australia's ageing population - NATSEM

8 downloads 0 Views 195KB Size Report
May 19, 1997 - between the University of Canberra and the then federal ...... continue contributing to a superannuation fund, and people aged above. 50 years ...
AUSTRALIA’S AGEING POPULATION: HOW IMPORTANT ARE FAMILY STRUCTURES?

Agnes Walker

Discussion Paper No. 19 May 1997

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

National Centre for Social and Economic Modelling • Faculty of Management • University of Canberra •

AUSTRALIA’S AGEING POPULATION: HOW IMPORTANT ARE FAMILY STRUCTURES?

Agnes Walker

Discussion Paper No. 19 May 1997

ISSN 1320-3398 ISBN 0 85889 614 1

© NATSEM, University of Canberra 1998 National Centre for Social and Economic Modelling GPO Box 563 Canberra ACT 2601 Australia Phone: + 61 6 275 4900

Fax: + 61 6 275 4875

Email: Client services General

[email protected] [email protected]

World Wide Web site

http://www.natsem.canberra.edu.au

Core funding for NATSEM is provided by the federal Department of Health and Family Services.

iii

Abstract In Australia, as in most Western economies, concerns about the impact of the ageing of the population have led to a series of policy initiatives that generally aim to assign greater responsibility to individuals for the costs associated with ageing. Analyses of the effects of ageing published to date have tended to rely on aggregate projections of population and age structures. Interdependencies between individuals have rarely been taken into account, and it is only in some recent studies that the complex relationship between family structures and the size of government social expenditure has been recognised. Such studies cite broad statistics on the significant support that older people offer their families and on the importance of family members as carers for the aged. In this paper traditional approaches to studying the effects of ageing in the population are reviewed and a complementary approach, using dynamic microsimulation, is discussed. The advantage of dynamic microsimulation is that, in projecting future population patterns, the life cycles of individuals can be tracked, along with their family characteristics (such as spouses, children and grandparents) and household characteristics (including finances and wealth accumulation). The paper concludes with an account of the development work on dynamic microsimulation modelling at NATSEM, indicating the areas where this tool will be best able to advance understanding of age related policy issues.

iv

Author note Agnes Walker is a Research Fellow at the National Centre for Social and Economic Modelling.

Acknowledgments The author is grateful to NATSEM research staff for their support for an earlier version of this paper (presented at the 1996 Conference of Economists, Canberra) and the encouragement of Professor Ann Harding for publication of this paper, as well as for further work in this area. Thanks are also due to Anthony King for his valuable comments on chapter 4 of the paper. The extensive subject knowledge of two external referees — Dr Ashok Tulpule, who co-authored the 1994 EPAC report on ageing, and Mr Warwick Bruen, who is head of the Community Care Branch of the Department of Health and Family Services — was much appreciated, as was their thoroughness when reviewing this paper. As usual, all errors and omissions remain solely attributable to the author.

General caveat NATSEM research findings are generally based on estimated characteristics of the population. Such estimates are usually derived from the application of microsimulation modelling techniques to microdata based on sample surveys. These estimates may be different from the actual characteristics of the population because of sampling and nonsampling errors in the microdata and because of the assumptions underlying the modelling techniques. The microdata do not contain any information that enables identification of the individuals or families to which they refer.

v

Contents Abstract Author note Acknowledgments General caveat

................... iii ....................iv ....................iv ....................iv

1 Why ageing is of concern

..................... 1

1.1 Caring for the aged — whose responsibility?

..........................1

1.2 Cost pressures

..........................3

1.3 Government revenue

..........................6

1.4 Policy responses to likely cost increases

........................10

1.5 Intended effects of age related policies

........................13

2 Importance of a family perspective

................... 18

2.1 The elderly: traditionally a family responsibility

........................18

2.2 Extent of family contributions

........................18

2.3 Policies with a family focus

........................23

3 Previous approaches to studying ageing

................... 26

3.1 Population projections

........................27

3.2 Family linkages

........................28

3.3 Ageing and social expenditure

........................29

3.4 Summary of selected studies

........................35

3.5 Impact of studies

........................39

4 Approaches using dynamic microsimulation

................... 41

4.1 Characteristics of dynamic microsimulation

........................41

4.2 Current status of dynamic microsimulation modelling at NATSEM

........................42

4.3 Microsimulation well-suited to distributional analysis

........................47

4.4 Examples of possible policy relevant analyses

........................47

5 Summary

................... 49

References

................... 55

1

1

Why ageing is of concern Would this be the beginning of an Australia of old people, in old houses, dreaming old ideas? … The inter-generational and extended family needs to be given greater political and economic support if our society is going to overcome its principal future population problem — decline and ageing. (R. Blandy in EPAC 1994b, Report 4, p. 95)

1.1 Caring for the aged — whose responsibility? One hundred years ago people did not expect to live much beyond 50 years of age.1 By 1994 life expectancies at birth had risen to 75 years for men and 81 years for women and by 2051 they are likely to be 81 years for men and 86 for women (ABS 1996b). Key reasons for continued increases in life expectancies include the considerable breakthroughs made by modern medicine in recent decades, as well as better nutrition and healthier lifestyles. Increased life expectancies, together with dramatic declines in birth rates, have led to the ageing of the populations — that is, the increase of the proportion of people aged over 65 years — in most developed countries.2 This is illustrated for Australia by figure 1. People aged over 65, currently around 12 per cent of Australia’s population, are projected to account for close to 23 per cent of the population by 2051. Over the same period, the proportion of the population accounted for by people aged over 85 years is projected to increase from around 1 per cent to 5 per cent (ABS 1996b). The age dependency ratio, defined as the proportion of the population aged 65 and over to the proportion aged between 18 and 64 years, is projected by the Retirement Income Modelling (RIM) Task Force to increase from around 20 per cent currently to close to 40 per cent by 2051 (National Commission of Audit 1996, p. 127). Most of these increases are projected to occur between 2010 and 2030.

1

At the beginning of the century, life expectancy at birth in Australia was 55 years for men and 59 years for women (Abraham, d’Espaignet and Stevenson 1995). 2 In Australia, the relatively high birth rate of 2.9 children per woman in the early 1970s had declined to 1.9 by 1980 and has remained around 1.8–1.9 since. As this is below the 2+ level needed to replace the population over the longer term, the most recent ABS projections are for a steady decline in population beyond 2031, assuming natural population growth and no net overseas migration (ABS 1996b).

2

Discussion Paper No. 19

Figure 1 Projected population of Australia by age and sex, 1995 and 2051

Source: ABS (1996b, p. 31) for series A/B — that is, high fertility and low overseas migration assumptions.

As the ratio of those working (that is, taxpayers) to those not working declines, significant additional pressures are imposed on taxpayers and thus on nations’ budgets. In countries with budget deficits — where they spend more than they collect in taxes — people’s expectations that government spending will continue to increase will probably not be met. Such expectations raise near universal concerns in the developed world because: • current budgetary pressures are expected not only to continue but to increase; • future cost increases will add to the already considerable pressures for fiscal consolidation worldwide; and • there are concerns that, if we do not act now, the financial burden on our children and grandchildren in the decades ahead may be too high.

Australia’s Ageing Population: How Important Are Family Structures?

The Budget Paper (1996a) noted that fiscal consolidation is now a world priority. Action is currently being taken by key OECD economies — particularly in Europe, Japan,3 the United States and Australia — to reduce budget deficits over the medium term so as to support stronger economic growth and reduce unemployment. 1.2 Cost pressures4 Key areas of cost pressures for Australia are similar to those in other developed economies — health care, aged care, family support payments, and age pensions. Health care

The rising costs of health care are of particular concern for a number of reasons. First, they represent a significant share of government expenditure — currently 15 per cent of the total federal government outlay. Second, health care costs are expected to increase significantly as the population ages. Goss et al. (1994) show that in 1989-90 those aged 65 and over were responsible for 33 per cent of Australia’s total health expenditure, although they accounted for only 11 per cent of the population. Third, health costs have tended to increase more rapidly than expenditure generally. Since 1985-86 the growth in the total federal government outlay on health has averaged 4 per cent a year in real terms (Budget Paper 1996d, pp. 3–94). The growth mainly reflects increases in per unit costs, arising from more frequent use of medical and pharmaceutical

3

In the case of European countries, implementation of the European Monetary Union requires budget deficits of 3 per cent or less of GDP. In Japan, the sizable budget deficit and debt flowing from the economic stimulus provided during the past four years was to be wound back, since growth was expected to rise above 1 per cent in 1996. The imperative for fiscal consolidation was highlighted by the cost pressures arising from Japan’s ageing population (Budget Paper 1996a).

4

Whenever possible, references in this paper to public costs in Australia involve expenditures by the state governments and the federal government. While recently proposed shifts of some social functions between levels of government may lead to efficiency gains, their consideration is beyond the scope of this paper.

3

4

Discussion Paper No. 19

services and a drift towards more costly drugs and medical services. Budget Paper (1996f, p. 19) notes that the annual increase in the cost of the Pharmaceutical Benefits Scheme has averaged over 14 per cent in recent years, and concludes that more of the costs must be born by users of the scheme. It also notes that older people are the heaviest users of health care services in the population. Table 1 details federal government outlays on health services in 1995-96. Table 1 Distribution of federal government health expenditure, 1995-96 Type of expenditure

Medical Hospitals Pharmaceuticals Nursing homes Other (includes administration) Total

Amount

Share of total

A$b

%

7.0 5.4 2.7 2.3 1.2

38 29 14 13 6

18.6

100

Source: Budget Paper (1996d, pp. 3-45, 3-94).

Age pensions

In 1993 close to 70 per cent of people aged 65 years and over and living in New South Wales relied on the age pension as their main source of income. While some 13 per cent cited investment and interest as their main source of income, only 5.8 per cent relied mainly on superannuation and 1.3 per cent on wages or salaries (ABS and Office on Ageing 1995). These figures are in line with 1992-93 Australia-wide statistics for retirees aged 60 and over, quoted by the Retirement Income Modelling Task Force (1996b). This information and the fact that, in 1994, only 25 per cent of the health care of the elderly was financed privately (EPAC 1994a) suggest that the elderly rely heavily on government support. EPAC noted that, in 198990, public expenditure on aged health care was comparable with expenditure on age pensions, the latter having amounted to around A$12 billion.

Australia’s Ageing Population: How Important Are Family Structures?

5

Expected future growth

Health and aged care costs have been projected by most analysts to grow significantly. Demographically projected total health expenditure was estimated by EPAC (1994, p. 36) to increase, in constant dollars, from A$29 billion in 1990 to around A$126 billion in 2051, up from 8.4 of GDP to 11.1 per cent. In deriving these projections, it was assumed that real growth in per person health expenditure would be 1 per cent a year, roughly in line with assumed growth in living standards over the projection period. EPAC noted that the resulting estimated increase in health expenditure of 2.4 per cent a year was due to the ageing of the population. Recent projections reported by the National Commission of Audit (1996, p. xv) indicate an increase (in real terms) of A$19 billion in federal government health expenditure between 1996 and 2041, assuming that pre-1996 budget policies continued. The Commission also refers to RIM Task Force estimates of increases between 1996 and 2031 in expenditure as a percentage of GDP — 6.1 per cent for health care (including the aged) and 1.1 per cent for age pensions (p. 134). The focus of this paper

Whether health care costs will continue to rise rapidly is a complex question. There are many influences at work other than population ageing. These include improvements in medical technology, doctor generated demand, changes in patient expectations, and the way costs are split between patients and governments. Of key concern is how to limit future rises in health costs, while ensuring that adequate services are available to all. In an international context, Australian policies have to date been relatively successful. For example, health care costs as a percentage of GDP are well below those in the United States. Also, Australian health policies are generally seen as having been effective in channelling public funds to those in need. The policy challenge is to maintain these successes. Consideration of future health care costs generally is beyond the scope of this paper. For our purposes the important points are that:

6

• •



Discussion Paper No. 19

in Australia there are already pressures on government budgets from health care expenditure; those aged 65 years and over are currently responsible for around a third of total health care expenditure, although they account for just over a tenth of the population; and Australia’s projected growing proportion of aged in the population (over 20 per cent by 2050) will place extra pressures on future budgets, over and above those that would have prevailed if Australia’s population had not been ageing.

These points can be made with some certainty. The extent to which such pressures on public funds will be met in the future is, however, less certain (see section 1.3).

1.3 Government revenue Revenues are related to living standards5

Whether taxation revenue can continue to support the young and the elderly to the same extent as in the past depends not only on the ratio of those working to those not working, but also on how well off those who are working are. The higher is per person national income, the greater is the taxation revenue at a given tax rate. Over the past 50 years, living standards in the Western world have risen dramatically. As Paul Krugman (1992, p. 9) put it for the United States: World War II veterans came home to an economy that doubled its productivity over the next 25 years; as a result, they found themselves achieving living standards their parents had never imagined.

Australia’s performance is below the OECD average

Although Australia’s experiences were similar after World War II, over the past 25 years its productivity performance has been significantly below that of the OECD average, and poor relative to the dynamic Asian economies (Productivity Commission 1996). This is despite Australia’s

5

Defined as GDP per person.

Australia’s Ageing Population: How Important Are Family Structures?

7

proximity to, and increasing trade with, the rapidly growing Asian region. Of particular concern is that Australia’s labour productivity, which was above 2 per cent a year during the 1960s and early 1970s, has declined steadily since the mid-1970s (EPAC 1994a). While there are indications of some improvement over the past 12 months6, it is by no means certain that Australia can replicate the sustained growth in per person GDP achieved in the 1960s. EPAC (1994a) has noted that productivity growth and income distribution are the keys to dealing with the ageing of Australia’s population — even more important than population policies. It projected aggregate GDP to grow from A$340 billion in 1994 to around A$1100 billion (constant prices) by 2051, assuming productivity growth of 1 per cent a year and that recent rates of net migration remain unchanged. The result, a near doubling of real GDP per person, would have significant implications for taxation revenue. There has been some discussion in Australia on the extent to which ageing might become more of a policy issue than it currently is. While 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 are of the view that the reason why ageing has not had ‘disastrous’ consequences so far is that in the past the supply of labour expanded considerably as more and more women entered the workforce (Walker 1997, section 2). That population ageing will increase the pressures on government budgets does not appear to be questioned. It is the extent of these pressures — whether of crisis proportions or unimportant compared with other policy matters — that is being debated. It is in this area that comprehensive quantitative analyses — as reported in chapters 3 and 4 — could contribute to the debate by highlighting the likely implications of various policy options under different demographic, economic and social scenarios.

6

See, for example, Institute of Management Development (1997), which rated Australia 16th in the world in terms of competitiveness in 1997, compared with 21st in 1996.

8

Discussion Paper No. 19

Possible improvements

So how can Australia’s economic performance be improved? Can we learn from the success of the dynamic East Asian nations? These economies were or are in a ‘developing’ phase, with considerable technological catch-up potential, which is not available to developed nations such as Australia. Between 1965 and 1990 they were able to achieve growth rates of close to 6 per cent a year in GNP per person compared with just over 2 per cent for OECD economies (World Bank 1993). According to the World Bank, what distinguished the successful East Asian economies from other developing countries was that they invariably got the fundamentals right: market friendly policies, macroeconomic stability, rapid export growth, a sharing of the benefits of growth across society, thereby reducing poverty and creating human capital, high savings and hence high investment rates, and resource allocation that fostered rapid productivity growth. Without these fundamentals, total factor productivity growth rates above those of key industrialised economies would not have been possible. In addition, a willingness to experiment and adapt policies to changing circumstances —limiting the costs and duration of inappropriately chosen interventions — was a key to the dynamic East Asian nations’ economic success (World Bank 1993).7 The findings reported by the Industry Commission (1990) were broadly similar. While recognising that Australia had very different characteristics from East Asian nations, the Industry Commission noted that the success of the economies in East Asia offered valuable lessons for Australia. Among these were the need to vigorously pursue policies that encouraged competition (the key to successful upgrading and innovation) and the need to improve labour markets through greater flexibility, a willingness to accept change and less industrial unrest. Other factors of importance to Australia identified in the literature include the need to improve national savings, the sharing of the benefits of growth across society and a more effective allocation of resources with a view to enhance, rather than hinder, productivity growth. In relation to

7

Root (1996) adds another factor essential to success: the ability of governments to implement the policies of their choice.

Australia’s Ageing Population: How Important Are Family Structures?

9

the last point, EPAC (1994a) called for focus on productivity growth rather than population policy, and the Productivity Commission (1996, p. 24) noted that: …technological improvements and the quality (and quantity) of human and physical capital ... are clearly important. But a more basic requirement ... is to have the right incentives in place for people to work harder and smarter, to acquire new skills and to seek new opportunities.

The Productivity Commission also cited the gains from selected microeconomic reforms in Australia as estimated by EPAC, the Bureau of Industry Economics, the Business Council of Australia and the Industry Commission. The reforms considered varied across studies, but included areas such as transport and communication, taxation, contracting out by governments, the labour market, labour productivity in the private sector and national competition policy (implementation of the Hilmer reforms). Estimates of long term gains in real GDP ranged between 5 and 20 per cent a year across the studies (Productivity Commission 1996, p. 31). In today’s dollars, a gain of 5 per cent of GDP is equivalent to around A$24 billion a year. While not directly comparable8, gains through microeconomic reform appear to be similar to (or greater than) projected increases in total health9 and age pension costs (see section 1.2). Therefore, as the country’s population ages, there is significant potential for Australians to maintain (or even improve) their living standards through productivity improvements that are already being publicly discussed. With technological breakthroughs this potential could be even greater.

8

Gains from microeconomic reform generally indicate the extent to which GDP would be higher with the reforms than without them. The gains are usually estimated under the assumption that all else remains unchanged, including Australia’s population. Estimates of gains from microeconomic reform are not directly comparable with projections of health and age pension costs, since the latter account for expected changes in population and other factors. 9 Around a third of the increase in total health costs would be accounted for by people aged 65 years and over.

10

Discussion Paper No. 19

1.4 Policy responses to likely cost increases Shifts towards greater responsibility by individuals

In many countries there has been a policy shift to encourage individuals to assume greater financial responsibility for their own (and their families’) wellbeing. This has tended to happen in response to concerns that the future demands on the public purse could be considerably higher than government revenue. However, attempts to moderate expectations about government support have been unpopular, especially in countries where there were expectations of increased future public support of both the young and the elderly. Policy shifts towards greater individual financial responsibility have sometimes led to disturbances as in, for example, France and Germany. Although the early signs of a similar reaction in Australia may not indicate longer term trends, it is worth noting that there has been strong resistance in some quarters to recent government policies — for example, those that aim to recover from university students a higher proportion of the cost of their tertiary education.10 Also, May 1996 had the highest monthly level of industrial disputation since November 1992. Working days lost in May amounted to 164 000, compared with monthly estimates of between 20 000 and 88 600 since March 1995 (ABS 1996a). These events could be seen as an indication of frustration as people realise that increases in living standards may no longer be a certain outcome for the future. Recent Australian initiatives

In Australia a range of policies have been introduced with a view to shifting greater responsibility to individuals for the costs of their

10

In Australia, currently 23 per cent of higher education course costs are recovered on average (through the Higher Education Contribution Scheme), compared with close to 50 per cent in the United States and 60 per cent in Japan (Budget Paper 1996g, p. 6).

Australia’s Ageing Population: How Important Are Family Structures?

11

ageing.11 One of the most important is the Superannuation Guarantee Charge, which essentially makes superannuation compulsory for most employed Australians.12 As a result, the proportion of elderly relying mainly on their superannuation income is expected to rise, and those on the age pension — close to 70 per cent in 1993 — to decline significantly. In its central projection, EPAC (1994a, p. 42) estimated that total social expenditure would rise from 21.3 per cent of GDP in 1990 to 25.2 per cent by 2051 without the Superannuation Guarantee Charge, but to only 23.8 per cent of GDP with retirement income reform. EPAC (1994a, p. 41) noted that the net increase in total social expenditure of 2.5 per cent of GDP between 1990 and 2051 — as under the retirement income reform scenario — would imply a rise of around 8 per cent in average taxes from the 1994 level. Recently announced policies of relevance are in box 1. Community concerns

Bringing about policy changes tends to be a hard task generally, because a changing environment causes uncertainties within the community, because some policies are extremely complex and cumbersome to implement, and because not all policy changes can achieve their stated aim. These difficulties tend to be exacerbated when the announced policy changes are likely to lower living standards for some. Recent examples of difficulties include community reaction to the entry fees to nursing homes and the taxation incentives available to those with private health insurance. In relation to nursing home entry fees, which were reported in the press to amount to as much as A$26 000, there was some concern that older people may be required to sell their houses and, in turn, leave their spouses without homes. Others expressed concern

11

For a charting of the evolution of Australian policies on health, housing and welfare services for the aged, see Sax (1993). 12 While the Superannuation Guarantee Charge is expected to shift responsibility for retirement incomes towards individuals, it is not clear that it will increase the nation’s total savings. It may simply affect the form in which savings take place (for example, superannuation instead of shares). However, EPAC (1994a, p. 67) notes that the Superannuation Guarantee Charge’s overall effect on savings in Australia has been estimated by some analysts to be positive.

12

Discussion Paper No. 19

Box 1 Selected policies in the 1996-97 budget affecting older people Encouraging people to make greater provision for their old age by: • providing greater choice of superannuation avenues by, for example, allowing banks, credit unions, building societies and life offices to provide superannuation through retirement savings accounts; • applying a surcharge of up to 15 per cent to tax-deductible contributions made to superannuation funds by high income earners; • asking people to pay a fair contribution to the cost of their aged care by: – increasing copayments for subsidised prescription drugs (from $17 to $20 generally, and from $2.70 to $3.20 for those on concessions); – allowing nursing homes to introduce upfront entry contributions, which are to become the principal source of capital funding (expected savings to government of A$479 million over four years); – introducing income tested subsidies for residential aged care (nursing homes and hostels); and – introducing a national client fees policy for home and community care (that is, home help, personal care, Meals on Wheels and home nursing services); and • providing taxation incentives of up to A$125 a year for people to have private health insurance.

Better targeting the age pension by: • removing the superannuation assets means test exemption for people over 55 years of age; and • removing the deeming ‘free area’ on certain bank deposits.

Encouraging people to do paid or voluntary work and have healthier lifestyles by: • allowing people aged between 65 and 70 years to continue contributing to a regulated superannuation fund, provided that they are in the paid workforce for at least 10 hours a week over the year; • allowing people aged 50 years and over to undertake unlimited full-time voluntary work and still remain qualified for social security allowances; and • providing grants for innovative health related projects (exercise regimes, leisure activities) and redeveloping social networks for those isolated.

Committing to provide greater assistance to carers by: • providing additional funds (A$36.7 million above 1995-96 expenditure over four years); and • improving the flexibility of caring arrangements by liberalising eligibility for carer payments. Sources: Budget Paper (1996a,f).

Australia’s Ageing Population: How Important Are Family Structures?

13

that nursing home entry fees may act as a de facto death duty, since fees could be charged on the estate. With regard to the health insurance taxation incentive, concerns have been expressed that the fee increases announced by several private health insurance companies shortly after the budget would reduce the extent to which the incentives would encourage people to take out or keep their private cover. In response, the government announced an inquiry into the private health insurance industry. These recent examples of community concerns illustrate why progress in adjusting to ageing populations has been slow worldwide.

1.5 Intended effects of age related policies Overall, the main aims of recent Australian policy initiatives fall into five categories: • to encourage people to provide for their old age during their working life; • to encourage the aged to contribute more towards the cost of their care; • to increase the age range within which people can work; • to increase the proportion of older people who remain in good health; and • to increase government support to carers. Encourage people to provide for their old age during their working life

Compulsory superannuation and greater choice in superannuation products (for example, retirement savings accounts) are examples of already introduced or announced policies to meet this aim. Other suggestions discussed in the public arena centre around making superannuation benefits available exclusively for old age related services. Among these is a proposal to remove the choice of taking part of superannuation benefits as a lump sum. The aim would be to discourage the practice of spending the lump sum early in retirement (for example, on overseas trips) and later applying for the age pension.

14

Discussion Paper No. 19

Encourage the aged to contribute more towards the cost of their care

The policies involve areas such as health services, pharmaceuticals, home care services and nursing homes. Greater targeting of the age pension and superannuation taxation concessions, entry fees to nursing homes and taxation incentives for private health insurance13, as announced in 1996-97 budget, fall into this category. Some of these have already caused community concern (see section 1.4). Against this trend, there have been experiments overseas that aim to spread the costs of ageing. Lloyd-Sherlock and Johnson (1996) reported on the 1989 pension reform in Germany, a country that has one of the most rapidly ageing populations and one of the most comprehensive public pension systems in Western Europe. Lloyd-Sherlock and Johnson noted that the reforms proved successful in spreading the pension costs of ageing across generations. Increase the age range within which people can work

Recently announced measures allow 65–70 year olds with paid jobs to continue contributing to a superannuation fund, and people aged above 50 years to undertake full-time voluntary work without losing their benefits. Such policy initiatives are in line with trends in most OECD countries, where policies aim to stimulate private, non-profit and voluntary sector responses to the needs of the ageing population, with the public sector setting standards and monitoring performance (OECD 1992). Such policy initiatives are also in line with experiments in some OECD countries that aim to lengthen the time people spend in the workforce. For example the OECD (1995, p. 121) notes that:

13

The private health insurance initiative involves all people with taxable incomes above set limits. However, the elderly are the heaviest users of health services in the population. The initiative aims to slow the trend away from private health insurance, which guarantees a hospital bed and the doctor of choice. Since the introduction of Medicare, the proportion of Australians with such insurance has plummeted from 60 per cent to less than 35 per cent of the population (Budget Paper 1996f).

Australia’s Ageing Population: How Important Are Family Structures?

15

Faced with a projected increase in the number of people over pension age and a stagnation in the number of people of working age … Member countries have taken steps to revise the conditions of pension receipt and/or to restrict those measures which support older people who have ceased working or which encourage them to leave.

The same report describes how Sweden has shown that a partial retirement scheme can be successfully built into a national social security system. According to the OECD the key to this success was the way the scheme was financed — by a levy on the total wage bill of each organisation. This diminished employer resistance to requests for part-time work, since the levy was paid even if the scheme was not used. The OECD (1995) concluded that the high take-up of the scheme was an indicator of its popularity. It is generally recognised in Australia that older people’s participation in the labour force is a complex and rapidly changing area. The greater availability of contract work, part-time work and paid voluntary work has made it difficult to accurately assess (through official statistics) the extent to which older Australians participate in the labour force. The available statistics suggest that a significant proportion of Australians retire from full-time work prior to the age pension age. For example, in 1994: • 42 per cent of retired men aged over 45 years were found to have retired from full-time work by the age of 59; and • 65 per cent of retired women aged over 45 years were found to have retired from full-time work by the age of 49 (Department of Social Security 1996, pp. 6, 8). The available statistics also show that: • part-time work for older Australians is increasing; • females tend to stay in the workforce longer than males; • those on high incomes retire later than those in low paid jobs; and • involuntary and family related retirements tend to outnumber voluntary retirements by three to one (Department of Social Security 1996, pp. v, 259–60).

16

Discussion Paper No. 19

Some have expressed concern that the eligibility ages for the age pension (60 for women and 65 for men), which were set in 1909, had not been changed despite significant increases in life expectancy (Department of Social Security 1996, p. 260). Whether people wish to retire early or work past retirement age depends on a range of economic, medical, cultural and social factors, as well as on government policies that may encourage or discourage early retirement (for example, labour market and taxation policies, and social security entitlements). Given the predominance of involuntary retirements, however, it appears that many who wish to stay in the workforce are not able to do so. With the ageing of Australia’s population, more and more people would like to see the development of a set of coordinated and creative policy initiatives regarding retirement (see, for example, Department of Social Security 1996, p. v). Increase the proportion of older people who remain in good health

The recently announced grants for innovative health related projects, such as exercise regimes, leisure activities and, for those isolated, the redevelopment of social networks are examples of policy initiatives in this category. If older people remain active, they can continue to contribute to society through paid or voluntary work and, if they lead healthy lifestyles, they are less likely to use health and age related services intensively. Many older people remain in relatively good health, using health services intensively only in the last year or two of their lives. Almost half of health expenditure on people over the age of 75 is on the 13 per cent who die within two years (EPAC 1994a, p. 39). Recent studies of the elderly, such as the one conducted jointly by the ABS and Office on Ageing (1995), report on a range of characteristics of importance to the aged, other than those traditionally considered. For example, their study shows that, in the early 1990s in New South Wales:

Australia’s Ageing Population: How Important Are Family Structures?

• • •



17

around half of those over 75 years of age exercised for recreation, sport or fitness; nearly two-thirds of those who helped sick or disabled adults were aged 55 years and over; voluntary work was much more important for the elderly than paid work (only a little over 1 per cent of those over 65 years of age relied on wages and salaries as their main income); and over half of all informal child care was provided by a grandparent.

Increase government support to carers

The recently announced additional funds for carers amounting to A$36.7 million over four years and the improved flexibility of caring arrangements fall into this category of policy initiatives. As already noted, nearly two-thirds of carers are elderly themselves (aged over 55) and tend to look after sick or disabled adults on a voluntary basis. So providing incentives to carers is an important way of tapping into the growing number of relatively healthy elderly people who are willing to help others. However, it has been generally recognised that there is a need to provide incentives to non-elderly carers so that they are not excessively disadvantaged relative to those with paid jobs. The OECD (1992) reports on a variety of policies used by its member countries that provide incentives for the informal care of elderly people. Some aim to safeguard the current earning power of the carer, while others minimise the negative consequences for the caregiver in old age. The measures include direct payments to family members to partly compensate them for loss of earnings, guarantees of job security after long term leave to care for dependent family members, and an allowance for caregiving years in calculations of pension benefits.

18

2

Discussion Paper No. 19

Importance of a family perspective

2.1 The elderly: traditionally a family responsibility Historically there has been an important material link between the number of children couples have had and their old-age security. While in many developing countries families are still the primary source of care given to older people, in the Western world their importance has declined considerably with the introduction of government funded social security ‘safety nets’. In developing countries without such ‘safety nets’, the growing number of older people now face uncertainty because their children are following the trends in the developed world (see Kendig, Hashimoto and Coppard 1992). The following questions therefore need to be addressed. • To what extent can families continue to fulfil their roles as caregivers to the elderly in the face of considerable demographic, social and economic change? • How can governments best complement families’ efforts under likely future scenarios?

2.2 Extent of family contributions Carers for the elderly

Studies in many developed countries show that the family remains the most important provider of care for the elderly, although its importance is declining. While there are fewer younger people relative to the number of older people, more family breakdowns and fewer potential carers (because many women — the traditional carers — have joined the workforce), there are more elderly people with a spouse and more elderly people with at least one child (OECD 1994, p. 9). Working women and, to a lesser extent, working men (sometimes with children) assume much of the responsibility of caring for the elderly. Such multiple responsibilities often place tremendous strain on caregivers (OECD 1992, p. 133). The ABS and Office on Ageing (1995) show that, in New South Wales, well over half of those aged over 65 in 1991 were married, most living

Australia’s Ageing Population: How Important Are Family Structures?

19

with their partner with no children present (table 2). The fact that in New South Wales nearly two-thirds of those who helped sick or disabled adults were aged 55 years and over (section 1.5) suggests that, in line with the findings of several other Australian and OECD studies, a high proportion of carers for the elderly are spouses. For example, Howe (1991) found that the spouses of the elderly needing care accounted for three out of four co-resident carers, while EPAC (1994a) reported that in the early 1990s wives accounted for around 40 per cent of the carer population, husbands 33 per cent, daughters 16 per cent and sons 5 per cent (see also ABS 1990a and 1993a, p. 23). Table 2 Living arrangements of people aged 65 years and over, New South Wales, 1991 Living arrangement

Member of a family Partner in couple, no children Partner in couple, with children Sole parent Lone person Non-private dwelling Other Total

Men

Women

People

%

%

%

71.0 53.5 10.9 1.8 14.3 7.2 7.5

50.5 31.1 4.4 5.4 31.3 1.1 7.1

59.2 40.6 7.2 3.9 24.1 9.4 7.3

100.0

100.0

100.0

Note: Among people aged 75 years and over, 63 per cent of men and 37 per cent of women lived as a member of a family, 18 and 37 per cent respectively lived alone and 12 and 20 per cent respectively lived in non-private dwellings. Source: ABS and Office on Ageing (1995, p. 12).

The second most common living arrangement among older people was to live alone (table 2).14 While many needing assistance in this group relied solely on formal services, others were helped mainly by their children, other relatives or elderly friends, with only occasional use of formal services. Evidence for this includes the extensive use of the Home and Community Care Program in Australia (section 2.3) and the studies

14

The proportion of elderly living alone in Australia, at around 20 per cent, is very low compared with most other OECD countries, where the proportions had risen from around 20 per cent in the 1950s to between 30 and 50 per cent by the late 1980s. Over the same period, reflecting a trend toward greater preference for independence by both the young and old, the proportion of elderly living with their children dropped considerably in most OECD countries (OECD 1994, ch. 2).

20

Discussion Paper No. 19

for Sydney, Denmark, France, Germany, Japan and the United States that show that, in the late 1980s, over half of elderly people living alone had adult children living within half an hour’s travelling distance (OECD 1994, p. 32). Family support offered by older people

As well as supporting their spouses and peers, older people have been providing support to younger members of their families. According to the ABS (1990b), in the late 1980s a third of children under 12 years of age were cared for by grandparents, aunts, other relatives and friends. Table 3 illustrates the range of services older people provide to their families. Table 3 Family support offered by older people Type of support

Proportion providing support %

Child minding Emotional support in crisis Care for others when sick Assisting with renovations Financial assistance with: Major purchases Tertiary education Deposit for house or flat

76 76 61 38 37 27 33

Source: EPAC (1994a, p. 71).

Inheritance

The Retirement Income Modelling Task Force (1996a, p. 20) estimated that the total value of assets of Australians of age pension age in 1992-93 was A$305 billion, of which some A$193 billion was home equity. Commenting on these assets, EPAC (1994a, p. 72) noted that ‘the biggest transfer of wealth ever seen in this country’s history will take place early in the next century’. However, the extent to which this wealth is transferred to the next generation will depend in part on government policies. For example, if the recently announced entry fees to nursing homes require the assets of

Australia’s Ageing Population: How Important Are Family Structures?

21

the elderly to be used, the amount of wealth transferred to the next generation will be lower than otherwise would be the case. On the other hand, because such a scenario would mean that the cost of aged care would rely less on taxation revenue, the younger generation’s capacity to save would be greater. Thus, while there is significant scope for shifting responsibility for the cost of aged care across generations, it is not certain that at the aggregate level this would affect the total lifetime assets available to people. For example, if the older generation used their assets to pay for the costs of their aged care, the younger generation, while receiving less inheritance, would face lower taxes. Assuming equal efficiencies in revenue collection methods and independence between the cost of aged care and the way it is financed, the amount lost by the younger generation through lower inheritances should roughly equal that gained through lower taxes. However, at the individual level there would be considerable differences, not only financially but also emotionally, as inheritance often concerns the family home. Other complications also need to be considered, such as the use of assets — for example, the family home — for financing the care of one partner in a couple when the other is still well (see section 1.4 and chapter 5). Proposals to have the elderly use their assets to cover some of their aged care costs are not new. As noted earlier, historically there has been an important material link between the number of children couples have had and their old-age security (section 2.1). In the late 1980s there was vigorous debate in the United States (and some discussion in Australia) about whether this material link should be re-established. An eminent US demographer, Paul Demeny, proposed that governments assign a proportion of the social security contributions of children directly to their parents. Demeny argued that, by cutting the link between material incentive and family size, the social security system discouraged couples from having children and that, by re-establishing the link, birth rates would increase (Demeny 1986; Dharmalingam 1991). The greater the financial pressures on governments, the more likely it is that the elderly will be required to dip into their wealth to pay for their aged care.

22

Discussion Paper No. 19

Ageing at home or in institutions?

Worldwide, elderly people have expressed a preference for being independent and remaining in their homes for as long as possible (OECD 1992). The increasing importance of home and community care services in Australia (section 2.3), combined with the cap placed on nursing home beds, is delaying the elderly’s entry into nursing homes. Since 1992 the elderly admitted to nursing homes has been a declining proportion of the total admitted to nursing homes, hostels or hostel equivalents, having declined from 61 to 54 per cent, while the elderly admitted to hostels or their equivalent has increased from 39 to 46 per cent (table 4). In 1995, of the 435 000 people aged over 70 with a handicap, 15 per cent were in nursing homes, 12 per cent in hostels or their equivalent, 37 per cent were receiving home and community care services and 36 per cent were not using government-provided services (Department of Human Services and Health 1995a, p. 126). As far as governments are concerned, the differences in outlays on nursing homes, hostels and home care are considerable. EPAC (1994a) noted that in the early 1990s the public cost of nursing home care was almost five times the cost of hostel care. With home care, the impact on the public purse is even lower, as costs are shared between the carers (family or friends), the government and voluntary workers.

Table 4 Admissions by type of care Type of care

Nursing homes Total hostel equivalent Hostels Aged care packagesa Total

1991-92

1992-93

1993-94

1994-95

%

%

%

%

61.4 38.6 38.5 0.1

60.9 39.1 39.0 0.1

57.8 42.2 41.7 0.5

54.1 45.9 40.1 5.8

100.0

100.0

100.0

100.0

a Care packages were introduced in 1992.

Source: Department of Human Services and Health (1995a, p. 123).

Australia’s Ageing Population: How Important Are Family Structures?

23

In 1994-95, nursing home costs were $24 162 per place, close to four times the cost of hostels or their equivalents (table 5).15 However, Goss et al. (1994) noted that a significant proportion of nursing home costs, although classified as ‘health’ costs, refer to food and accommodation costs. The same applies to hospital costs.

Table 5 Federal government costs per person using nursing homes, hostels and home and community care services Type of care

Nursing home beds Hostel (or equivalent) placesa Hostels Aged care packages Home and community care services

1991

1992-93

1993-94

1994-95

$

$

$

$

22 573 4 427 4 427 – 2 421

22 727 5 048 5 048 5 338 na

23 038 5 479 5 479 5 464 na

24 162 6 169 6 060 6 924 na

aCalculated as weighted average, using shares in table 4. – Not applicable as the scheme was introduced in 1992. na Not available.

Sources: EPAC (1994a, p. 77) for 1991; Department of Human Services and Health (1995a, p. 138) for 1992-93 onwards.

Overall, the trend away from institutionalisation for aged people is likely to continue, partly because it is preferred by the elderly and partly because it is less demanding of public funds.

2.3 Policies with a family focus Aimed at strengthening families

Several recently announced policies aim to relieve the financial pressure on low and middle income families bringing up children (box 2). Such measures may reduce the pressures of child raising, either financially for low income families or in terms of stress for families where parents

15

The cost of home and community care services per person is not available as yet, except for 1991. However, the Department of Human Services and Health (1995b) provided examples of unit costs per service provided — for example, $30–40 an hour for aged care services in 1991-92 and $4 a meal for the Meals on Wheels service in 1991.

24

Discussion Paper No. 19

juggle the responsibilities of work and children. However, if the measures also influence people’s decisions to have a first, second or subsequent child, Australia’s natural birth rate may rise. As a result, the proportion of aged in the total population will decline. Studies have found that, for Australia, increasing the natural birth rate would be more effective in reducing the ageing of the population than would higher immigration (see section 3.1 and Young 1990). However, as seen in section 1.3, EPAC (1994a) noted that productivity growth, even more than population policy, was a key to dealing with the ageing of Australia’s population.

Box 2 Selected policies in the 1996-97 budget affecting families Providing financial assistance to low and middle income families through: • the family tax initiative, which aims to assist over 2 million Australian families to the extent of $700–1300 a year for those with one to four children; • incentives of up to $450 a year for families with private health insurance; and • $452 million allocated to retain operational subsidies to family day care, occasional care and outside school hours care;

Increasing the financial responsibility of high income families through: • the imposition of a 1 per cent surcharge on the Medicare levy on families with incomes over $100 000 a year and without private health insurance; • a lower cut-off for child care assistance for high income families; and • a reduction from 30 to 20 per cent in the child care cash rebate for families with incomes over $70 000 a year.

Providing preventive health initiatives through: • the Health Throughout Life Program, which includes a range of preventive health initiatives of direct benefit to families (maternal and child health, injury prevention and diabetes).

Strengthening families through: • an additional $11.85 million over the next three years to expand marriage and relationship education services. Source: Budget Paper (1996e).

Australia’s Ageing Population: How Important Are Family Structures?

25

Aimed at the aged … patterns of care for the elderly are changing and … we are only beginning to understand and adapt to these. (OECD 1994, p. 4)

In a comprehensive review of age related policies among its member countries, the OECD (1994) found that, while there was a growing understanding in many countries of the need to share informal care of the aged between families and outside services, this was rarely reflected in government policies, except in Australia, Sweden and the United Kingdom. The range of possible policy initiatives mentioned by the OECD (1992) include: • respite centres on a day or longer term basis, so as to provide affordable relief for family carers; • home monitoring and alarm systems (backed up by around-the-clock home care and ambulance services), the installation or running costs of which may be subsidised; • the extension of outreach services, such as home health care and meal delivery to elderly residents in the community; • greater use of nonprofit and voluntary organisations, especially for contracted out and complementary services; and • financial support for informal caregivers. Other policy options mentioned by the OECD (1992) with indirect implications for families include a greater involvement of the private sector by, for example, offering new insurance schemes covering institutional and home care, and the conversion of home equity into a source of income. The OECD noted that several member countries were supporting experiments that included reverse mortgages, home sales with lease-back provisions and the use of home equity to cover the costs of publicly provided social services. In Australia significant progress has been made on policies with direct implications for families. Reports in the early 1980s that highlighted the dominance of expenditure on residential care compared with home care, and stressed the incapacity of the then community care services to respond to individual needs, were acted on. The Home and Community Care Program was introduced in 1985 and is now seen as having had a

26

Discussion Paper No. 19

significant and generally positive impact on community care in Australia. A recent report on the program by the House of Representatives Standing Committee on Community Affairs (1994), noted that: • the range of services has expanded substantially to include day care centres, other forms of respite care, referral and advocacy services, support groups for carers and services for special need groups; • by 1994, around 215 000 people were using the services of the Home and Community Care Program each month; • joint federal and state funding levels increased from A$152 million in 1984-85 to an estimated A$564 million (current prices) in 1992-93; and • coordination of community care services had improved substantially. The coordination and integration of services across government agencies was highlighted by the OECD (1994) as generally needing considerable improvement. Several interesting questions arise about likely responses to Australia’s recent policy changes. • To what extent are the Home and Community Care Program and the carer support programs substitutes for residential care, given that it is people living alone and without carers who are most likely to need residential care? • To what extent will entry fees and means-tested recurrent fees for aged care facilities increase the aged’s use of public hospitals? • What incentives do recent policy changes provide for people to rearrange their affairs so that they remain (or become) eligible for the age pension?

3

Previous approaches to studying ageing

The quantitative studies referred to in chapters 1 and 2 generally relied on traditional methodologies. In this chapter traditional methodologies used in population projections, in the study of family linkages and in linking ageing with social expenditure are described briefly.

Australia’s Ageing Population: How Important Are Family Structures?

27

3.1 Population projections Population projections are an essential first step in determining the rate at which a population is going to age. There are many techniques that can be used, ranging from simple extrapolation to broad economic, social and time series analyses to detailed component methods. The bases of many of the Australian studies reported earlier are the population projections published by the Australian Bureau of Statistics, which use the cohort-component method.16 This method begins with base populations of males and females, by single years of age. By applying assumptions about future migration and mortality rates, the population can then be advanced year by year. Assumed age-specific fertility rates are applied to the female population of child bearing ages to provide the new cohort of births (ABS 1996b, p. 136). The ABS projections are useful for studying issues associated with ageing, as they provide year-by-year profiles, usually over a 50-year period, of Australia’s male and female populations. However, they do not contain information on households, family linkages and income units.17 Such information is important if the housing needs of the aged, their wealth transfers to the next generation, and the availability of family members as carers are to be studied. In some previous studies such information was imputed or superimposed on the basic population projections. In addition, the ABS projections do not identify immigrants by the visa category under which they entered Australia. Thus, the effects of shifts in government immigration quotas in favour of particular categories cannot be analysed through ABS data alone. Such effects, however, could be important for projecting the pace of the ageing of the population, the rate of unemployment and changes in the pattern of family linkages. NATSEM analyses using data from the Longitudinal

16

For example, several studies by the Retirement Income Modelling Task Force used either ABS population projections or projections obtained through similar techniques. EPAC (1994a) derived its population projections using techniques similar to those of the Australian Bureau of Statistics. 17 Although not used in its population projections, the ABS has published statistics on families from its family survey conducted in 1992 (see ABS 1993a and ABS 1995b).

28

Discussion Paper No. 19

Survey of Immigrants to Australia (BIMPR 1996) show that the family compositions of people entering Australia under the ‘Business’ visa category are very different from those entering under the ‘Family Reunion’ categories. Within the former 22 per cent of immigrants entered Australia alone and 78 per cent as families of two or more while, within the latter, 70 per cent came alone and 30 per cent as families of two or more. Also, six months after arrival, around half of those who entered as ‘Business’ or ‘Independent’ immigrants had a job, compared with a third of ‘Family’ and around 5 per cent of ‘Humanitarian’ immigrants. These differences could have significant implications for ageing and public expenditure, especially under a ‘high’ immigration scenario. Young (1990), using similar projection methods to those of the ABS, addressed the question of whether it was a higher birth rate or immigration that was more effective in retarding the ageing of the population. Two immigration scenarios were considered — 50 000 and 150 000 a year over a 60-year period. In line with ABS assumptions, the age distribution of immigrants was assumed to continue unchanged from its average over the years 1985 and 1986. Young found that an increase in the birth rate was more effective in retarding ageing than an increase in immigration levels. Immigration, while having a significant effect on Australia’s total population, had little impact on its age structure either in the short or the long run. Young also found that, for a given target population, the best way to produce the youngest population was to maintain the birth rate at near replacement and have a low level of immigration. Young also noted that most European countries chose assistance to families rather than higher immigration. However, as noted earlier, EPAC (1994a) considered that productivity growth and income distribution were more important than population policies in dealing with the ageing of Australia’s population (section 1.3).

3.2 Family linkages ‘Although kin availability has been much discussed, it has been poorly documented …’ (Myers 1992, p. 58). This is true not only for kin

Australia’s Ageing Population: How Important Are Family Structures?

29

availability, but also for the extent to which family members assist each other within and across generations. Information tends to be most readily available on spouses of the elderly and on those living alone because such information can generally be obtained from censuses. Myers (1992) reports on demographic models — mathematical or simulation — that have been used to estimate the number and types of relative that may be available to older persons in the future.18 Myers cites several studies that projected kin availability under stable population conditions — that is, fixed regimes of mortality and fertility. These have been used to estimate the number of daughters that would still be alive at various older ages of a mother, as well as siblings, granddaughters, nieces and cousins. A key finding was that, under a scenario of an exact replacement birth rate (that is, a static population), many people would have no (or few) siblings, aunts, uncles, cousins or grandchildren. Myers also reported on a model of household development that had been used to indicate the joint effect of both declining fertility and mortality. Using this model, the ratio of parental dependence over a 75year period was shown to increase more than fivefold for 40–50 year old offspring. Myers concluded that, while dynamic demographic developments constrained the conditions under which societies evolved, the nature of the support systems was influenced by many social and cultural factors as well. Overall, studies of the support that family members are able and willing to provide to each other have been few and have tended to be confined to the effects of likely demographic developments.

3.3 Ageing and social expenditure It is apparent from the preceding discussion that projections of the social expenditure associated with Australia’s ageing population need to be carried out in a broad, nationwide framework. Demographics can throw some light on the extent to which ageing is likely to occur, as well as on the likely size of the working population, which would be one factor in

18

See also King and Singh (1994), who studied family structures in Australia.

30

Discussion Paper No. 19

determining the amount of taxation revenue government would be able to collect. Another key factor that could lead to higher taxation revenue is higher living standards, achieved through productivity improvements. A further factor is the age at which people tend to retire. The longer older people remain in the workforce, the longer they contribute to taxation revenue and the less likely they are to rely on the social safety net.19 However, projecting social expenditure (health, age pensions, housing assistance, unemployment benefits, education, etc.) within a broad framework would be very complex. Because of this, much of the literature considers one element of social expenditure only, assuming that past trends will continue for other variables (for example, that GDP and taxation revenue will continue to grow at rates observed in the past). Projections of elements of social costs

The Eckermann (1992) and Goss et al. (1994) studies are of one element of social expenditure. They compared two ways of projecting health expenditure — the conventional method, which assumes that current real health expenditure by age group remains constant over time, and an adjustment to the conventional methodology, which differentiates the health expenditure of survivors from the expenditure of those who are in their last two years of life. The adjustment is desirable because life expectancies are likely to continue to rise and because health expenditure among the elderly is more a function of the time of death than of chronological age. For example, in 1989-90, health expenditure per 70–80 year old who survived more than two years ranged between $4000 and $7000; for those who had less than two years to live it was between $15 000 and $22 000 (Goss at al., fig. 2.8). Eckermann concluded that the

19

Note that the issue of older people in the labour force is a complex one. On the one hand, it is common for older people who wish to work not to be able to find jobs (see section 1.5). On the other hand, it is sometimes argued that, in a tight labour market with high unemployment, a job for an older person is likely to mean unemployment for a younger one. Some evidence that this view is fairly widely held in Australia (that is, that early retirement would free up jobs for younger workers) was found in a study of the attitudes of pre-pension age persons (see Department of Social Security 1996, p. 31).

Australia’s Ageing Population: How Important Are Family Structures?

31

traditional methodology would overestimate the health expenditure per aged person by 4–8 per cent by 2030. Cost projections relative to government budgets

A study with a much wider scope, covering 20 countries, is reported by OECD (1996). It simulated the impact of ageing on public pensions and health expenditure between 1995 and 2030, assuming that current government taxation and expenditure policies continued. The impact of ageing on fiscal balances and on national savings was also assessed for the 20 OECD countries studied. The study made use of demographic projections prepared by the World Bank. The key aim of the study was to provide some idea of the magnitude of likely future problems if no policy action were taken. The modelling method used, although close to traditional methods, was based on more aggregated data than in other studies reported in this paper, and relied on a greater number of broad simplifying assumptions. The underlying economic scenarios were derived on the basis of the following assumptions: • economies followed their medium term growth paths (no cyclical unemployment); • medium term growth was determined by the projected growth of the working-age population and an assumed labour productivity growth rate of 1.5 per cent a year; and • participation rates were assumed to remain unchanged. For each country, separate models were developed to simulate public pension and health expenditure. For pensions in Australia, for example, an eligibility ratio and average benefit were calculated from the total number of pensions paid and total pension payments reported for 199495 in the budget papers. Future retirees were then assumed to receive the same initial benefit as a proportion of future average wages, with pensions for each new cohort of retirees being indexed to inflation. The announced increases in the retirement age for women were also incorporated in the model. Of the four pension scenarios considered, two were of particular relevance to Australia. In the first, the ‘later retirement scenario’, the age

32

Discussion Paper No. 19

of entitlement to benefits was gradually raised to 70 years and it was assumed that every year before that was spent working and paying contributions. Under this scenario, the upward pressure on pensions due to ageing could be largely offset in most countries. In the second case, the ‘targeting scenario’, the ratio of benefits to wages was held constant from 2010 onwards, but the proportion of the elderly population qualifying for a pension was gradually reduced to 30 per cent. This scenario resulted in a large reduction in public expenditure. For Australia, pension expenditure by 2030 as a percentage of GDP was simulated to be 2.4 per cent under the ‘later retirement scenario’ and 1.7 per cent under the ‘targeting scenario’, compared with 3.8 per cent if policies did not change. The actual ratio for 1995 was 2.6 per cent. For health expenditure, two approaches were modelled. The first assumed that as people grew older they consumed more health care. For this case the cost to government of providing health care services to the elderly was obtained by multiplying per person public expenditure on health by the total number of elderly people. For Australia, public health care costs, which amounted to 5.8 per cent of GDP in 1995, were simulated to be 7.6 per cent by 2030, assuming that health expenditure grew at the same rate as GDP. The second approach assumed that consumption of health care was concentrated in the period immediately before death. For this case per person public expenditure on health care was multiplied by the number of deaths among the elderly population. For Australia, public health care costs were simulated to rise to 6.2 per cent of GDP by 2030, assuming that health costs grew at the same rate as GDP. For most countries, including Australia, the first assumption generated the higher total health care cost. The fiscal framework for the simulations was provided by the medium term scenarios published by the OECD in its Economic Outlook 59. The findings were that countries with relatively good fiscal positions were able to offset some of the effects of an ageing population, while for others, the ‘pure ageing effect’ led to an even larger accumulation of net debt. At around 35 per cent of GDP, the ‘pure ageing effect’ on net financial liabilities for Australia was the third lowest among the 20 countries studied, the highest being close to 200 per cent of GDP for Finland and Japan. To estimate the effects of ageing on private savings, the OECD study used the IMF coefficient for industrialised countries for demographic

Australia’s Ageing Population: How Important Are Family Structures?

33

effect. For government savings, the long term scenario for government budgets developed by the OECD Secretariat was used. Then, assuming a 50 per cent Ricardian equivalence effect for both government and private savings, national savings rates were simulated to decline between 2000 and 2030 by around 8 percentage points for the OECD as a whole. This result is considered by the authors ‘inconclusive’ (in terms of the size of the effects) since it is significantly lower than those predicted by macroeconomic models based on life cycle saving and neoclassical growth theories. For Australia, the OECD (1996) found that net national saving, which was 1 per cent of GDP in 1995, would be between -1 and -2 per cent in 2030, depending on the assumed change in the dependency ratio. The study concludes that the higher the savings, the greater the scope for dealing with the adverse effects of an ageing population. Social cost projections relative to GDP

Australian studies using traditional methodologies with a wider scope include analyses by EPAC (1988, 1994a), Kelley (1988) and Creedy and Taylor (1993). They generated projections for broad categories of federal and state social expenditure (by age and sex, based on ABS population projections) and compared these with projections of GDP, taking into account labour force participation and unemployment rates, as well as labour productivity (that is, the ratio of GDP to employment). Creedy and Taylor, who used a more detailed breakdown of expenditure categories, described the complex set of data matrices and the complex relationships between the demographic, GDP and social expenditure variables that were modelled.20 With the relatively wide perspective of this study, Creedy and Taylor were able to estimate the net effect of ageing on social expenditure. For example, the higher proportion of older people in the population was associated with lower spending on unemployment benefits, employment and education as a proportion of

20

Creedy and Taylor (1993) considered the following breakdown for combined federal and state social expenditures, by nine age groups: age pensions, other aged assistance, unemployment benefits, other social security, health, education and employment. They noted that the projections were severely constrained by data availability. Social expenditure data could not be disaggregated by sex, various sources defined age groups differently, and it was difficult to find a detailed breakdown of state government expenditure.

34

Discussion Paper No. 19

total social expenditure and higher spending on health, age pensions and other age related assistance. Difficulties with disaggregation

Creedy and Taylor (1993) noted that further disaggregation would have been desirable in order to include, for example, consideration of changes in components of per person unemployment expenditure. This would involve allowing growth in total unemployment expenditure per person to vary with unemployment benefits, aggregate unemployment and the labour force participation rate. Creedy and Taylor also noted that growth rates of the ‘total’ and of its elements could not be set in isolation, as they were interrelated — hence the dilemma with this type of analysis, in part due to the use of traditional methods. If the effects of specific age related policy proposals are to be considered (for example, entry fees to nursing homes), there is a need for a fine breakdown of expenditure. The budgetary effects of the policies need to be considered in a nationwide framework. However, as the data matrices and their interrelationships are already complex, further disaggregation could make the analyses virtually unmanageable. On this point Butler (1996) noted that allowing for consideration of more individual characteristics and for all possible combinations of these would result in exponential increases in the number of cells in the data matrices. Importance of well designed sensitivity testing

Creedy and Taylor (1993) used sensitivity analysis to assess the effects of varying growth rates (which in the benchmark case were a constant 2 per cent a year for productivity growth and for real per person outlays in each social expenditure category). In the benchmark simulation the social expenditure to GDP ratio increased from 0.204 in 1988 to 0.274 by 2031, worth A$34 billion in 1996 dollars. When productivity growth was increased to 2.2 per cent a year and the growth in age related expenditure was reduced to 1.8 per cent, together with lower unemployment rates, the social expenditure to GDP ratio was reduced to 0.231 in 2031. In the reverse case (productivity growth reduced to 1.8 per cent and growth in age related expenditure increased to 2.2 per cent, together with higher unemployment rates) the ratio increased to 0.320. In 1996

Australia’s Ageing Population: How Important Are Family Structures?

35

dollars the projected increases in the social expenditure to GDP ratio between 1988 and 2031 would amount to around A$13 billion under the ‘optimistic’ scenario and to A$56 billion under the ‘pessimistic’ scenario. Creedy and Taylor noted the importance of the assumed productivity growth rate, it having contributed close to half of the difference between the ratios of the two scenarios. They also noted that, while the differences between the two scenarios were relatively minor in terms of the assumptions underlying the projections, the difference in their effects on government revenue requirements was considerable. Considering that, in recent years, growth in federal government outlays averaged 4 per cent a year in real terms for health and 14 per cent for pharmaceutical benefits (section 1.2), and that health costs in the United States where the private sector dominates are much higher than in Australia, the growth rates considered by Creedy and Taylor for age related costs (that is, 1.8 and 2.2 per cent a year) seem low. Also, the assumed productivity growth rates need to be considered within the context of historically achieved rates, as well as expected future rates. Studies published to date do not consider the total costs associated with the care of the elderly. That is, they do not combine the cost to government with private costs, such as care by family members, other elderly, volunteers and community organisations. Reasons for this include the difficulties of analysis and the lack of suitable financial and other data. In this regard, the mid-term review of the Aged Care Reform Strategy (Department of Health, Housing and Community Services 1991) recommended in 1991 that consideration be given to a single budget structure for aged care, integrating the control of the financing of both residential and community care programs. In general, recent policy initiatives have moved towards greater integration, at least of services (see sections 1.4, 2.2 and 2.3).

3.4 Summary of selected studies The findings of selected studies on ageing mentioned in this paper are summarised in table 6. Publications that report purely on statistical collections — such as most ABS publications — and government statements on ageing are not listed in the table. The list of studies starts

36

Discussion Paper No. 19

with the most recent, citing author and year. The ‘Section’ column in table 6 identifies the sections of this paper where the study is mentioned. Table 6 Summary of selected studies on ageing Study

Section Main subject of analysis

Method used

ABS (1996b)

1.1 3.1

Projections of the population of Australia: 1995 to 2051.

Derived projections using the cohort-component method, providing year-by-year profiles of the ageing male and female populations.

OECD (1996)

3.3

Ageing populations, pension systems and government budgets for 20 OECD countries: 1995 to 2030.

Simulated the impact of ageing on public pensions and health expenditure. Developed separate models for each country and used traditional methods of analysis with highly aggregated data.

Lloyd-Sherlock & Johnson (1996)

1.5

Global comparisons of ageing issues and related social policies.

Used various approaches for one developed country (Germany) and a number of developing countries (Chile, Thailand, China, India).

Department of Social Security (1996)

1.5

Early retirement.

Used various methods.

Retirement 2.2 Income Modelling Task Force (1996b)

Distributional analysis of Australian Used a comprehensive cohort superannuation. projection group model.

ABS & Office on Ageing (1995)

1.2

A statistical profile of older people in New South Wales.

Statistically analysed the older NSW population, reporting on a broader range of characteristics than most other studies.

OECD (1995)

1.5

Transition from work to retirement.

Described policy changes by OECD member countries that aim to improve the possibilities for older people to remain in paid jobs.

EPAC (1994a)

1.2 1.4 2.2

A series of comprehensive and interlinked analyses relating to Australia’s ageing society (some economywide, others detailing particular issues).

Derived demographically projected public expenditure arising from the ageing of the population, generally using traditional methodologies. Included statistics and discussions on the social aspects of ageing.

Goss, Eckermann, Pinyopusarerk

1.2 3.3

Health impact of ageing.

Used two methods to project health costs: the conventional (assuming constant real health costs by age) and a method that differentiated the

Australia’s Ageing Population: How Important Are Family Structures?

and Wen (1994)

37

health costs of aged ‘survivors’ from those close to death.

38

Discussion Paper No. 19

Key findings (as mentioned in this paper) Natural population growth (that is, no net overseas migration) was projected to steadily decline beyond 2031. The number of people aged over 65 years was projected to rise from 12 per cent of the population in 1995 to 23 per cent by 2051, and of those aged over 85 from 1 per cent to 5 per cent of the population. Greater targeting of pensions (that is, only 30 per cent of elderly qualifying) were projected to result in large reductions in pension expenditure (for Australia, assuming no other policy changes, the reductions were from 3.8 per cent of GDP in 1995 to 1.7 per cent by 2030). Health expenditure for Australia was projected to rise from 5.8 per cent of GDP in 1995 to 7.6 per cent by 2030, assuming that health costs increased with age (with lower estimates being obtained when concentrating health costs in 1–2 years before death). In relation to the 1989 pension reform in Germany (a country that has one of the most rapidly ageing populations, together with one of the most comprehensive public pension systems in Western Europe), the editors found that the 1989 reforms proved successful in spreading the pension costs of ageing across generations. A significant proportion of Australians retire from full-time work prior to the age pension age; part-time work for older Australians is increasing; females tend to stay in the workforce longer than males; and those on high incomes retire later than those in low paid jobs. Involuntary and family retirements were shown to outnumber voluntary retirements by three to one. The total value of assets of age pension age Australians in 1992-93 was estimated to have been A$305 billion, some A$193 billion of this being home equity.

In 1993, close to 70 per cent of people aged 65 and over and living in New South Wales relied on the age pension as their main source of income; some two-thirds of those who helped sick or disabled adults were aged 55 and over; voluntary work was much more important for the elderly than paid work; and over half of all informal child care was provided by a grandparent. Some of the policy ‘experiments’ appear to have been successful (for example, in Sweden, where a partial retirement scheme has been successfully built into the national social security system).

Total social expenditure was projected to rise from 21.3 per cent of GDP in 1990 to 25.2 per cent by 2051 without the Superannuation Guarantee Charge, and to 23.8 per cent of GDP with the charge. Health expenditure was predicted to rise from 8 per cent of GDP in 1990 to 11 per cent by 2051. In 1989-90, public expenditure on aged health care was comparable to expenditure on age pensions. Over 60 per cent of older people offered support to family in the forms of child minding and care for others when sick. In 1989-90 those aged 65 and over were responsible for 33 per cent of Australia’s total health expenditure, but accounted for only 11 per cent of the population; and health expenditure per 70–80 year old who survived for more than two years ranged between $4000 and $7000, compared with $15 000 to $22 000 for those who lived for less than two years.

(Continued on next page)

Australia’s Ageing Population: How Important Are Family Structures?

39

Table 6 Summary of selected studies on ageing (continued) Study

Section Main subject of analysis

Method used

OECD (1994)

2.3

Caring for frail elderly people.

Comprehensively reviewed age related policies of OECD countries.

Creedy & Taylor (1993)

3.3

Ageing and social expenditure in Australia.

Based on ABS population projections, estimated federal and state social expenditures at a detailed level, and compared these with projections of GDP, taking into account labour productivity, labour force participation rates and unemployment rates.

Eckermann (1992)

3.3

Health expenditure and ageing: revised methodologies.

See Goss et al. (1994) above.

Myers (1992)

3.2

Demographic ageing and family support for the elderly.

Reported on models that estimate the number and type of relatives that may be available to provide support for the elderly.

Young (1990)

2.3 3.1

Policy options for Australia’s ageing Derived projections using the population. cohort-component method, similar to ABS (1996b).

3.5 Impact of studies The brief overview of the extensive literature on ageing presented in this paper, and of the methods of analysis generally used, suggests that previous studies have not only proved useful to policy makers, but also contributed to the introduction and subsequent improvement of age related policies. This is despite the fact that analyses of complex real life problems invariably require simplification and assumptions, and that attempting projections several decades ahead is particularly ‘hazardous’. The overview also suggests that although Australia has a relatively young population it has moved in line with, if not ahead of, most OECD countries in implementing age related policies. However, while recognising this, both EPAC (1994a) and the Productivity Commission (1996) noted that there is considerable scope for improvement. In its stocktake report the Productivity Commission (1996) described a hierarchy of broad reforms in several areas, including health and

40

Discussion Paper No. 19

Key findings (as mentioned in this paper) While understanding of the need to share informal care between families and services was growing in many countries, this was rarely reflected in government policies, except in Australia, Sweden and the United Kingdom. With growth in productivity and real per person outlays in each social expenditure class set at 2 per cent a year, the social expenditure to GDP ratio was projected to rise from 0.204 in 1988 to 0.274 by 2031. Sensitivity testing highlighted the importance of the assumed productivity growth rate. Small differences between scenarios (that is, relatively minor changes in the assumptions underlying the projections) were found to result in relatively large effects on governments’ revenue requirements.

See Goss et al. (1994) above. Eckermann concluded that the traditional methodology would overestimate health expenditure per aged person by 4–8 per cent by 2030. Under a scenario assuming a birth rate giving a static population, many people would have no or few siblings, aunts, uncles, cousins or grandchildren. When considering the joint effect of both declining fertility and mortality over a 75-year period, it was found that 40–50 year old offspring would see a more than fivefold increase in the ratio of parental dependence. Increases in the birth rate were more effective in retarding ageing of the population than higher levels of immigration. Immigration, while having a significant effect on Australia’s total population, had little impact on its age structure.

community services. In relation to aged care, the Commission noted that the proposals currently being developed by the Coalition of Australian Governments to draw together health and related community services so that they are more responsive to clients’ needs and provide better continuity of care should lead to more effective and coordinated delivery of aged care services. It also referred to a suite of measures proposed to strengthen the contributions made by charitable organisations in delivering community services.21 It stressed, however, that such reforms were only illustrative of the directions that should be considered, noting the need for a more detailed review. The Productivity Commission (1996, p. 100) stated that: ‘A detailed review is particularly important to ensure adequate recognition of the need for all Australians to have access to quality health care’. 21

The Industry Commission (1995) lists these as greater focus on service quality and on community involvement, and closer partnership with governments.

Australia’s Ageing Population: How Important Are Family Structures?

41

4 Approaches using dynamic microsimulation 4.1 Characteristics of dynamic microsimulation It is its ability to consider the life cycles of Australians individually, rather than in broad statistical groupings, that most distinguishes dynamic microsimulation from more traditional methods. Through its capacity for detailed analyses of policy initiatives that affect people — not only now, but over their and their children’s lifetimes — the technique of dynamic microsimulation has the potential to provide additional insights to those gained using traditional techniques. Carrying out analyses at the individual level has the advantage of allowing assessment of the distributional and other effects of policies at any level of aggregation — the individual; families; groupings by age, sex, work status, incomes or assets; and the population. In addition, an intergenerational perspective is important when considering the impact of policies related to ageing because issues such as the capacity of the younger generation to pay taxes, whether the elderly have living spouses, children or relatives, the extent to which family members can or are willing to provide caring, and whether assets are used to pay for aged care or are inherited would need to be included in the analyses. When considering complex policies, being able to take into account a range of factors simultaneously becomes important. For example, when analysing the effects of introducing entry fees to nursing homes, being able to use an integrated model of key life cycle choices and events (such as marriage, births, study, work, earnings, savings, retirement, health, disability and death) would be preferable to using the traditional approach of examining each of the factors affecting the new policy separately. While attempts had been made in some earlier studies to consider several factors simultaneously, the computations tended to become very complex, especially when disaggregation was required so as to allow analysis of the effects of specific policy proposals (section 3.3). In describing a proposed dynamic microsimulation model for Australia’s health sector, Butler (1996) made similar points. When analysing policies related to ageing, dynamic microsimulation offers the following advantages over traditional approaches:

42

• • • •

Discussion Paper No. 19

a greater number of factors affecting policy can be taken into account simultaneously; considerably greater disaggregation at the policy and individual characteristic levels is possible; policies that have impacts across generations can be considered; and the distributional effects of the policies can be assessed, both at the level of the individual and across generations.

Disadvantages of dynamic microsimulation include the limited data available for this type of analysis, the computational difficulties associated with the size of such a detailed nationwide model, and the time needed to develop such models. A comprehensive discussion of the problems and prospects of dynamic microsimulation models can be found in Harding (1990) and an example of a recent policy application of such a model in Harding (1996a). A range of other publications are also relevant — see, for example, Harding (1996b, 1993) and Orcutt, Merz and Quinke (1986). 4.2 Current status of dynamic microsimulation modelling at NATSEM In its present form, NATSEM’s dynamic microsimulation model, DYNAMOD-2, generates life cycle profiles and cross-sections of Australia’s resident population at the individual and family levels. Starting with a 1 per cent sample of Australia’s population in 198622, DYNAMOD-2 simulates the life cycle events of individuals, in terms of: • education at primary, secondary and tertiary levels; • labour market experiences, together with associated earnings; • first and subsequent marriages, and de jure or de facto unions; • first and subsequent marital births, as well as premarital births; • couple dissolution and divorce after separation; • onset of and recovery from disability23; and • mortality. 22

The 1 per cent unit record sample file from the 1986 census, which comprises some 156 000 individuals, is complemented with imputed information for variables that are not in the census (for example, certain education and earnings variables). 23 Disability rates increase markedly after age 50.

Australia’s Ageing Population: How Important Are Family Structures?

43

The model allows for immigrants and emigrants. Links between spouses and children living with their parents are maintained, and the number of children each individual parent has had is known. An extensive range of output variables is produced at the individual and family levels (box 3) and can be drawn on, depending on the policy application to be considered. This is one of the key advantages of dynamic microsimulation over traditional methods. For example, it can be used for a range of analyses, such as changes in the age profile of the population, changes in family patterns, or the budgetary effects of various options for the Higher Education Contribution Scheme. It is much easier to extract variables of relevance to a particular application, and aggregate from the individual to the level desired, than to disaggregate from projections carried out at a highly aggregated level, as is the case with traditional analyses (section 3.3). However, dynamic microsimulation models require considerably longer development time and much greater computational power than traditional models. For example, development of NATSEM’s dynamic microsimulation model commenced in 1993 and simulations currently run for around 8 hours on a large scientific work station. 24 DYNAMOD-2 has been designed primarily to allow analysis of possible changes to the Higher Education Contribution Scheme and AUSTUDY (means-tested federal government grants available to people studying). NATSEM has provided preliminary model output to the Department of Employment, Education, Training and Youth Affairs, with variables specially selected for a HECS application25, together with a HECS– AUSTUDY interface, so that the effects of various policy alternatives can be analysed. While Harding (1996a) used a model for a single cohort, DYNAMOD-2 covers all cohorts.

24

NATSEM’s scientific work station has two 150 MHz processors. DYNAMOD-2 requires 128 MB of RAM. 25 Variables of interest included study status, institution type (where studying), highest educational qualification, earnings profile and date of death (if applicable) (see box 3).

44

Discussion Paper No. 19

Box 3 Selected output variables in DYNAMOD Description

Comments

Demographics — personal characteristics Date of birth Date of death If applicable at end of simulation Identification numbers For person, family and income unit Sex Male, female Country of birth New Zealand, United Kingdom, Other Europe and the former USSR, West Asia and Middle East, Other Asia, South America, Other America, Africa State of residence NSW, Vic., Qld, SA, WA, Tas., NT, ACT Disabled Whether disabled or not Demographics — family characteristics Relationship in family Reference person, spouse, child (dependent or not), other family Child status Natural, step (reference person or spouse), other Marital status Married, separated, divorced, widowed, never married Previous marital status Married, separated, divorced, widowed, never married Number of de jure unions Number of de facto unions Number of births At end of simulation period Education Highest qualification

Study status School level Institution type

Labour force Labour force status Employment status Occupation Employment sector Industry sector Hours worked Earnings

Not yet at school, still at school, didn’t finish secondary, year 12 certificate or equivalent, trade certificate, trade diploma, bachelors degree, graduate diploma, masters, higher Full time, part time Pre-year 1, years 1, 2, …,12 School (government, catholic, other), TAFE, university (bachelor, graduate diploma, masters or higher) Employee, self-employed, employer, unemployed, not in labour force Full time, part time Manager, professional & paraprofessional, tradesperson and clerk, sale, labourer Private, government Agriculture, manufacturing, services Per week Weekly wages

Australia’s Ageing Population: How Important Are Family Structures?

45

Considerable work to extend DYNAMOD-2 and further develop NATSEM’s capability in dynamic microsimulation is either under way or planned. The extensions involve completing DYNAMOD-2 by allowing for a greater range of family linkages, and modelling the employment experience of full-time students. Subsequent development will add the capacity to model asset accumulation (including housing wealth) and intergenerational issues, as well as an improved ability to take account of overall economic circumstances. Development work is planned in three stages (table 7). Table 7 Plans for further development Stages

Development work

Examples of age related applications

1 DYNAMOD-2

Extend family linkages (for example, Study extended family linkages to parents of the reference person). and their implications for Model employment patterns of full- ageing. time students.

2 GENSIM

Model asset accumulation, including Study issues relating to superannuation, health, wealth, retirement incomes and housing. housing and inheritance. Model health status. Allow for greater user flexibility of demographic processes.

3 Macro–micro link

Account for intergenerational effects (for example, in studies of ageing).

Link dynamic microsimulation model Study the overall effects on to economywide model. budgets (estimate revenue in conjunction with expenditure, under different demographic and economic growth scenarios).

Broadly, stage one aims to bring DYNAMOD-2 to a stage where it will be useful for analysing a range of specific policy issues. At this stage, as well as analysing the effects of policy options concerning the Higher Education Contribution Scheme, AUSTUDY and international migration, the pattern of family linkages across generations could be estimated under different demographic scenarios, shedding some light on the extent to which family members could be available to support the elderly when needed. Stage two would tap into the greatest advantage dynamic microsimulation has over traditional methodology: the ability to account for wealth

46

Discussion Paper No. 19

accumulation and the transfer of wealth across generations, and to explore interactions to a far greater extent than is possible with DYNAMOD-2. Work to date on this stage is described in Bækgaard and King (1996) and King and Bækgaard (1996). With GENSIM the combined effects of, for example, various superannuation options, savings incentives and housing policies could be analysed simultaneously and across generations. For example, it would be possible to study the combined effects on the future aged of a number of 1996-97 budget announcements, such as the changes in HECS and AUSTUDY conditions, the incentives for people to take out private health insurance, the incentives for older people to have healthier lifestyles and stay in the workforce longer, the greater incentives for carers, and allowing nursing homes to charge an upfront entry fee. Such an analysis would be able to take account of a wide range of interrelated factors. For example, as younger people make greater contributions to the costs of their health and education, they are less able to save; if older people take greater responsibility for the costs of their aged care, the size of the inheritance they transfer to the next generation will be lower; and when poorer people enter a nursing home, they may no longer require housing assistance from government. GENSIM will also allow users to have greater control over the assumptions underlying its demographic projections. While the survival functions on which DYNAMOD-2’s couple formation and fertility processes are based have the advantage of considering several individual characteristics simultaneously, they do not allow for exact matching of traditional assumptions in projections, such as changes over time in fertility rates. Thus, simulations that adopt exactly the same assumptions as, for example, ABS (1996b) are not easily obtained with DYNAMOD2.26 To improve the flexibility of user control, GENSIM will make greater use of conventional transition probabilities in its demographic module. Stage three involves linking NATSEM’s dynamic microsimulation model to an economywide model. NATSEM currently has a grant from the Australian Research Council for a project on macro–micro model linkages.

26

However, higher or lower fertility in a broad sense can be modelled by altering the relevant parameter in the survival functions.

Australia’s Ageing Population: How Important Are Family Structures?

47

With a combined macro–micro model, it would be possible to extend the analyses mentioned in relation to stage two. For example, it would be possible to directly account for likely future increases in living standards (as measured by GDP per person) and for the impact this would have on government revenue. For example, if people took greater responsibility for the costs of their own health, education and aged care, government spending in these areas would be lower and tax rates may decline as a result. Alternatively, the extent to which taxation rates would need to increase to finance certain welfare initiatives could be directly estimated under different demographic and economic growth scenarios.

4.3 Microsimulation well-suited to distributional analysis Regardless of the stage of development of the dynamic microsimulation model, its results could be used for distributional analyses at various levels of aggregation. As seen in section 1.3, income distribution is considered one of the key policy tools in dealing with the ageing of Australia’s population. With dynamic microsimulation, the immediate cross-sectional income distributional effects of policies can be compared with their impact on people’s lifetime incomes, as well as with lifetime income patterns across generations. Such comparisons would be particularly valuable when analysing policy options that cut across generations, such as the financing of nursing home costs either through the assets of the elderly or through increased taxes.

4.4 Examples of possible policy relevant analyses So what practical value would a dynamic microsimulation model similar to the one described above have for policy makers concerned with the ageing of Australia’s population? Based on sections 4.1–4.3, dynamic microsimulation would be particularly suited to addressing the following policy relevant questions.

48

Discussion Paper No. 19

1. How would various forms of superannuation legislation affect people’s wealth after retirement, the inheritances they are likely to pass on to the next generation, and the taxation rates required to balance government budgets? 2. What would be the long term impact of policies encouraging later retirement? 3. What impact would the answers to questions 1 and 2 have on taxation revenue and social expenditure? 4. What proportion of elderly people are likely to require aged care in 20 or 30 years time? 5. What proportion of people requiring aged care are likely to have family members available as potential carers? 6. How would policies that assist carers affect the number of carers available, as well as the demand for nursing home places? 7. What proportion of spouses are likely to be affected by policies that consider the family home as a source for financing nursing home care? 8. What would be the impact in 10, 20 and 30 years of policies that successfully encourage healthier lifestyles? 9. How would the policies cited in the above questions affect the distribution of income generally, as well as across generations? 10. What would be the long term effect of implementing all of the policies cited in the above questions, either simultaneously or in a phased fashion? These are only a selection of possible age related policy questions that dynamic microsimulation would have advantages over traditional methods in addressing. As with all analyses of possible future policy options, assumptions would need to be made and the results would be in the form of projections of what might happen given an assumed scenario. While new techniques would not overcome the uncertainties associated with analyses concerning the future, microsimulation could significantly broaden the range of age related policy issues that analysts are able to quantitatively explore.

Australia’s Ageing Population: How Important Are Family Structures?

5

49

Summary

Although proceeding more slowly than in other OECD economies, the ageing of Australia’s population is causing concern. The proportion of the population over 65 years old is expected to double by 2051, and the proportion over 75 to increase fivefold. The age dependency ratio (that is, the ratio of elderly to those of working age) is projected to increase from 20 to 40 per cent, with much of this occurring between 2010 and 2030. There are fears that, if we do not act now, the financial burden on our children and grandchildren in the decades ahead will be too high. Currently, Australia’s elderly rely heavily on governments for financial support. Close to 70 per cent rely on the age pension as their main source of income and they are responsible for around a third of Australia’s total health expenditure. Of immediate concern is that, in recent years, federal government outlays on health have increased more rapidly than living standards generally, and that it is difficult to meet such increases when fiscal constraint is required to stimulate growth and reduce unemployment. While past studies have tended to focus on the cost implications of ageing, some have highlighted the importance of economic growth in being able to cope with such costs. They pointed out that it is much easier to adequately care for a growing number of elderly when productivity is growing rapidly and incomes are rising. EPAC predicted a near doubling of real GDP per person between 1994 and 2051, assuming productivity growth of 1 per cent a year. Such an increase in GDP per person — an indicator of living standards — would result in similar increases in taxation revenue without any changes in taxation rates. Over the past 50 years, living standards in Australia, as in the Western world generally, have risen dramatically. Young Australians achieved living standards their parents had never imagined. However, over the past couple of decades Australia’s economic performance has declined relative to the performances of other developed countries, and is now below the OECD average. But there is scope for significant improvement through already proposed productivity related reforms. Estimates suggest possible real GDP gains of between A$24 billion and $97 billion a year. With technological breakthroughs this potential could be considerably greater.

50

Discussion Paper No. 19

Productivity growth and income distribution, even more than population policies, have been identified as key variables when addressing the issues associated with the ageing of Australia’s population. Looking at the cost side, Australia is following the OECD-wide trend to shift the financial responsibility of ageing from governments to individuals. The Superannuation Guarantee Charge, the greater targeting of the age pension and of superannuation taxation concessions, entry fees to nursing homes, taxation incentives for those with private health insurance, and the increase in the age range within which people can work, all support this shift. Other policies of relevance adopted in some OECD countries include new insurance schemes offered by the private sector to cover institutional and home care, the conversion of home equity into a source of income, reverse mortgages, home sales with lease-back provisions and the use of home equity for the cost of publicly provided social services. Although not generally taken up in Australia, measures of this kind allow, for example, the use of patients’ assets to cover nursing home entry fees without the fear of having to leave a spouse homeless. Other cost related policies focus on prevention since, if older people remain active and lead healthy lifestyles, they are more likely to undertake paid or voluntary work, be better able to remain independent and continue living in their homes, and be less likely to use health and age related services intensively. Yet other incentives provide greater support for carers (mainly elderly spouses, family and friends) so that they are not excessively disadvantaged compared with those in paid jobs and the need for institutional care of the elderly is reduced. These latter policy initiatives are in line with trends in most OECD countries, where policies aim to stimulate private, non-profit and voluntary sector responses. Studies in many countries show that, although the family is declining in importance, it remains the most important provider of care for the elderly. In Australia in the early 1990s wives accounted for 40 per cent of the carer population, while husbands, daughters and sons accounted for 33, 16 and 5 per cent respectively. In return, older people provide support to younger members of their families, with a third of Australian children under 12 in the late 1980s being cared for by grandparents, aunts, other relatives and friends. Of those elderly living alone, over half had adult children living within half-an-hour’s travelling time.

Australia’s Ageing Population: How Important Are Family Structures?

51

Continued rises in living standards since World War II have allowed the elderly to accumulate wealth of significant proportions. Recent estimates indicate that the assets of age pension age Australians are worth around A$305 billion, some A$193 billion being in home equity. Some expect that the biggest transfer of wealth ever seen in this country’s history will take place early in the next century. However, the extent to which wealth is transferred will depend on how successful governments are in shifting onto individuals the responsibility for the cost of their aged care. The greater the financial pressures on governments, the more likely it is that the elderly will be required to use their wealth to pay for their aged care. While total assets may not be affected by this at the aggregate level, there will be important financial and emotional effects at the individual level, as inheritance often involves the family home. Worldwide, the elderly have expressed a preference for being independent and remaining in their homes for as long as possible. In Australia, the increased importance home and community aged care services, combined with the cap placed on nursing home beds, has resulted in a decline from 61 to 54 per cent in the proportion of elderly admitted to nursing homes since 1992. In 1995, among people aged over 70 with a handicap, 15 per cent were in nursing homes, 12 per cent in hostels, 37 per cent received home and community care services and 36 per cent did not use government provided services at all. As far as governments are concerned, the cost differences between nursing homes, hostels and home care are considerable. Nursing home costs, at around $24 000 per place a year, are now close to four times the per person cost of hostels. With home care, the cost to government is even lower, as the costs are shared between the carers, the government and voluntary workers. Among OECD countries, Australia was found to be exceptional in the progress made in policy shifts toward informal and community based care, away from residential care. However, much still needs to be done, including improved coordination and integration of services across government agencies. Looking at the impact of the extensive literature on ageing, it is clear that past studies had proved useful to policy makers, not only in terms of the broad direction in which age related initiatives have developed, but also in formulating the details of particular policy measures. Thus, the traditional methods of analysis — based on the projection of aggregates

52

Discussion Paper No. 19

and only rarely allowing for interactions between variables — have served the growing number of elderly well. New methodologies, which have become possible only since the emergence of more powerful computing equipment, can extend the scope of earlier studies into areas of particular relevance to the ageing. One such methodology is dynamic microsimulation, which can extend previous analyses into areas where families, nonprofit organisations and governments jointly deliver aged care services, and where policies affect not only current but future generations. Dynamic microsimulation can track family linkages across generations, while maintaining an information base on households and income units. Such information is important if the housing needs of the aged, their wealth transfers to the next generation, and the availability of family members as carers are to be studied. Overall, while dynamic microsimulation has disadvantages in terms of development time and computing requirements, it can offer extensions to traditional approaches: • a greater number of factors affecting policy can be taken into account simultaneously; • considerably greater disaggregation at the policy and individual characteristic levels is possible; • policies that have an impact across generations can be considered; and • the distributional effects of the policies can be assessed at the level of the individual, as well as across generations. In its present form, NATSEM’s dynamic microsimulation model generates life cycle profiles of Australia’s resident population at the individual and family levels. Links between spouses and children living with their parents are maintained, and the number of children each individual parent has had is known. An extensive range of individual characteristics are considered and can be drawn on, depending on the policy application to be considered. The model, in its present form, has been designed primarily to analyse the Higher Education Contribution Scheme and AUSTUDY policy options. Further developments currently under way or planned are in three stages. Completion of the first stage will allow analysis of family linkages within and across generations and of how these would vary under different demographic scenarios. This would shed some light on the

Australia’s Ageing Population: How Important Are Family Structures?

53

extent to which family members could be available to support the elderly when needed. The second stage will tap into the greatest advantage dynamic microsimulation has over traditional methods: the ability to account for wealth accumulation and the transfer of wealth across generations. However, once the intergenerational impact of age related policies is considered, the effects of policies applying to the working age population directly can no longer be ignored. For example, as younger people make greater contributions to the costs of their health and education, they are less able to save; if older people stay in the workforce longer, they will rely less on the age pension, but government spending on unemployment benefits could increase; if older people take greater responsibility for the costs of their aged care without changes in workforce participation rates, the size of the inheritance transferred to the next generation could be lower; and when poorer people enter a nursing home, they no longer require housing assistance from government. In an intergenerational setting there is a need to study the combined effects of a range of policy initiatives. One example would be a study of the combined effects on the future aged of selected 1996-97 budget announcements, such as the changes in the Higher Education Contribution Scheme and AUSTUDY conditions, the incentives for people to take out private health insurance, the incentives for older people to have healthier lifestyles and stay in the workforce longer, the greater incentives for carers, and the provision for nursing homes to charge an upfront entry fee. Stage three involves linking NATSEM’s dynamic microsimulation model with an economywide model that is able to account for a range of factors, such as economic growth through productivity gains. With a combined model, it would be possible to extend the analyses mentioned above. For example, it would be possible to directly account for likely future increases in living standards and for the impact this would have on government revenue. If people took greater responsibility for the costs of their own health, education and aged care, government spending in these areas would be lower and taxation rates may decline as a result. Alternatively, the extent to which taxation rates would need to increase to finance certain welfare initiatives could be directly estimated under different demographic and economic growth scenarios.

54

Discussion Paper No. 19

Regardless of the stage of model development, simulation results could be used for distributional analyses at various levels of aggregation. The immediate cross-sectional income distributional effects of policies could be compared with their impact on people’s lifetime incomes, as well as lifetime income patterns across generations. Such comparisons would be particularly valuable when analysing policy options that cut across generations, such as the financing of nursing home costs either through the assets of the elderly or increased taxes. Overall, dynamic microsimulation could provide valuable additional insights to those arising from previous studies on ageing. It could also provide a systematic framework for analysing policy options that affect the provision of joint services (for example, by families and the voluntary, nonprofit and government sectors), as well as policy initiatives that have intergenerational implications.

Australia’s Ageing Population: How Important Are Family Structures?

55

References Abraham, B., d’Espaignet, E. and Stevenson, C. 1995, Australian Health Trends: 1995, Australian Institute of Health and Welfare, Canberra. ABS (Australian Bureau of Statistics) 1990a, Domestic Care of the Aged, Australia, 1988, Cat. no. 4121.0, Canberra. —— 1990b, Child Care Arrangements: Australia, Cat. no. 4402.0, Canberra. —— 1993a, Australia’s Families: Selected Findings from the Survey of Families in Australia, March 1992 to May 1992, Cat. no. 4418.0, Canberra. ——1993b, Disability, Ageing and Carers: Australia, Cat. no. 4430.0, Canberra. —— 1995, Focus on Families: Income and Housing, Cat. no. 4424.0, Canberra. —— 1996a, Industrial Disputes, May, Cat. no. 6321.0, Canberra. —— 1996b, Projections of the Populations of Australia, States and Territories: 1995– 2051, Cat. no. 3222.0, Canberra. —— and Office on Ageing (NSW) 1995a, Older People in New South Wales: A Profile, Cat. no. 4108.1, Canberra. Antcliff, S. 1993, An Introduction to DYNAMOD: A Dynamic Microsimulation Model, DYNAMOD Technical Paper no. 1, National Centre for Social and Economic Modelling, University of Canberra. ——, Bracher, M., Gruskin, A., Hardin, A. and Kapuscinski, C. 1996, Development of DYNAMOD, 1993 and 1994, Dynamic Modelling Working Paper no. 1, National Centre for Social and Economic Modelling, University of Canberra. Arnold, E. 1978, Population Decline in Europe: Implications of a Declining or Stationary Population, Council of Europe, Butler and Tanner Ltd, London. Baekgaard, H. and King, A. 1996, Modelling the accumulation and distribution of Australian household assets, NATSEM paper presented at the 24th General Conference of the International Association for Research in Income and Wealth, Lillehammer, 18–24 August. BIMPR (Bureau of Immigration and Population Research) 1996, Longitudinal Survey of Immigrants to Australia, Revised version of ‘Full Wave’, Canberra.

56

Discussion Paper No. 19

Budget Paper 1996a, Budget Overview and Economic Outlook 1996-97, Commonwealth of Australia, Canberra, 20 August. —— 1996b, Budget Statements 1996-97, Budget Paper no. 1, Circulated by the Treasurer and the Minister for Finance, Commonwealth of Australia, Canberra, 20 August. —— 1996c, ‘Statement 2: Economic conditions, prospects and policy’, Budget Statements 1996-97, Budget Paper no. 1, Circulated by the Treasurer and the Minister for Finance, Commonwealth of Australia, Canberra, 20 August. —— 1996d, ‘Statement 3: Outlays’, Budget Statements 1996-97, Budget Paper no. 1, Circulated by the Treasurer and the Minister for Finance, Commonwealth of Australia, Canberra, 20 August. —— 1996e, Strengthening Families, Statement by the Minister for Family Services, Commonwealth of Australia, Canberra, 20 August. —— 1996f, Recognising Older Australians, Statement by the Treasurer and the Ministers for Social Security, Family Services and Veterans’ Affairs, Commonwealth of Australia, Canberra, 20 August. —— 1996g, Higher Education Budget Statement, Statement by the Minister for Employment, Education, Training and Youth Affairs, Commonwealth of Australia, Canberra, 9 August. Butler, J.R. 1996, A microsimulation model of the Australian health sector: design issues, Paper presented at the Australian Conference of Health Economists, Coffs Harbour, July. Creedy, J. and Taylor, P. 1993, ‘Population ageing and social expenditure in Australia’, Australian Economic Review, no. 3, July/August, pp. 56–68. Demeny, P. 1986, ‘Pronatalist policies in low-fertility countries: patterns, performance and prospects’, in Davies, K. et al., Below-Replacement Fertility in Industrial Societies: Causes, Consequences, Policies (supplement to Population and Development Review, no. 12), Population Council, New York. Department of Health, Housing and Community Services 1991, Aged Care Reform Strategy Mid-Term Review: Report, AGPS, Canberra. Department of Human Services and Health 1995a, Annual Report 1994-95, AGPS, Canberra. —— 1995b, Finding the Unit Cost, Home and Community Care: a Joint Commonwealth and State/Territory Program, Canberra.

Australia’s Ageing Population: How Important Are Family Structures?

57

Department of Social Security 1996, Early Retirement Seminar, AGPS, Canberra. Dharmalingam, A. 1991, ‘Re-linking fertility behaviour and old age economic security: a delayed note on Demeny’, Journal of the Australian Population Association, vol. 8, no. 1, pp. 23–32. EPAC 1988, Economic Effects of an Ageing Population, Economic Planning Advisory Council Paper no. 29, AGPS, Canberra. —— 1994a, Australia’s Ageing Society, by R. Clare and A. Tulpule, Economic Planning Advisory Council Background Paper no. 37, AGPS, Canberra. —— 1994b, Shaping Our Future: National Strategies Conference, Economic Planning Advisory Commission Conference Proceedings, Reports 1, 2, 3 and 4, AGPS, Canberra. Eckermann, S. 1992, Projected health expenditure and ageing: revised methodologies, Paper presented at the PHA Conference, Canberra, September. Goss, J., Eckermann, S., Pinyopusarerk, M. and Wen, X. 1994, Economic perspective on the health impact of the ageing of the Australian population in the 21st Century, Paper presented at the Conference of the Australian Population Association, Australian National University, Canberra, September. Harding, A. 1990, Dynamic Microsimulation Models: Problems and Prospects, Discussion Paper no. WSP/48, Suntory and Toyota International Centres for Economics and Related Disciplines, London School of Economics. —— 1993, Lifetime Income Distribution and Redistribution, North Holland, Amsterdam. —— 1996a, Financial Implications of HECS Reform Options, Policy Paper no. 2, National Centre for Social and Economic Modelling, University of Canberra. —— (ed.) 1996b, Microsimulation and Public Policy, North Holland, Amsterdam. House of Representatives Standing Committee on Community Affairs 1994, Home but Not Alone: A Report on the Home and Community Care Program, AGPS, Canberra. Howe, A. 1991, Housing for Older Australians: Affordability, Adjustment and Care, National Housing Strategy, AGPS, Canberra. Hugo, G. 1986, Australia’s Changing Population: Trends and Implications, Department of Immigration and Ethnic Affairs, Oxford University Press.

58

Discussion Paper No. 19

—— and Wood, D. 1984, Ageing of the Australian Population: Changing Distribution and Characteristics of the Aged Population, Census Project Paper 8, Working Paper Series no. 63, National Institute of Labour Studies, Flinders University, Adelaide. Industry Commission 1990, Strategic Trade Theory: The East Asian Experience, Information Paper, AGPS, Canberra. ——1995, Charitable Organisations in Australia: An Inquiry into Community Social Welfare Organisations, AGPS, Melbourne. Institute for Management Development 1997, The World Competitiveness Report 1997, Switzerland (and previous issues). Kelley, A.C. 1988, ‘Australia: the coming of age’, Australian Economic Review, 2nd Quarter, pp. 27–44. Kendig, H., Hashimoto, A. and Coppard, L. (eds) 1992, Family Support for the Elderly: The International Experience, Oxford University Press on behalf of the World Health Organisation. King, A. and Bækgaard, H. 1996, The dynamics of housing wealth, NATSEM paper presented at the 8th National Conference of the Australian Population Association, December. —— and Singh, M. 1994, Australia’s Future Aged Population: A Projection of Selected Aspects, Report prepared for the Department of Human Services and Health, Social Policy Research Centre, University of New South Wales, Sydney. Krugman, P. 1992, Age of Diminished Expectations: US Economic Policy in the 1980s, MIT Press, Cambridge. Lloyd-Sherlock, P. and Johnson, P. (eds) 1996, Ageing and Social Policy: Global Comparisons, Suntory and Toyota International Centres for Economics and Related Disciplines, London School of Economics. Myers, G.C. 1992, ‘Demographic ageing and family support for the elderly’, in Kendig, H., Hashimoto, A. and Coppard, L. (eds), Family Support for the Elderly: The International Experience, Oxford University Press on behalf of the World Health Organisation. National Commission of Audit 1996, National Commission of Audit Report to the Commonwealth Government, AGPS, Canberra. OECD (Organisation for Economic Cooperation and Development) 1988, Ageing Populations: The Social Policy Implications, Paris.

Australia’s Ageing Population: How Important Are Family Structures?

59

—— 1992, Urban Policies for Ageing Populations, Paris. —— 1994, Caring for Frail Elderly People: New Directions in Care, Social Policy Studies no. 14, Paris. —— 1995, The Transition from Work to Retirement, Social Policy Studies no. 16, Paris. —— 1996, Ageing Populations, Pension Systems and Government Budgets: Simulations for 20 OECD Countries, Economic Department Working Papers no. 168, Paris. Orcutt, G., Merz, J. and Quinke, H. 1986, Microanalytic Simulation Models to Support Social and Financial Policy, North Holland, New York. Productivity Commission 1996, Stocktake of Progress in Microeconomic Reform, Canberra. Retirement Income Modelling Task Force 1996a, Sources of Income and the Assets of Older Australians, by C. Brown, Department of the Treasury, Canberra. —— 1996b, Aggregate and Distributional Analysis of Australian Superannuation Using the RIMGROUP Model, by G. Rothman, Department of the Treasury, Canberra. Root, H. 1996, Small Countries, Big Lessons, Governance and the Rise of East Asia, Oxford University Press. Sax, S. 1993, Ageing and Public Policy in Australia, Allen & Unwin, Sydney. Walker, A. 1997, Economic effects of Australia’s ageing society: key issues and methods of analysis, NATSEM paper to be presented at the FISS (Foundation for International Studies on Social Security) International Conference ‘Issues in Social Security’, Sweden, 14–17 June. World Bank 1993, The East Asian Miracle: Economic Growth and Public Policy, World Bank Policy Research Report, Oxford University Press. —— 1994, Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press. Young, C. 1989, ‘Population policies in developed countries: how do Australia’s policies compare?’, Journal of the Australian Population Association, vol. 6, no. 1., pp. 38–56. —— 1990, Australia’s Ageing Population — Policy Options, Bureau of Immigration Research, AGPS, Canberra.

NATSEM publications Copies of NATSEM publications and information about NATSEM may be obtained from: Publications Officer National Centre for Social and Economic Modelling University of Canberra GPO Box 563 Canberra City ACT 2601 Australia Ph: + 61 6 275 4900 Fax: + 61 6 275 4875 Email: [email protected] See also NATSEM’s World Wide Web site: http://www.natsem.canberra.edu.au Periodic publications NATSEM News keeps the general community up to date with the developments and activities at NATSEM, including product and publication releases, staffing and major events such as conferences. This newsletter is produced twice a year. The Income Distribution Report (IDR), which is also produced twice a year, provides information and comment on the average incomes of Australian families, covering the incidence of taxation for different family types, the income support provided by the government and how different family groups are faring. The IDR, which is available on subscription, presents this information in a simple, easy-to-follow format. NATSEM’s Annual Report gives the reader an historical perspective of the centre and its achievements for the year. DYNAMOD Technical Paper series No.

Authors

Title

1

Antcliff, S.

An Introduction to DYNAMOD: A Dynamic Microsimulation Model, September 1993

STINMOD/96A STINMOD is an easy-to-use microsimulation model for better understanding the impact of social and economic change on Australian families. It is a cutting edge tool for exploring the distributional impact of possible government policy changes. Microsimulation models are built from the bottom up, using samples of microdata based on real Australians. From such a foundation, projections can then be made for the wider community. The fourth release of STINMOD is now available. This release incorporates several of the key 1996-97 budget announcements, including the introduction of the family tax initiative and the Medicare levy surcharge for individuals and families without private hospital cover. Also included is the simplified calculation of the beneficiary rebate and the new income tax for self-funded retirees. STINMOD Technical Paper series No.

Authors

Title

1

Lambert, S., Percival, R., Schofield, D. and Paul, S.

An Introduction to STINMOD: A Static Microsimulation Model, October 1994

2

Percival, R.

Building STINMOD’s Base Population, November 1994

3

Schofield, D. and Paul, S.

Modelling Social Security and Veterans’ Payments, December 1994

4

Lambert, S.

Modelling Income Tax and the Medicare Levy, December 1994

5

Percival, R.

Modelling AUSTUDY, December 1994

6

Landt, J.

Modelling Housing Costs and Benefits, December 1994

7

Schofield, D.

Designing a User Interface for a Microsimulation Model, March 1995

8

Percival, R. and Schofield, D.

Modelling Australian Public Health Expenditure, May 1995

9

Paul, S.

Modelling Government Education Outlays, September 1995

10

Schofield, D., Polette, J. and Hardin, A.

Modelling Child Care Services and Subsidies, January 1996

Discussion Paper series No.

Authors

Title

1

Harding, A.

Lifetime Repayment Patterns for HECS and AUSTUDY Loans, July 1993

2

Mitchell, D. and Harding, A.

Changes in Poverty among Families during the 1980s: Poverty Gap Versus Poverty Head-count Approaches, October 1993

3

Landt, J., Harding, A., Percival, R. and Sadkowsky, K.

Reweighting a Base Population for a Microsimulation Model, January 1994

4

Harding, A.

Income Inequality in Australia from 1982 to 1993: An Assessment of the Impact of Family, Demographic and Labour Force Change, November 1994

5

Landt, J., Percival, R., Schofield, D. and Wilson, D.

Income Inequality in Australia: The Impact of Non-Cash Subsidies for Health and Housing, March 1995

6

Polette, J.

Distribution of Effective Marginal Tax Rates Across the Australian Labour Force, August 1995

7

Harding, A.

The Impact of Health, Education and Housing Outlays on Income Distribution in Australia in the 1990s, August 1995

8

Beer, G.

Impact of Changes in the Personal Income Tax and Family Payment Systems on Australian Families: 1964 to 1994, September 1995

9

Paul, S. and Percival, R.

Distribution of Non-Cash Education Subsidies in Australia in 1994, September 1995

10

Schofield, D., Polette, J. and Hardin, A.

Australia’s Child Care Subsidies: A Distributional Analysis, January 1996

11

Schofield, D.

The Impact of Employment and Hours of Work on Health Status and Health Service Use, March 1996

12

Falkingham, J. and Harding, A.

Poverty Alleviation Versus Social Insurance Systems: A Comparison of Lifetime Redistribution, April 1996

13

Schofield, D. and Polette, J. How Effective Are Child Care Subsidies in Reducing a Barrier to Work?, May 1996

14

Schofield, D.

Who Uses Sunscreen?: A Comparison of the Use of Sunscreen with the Use of Prescribed Pharmaceuticals, May 1996

Discussion Paper series (continued) No.

Authors

Title

15

Lambert, S., Beer, G. and Smith, J.

Taxing the Individual or the Couple: A Distributional Analysis, October 1996

16

Landt, J. and Bray, J.

Alternative Approaches to Measuring Rental Housing Affordability in Australia, April 1997

17

Schofield, D.

The Distribution and Determinants of Private Health Insurance in Australia, 1990, May 1997

18

Schofield, D., Fischer, S. and Percival, R.

Behind the Decline: The Changing Composition of Private Health Insurance in Australia, 1983–95, May 1997

Policy Paper series No.

Authors

Title

1

Harding, A. and Polette, J.

The Distributional Impact of a Guns Levy, May 1996

2

Harding, A.

Lifetime Impact of HECS Reform Options, May 1996

3

Beer, G.

An Examination of the Impact of the Family Tax Initiative, September 1996

Dynamic Modelling Working Paper series No.

Authors

Title

1

Antcliff, S., Bracher, M., Development of DYNAMOD: 1993 and 1994, Gruskin, A., Hardin, A. and June 1996 Kapuscinski, C.