government pensions and allowances as their principal source of income. On average there were 0.9 employed persons in the household (Table 11).
States and territories
There were differences in the average levels of income between the states and territories (see Table 16). However, not all the differences were large enough to be regarded as statistically significant at the 95% confidence level (see Appendix 5).
Tasmania's mean equivalised disposable household weekly income was 19% below the national average and South Australia was 8% below. In table 16, the Australian Capital Territory, the Northern Territory and Western Australia are shown to have the highest mean incomes (27%, 8%, and 6% above the national average respectively). The high income levels reflect in part the younger age profile of the ACT and NT and the greater number of employed persons per household. The results for the Northern Territory also reflects the exclusion from the results of households in collection districts in the NT defined as very remote which, if included, would be likely to reduce the mean income in that territory. This potential for an overestimated mean income in the NT is based on the large relative size of the very remote population for that territory.
New South Wales, with the largest State population, recorded a mean equivalised disposable household weekly income only 1% above the national average (Table 16).
There are also differences between the equivalised disposable household incomes recorded in capital cities compared to those earned elsewhere in Australia. At the national level, mean incomes in the capital cities were 25% above those in the balance of state (Table 14), with all states (separate information is not available for the ACT and NT) recording capital city mean incomes above those in the balance of state. The largest differences recorded were for NSW and Queensland where the capital city incomes were 37% and 26% respectively, above the mean incomes across the rest of the state (Tables 14 and 15) .
While the mean equivalised disposable household income of all households in Australia in 2007-08 was $811 per week, the median (i.e. the midpoint when all people are ranked in ascending order of income) was somewhat lower at $692 (shown as P50 in Table 1). This difference reflects the typically asymmetric distribution of income where a relatively small number of people have relatively very high household incomes, and a large number of people have relatively lower household incomes, as illustrated in the following frequency distribution graph.
Percentile ratios are one measure of the spread of incomes across the population. P90 (i.e. the income level dividing the bottom 90% of the population from the top 10%) and P10 (i.e. dividing the bottom 10% of the population from the rest) are shown on the above graph. In 2007-08, P90 was $1360 per week and P10 was $317 per week, giving a P90/P10 ratio of 4.30. Changes in these ratios can provide a picture of changing income distribution over time (Table 1).
Another measure of income distribution is provided by the income shares going to groups of people at different points in the income distribution. The following table (S6) shows that, in 2007-08, 10.1% of total equivalised disposable household income went to people in the 'low income' group (i.e those people with household income in the second and third deciles) with 40.5% going to the 'high income' group (i.e. the 20% of the population in the highest income quintile) (Table 1).
The Gini coefficient is a single statistic that lies between 0 and 1 and is a summary indicator of the degree of inequality, with values closer to 0 representing a lesser degree of inequality, and values closer to 1 representing greater inequality. For 2007-08, the Gini coefficient was 0.331.
Some of the change in some of the income distribution measures between 2005-06 and 2007-08 reflects the most recent improvements made with the 2007-08 cycle. See Appendix 4 for further information and analysis of the impacts. The estimates presented in tables 1-3 for 2003-04 and 2005-06 have been revised to be as comparable as possible with 2007-08.
For more information on analysing income distribution please refer to Appendix 1.
In this publication, the child care use and cost estimates are based on data collected from child care questions being asked of households in the survey where children 12 years of age or less were resident. Table 7, Child care, provides key child care information by specific household characteristics. The full list of data items is available as a part of the Confidentialised Unit Record File (CURF), due to be released concurrently with this publication (see explanatory notes paragraph 11 for details).
In SIH 2007-08, respondents were asked to report child care usage in the previous month. On this basis, the proportion of children using care may be smaller than a measure based on a usual (or regular) attendance basis due to temporary absences, and larger than the proportion attending in a shorter reference period (such as a school term week). The largest difference will reflect the numbers of school children who will attend vacation care but no other formal care during a school term. For example, the number of children aged 0-12 years using formal care in the last month was estimated in the 2007-08 SIH at 836,000 (24% of children of this age) whereas the 2008 ABS Childhood Education and Care Survey (CEaCS) estimated 756,000 children (22% of this age group) usually in formal care, and a further 137,000 children attending formal care in the previous school holidays for whom no usual formal care arrangements were reported.
For informal care, the 2007-08 SIH the estimate was 1,787,000 children aged 0-12 years using care in the previous month compared with a CEaCS estimate 1,008,000 children with usual informal care arrangements. The differences were smaller for informal care arrangements that are likely to be dominated with usual care patterns e.g., for care provided by a parent living elsewhere the estimate from SIH was 207,000 children compared with CEaCS estimate of 162,000 children usually receiving informal care from a parent living elsewhere. However, where much of the care may be provided on an ad hoc basis, such as other relative care (other than grandparent ), the SIH estimate was 513,000 children, compared with CEaCS estimate of 174,000.
The cost of care estimates in table 7 are shown, on a household basis, before the Child Care Benefit (or CCB which is shown separately) is deducted. For formal child care costs, the after CCB average weekly cost per child was $68, compared with $73 estimated from the CEaCS.
Of households with children aged 0-12 years, 67% (Table 7) used either formal or informal child care. Informal care was used by 58% of households with children of this age and 31% using formal care.
These documents will be presented in a new window.
This page last updated 29 August 2011