UBMISSION TO THE
ESEARCH ON INCOME AN
D ASSETS OF THE
Professor John McCallum
Dean, College of Social and Health Sciences,
University of Western Sydney
own NSW 2560
The connection between superannuation and other forms of retirement
income and living standards is a critical issue for an ageing society.
We are currently looking in detail at the impacts of health and fam
changes on the income and assets of the ‘old’ old in a study entitled
the Asset and Health Dynamics of the ‘Old’ Old (AHEAD).
The study is a collaboration between USA, Japan and other countries
which will allow observation of effects of different poli
The asset and health dynamics of the aged provides a guide for the
current generation of elderly and it allows us to model potential
futures of the babyboom generation.
More than half of our population 70+ received income from only the
and less than 10% received superannuation income,
while about 25% received some income from assets.
Around 9 out of 10 were able to manage on their income but this often
was with resignation and ‘out of necessity’.
Most older people have had major chang
es and crises in the last 10
years of their lives and these involve dissaving to manage the changes.
We would support the immediate development of a national study, of
a similar type as AHEAD, to monitor the asset and health dynamics of
the babyboom gener
ation over the next 10 years.
Population ageing is the product of lower fertility which increases the relative number of
older to younger persons [slowed in Australia by high post World War II immigration
and lower mortality which increases gradually the number of years that an
individual expects to live beyond “normal” retirement age. Consequently the Australian
population 70+ is increasing much faster than younger groups both in numbers and
the population, and the group aged 80+ is increasing the fastest of all
(McCallum 1999). The rate of growth of the ‘old’ old population is double peaked at
6 and 2026
31. So there is an immediate need for research on the behaviour of the
’ old. This can be an initial guide to the larger issue of the greying of the
babyboom and our long
term interest in modelling the future implications.
The debate on the costs of ageing was discussed in the OECD Country Report on
Australia (1998), the Pr
oductivity Commission Conference ‘The Policy Implications of
the Ageing of Australia’s Population’ in March 1999 and in the Ensuring Quality of Later
Life (EQOLL) model (McCallum et al 1998) which projected ‘aged care services’
excluding health costs using
age, sex and service specific historical data to create a model
term care insurance. In 1996 the OECD modelled health costs for ageing in 20
countries. In Scenario One it was assumed that as people grew older they consumed more
health care so cur
rent per person health care expenditure was multiplied by the total
number of older people. In this model health care costs in Australia grew from 5.8
percent of GDP in 1995 to 7.6 percent in 2030, assuming health care cost grew at the
same rate as GDP. In
Scenario Two costs of health care in the period prior to death were
modelled. Current expenditure per person was multiplied by the number of deaths in the
elderly population and health care costs were assumed to grow at the same rate as GDP.
Under this sc
enario costs rose to 6.2 percent of GDP in 2030. Australia and other
countries had higher projected costs under the first method compared to the second. In
the Dubbo longitudinal study, people 60+ who died over the first 5 years were 8 times
in health care use, mostly in the last 6 months of life, than those who
survived (McCallum and Geiselhart 1996). McCallum (2000) reviewed the ‘waxing and
waning’ debate about costs of ageing and concluded that, whatever side of the debate one
increases from ageing will need to be met from public and/or private sources.
This requires better analytical models to inform significant service reforms.
We identified a gap in Australian ageing research which we are now bridging. Various
reau of Statistics (ABS) national surveys contain data about the resources
and health of Australia’s ‘old’ old notably the health, disability and household
Limited links are possible between these surveys but it is not possible
detailed economic status information with health and ability to function for the
‘old’ old. Moreover ABS surveys are cross
sectional rather than longitudinal and the
important questions require data from the same people who are followed over time.
are a series of cross
sectional and longitudinal studies that have concentrated on
social and health aspects of older Australians but have limited economic information.
sectional studies were completed in the early 1980s: Ageing and the Family
Project (Kendig 1986)
which became a longitudinal study when the survivors were
up at 7 years (McCallum et al 1991); ACOTA/DCS Survey of Older People at
Home (ACOTA/DCS 1985); and the AIMA Community and Institutional Care for Aged
Migrants in A
ustralia (AIMA 1986). The studies focused on social support, health status
and aged care services but not on assets and income. Similarly the Australian
Longitudinal Study (Andrews and Myers in press) is focused on health and social
transitions but limited
in economic and asset areas, as is the aged cohort 70+ in Womens’
Health Australia’s Womens’ Longitudinal Study (Brown et al 1998) and the Victorian
Carers Program (Schofield et al 1998). The Dubbo Longitudinal Study in a population
born before 1930 (Simo
ns, McCallum et al 1990 & 1991), continuing since 1988, has a
strong health and health service focus along with socio
demographics but as yet only
basic economic information. Australian surveys do not have detailed questions on income
and assets nor do the
y allow exact links of these to life and health transitions and
international comparisons of these relationships. We have added these items to the
existing Dubbo cohort (Simons, McCallum et al 1990, 1991) which from baseline
measurement has been matched to
an existing Japanese survey.
We are conducting a longitudinal study of the health and economic behaviour of very old
people (AHEAD) which will fill a gap in Australian and intern
ational research. The study
addresses important behavioural questions related to work, saving, and health for the
first century. We have added a longitudinal study of economic behaviour and
ageing to a successful cohort in the Dubbo community and co
mpare results with
comparable studies in Japan and USA. This provides a database for modelling the costs
of support for the elderly through public programs such as Age Pensions, Medicare and
It has collected longitudinal data abo
ut resources of three types: economic, public
programs and families, and is constructing combined models of transitions in wealth,
resources and health status for the population 70+.
are those from
past and present employment, consumptio
n and saving decisions of individuals and
households including housing equity, superannuation and consumer durables.
cover individual claims on age and veterans’ pensions, health and aged care
(particularly Medicare and Pharmaceutical Benef
its funded services), and other
are the combined stocks of time, skills and
financial reserves held by the older person’s extended family. AHEAD has measured
current amounts and sources of income, assets, housing, and
employment; use of health
and aged care services, out
pocket expenses, health and long
term care insurance; and
age pension, primary health care and other program use through linkage to administrative
data (subject to privacy constraints), from the Heal
th Insurance Commission (HIC) and
Department of Health and Aged Care (DHAC), Department Family and Community
Services (DFaCS) and Department of Veterans’ Affairs (DVA).
AHEAD tests the general hypothesis that the course of age
related changes in health ca
be offset by three broad types of "resources": economic resources which result from past
employment, consumption, and savings; claims on structured programs such as Age
Pensions, Medicare, and publicly funded aged care programs; and family resources of
ime and money which can be made available to an older relative for caregiving or the
purchase of services. AHEAD provides data to address a broad range of scientific
questions focused on the interplay of resources and late life health transitions. These
clude: the costs of illness borne by the family; differences in how resources are used to
offset cognitive, physical, and functional losses; the effectiveness of various care
arrangements in preserving function and delaying institutionalisation; the extent
transfers from family buffer the assets of older persons and slow transitions to late life
impoverishment; and the pattern of income flows and the extent and mechanisms for
dissaving and ‘spend down’.
Australia is a partner in an international collaborative project. The Australian
study links with the national AHEAD
USA study now in its 7
wave and a Japanese
cohort study which paralleled the progress of the Dubbo Study, beginning in 1987 wit
2200 people 60+, and in November 1999, the 5
Wave, added the AHEAD questions to
its ongoing cohort. The original US study is proceeding with record linkage and is now
covering all people born before 1953. [Note: Our study covers all people born before
1930.] It has had links to Social Security for earnings and benefits records for 5 years and
it has links to Medicare/Medicaid. It will also connect to records on Employer Health
Insurance and Pension Plans. Americans generally only have their health insur
through their employer. They are considering looking at property tax records to get
values of properties. AHEAD
Australia has common questions with the US Health and
Retirement Study of persons born between 1931 and 1941 (the partner study to AHEAD)
hich is also in progress in Japan, Holland and Mexico. New collaborative studies are
developing in the English Longitudinal Study on Ageing (ELSA), Study of Health, Aging
& Retirement Europe (SHARE). Full survey instruments, study methods and publications
are disseminated on a web site supported by the US National Institute on Aging (NIA):
The US data indicate dissaving among ‘old’ old due to ‘spend down’ of a
ssets to gain
means tested access to the US Medicaid program. A comparison with Australia may
indicate the degree of dissaving due to the different eligibility criteria for access to
medical and aged care services. Japanese elderly differ from Australians
in having high
savings, high income relative to adult earnings, relatively high rates of employment, and
high rates of co
residence with family in 3 and 4 generation households. Japan is also
implementing public long
term care insurance. All these comparis
ons provide useful
information and models for Australian policy makers.
A major feature of this project is the extensive collaboration between the University of
Western Sydney, four major Universities (
UNSW, ANU, Michigan and Tokyo
) and four
public and pr
ivate industry stakeholders. The strategic importance of the project to the
partners indicates they will benefit from early access to survey data i.e. it is not just an
academic exercise. Briefly, the linkage of survey data with their administrative datase
will enhance the capabilities of the
Department of Health and Aged care
Department of Family and Community Services
to formulate effective policies for the
ageing Australian population; and
Southern Cross Homes
are committed to ensuring
cilities and services are responsive to the needs and expectations of older
A much debated policy risk of the new millennium is that Australia’s economic and
social advancement will be slowed by the need to devote i
sources to maintain
the well being of the retired elderly. The post
war Baby Boom has provided transitory
relief from these trade
off decisions by i
creasing the share of workers in the population
until about 2010 when the first of these large c
ohorts turns 65; thereafter the pace of
population aging will accele
ate, increasing the aggregate transfer burden on the working
tion. AHEAD addresses these macro
level policy questions by posing a large
number of micro
level scientific question
s which allow social and health scientists to
formulate and test theories of determinants and consequences of individual and family
behaviour over the latter part of the life cycle. The broad scope of inform
in AHEAD creates new opportunitie
s for interdisciplinary and multi
works to emerge as outcomes of the study. For one example, the most evocative
early finding from the US AHEAD study, is the strong positive correlation between
health and wealth (Smith, 1995, 1997) also e
vident in unpublished Dubbo data. Sorting
out the causal factors that underlie this correlation is one of the greatest challenges
Australian (Dixon 1999) and international researc
ers face, a task that has important
implications for po
icy and will require
insights from economics, biomed
psychology, sociology and demography (Smith and Kington, 1997).
No existing Australian study, public or private, provides the data necessary to know the
links between health, social relations and economic well
being. Consequently reforms in
health and nursing home funding have been developed in relative ignorance of behaviour
of older people. These practices are also of interest for older people themselves, service
providers, financial institutions and insuranc
e companies as well as for governments.
AHEAD can contribute to economic and social advancement by addressing urgent issues
in aged care policy: how much of the responsibility to fund care of the aged falls to the
individual and how much is shared by socie
ty as a whole; what circumstances cause
people to move from private savings and assets to public dependency; the actual practice
of the principle of “mutual obligation” and its variation across countries; the financial
arrangements needed to cover the “bab
y boomer” ageing and intergenerational equity
(OECD 1998); the viability of options such as continuing care retirement communities;
and the potential of different policy mixes across the countries in the AHEAD study
(currently Australia, Japan, USA).
: Dec 2000
Questionnaire design, ethics approvals & staff recruitment & training
Conduct Wave 1 interviews in Dubbo
stages 1 & 2 completed on time
Preparation of data and first cut data analysis
b. Nov/Dec 2002
Early results workshop in Perth
c. Jan 2002
Dissemination: data reports, peer reviewed publications, webs
: Dec 2002
Conduct interviews Wave 2
Preparation of data Wave 2
Wave 2 data analysis
The survey work for Wave 1 of the AHEAD Austral
ia Study has been completed. It was
conducted in two steps: the “in
Dubbo” surveys and the “out
Dubbo” surveys for
respondents who had moved.
Table 1: Survey and Follow
Brief telephone questionnaire
Could not be located
Could not be contacted
Prior nursing home classification
Moved into nursing home
Refused to participate
Too sick to be interviewed
Moved out of Dubbo
* Of these, 921 gave full consent, 30 gave partial
consent and 69 refused to give consent.
** Of these, 94 gave full consent, 1 gave partial consent and 2 refused to give consent.
The two year follow
up will begin in November 2002 to observe changes across the 2
Less than 10% received income
from superannuation while a quarter had some asset
incomes. Generally the group have a high dependency on the age pension:
55% of people received money only from the age pension
6.3% of people received some income from current superannuation and
people received some income from savings and assets.
2.2% of people received some income from their business(es)
Most people agree that they are able to manage on their current income.
“I would like to ask you about managing your money.”
manage my money so I do
not have problems living day to
Neither agree or disagree
I can decide on the amount of
money I spend over and above
daily living expenses
Neither agree or disagree
I am financially prepared for
d medical expenses
Neither agree or disagree
t桩he a牬y a汬⁰ 潰汥ee洠m漠扥扬攠瑯慮age渠瑨攠 ge⁰ 湳楯n
people who have income other than the pension are more likely to score ‘strongly agree’
1/3 of all people with pension plus other income compared to just above 1/10 for people
who have only the pension income. However about 90% of both groups
agree that they
are coping on their current income. This is quite often reported with resignation e.g. ‘you
just have to don’t you’. A few people did express considerable financial distress.
“We’re interested in some of the things that may have
happened to you since
1988. Have you …”
Were you worse off
financially because of it?
Did you need to spend
savings or assets?
a. moved house?
b. had a major health
c. had a spouse go into
d. had a major family
e. moved to live with
family move in
People spend down their ass
ets to cope with crises in old age not just in maintain levels
of consumption. Multiple generations of family benefit from these expenditures not just
the older individuals and their spouses.
“What do you think about children looking after their
Your answer does not need to reflect what would actually happen.”
Children should take care of their parent/s even if it is difficult for
Children should take care of their parent/s if they are able to
The parent/s should take care of themselves
“What would you prefer to happen if you were unable to care for yourself?
Your answer does not need to reflect what would actually happen.”
would like to be taken care of by my family in my own home
I would like my family to care for me in their home
I would like to be taken care of at home with outside help for
housework and personal care services
I would like to be
admitted to a hospital
I would like to reside in a nursing home/hostel
併l 晩f獴sc潭灡o楳潮i睩瑨w瑨攠ga灡湥獥 獴畤y 睩汬 扥 瑯tex灬潲p 摩晦e牥湣e猠潮o瑨敳t 污獴
The early results from the AHEAD study indicate that it can provide essential data for the
current generations of people 70+ which will allow us to model future experiences for
babyboomers. However the new data from NATSEM (Harding et al 2001) indica
the assets of the babyboomers will be larger than those in current generations. Hence it
would be desirable to survey the babyboom generation as it passes through to retirement
ages. This would allow real time monitoring of changes in behaviour fr
om the patterns
previously established in the AHEAD work. The experience of the USA studies of this
age cohort indicates that substantial resources are required to do the work. Hence there is
a need for both public and private funds from outside peer
wed granting bodies to
allow this work to proceed.
Andrews, G.R. & Myers,G.C.(in press). The Australian Longitudinal Study of Ageing: A
multidimensional prospective study of a 70 years & over population sample.
Clinical & Experimenta
Australian Council on the Ageing & Department of Community Services
Older People at
Home: Survey Results.
1985. Canberra: AGPS.
Australian Institute of Multicultural Affairs
Community & Institutional Care for Aged
Migrants in Australia.
Brown,W.J.,Bryson, L., Byles, J.E., Dobson, A.J., Lee, C., Mishra, G., & Schofield, M.
Women’s Health Australia: Recruitment for a National Longitudinal Cohort Study,
Women & Health,
A National R&D Collaboratio
n on Health & Socio
Economic Status for
1999. Canberra: NCEPH.
Harding, A., King, A., and Kelly, S.
Trends in the Assets and Incomes of Older
. Canberra: NATSEM November, 2001
Kendig, H.L, et al.
Health Status of Older People: Data
.. Melbourne: Victorian
Health Promotion Foundation, Lincoln Gerontology Centre, & National Ageing Research
Ageing & Families.
1986. Sydney: Allen & Unwin.
McCallum J, Botsman P, Clarke A, Pearse W, and Keogh S,
ality of Later
Life. Financing Aged Care Services in the Context of an Ageing Society
Centre for Health Outcomes and Innovations Research, South West Health Papers No. 2,
McCallum J. Australian Ageing: Policy Implications of
Growing Old in a Young Society.
Journal of Sociology & Social Welfare,
1999, 26: 87
Health in an Ageing Society: Cost Rhetoric versus Service Reform, In A.L.Bloom
(ed) Health Sector Reform in Australia & New Zealand.. Melbourne: Oxford University
Press, 2000 pp 325
Geiselhart, K. Australia’s New Aged. Issues for Young & Old. 1996. Sydney:
Allen & Unwin.
Ageing & Families 7 Years After: Data from a 7
up of Older
1991.Canberra: A joint NCEPH/Lincoln Ger
ontology Centre Monograph.
Economic Surveys: Australia
Schofield H., et al
Family Caregivers : Disability. Illness & Ageing
1998 Sydney: Allen
Simons, L. McCallum. J, et al., The Dubbo Study of the Health of the Elderly
Sociological & Cardiovascular Risk Factors at Entry,
Australian & New Zealand Journal
, 1991, 21: 701
___ et al, The Dubbo Study: An Australian Prospective Study of the Health of the
Australian & New Zealand Journal of Medicine
1990, 20, 783
Smith, J. P. Racial & Ethnic Diffe
ences in Wealth.
Journal of Human Resources
___Wealth Inequality among Older Americans.
The Journal of Gerontology Series
& R. Kington.. Demograph
ic & Economic Correlates of Health in Old Age.
, 1997, 34(1):159