AUSTRALIA S AGEING POPULATION WHAT ARE THE KEY ISSUES AND THE AVAILABLE METHODS OF ANALYSIS? Agnes Walker

Transcription

1 AUSTRALIA S AGEING POPULATION WHAT ARE THE KEY ISSUES AND THE AVAILABLE METHODS OF ANALYSIS? Agnes Walker Discussion Paper No. 27 February 1998

2 National Centre for Social and Economic Modelling Faculty of Management University of Canberra The National Centre for Social and Economic Modelling was established on 1 January 1993, following a contract between the University of Canberra and the then federal Department of Health, Housing, Local Government and Community Services (now Health and Family Services). NATSEM aims to enhance social and economic policy debate and analysis by developing high quality models, applying them in relevant research and supplying consultancy services. NATSEM s key area of expertise lies in developing and using microdata and microsimulation models for a range of purposes, including analysing the distributional impact of social and economic policy. The NATSEM models are usually based on individual records of real (but unidentifiable) Australians. This base produces great flexibility, as results can be derived for small subgroups of the population or for all of Australia. NATSEM ensures that the results of its work are made widely available by publishing details of its products and research findings. Its technical and discussion papers are produced by NATSEM s research staff or visitors to the centre, are the product of collaborative efforts with other organisations and individuals, or arise from commissioned research (such as conferences). Discussion papers present preliminary research findings and are only lightly refereed. Its policy papers are designed to provide rapid input to current policy debates and are not externally refereed. It must be emphasised that NATSEM does not have views on policy and that all opinions are the authors own. Director: Ann Harding

3 National Centre for Social and Economic Modelling Faculty of Management University of Canberra AUSTRALIA S AGEING POPULATION WHAT ARE THE KEY ISSUES AND THE AVAILABLE METHODS OF ANALYSIS? Agnes Walker Discussion Paper No. 27 February 1998

4 ISSN ISBN NATSEM, University of Canberra 1998 National Centre for Social and Economic Modelling University of Canberra ACT 2601 Australia 170 Haydon Drive Bruce ACT 2617 Phone: Fax: Client services hotline@natsem.canberra.edu.au General natsem@natsem.canberra.edu.au Core funding for NATSEM is provided by the University of Canberra and. the federal departments of Social Security, Health and Family Services, and Employment, Education, Training and Youth Affairs.

5 iii Abstract Australia s population will age significantly from around 2010 onwards. While we are able to draw on the experiences of other developed countries where population ageing commenced earlier, recent trends suggest that the economic and social conditions underpinning Australia s ageing processes will be different in many respects. Drawing on the extensive Australian and international literature on ageing, this paper first identifies key economic and social factors that are likely to have a strong impact on age related public expenditure in Australia. These include likely changes in life expectancies and the health status of the elderly; social attitudes and government regulations regarding retirement; wealth accumulation and wealth transfers across generations; caring for the aged by family, governments or voluntary workers; and the extent to which economic growth and the consequent increases in tax revenue can be relied on in the future to finance the additional public costs associated with ageing. The paper then reviews the tools available to analyse these key factors, either individually or simultaneously. Traditional as well as emerging analytical methods such as microsimulation are considered. The changes in data requirements arising from greater flexibility in the labour market (for example, part-time or contract work) and the loosening of family ties (for example, the availability of carers) are also mentioned. The paper concludes with a discussion of how the static and dynamic microsimulation models developed at NATSEM could provide additional insights when studying the effects of the ageing of Australia s population.

6 iv Author note Agnes Walker is a Research Fellow at the National Centre for Social and Economic Modelling. Acknowledgments An earlier version of this paper was presented at the Foundation for International Studies on Social Security (FISS) Conference: Issues in Social Security, Sigtuna, Sweden, June The author is grateful to Maren Child for her valuable assistance with library searches. Glossary ABS DSS NATSEM OECD EPAC Australian Bureau of Statistics Department of Social Security, Australia National Centre for Social and Economic Modelling, University of Canberra, Australia Organisation for Economic Co-operation and Development Economic Planning Advisory Council

7 v Contents Abstract Author note Acknowledgments... iii...iv...iv 1 Introduction Population ageing Australia Experiences of other countries Limits to what Australia can learn from others Key issues Life expectancy, health status and health costs Age of retirement Saving for old age Caring for the aged family, governments, voluntary workers Key factors affecting age related public expenditure Life cycle issues Methods of analysis and data requirements Traditional analyses Microsimulation modelling Data requirements for additional analyses Summary References Other reading... 49

8 1 1 Introduction Over the next few decades living beyond the age of 80 years is expected to become the norm. 1 While the good news of long life is generally welcome, it also brings with it uncertainties about health, financial dependence, usefulness to the community, and remaining part of a family. In many countries, including Australia, planning for an ageing population is made more difficult by chronic budgetary pressures as well as by community concern about slow economic growth, high levels of unemployment and a growing polarisation of income (OECD 1996a). While population ageing encompasses many social, personal and economic issues, public policy is expected to continue to focus on the additional expenditure that future governments are likely to face. For this reason, this paper focuses on those factors likely to have a significant impact on age related public expenditure, and on the analytical methods most likely to lead to policy relevant conclusions. While most previous studies tended to focus on one or two aspects of ageing, what is of greatest relevance to policy is the combined effect of many factors. This paper considers: individual ageing (the trend towards increased life expectancy); population ageing (the increased proportion of older age groups in the population); private savings and their use to finance aged care; work and lifestyle patterns that influence personal skill levels, income, health and eventually the need for care in old age; economic growth and the extent to which governments will be able to finance age related public costs; and life cycle effects, recognising that what happens at one stage in life often has a large impact on later stages 2. 1 The ABS predicts that by 2051 the average newborn Australian male can expect to live for 81 years and the average female for 86 years. People aged over 65 currently account for around 12 per cent of Australia s population and are projected to make up close to 23 per cent of the population by 2051 (ABS 1996). 2 A comprehensive account of recent research in medicine, biology and psychology with explanations as to why experiences early in life have a strong influence on patterns in later life can be found in Goleman (1996).

9 2 Discussion Paper No Population ageing 2.1 Australia United Nations projections show that, although trends similar to Australia s are projected for many countries, people aged 65 years and over in Australia account for a smaller proportion of the total population than in most other developed countries. By 2050, Australia is expected to have about 23 per cent of its population aged over 65, with Japan and Germany having close to 30 per cent in this age group (figure 1). 3 Figure 1 Projected proportion of population aged 65 years and over, selected countries Proportion of population (%) Australia UK Germany USA Japan China Source: For Australia, ABS (1996, p. 37) with series A/B projections that is, high fertility and low overseas migration. For other countries, United Nations projections obtained from the ABS in a format comparable with Australia s. ABS projections for Australia suggest that the highest annual rates of increase in the size of the over 65 population will occur between 2010 and 2020, when the peak of the post-world War II baby boom generation enters this age group. The ABS (1996) projections also suggest that, by 2050, the proportion of the population aged 85 years and over will have risen from 1 to 5 per cent. 3 Developing countries, such as Indonesia, Korea, China and Malaysia, are expected to experience an even more rapid ageing of their populations, with at least a tripling of the proportion aged 65 and over (ABS 1996).

10 Australia s Ageing Population 3 In a recent paper (Walker 1997) key aspects of the literature on population ageing were reviewed, and policy responses in Australia and in some other developed countries were reported. In that paper it was noted that the impact of population ageing has generally led to policy initiatives aimed at assigning greater responsibility to individuals for the costs associated with their ageing. It was also noted that in Australia there has been some discussion of the extent to which ageing may become more of a policy issue in the future than it is at present. In this respect some believe that the existing social arrangements will become unsustainable as the population ages. Others note that many European countries with older populations than in Australia have managed without a crisis. Yet others (OECD 1996e, p. 14) are of the view that the reason why ageing has not had disastrous consequences to date is that the supply of labour has expanded considerably as more women have entered the workforce. However, since in each country there are limits as to the number of women who are able to join the workforce, future governments may find it impossible to increase social spending at the same rate as in the past. That population ageing will place increased pressures on government budgets does not appear to be questioned. It is the likely extent of these pressures whether of crisis proportions, or not that is being debated. It is in this area that comprehensive quantitative analyses (referred to in chapter 4) could contribute to the debate by highlighting the likely implications of various policy options under different demographic, economic and social scenarios. Some have argued that in a rapidly changing environment it does not make sense to project decades ahead (see, for example, McCallum and Geiselhart 1996, p. viii). However, while there are uncertainties about future economic and social conditions, there is general agreement that from 2010 onwards Australia s population will age significantly more rapidly. In essence, the baby boomers are alive and well now, and unless there is a major war or a natural catastrophe, a high proportion of Australians currently alive will reach age 65 early in the next century. In view of the relative certainty of the demographic situation, successive governments have introduced major policy changes regarding older Australians. The question for the future is whether further policy changes will be needed and whether their effect will be felt rapidly (as in

11 4 Discussion Paper No. 27 the case of greater targeting of the age pension) or slowly (as in the case of the Superannuation Guarantee Charge 4 ). The time required for policies to take effect will no doubt influence whether action is taken now or later. In this respect the OECD (1996e, pp ) notes that: It is hard to build consensus for changes, even if the change takes effect a decade or more in the future Long lead times and advance warning are needed to allow people to plan and adjust 2.2 Experiences of other countries In most developed countries age related pensions, health and long term care costs are partly publicly and partly privately funded. At one extreme are the Scandinavian countries, where universal basic pensions, health care and welfare services tend to be the norm. In these countries such services are usually provided by the public sector and are financed through general taxation. At the other extreme is the United States, where a high proportion of services are provided by the private sector and most individuals are expected to provide for the costs of their own health and old age care. However, there is public financing for the needy. For example, the nursing home costs of low income elderly US citizens are publicly financed through Medicaid (OECD 1996c, p. 40). The benefit of the Scandinavian model is that it provides all people with a relatively generous minimum living standard. As a result, in these countries poverty in old age is now a marginal phenomenon (OECD 1996c, p. 251). The disadvantage is that it places a heavy burden on the public purse. It thus relies on Scandinavians accepting very high levels of taxation. With tighter budgets and higher unemployment in recent years Scandinavians, as well as people in other countries, have questioned the future affordability of existing levels of support (see OECD 1996c, p. 38 in relation to the Nordic countries and the Netherlands, and p. 165 in relation to Sweden). 4 Essentially, through the Superannuation Guarantee Charge (SGC) the government ensures that individuals savings are not accessed before the ages 55 or 65 years. In exchange, the SGC scheme offers tax concessions and is partly funded by employers.

12 Australia s Ageing Population 5 The benefit of the US model is that the demand for health and aged care services is generally market driven, since most individuals are responsible for the costs of the services they consume. Demand is thus rationed by what people can afford. The disadvantages are that the gap between the rich and the poor is widened, that minimum living standards are lower than in other developed countries and that a relatively high proportion of people live in poverty, with resulting high levels of crime, etc. In the United States the private provision of services has led to high health expenditure, which currently accounts for around 14 per cent of GDP. Also, in the United States aged care is unavoidably linked to the broader health reform effort, which aims to expand medical cost coverage for the uninsured while controlling the rise of health expenditure (OECD 1996c, p. 214). Overall, many developed countries are questioning the affordability of existing systems, seeking alternatives that are able to contain costs without affecting the quality of care. Responding to these concerns is seen by the OECD (1996d) as a critical policy challenge. While asking individuals to bear a greater share of their age related costs seems an attractive solution, the experience of the United States suggests that there could be significant disadvantages in moving too far in the private provision direction. The general view is that there is no correct balance in public versus private financing. For each country the final compromise will depend on what its people accept as being both socially desirable and financially responsible. The following describes those experiences of individual countries that are of relevance to identifying the key factors worth including in future studies of population ageing in Australia. Retirement policies Early retirement In the early 1970s, rising unemployment and massive job cuts led the governments of many OECD countries to set up attractive early retirement schemes (see Okba 1993). For example, in the United Kingdom there was the Job Release Program, which offered workers a benefit if they left the workforce before pension age, plus a subsidy for

13 6 Discussion Paper No. 27 their employers if they took on younger workers in their place (Encel 1996, p. 115). Later retirement The World Bank (1994) noted that using early retirement as a short term policy measure to reduce unemployment could prove extremely costly in the long run, in that: the cost of pensions and benefits would rise considerably as more and more people lived longer and longer; and there would be hidden economic costs due to the loss of many people with valuable skills. A similar point was made by the International Labour Office (1995), concluding that the very concept of retirement was obsolete and needed fundamental rethinking. The need for greater flexibility in job sharing and a spread of employment opportunities over the life span were stressed. An example of a recent move towards greater flexibility is Sweden s increase in the age pension age from 65 to 67 years. Other changes announced (to be phased in over 20 years) included actuarially lower pensions for those retiring early (from age 61 years) and higher pensions for those remaining at work until age 67 (OECD 1996c, p. 175). Institutional versus community care Since the early 1960s many countries have put in place policies that allow the elderly to stay in their own homes for as long as possible. The OECD (1996c, pp. 61 2) grouped countries into those where: no more than 5 per cent of people aged 65 years or over received a home help service (nine countries, including Canada, the United States, Germany and Japan); between 5 and 10 per cent received a home help service (five countries Australia, the United Kingdom, France, the Netherlands and Belgium); and more than 10 per cent received a home help service (four countries Sweden, Denmark, Norway and Finland).

14 Australia s Ageing Population 7 When considering both institutional and home based care, the OECD (1996c, p. 61) found Australia to be in the middle, providing an average level of service on both dimensions. Contrary to expectations, the proportion of elderly living in institutions has not declined in most countries with policies favouring home based care. In a study of eight countries with such policies since the 1960s Canada, the United Kingdom, the United States, New Zealand, Sweden, Denmark, Finland and Norway the OECD (1996c, pp. 56 7) found that only in three of the eight countries did the proportion of elderly living in institutions decline and this was mainly in the younger years age group. The OECD concluded that, for people aged 80 years and over, the development of home based care failed to keep pace with the growth in their numbers. It also noted that, given the sheer scale of this growth, improving the quality of institutions seemed a more realistic policy goal for the very elderly than home based care. As seen in section 2.1, ABS projections for Australia suggest that there could be a fivefold increase in the proportion of the population aged 85 years and over by Overall, while community attitudes made it possible to expand home help services in some countries for example, Australia, the United Kingdom, the Netherlands and Scandinavia the available evidence suggests that such services developed more in parallel rather than as a substitute for institutional care. In Australia s case there appears to have been too rapid change in residential care and inadequate progress in community care (OECD 1996c, pp. 69, 229). Caring for the very old Unpaid carers In most developed countries caring for the elderly has traditionally depended on a substantial input of unpaid, informal care largely provided by family members, particularly women in the age group (OECD 1996c, p. 19). While there are well-documented costs to informal care in terms of time, stress and ultimately the carers own health, the OECD (1996c, p. 63) notes that the efforts of family carers tend to be seen as a free good. It also notes that in most member countries it was only gradually and with

15 8 Discussion Paper No. 27 some reluctance that governments accepted a greater share of the task of caring for the elderly (OECD 1996c, p. 33). While there is general recognition that informal carers need support if they are not to relinquish their role, studies to date have rarely considered this issue in any depth (OECD 1996c, pp. 76, 89). Paid carers In most countries carers employed as primary carers account for a small proportion of all carers. The exception is in Denmark where they make up just over 40 per cent of primary carers (OECD 1996c, p. 63). In Sweden, people who give up their job to care for relatives may receive from the government a salary equivalent to that of home helpers. The vast majority taking up this option are women aged years, many combining care giving with other part-time employment. However, as home help services expanded in Sweden between 1970 and 1990, the number of paid caregivers declined dramatically by close to 70 per cent (OECD 1996c, p. 173). The potential for further expansion of publicly provided home help services is considered limited, as female employment already stands at close to 80 per cent. The OECD (1996c, p. 175) notes that in future more of these functions will have to fall on volunteers or family members, with greater emphasis on support being provided to such people. Financing long term care In most developed countries long term care tends to be mainly publicly funded. Although data on the share of public versus private financing is hard to come by, for the United Kingdom and the United States it has been estimated that around half of nursing home costs is met by private payments. The requirement of private contributions in several other countries, such as France, Germany and Japan, has also resulted in significant levels of private funding. In some countries for example, Germany and the United States the growing burden of these costs on middle class families has led to requests for extending public funding (see OECD 1996c, p. 41). There have been expectations that the increased availability of home care services would lead to lower public outlays on long term care. However,

16 Australia s Ageing Population 9 studies in Australia, the United Kingdom, the United States, Denmark and the Netherlands suggest that, while community care options can enable the elderly to stay in their homes longer, it is doubtful that such options lead to lower public outlays than the alternatives would. Such studies, however, do not take account of the costs borne by informal carers (OECD 1996c, pp. 74, 76). In a number of countries there has been considerable interest in long term private care insurance as a way of limiting public outlays while ensuring coverage against the substantial risk posed by long term care costs. The focus has been on the United States, as the only OECD country with sufficient experience with this type of insurance product. An initial review of private care insurance in the United States revealed that there have been considerable problems. For example, on the demand side such insurance was being taken up by only about 3 per cent of the elderly; on the product design side those who were forced to quit the scheme due to unforeseen circumstances have, in practice, ended up subsidising those who were able to remain in the scheme (OECD 1996c, pp. 41 2). In a number of countries products are available which allow the elderly to access their housing equity in order to finance long term care. However, so far such housing finance initiatives have been taken up by few people and have had very little impact on public expenditure (OECD 1996c, p. 41). France has also experimented with part recovery from inheritances exceeding Fr of its dependency allowances (OECD 1996c, p. 140). Glennerster (1996) reports on further experiences in selected countries. The OECD (1996c, p. 42) concluded from these experiments that in the future public sources were likely to continue to bear the major part of the risk of long term aged care. Partnership In several countries, including Australia, Canada, the United States, France and Japan, the role to be played by services is being redefined. In general there is lesser emphasis on services as an alternative to absent family care and greater emphasis on seeking new forms of partnership between families and services provided by government or nonprofit organisations.

17 10 Discussion Paper No Limits to what Australia can learn from others Since population ageing in most developed countries commenced earlier than in Australia, this country has the advantage of being able to draw on the experiences of other countries. However, because no single solution has emerged out of past experiences and because most countries are currently reviewing their policies so as to make them more relevant to changing economic, social and demographic circumstances, the extent to which Australia can learn from others is likely to be below expectations. The greatest relevance of other countries experiences is probably that they highlight the practical advantages and disadvantages of policy options that have been tried. Even if one country s policies seemed especially relevant, there are a number of reasons why they may not successfully transplant to Australia. First, there is the generally accepted point that each country has unique social security, health and retirement systems, and that changes effective in (or acceptable to) one nation may be ineffective in (or unacceptable to) another. Second, many of the documented experiences of other countries took place some years ago, with circumstances being significantly different today both in those countries and in Australia. For example, while many earlier experiences took place in a more traditional environment, where education was mainly for the young, full-time jobs were the norm for men, and most women either stayed at home or worked in casual, low paid jobs, the current situation in Australia is different. Traditional family ties no longer hold to the extent that they did. Fewer men work in full-time career-based jobs, and many retire before the age of 65 years. More women are well-educated and work until retirement, many preferring flexible working arrangements. Work arrangements have been and are still changing rapidly. The number of part-time, contract and casual jobs has grown rapidly while the number of full-time jobs has declined. Currently, Australia has one of the highest proportions of part-time workers in the Western world (around a quarter of the workforce). Some three-

18 Australia s Ageing Population 11 quarters of these want to work part-time. Also, working from home is becoming more common, especially for older people. 5 There is an emerging expectation among older people that something better than a life filled with inactivity and/or doctor visits may be possible. Third, as in other countries the situation in Australia is changing rapidly, so policy making requires vision and experimentation. Overall, consideration of Australia s rapidly changing economic, demographic and social structures, in conjunction with the experiences of other countries, may help policy makers to better focus on those government initiatives that are most likely to deliver the desired outcomes. 3 Key issues The ageing of Australia s population will inevitably bring with it profound changes in the way we live and in the traditional social structures. As already noted, the extent to which other countries experiences may help us adapt to these changes is likely to be limited. For this reason, rigorous and state-of-the-art evaluations of both traditional and novel policy alternatives could be particularly useful to the policy development process. Previous studies provide some indication as to which economic and social factors are particularly important in analyses of ageing. For example, Rothman (1996, pp. 5 6) notes that, in relation to pension costs, demographic factors (such as longevity), retirement rates and the assets of the retired are particularly important. He also notes that, in relation to projecting superannuation assets and national savings, economic factors, especially the rate of return on invested funds, dominate. In addition, Walker (1997) reported on Australian and international work that showed the crucial importance of economic growth to a country s ability to accommodate an ageing population. 5 Between 1989 and 1995 the number of Australians who worked at home grew by 29 per cent, with those aged 55 years and over accounting for 18 per cent of all home workers compared with only 9 per cent of all employed persons (see Hall et al. 1997).

19 12 Discussion Paper No. 27 In Australia policy makers need to consider not only those factors that have been shown to be important, but also those that are likely to become important. The key issues that seem important in the context of analyses of population ageing when looking ahead some years are now considered. 3.1 Life expectancy, health status and health costs Life expectancy Since the beginning of the twentieth century, life expectancy at birth has increased by some 20 years. While the average person born early this century could look forward to living until age 55 (for men) and until age 59 (for women), a person born in 1996 can expect to live years. For those born half a century from now life is likely to be some five years longer (table 1). 6 Table 1 Past and projected life expectancies at birth years years years Male Female Sources: Abraham, d Espaignet and Stevenson (1995); ABS (1996). A key reason for longer lives has been the increased scope of medical science and related technologies. Some of the breakthroughs have resulted in dramatic declines in deaths from particular causes, such cardiovascular disease 7. Also, since the 1960s, gains in life expectancy have tended to be concentrated among the middle aged and older population (Goss et al. 1994, p. 2). This has resulted in a very rapid increase in the number of people aged over 75 years. 6 It is generally thought that there is a physical limit to life. Such a limit would be determined by the length of life a person would have if he or she did not suffer any serious diseases. In recent years the oldest known living persons reached ages of around 120 years. 7 Between 1968 and 1988 age-adjusted death rates from cardiovascular disease in Australia declined by close 50 per cent (d Espaignet et al. 1991).

20 Australia s Ageing Population 13 Health status Today people tend to view longer lives with mixed feelings. One of the main reasons for this is that much of the increased longevity tends to be spent in a state of ill health and disability. An Australian study noted that improved life expectancy, at least since the 1980s, has not so much been the result of better health than the ability of the medical profession to lengthen the lives of the seriously and chronically ill (Mathers 1991, p. xii). After reviewing the international literature on this topic, the OECD (1996c, p. 248) concluded that the extension of life tended to prolong both the independent and dependent periods, and that it was realistic to expect that the need for care was most likely to increase in line with demographic developments. Health costs and lifestyle risk factors Health costs In Australia, health costs are mainly publicly funded. Although a small proportion of costs are funded through the Medicare levy (paid by workers as a proportion of salary) and around a quarter of the population has private health insurance, the bulk of health costs are borne by governments. Those aged 65 and over account for about a third of Australia s total health expenditure, although they make up only 11 per cent of the population. Public health expenditure, around $19 billion a year, represents some 15 per cent of the total Commonwealth government outlay. There are continued concerns about health costs increasing considerably more rapidly than expenditure generally (see Walker 1997, section 1). Lifestyle risk factors Mathers (1991) noted that a longer life spent in a state of ill health was expected to result in increased demand for social and health services. However, this trend could be countered in the future if significant advances in primary prevention could be made. This is because primary prevention has the capacity to lower the incidence of chronic and disabling diseases (Mathers 1991, p. xii).

21 14 Discussion Paper No. 27 Table 2 illustrates the potential for lowering the demand for health services by older people through one form of prevention reducing lifestyle risk factors. These factors include obesity, inactivity, tobacco smoking and alcohol consumption. The table shows, for example, that hypertension was experienced more frequently by overweight people aged 65 and over than by those who were not overweight 64 per cent more frequently by men and 52 per cent by women. 8 Key findings from table 2 are summarised in box 1. Table 2 Effect of older persons lifestyle risk factors on chronic illness and on the demand for health services Aged 65 years and over Overweight, obese a Inactive a Smoker (tobacco) a Drinker (alcohol) a Specific chronic illnesses b Diabetes Asthma (men 1.73, women 2.16) Hypertension (men 1.64, women 1.52) Arthritis (men 1.34, women 1.18) Varicose veins (men 1.51, women 1.24) (men 1.34, women 1.32) Bronchitis (men 1.34, women 1.25) Heart disease (men 1.23, women 1.22) Arthritis (men 1.22, women 1.15) Health service usage b Hospital episodes Hospital episodes (men 0.92, women 0.87) Outpatient visits (men 1.19, women 1.05) Doctor visits (men 1.00, women 1.10) Other health professionals (men 0.82 women 1.17) (men 1.34, women 1.48) Outpatient visits (men 1.29, women 1.50) Doctor visits (men 1.12, women 1.28) Other health professionals (men 1.14, women 1.46) Bronchitis (men 1.52, women 1.74) Asthma (men 1.25, women 1.03) Hypertension (men 0.67, women 0.60) Diabetes (men 0.49, women 0.57) Hospital episodes (men 0.85, women 0.92) Outpatient visits (men 0.34, women 1.01) Doctor visits (men 0.78, women 0.91) Other health professionals (men 0.65, women 1.51) Hernia (men 1.59, women 2.24) Bronchitis (men 1.10, women 1.97) Asthma (men 1.25, women 1.14) Heart disease (men 0.78, women 0.31) Hospital episodes (men 0.64, women 1.06) Outpatient visits (men 2.91, women 1.10) Doctor visits (men 0.90 women 0.80) Other health professionals (men 0.79, women 0.76) a Overweight: if body mass index is above 25 kg/m 2. Inactive: if reported no physical exercise for recreation, sport or health/fitness reasons. Smoker: if currently smokes tobacco. Drinker: was considered medium or high risk if reported average daily consumption of ethanol was above 50 ml for men and above 25 ml for women. b Age standardised rates, data. Comparisons are with the rest of the population (for example, the obese are compared with the not obese, the inactive with the active, etc.). Source: Mathers (1994b, pp ). 8 A comprehensive study of health differentials among Australians of various age groups concluded that inequalities in the health of younger Australians continued into older ages (Mathers 1994b, p. v). Also see Mathers (1994a, 1995, 1996).

22 Australia s Ageing Population 15 Box 1 Key findings from table 2 Overweight people aged 65 years and over reported having the chronic illnesses of diabetes, hypertension, arthritis and varicose veins much more frequently than non-overweight people. While they reported hospital episodes less frequently, their outpatient visits were higher than for non-overweight people. Inactive older people reported having the chronic illnesses of asthma, bronchitis, heart disease and arthritis considerably more frequently than active people did. They were intensive users of health services, having reported using hospitals, doctors and other health professionals considerably more frequently than active persons. Smokers aged 65 years and over reported having the respiratory diseases of bronchitis/emphysema and asthma considerably more frequently than nonsmokers. However, they experienced hypertension and diabetes considerably less frequently than nonsmokers and reported significantly lower cholesterol levels (men 0.74, women 0.89). Smokers generally reported using health services less frequently than nonsmokers did. Moderate to high risk drinkers aged 65 years and over reported having the chronic diseases of hernia, bronchitis/emphysema and asthma more frequently than low risk drinkers. However, they reported much lower cholesterol levels than the control group (men 0.39, women 0.30) and experienced heart disease considerably less frequently. Moderate to high risk drinkers generally reported using health services less frequently than low risk drinkers. The exceptions were outpatient visits, especially for men, who reported using such services close to three times the rate of low risk drinkers. Apart from lifestyle factors, many other socioeconomic factors affect health. For example, older people with low incomes or low education levels are likely to have health risks from inactivity, smoking and obesity concurrently. People with such socioeconomic characteristics also tend to report higher levels of health service use. Different attitudes between the sexes provide some explanation for the fact that older men tend to report fewer illnesses and visits to the doctor than do women, although they die earlier and at a greater rate than women do in the same age group. Older immigrants from continental Europe are more likely to be overweight, and non-english speaking older immigrants are more likely to report diabetes than the Australian-born older population. However, there were no significant differences between the use of health services by non-english speaking and other older Australians (Mathers 1994b, pp. v, vi, viii).

23 16 Discussion Paper No Age of retirement As long out-of-the workforce as in it? In Australia, the age pension ages of 60 years for women and 65 years for men were set in 1909 (Department of Social Security 1996a, p. 260), when life expectancy at birth was just under 60 years of age. As a result, at the beginning of the century only a very small proportion of people were able to rely on government support in their old age. Although the age pension age for women is now gradually being increased to 65 years, no other changes have been made since the early 1900s. This is despite the fact that life expectancies are now some 20 years longer than they were then (table 1). Assuming full-time employment until the age of 65, people today can expect on average to be out of the workforce for a period of years. By 2051, this period is likely to have increased to close to 20 years. In addition, today many retire at age 55. In Australia there is evidence of a strong preference by employers for younger workers (see, for example, Department of Social Security 1992). Surveys have found that employers see people who are years of age as older workers. Also, many respondents considered that age discrimination commenced at or below age 45. In addition, older workers tended to believe that the government supported early retirement. Thus without changes, by the middle of the next century many are likely to be out of the workforce for up to 40 years for as long as they had been in it. The trend towards early retirement Despite longer periods being spent in retirement, retiring before the age pension age is an increasing phenomenon for both Australian men and women. Of concern is that early retirement leads to substantial use of social security payments (Bacon and Gallagher 1996, pp. 65 6). 9 9 Those with larger superannuation benefits tend to invest these and not rely on social security as their main source of income.

24 Australia s Ageing Population 17 So why are people retiring before the age pension age? Indeed, given the increase in life expectancy, why is not there a trend towards people continuing to work after the age pension age, possibly in casual or parttime capacities? Because of the dramatic changes in work patterns, in workforce participation rates (especially for females) and in laws governing superannuation in recent years, research on these questions has not provided definitive answers. There are problems with data definitions 10 and data availability, as well as with the complexities of analyses in this area. However, research by Bacon and Gallagher (1996, pp. 65 6) suggests that: an unexpectedly high proportion of retired people would have liked to have remained in the workforce (involuntary and family retirements 11 were found to have outnumbered voluntary retirements by more than three to one); and people tended to respond to the incentives embedded in government policies, since rates of early retirement were found to have increased dramatically at the superannuation preservation age. This was age 55 for all persons at the time of this study. 12 Maintaining an active, meaningful life From the point of view of those over 55 years of age, the challenge is to maintain an active and meaningful life. Some prefer to retire early, others to retire gradually, while others work full-time for as long as they can. However, the little evidence available suggests that despite much increased longevity there are strong pressures to retire early. As already 10 For example, currently there are problems with finding a meaningful official definition of being retired. 11 Involuntary reasons included retrenched, job was temporary, own ill health, business closed down, unsatisfactory work arrangements and employer thought too old. Family reasons, nearly all cited by females, included to look after family, house, or someone else, and to have children (Bacon and Gallagher 1996, p. 59). 12 Currently only applies to people born before 1 July For those born after that date the superannuation preservation age increases gradually to age 60.

25 18 Discussion Paper No. 27 noted, these arise from social attitudes, employer preferences and age pension and superannuation policies. How can governments develop a policy framework that would give a broader range of work-to-retirement choices to older people? As with most age related issues, there is a need to look well beyond the jurisdiction of individual departments. In relation to early retirement the Department of Social Security (1996a, pp ) notes that: there is not enough work being done in this area. Perhaps we currently have a social security system that has not entirely kept up with the times also we need to address early retirement in other forums such as in labour market or in taxation policy. We have to take a wide-ranging view of how we deal with early retirement. 3.3 Saving for old age At the national level, savings and investment in Australia have been relatively low in recent years. Indeed, many are concerned that currently they are too low (see Sheehan, Grewal and Kumnick 1996, part 2). In this section, however, it is not the overall level of savings that is at issue, but the extent to which individuals personal savings are put aside for their aged care rather than being locked into non-age related investments or consumed before the need for aged care arises. Public or private financing? In the age pension was the most important source of income for retired Australians aged 60 and over. In that year, 73 per cent of all Australians aged 60 or more received an age pension benefit, with 49 per cent of these receiving benefits at the full rate (Brown 1996, p. 4). The Commonwealth government outlay on the age pension was $12 billion in , following considerable increases in public funding of the age pension over several decades (Department of Social Security 1996b, p. 84). On this issue Tapper (1993, p. 3) noted that: Net lifetime transfers to the generation which is now enjoying retirement have been massive. This generation benefited in the 1950s and 1960s, when the State favoured the young, and benefited again when the State later transferred its favours to the elderly The lifetime personal income taxes paid by an average couple from that generation will amount to only a

26 Australia s Ageing Population 19 fraction probably less than half of the value of the age pension they will receive. Without policy changes, age pension expenditure will in future increase even more rapidly, due to greater longevity and an increased rate of population ageing. Concerned about this, the government recently indicated a redirection, stating its objective as: to encourage and assist retirees to achieve financial independence (Department of Social Security 1996b, p. 83). Wealth accumulation Those currently retired tended to save in their youth so that they could own the family home. Brown (1996) estimated that in the total assets of age pension age Australians amounted to $305 billion. Some $193 billion of this was home equity. Current trends suggest that the relative importance of home ownership in the savings of retired people will diminish in the future. One reason is that, with increased rates of family breakdown and greater uncertainty in the job market, people are more hesitant to commit themselves financially to buying their own home. In this respect Percival (1998) shows that, between and 1997, there was a significant decline in the proportion of those aged 35 years or less who had purchased their own home. Another reason is that because legislation has made saving for old age compulsory through the Superannuation Guarantee Charge it is likely that some or all of the savings that would have gone into other forms of investment (for example, the family home) are now being diverted towards the SGC. Table 3 shows that superannuation assets have been projected to increase eightfold between 1995 and 2020 (from $222 billion to around $1800 billion in current prices). The fastest growing element is projected to be compulsory savings 13, SGC funds increasing from around $26 13 In carrying out these analyses, Rothman (1996, p. 4) assumed that the spread of superannuation, especially at the basic SGC level, continued up to June 1995 and that coverage proportions stayed constant thereafter. The number of people with superannuation coverage followed demographic and labour force changes.

27 20 Discussion Paper No. 27 Table 3 Superannuation assets by type of fund In current prices Total SGC funds Personal and rollover funds Other funds a All superannuation assets $ billion $ billion $ billion $ billion a Includes public and private defined benefit funds, private defined contribution funds and superannuation funds for the self-employed. Source: Rothman (1996, p. 7). billion to $572 billion over the period more than a twentyfold increase in the amount of savings earmarked for post-retirement use. 14 Table 3 also suggests that the funds preserved for those retiring early 15 will increase more than fivefold over the period less rapidly than all superannuation assets, which have been projected to increase eightfold. Superannuation assets of around $1800 billion in 2020 would be far greater than the total assets of age pension age Australians in (around $300 billion). Over the past few years the exceptionally high and virtually government guaranteed growth in superannuation has attracted many new entrants to the superannuation industry. The vigorous competition that has followed has led to loss of market share for the few established superannuation giants, with subsequent declines in their performance. The result is that the average Australian worker is currently both concerned and bewildered. With many superannuation funds charging a 14 The projections were, of necessity, based on a set of simplifying assumptions. Consequently they should not be seen as accurate estimates of what will happen in 10 or 20 years time. For example, if employment arrangements continue to move from full-time to part-time jobs or casual and contract work, the SGC funds can be expected to grow less rapidly because more people will move in and out of work and fewer will be in continuous full-time employment. Nevertheless, with a projected twentyfold increase by 2020, total SGC funds can still be expected to grow rapidly. 15 Combining amounts in personal and eligible rollover accounts, and assuming that current trends in early retirement remain unchanged.

28 Australia s Ageing Population 21 fixed fee plus a percentage fee 16 say, 2 per cent a year on funds invested, once a fixed entry fee had been paid the value of workers superannuation investments can decline significantly. In the example, fund performance needs to be better than 2 per cent plus inflation a year during the workers contributory period (that is, better than 5 per cent if the average annual inflation rate is 3 per cent over the period), if investments are to at least maintain their value in real terms. So the choice of a well-performing fund is very important. However, in the current environment the performance of established firms can no longer be guaranteed. In addition, how the many new dynamic firms will perform in the future is anybody s guess. Such firms have already shown that, while they can outperform the larger established companies in boom times, they also face greater risks of bankruptcy in difficult periods. So there are valid concerns about the very real possibility, given current policies, that some workers may see either erosion of the value of their superannuation savings over time or lose their savings altogether. Additional concerns arise from the considerable teething problems associated with the implementation of the SGC scheme. For example, people do not appear to be using the tax advantage offered to the extent expected; superannuation funds are complaining about having to bear the high costs of administering the scheme; low income earners (for example, those in part-time and casual jobs) find that their small, forced superannuation savings can barely meet even the fixed parts of fund fees; there is an incentive for commission-based financial advisers not to fully disclose information about risk, since such a practice can greatly increase their earnings; and people are unclear about their superannuation payouts due to uncertainties surrounding the way the SGC regulations will be interpreted in practice. According to the Department of Social Security (1996a, p. 112), Confusion rather than certainty reigns over retirement income provisions, particularly with respect to superannuation payouts. 16 By charging fees in this manner, superannuation funds are essentially ensuring that the risks associated with investing are mainly borne by their clients. While the funds offer low risk options, the returns on these are also low, often barely covering the fees and compensating for inflation.