Bulletin – March 2020 Australian Economy Demographic Trends, Household Finances and Spending

Photo: xavierarnau, filadendron and Thurtell – Getty Images

Abstract

The share of the population in their peak earning and spending years (ages 35–54) has decreased over the past decade, while the share aged 65 and above has increased. Demographic change has tended to reduce aggregate growth in household income and consumption, but by less than what previous patterns of household spending would suggest. This is because older households have earned and consumed more than in the past, and they have become wealthier. By contrast, growth in spending by younger households has been subdued, consistent with their weak income growth. The different earning and spending behaviour of households across different age groups will continue to affect trends in aggregate household consumption and income as the population ages further.

Background

The Australian population is getting older. Over the past decade, the share of the population aged 65 and above has risen sharply and the share of the population in the peak stage of their lives for earning income and consuming (those aged 35–54) has declined noticeably (Graph 1). The demographic shift to an older population has been shaped by the large ‘baby boomer’ generation (those born between 1946 and 1964), who have driven changes to the age composition of the Australian population for five decades. The baby boomer generation began reaching the retirement age in the past decade, significantly increasing the share of the population aged over 65.

Graph 1
Graph 1: Population Share by Age

Large inflows of relatively young migrants have supported population growth and increased the share of the population aged 25–34 over the 2010s (Graph 2).[1] Despite this, the population has continued to grow older on average because the large baby boomer generation have begun to move into the 65 and over category and because of the ongoing trend increase in life expectancy. By contrast, people aged 35–54 have made only a very modest contribution to growth in the population, unlike in prior decades when the baby boomers moved through this age group. As a result, the number of people at retirement age per 100 working-age people (those aged between 15 and 64) has risen from around 20 to 25 over the past decade and is expected to rise further over the next decade.

Graph 2
Graph 2: Changes in Population Age Segments

Alongside these demographic developments over the past decade, there have been noticeable changes in key economic indicators for the household sector. Disposable income grew at a reasonably strong pace over the 1990s and 2000s, but income growth has been subdued over the 2010s (Graph 3). Growth in household consumption has also slowed. The saving ratio increased noticeably starting from the mid 2000s, but has declined more recently. A range of structural and cyclical factors have contributed to these trends. These include: the global financial crisis; the mining boom and its unwinding; higher levels of household debt; weakness in non-labour income; growth in household income tax revenue; and the recent downturn in the housing market.[2] Considering the contribution demographics have made to these trends, alongside these other factors, can help us further understand developments to date and how they might evolve in the future.

Graph 3
Graph 3: Household Consumption and Income

This article uses an Australian Bureau of Statistics (ABS) dataset that combines household-level information from the Household Expenditure Survey (HES) and the Survey of Income and Housing (SIH) with the national accounts to explore what effect demographic changes are likely to have had on household consumption and income over recent decades.[3] In this dataset, the relevant household information, such as age, is grouped according to a designated ‘household reference person’.[4] This dataset is broadly representative of the Australian household sector and includes persons living in non-private dwellings (such as nursing homes) and persons in very remote communities, who are often out-of-scope in micro datasets.[5] Consistent with the changes in the aggregate population, the share of households with reference persons' aged 55 and above in this dataset has increased from around one-third in 2004 to nearly half in 2018. The descriptive statistics offered by these data do not allow demographic effects to be isolated from other variations between households that may have been correlated with age. Nonetheless, they allow an examination of the average differences between households of different ages and how those differences contribute to the aggregate household sector and economy.

Income and consumption tend to decline with age …

Some portion of the slowing in consumption growth over the past decade is likely to be due to the ageing of the population. This is because the distribution of average household consumption by age in Australia follows a hump-shaped pattern: spending generally increases through the working life of a household, rising noticeably for households aged 35–54 in the stage of their life where many households support children and then declines once they retire. The decline in consumption is particularly pronounced for households aged 65 and over (Graph 4).[6] The reasons for a reduction in average spending for the 65 and above category would vary by household, but some likely explanations include a reduction in income as people enter retirement (discussed further below), lower weekly expenses (such as transport and eating out) after exit from the workforce, no longer needing to support children and actual or perceived inadequate savings for retirement.[7] The decline in the consumption of older households is not as large when incorporating social transfers provided by the government (discussed further below).

Graph 4
Graph 4: Real Household Consumption

The ageing of the population is also likely to have made some contribution to the slowing in income growth over the past decade. Similar to the distribution of consumption by age, household income generally increases through the working life of a household and then declines as they approach retirement (Graph 5). Households aged 65 and above typically have lower levels of income than those aged 25–54, and around 40 per cent of households 65 and over are in the lowest income quintile.[8] However, these older households are typically wealthier; only 25 per cent are also in the lowest two wealth quintiles.

Graph 5
Graph 5: Real Household Income

However, stronger income growth over the past decade has supported higher consumption growth for older households

The impact of the growing number of older households on aggregate consumption has been partly mitigated by a shift in the shape of the lifetime consumption distribution over the past decade. Average consumption per household aged 55 years and above has grown nearly twice as fast as the average of other households (Graph 6). In 2017/18, households aged 55–64 consumed more on average ($123,000) than those aged 35–44 ($111,000), although less than those aged 45–55 ($128,000). Stronger consumption growth for older households has partly reflected stronger growth in their incomes, compared with other age groups. Changes in household composition (such as children staying at home for longer or increased life expectancy) may be another contributing factor to consumption growth per household.

By contrast, the youngest households (those aged 24 and below) on average saw no growth at all in their consumption and a decline in their real income in the six years following the global financial crisis. The smaller share of the population comprised of these households, relative to prior decades, means that the effect on aggregate consumption has been smaller than it otherwise would have been.

Graph 6
Graph 6: Real Household Consumption

Aggregate growth in household disposable income has slowed noticeably over the past decade and has been lower than consumption growth for most of the past five years. But, as noted above, income growth across the age distribution has varied. Income growth was stronger for older households than for younger ones (Graph 7).[9] For example, households aged 15–24 had more disposable income than households aged 65 and over in 2003/04, but in 2017/18 earned about the same, and they still had significantly less wealth. As in the case of consumption, stronger growth in older households' incomes has reduced the effect of the ageing population on aggregate income.

For households aged 55–64 years, a key driver of stronger growth in overall income relative to other households has been strong growth in labour income. This has been supported by increased participation in the labour force, mainly by women. The female participation rate for those aged 55–64 years has increased from around 30 per cent in 1999 to a little more than 60 per cent in 2020.[10] For households aged 65 and above, growth in income over the past 15 years has also been supported by strong growth in non-labour sources of income such as financial income, rental income and social assistance income.[11]

Graph 7
Graph 7: Household Real Income Growth by Age

Financial income has grown more strongly than any other source of household income over the past 15 years. All age groups have benefited, but older households benefited the most because they hold more financial wealth, on average. Total financial returns for households over this period would be even larger if capital gains were included, but financial income in the national accounts excludes capital gains. This understates the total financial returns of households, particularly for older households because they receive the largest share of their income from financial wealth.

Older households save less of their current income than younger households

Households across most age groups increased their rate of saving in the mid 2000s, likely driven by precautionary motives, lower expectations for future income growth and declines in wealth.[12] Over the following six years, households aged 35–44 years increased their rate of saving further while the rate of saving for older and younger households was relatively unchanged (Graph 8). While older households generally save less than younger households, older households still had positive savings over the past 15 years, on average.

Graph 8
Graph 8: Household Saving Ratio

Since 2015/16, the aggregate saving rate in Australia has declined, as disposable income growth has been weaker than consumption growth. While distributional data on saving are not available for the past couple of years, historical experience suggests that demographics are likely to have contributed in some part to the further decline in the saving rate since 2016, as the share of older households, who save less, has increased. The relative increase in the saving rates of younger households over this time has mitigated this effect on the aggregate saving rate. A simple scenario that uses 2015/16 saving per household and population shares from 2003/04 suggests that in the absence of changes in demographics over this time, the saving rate would have been 1 percentage point higher in 2015/16. As the population continues to age this may weigh further on the saving rate.

Superannuation has also supported consumption by older households

Superannuation has played an important role in households' choices for smoothing consumption as they approach and enter retirement, giving them the option of drawing down their superannuation to fund spending above their income. The drawdown of super has supported older households to consume more on average (Graph 9).

Graph 9
Graph 9: Household Finances

Growth in household consumption has been supported by strong growth in asset prices, particularly for older households

Past research has identified a relationship between household wealth and consumption.[13] Net wealth has increased for all age groups, although the largest gains in dollar terms have accrued to older households (Graph 10). The average Australian household's wealth – under the definitions in the national accounts – increased in nominal terms from around $500,000 in 2004 to close to $1.1 million in 2015/16. The average wealth of households aged 15–34 increased by around $90,000 over this period, while for households aged 55 and above it increased by $630,000. Older households have accumulated considerably more wealth than households of the same age in the past, consistent with the increase in their consumption.[14]

Graph 10
Graph 10: Net Wealth by Age

Housing wealth increased strongly from 2003/04 to 2017/18, but debt owed by households grew even more strongly. While households aged 65 and above hold the least debt on average, these households (and those aged 55–64) have also seen a trend increase in the average housing debt per household relative to households of their age in the past, meaning that older households are now approaching or in retirement with more debt, on average (Graph 11).

Graph 11
Graph 11: Real Household Debt

The growth in housing wealth and debt in part reflects increased ownership of investment properties by older households. For older households, housing debt is roughly evenly split between owner-occupied and other properties, while for households aged 54 and below housing debt is largely for the property they live in. Data from the Australian Taxation Office indicate that increased ownership of investment properties over the past two decades has been driven by those aged 50 and above (Graph 12).

Graph 12
Graph 12: Residential Rental Properties

Social welfare has also supported consumption by older households

Households across all age groups are supported by sizeable social transfers from the state. Consideration of these public transfers gives a more complete picture of the set of resources available to households and helps explain the relatively resilient private consumption of older households because private income and consumption has been supplemented by support from the state.

Social assistance income provided to households aged 65 and above has increased around 30 per cent in real terms over the period 2003/04 to 2017/18 (Graph 13). Pension income has grown in excess of both the consumer price index and the wage price index since 2003, partly reflecting a number of policy changes.[15] Social assistance income declined a little in 2017/18 for older households, on average. This seems to reflect, at least in part, a larger share of part pensions.

Growth in nominal social assistance income has been subdued for all other households since 2003/04; in real terms, it has declined a little. The typical household aged 64 and below receives no social assistance income from the state. While unemployment benefits did increase a little towards the end of the mining boom, these only account for 15 per cent of social assistance.

Graph 13
Graph 13: Social Assistance Income by Age

Once other transfers, such as child care and education benefits (for example, subsidies for education), are included, the social welfare benefits are a little more evenly distributed across age groups in nominal dollar terms (Graph 14). These ‘transfers in kind’ also include aged care and benefits received through the National Disability Insurance Scheme. Social transfers in kind are captured by measures of government spending and are not included in household consumption growth. Total spending on these transfers has grown significantly over the past 15 years, which has been an important driver of growth in public consumption and economic activity.

Graph 14
Graph 14: Social Welfare Benefits by Age

Conclusions and considerations for the outlook

Australia, as in many countries, is experiencing large demographic shifts. Some portion of the slowing in aggregate consumption and household disposable income growth over the past decade is likely because of demographic changes as more households have moved into a stage of their lives where they earnt and spent less, on average. These effects have been smaller than what previous patterns of household spending would suggest because older households are spending more than in the past. This expenditure has been supported by relatively strong growth in income, large increases in wealth and withdrawals from superannuation.

Over the coming decade, a further strong increase in the share of households aged 65 and above is expected. Further impacts on consumption and income are likely, although these are likely to happen over a number of years. The increase in young overseas migrants over the past decade should support the share of the population that are of working age over the coming decade. This has made Australia relatively well placed, compared with many other advanced economies, to adjust to the effects of an ageing population.

Footnotes

The authors are from Economic Analysis Department. Sincere thanks are extended to Adam Sinclair who provided substantive analysis and drafting in the earlier stages of this work. The authors would also like to thank Iris Day, Fiona Price, Diego May and Peter Tulip for their helpful suggestions. [*]

Australia has experienced strong net overseas migration relative to most other advanced economies, most of which has been accounted for by people under the age of 35. For more information on net migration in Australia and Australia's demographic position relative to other advanced economies see Lowe (2018) Demographic Change and Recent Monetary Policy. [1]

For further discussion see Lowe (2011), Stevens (2011) and Ellis (2019). [2]

Adjustments have been made to this dataset for 2017/18 for select components to incorporate information from the latest Survey of Income and Housing, which was not available when the dataset was compiled. However, these adjustments are less sophisticated than the original methodology: the ABS is expected to publish an update later this year, which will include data from the latest survey. [3]

The household reference person is selected by the ABS based on a range of factors including home ownership status, income and age of the individuals within a household. [4]

The household sector in this data includes unincorporated enterprises and excludes the not-for-profit sector. Throughout the article, disposable income refers income after tax and interest payments. [5]

Even once changes in the composition of households, such as children leaving home, are accounted for, consumption for the average household still declines for households 65 and above. [6]

Increased average life spans mean that the 65 and above category now includes more households of a significantly older age, making it difficult to infer from these data how the decline in consumption for those aged 65 and above may vary as these households age further. [7]

Based on more detailed data from the Survey of Income and Housing. The shares of households aged 65 and above in the lowest income quintile are 38 per cent for equivalised disposable income and 44 per cent for unequivalised in 2017/18. [8]

Income growth has been stronger for older households even after controlling for changes in household composition by using equivalised disposable income from the survey of income and housing. [9]

For further discussion see Debelle (2019). [10]

Financial income includes interest receipts, dividends from financial institutions, private corporations and governments and a measure of estimated interest and dividend earnings on households' superannuation balances in each quarter. [11]

For further discussion see Price and Finlay (2014). [12]

May G, G Nodari and D Rees (2019). [13]

For further discussion on this topic see Wood D and K Griffiths (2019). [14]

A number of policy changes have boosted pension income, including the 10 per cent one-off increase to the pension rate for singles in 2009 and changes to eligibility; for example, the introduction of the work bonus, which excludes a certain amount of income from the pension income test. [15]

References

Debelle G (2019), ‘Employment and Wages’, Speech at the Australian Council of Social Service (ACOSS) National Conference 2019, Canberra, 26 November.

Ellis L (2019), ‘What's Up (and Down) With Households?’, Speech at the Housing Industry Association March Industry Outlook Breakfast, Sydney, 26 March.

Lowe P (2011), ‘Changing Patterns in Household Saving and Spending’, Speech at the Australian Economic Forum 2011, Sydney, 22 September.

Lowe P (2018), ‘Demographic Change and Recent Monetary Policy’, Address to Anika Foundation Luncheon, Sydney, 8 August.

May G, G Nodari and D Rees (2019), ‘Wealth and Consumption’, RBA Bulletin, March, viewed 11 March 2019 Available at <https://www.rba.gov.au/publications/bulletin/2019/mar/wealth-and-consumption.html>.

Price F and R Finlay (2014), ‘Household Saving in Australia’, RBA Research Discussion Paper No 2014-03.

Stevens G (2011), ‘The Cautious Consumer’, Address to The Anika Foundation Luncheon, Sydney, 26 July.

Wood D and K Griffiths (2019), ‘Generation gap: ensuring a fair go for younger Australians’, Grattan Institute, Grattan.edu.au site, 18 August, viewed 11 March 2020. Available at <https://grattan.edu.au/report/generation-gap/>.