Assessing the Adequacy of Private Provision for Retirement: A Living Standards Perspective

Bev Hong ¹
Light-shed Research & Evaluation
Wellington

John Jensen
Centre for Social Research and Evaluation
Ministry of Social Development

Abstract

Private provision for retirement in New Zealand is made on a voluntary basis without tax concessions or other direct concessions to encourage savings. It enables individuals to enhance their standard of living above that provided by New Zealand Superannuation and potentially maintain their living standards in retirement at the level experienced during their working lives.

This report examines the adequacy of private income provision in retirement within the context of working-life asset accumulation and living standards outcomes. The report considers the living standards of the current cohort of older New Zealanders in terms of whether minimum adequate living standards have been achieved and whether pre-retirement living standards have been maintained in retirement. Income and other factors that contribute to the living standards experienced by older New Zealanders are also discussed.

The outlook for future cohorts of older New Zealanders is then examined based on emerging trends for older New Zealanders and the factors found to contribute to variation in living standards.

Introduction

Preparation for Retirement

The living standards that people experience in their retirement are based on both the public provision of income and assistance through New Zealand Superannuation, subsidies and services (for example, health subsidies and public health services), and the private assets they have accumulated and related income from these assets. Some assets (such as rental properties and financial investments) can provide an income during retirement and other assets can contribute to retirement living standards by reducing the income required (for example, mortgage-free home ownership contributing to low accommodation costs). Retirement living standards will, therefore, be affected by private provision in the form of assets that offset living expenses (and the requirement for private income) as well as income-generating assets. For this reason, this report focuses on overall asset accumulation rather than solely on retirement income per se, because asset accumulation provides a stronger basis for assessing the adequacy of people's preparation for retirement.

People's asset accumulation for retirement needs to be considered within a life-cycle context. Circumstances, such as income levels and family commitments, change over people's lives, and attitudes towards saving for retirement and expectations about retirement may change as people get older and retirement age gets closer. The pattern of asset accumulation over the life cycle will vary between people. Generally, it is expected that people will have lumpy asset accumulation, with events such as personal partnerships and separations, children, inheritances, redundancies, business and economic upturns and downturns, and ill health, among other factors, producing fluctuations in asset accumulation over the working life.

A longitudinal approach is needed to achieve an authoritative understanding of people's saving behaviour over the life cycle and their preparation for retirement. This would help to provide information about levels of asset accumulation at different life stages and to assess whether these were considered sufficient for trajectories that resulted in at least adequate living standards in retirement. However, to date New Zealand research has involved "point in time" data, which are unable to inform us about the pattern of accumulation over an individual's life cycle.

"Point in time" research on the living standards of older New Zealanders in 2000 assessed the living standards of older New Zealanders and examined the factors that contributed to variation in living standards (Fergusson et al. 2001). This research was initiated by the Super 2000 Taskforce,² continued by the Ministry of Social Development³  and conducted by Statistics New Zealand. A nation-wide survey was undertaken involving 3,060 face-to-face interviews with older New Zealanders living in private dwellings, and including a supplementary survey of 500 Maori aged 65-69 years. The living standards research programme also involved a nation-wide survey of 3,682 people aged 18-64 years in 2000. This survey enabled the living standards across different groups in the New Zealand population to be compared.

The research involved the use of a measure of living standards based on ownership restrictions, social participation restrictions, economising behaviour, prevalence of severe financial problems, and self-assessments. Further analysis of the survey data resulted in refinement of this measure with the creation of the Economic Living Standard Index (ELSI) by Jensen et al. (2002), which has enabled a comparison of groups across the total population of New Zealanders to be described and compared (Krishnan et al. 2002).

The findings from this research can provide us with information about how older people in New Zealand are currently faring and the contribution of income, assets and other factors to their living standards. This information can also be used to explore potential outcomes for future cohorts of older people in New Zealand in terms of income, asset accumulation, and other factors likely to have an impact on what those living standards will be.

Statistics New Zealand has begun a longitudinal Survey of Family Income and Employment (SoFIE),⁴  which will provide life-cycle information about this issue in the future (Statistics New Zealand 2002b). A key objective of this survey is to determine patterns of saving for retirement and relate these to income dynamics and life-cycle stages. The survey will involve a starting sample of about 10,000 households (20,000 adult individuals) and is of the usually resident population in New Zealand living in private dwellings at the start of the survey. Interviewing for the first of eight planned interview waves began on 1 October 2002. First results are expected to be published about March 2004 and second-wave results in March 2005.

At this stage, living standards information is not included in the longitudinal data collection for SoFIE. It would be very useful for living standards questions to be included in future interview waves to measure changes in living standards over time. This would enable some analysis of the link between patterns of asset accumulation, achievement of minimum adequate levels of living standards and the maintenance of working-life living standards in retirement to be undertaken.

The analysis in this paper has been developed on the assumption that New Zealand Superannuation will be maintained in its current form, with age of eligibility remaining at 65 years and the payment rates being set according to the current formula. Reference is also made in various parts of this report to "pre-retirement" living standards. For many people, living standards during the later part of their working lives will be higher than during the earlier part. Pre-retirement living standards should be interpreted as relating to the later part of the working life (for example, from around age 50 to retirement).

Adequacy of Asset Accumulation and Income for Retirement

The vast majority of research into living standards is framed in terms of income and expenditure or consumption (Fergusson et al. 2001). Various ways of conceptualising "adequacy" have been proposed. They include (Scobie and Gibson 2003) thresholds based on:

• a proportion of an individual's pre-retirement income
• a proportion of the average pre-retirement incomes of the working-age population (relative measure)
• an income that enables people to sustain their pre-retirement level of consumption (a consumption smoothing approach)
• an income that permits people to have the same marginal utility of consumption over time.

Although research using income and expenditure is useful, it only provides information about people's access to resources. Inherent in this approach is the translation of income into the actual material circumstances of individuals. In particular, one of the issues that arises for analysis using pre- and post-retirement income and consumption is the different focus that pre-retirement expenditure may have in comparison with post-retirement needs. Work-related costs provide an obvious illustration, but many people will also have various other types of costs that are significant during their working years but less so after retirement. Examples are expenditure on bringing up and educating children and paying interest on home mortgages.

The development of the living standards measure by Fergusson et al. (2001) and more recently the ELSI by Jensen et al. (2002) provides a much more informative and direct approach for considering pre-retirement and post-retirement living standards.

ELSI combines information about whether people are able to afford "basic" items, "comfort or luxury" items, and self-assessments related to living standards. Items referred to as "basic" include a washing machine, heating for all main rooms, warm bedding, fresh fruits and vegetables, and doctor's visits. Items referred to as "comforts or luxuries" include overseas holidays or never having to cut back on shopping for clothes because of cost. This scale is able to measure deprivation because a low score indicates the lack of basics that the respondents want but are unable to have because of the cost, together with a paucity of "comforts or luxuries".

The scale has seven levels, with Level One indicating "very restricted" living standards and Level Seven indicating "very good" living standards. The first three ELSI levels (One, Two and Three) are characterised as indicating some degree of hardship and are labelled as representing "very restricted", "restricted" and "somewhat restricted" living standards respectively. People with living standards in these levels lack "basic" items, have few "comforts or luxuries" and have financial or accommodation problems. For the purposes of this report, the term "hardship" is used to refer to people whose living standards are typified by the first three levels of the ELSI distribution.

The adequacy thresholds relating to the measures of living standards that will be used in this report are:

• a minimum adequate living standard, defined by setting the distribution of living standards of older New Zealanders in 2000 as a benchmark standard for future generations of older people, where a minority were in hardship and the majority were doing quite well
• ensuring that post-retirement living standards are maintained relative to pre-retirement living standards in the decade prior to retirement.

Asset Accumulation and Retirement Outcome Frameworks

A useful way of examining asset accumulation and retirement outcomes is through taking a stylised approach. One way of conceptualising working-life incomes and different asset accumulation patterns is presented in Table 1. This table broadly categorises people into three groups relating to stylised earnings and savings patterns, and shows links between level of earnings and the capacity to save and accumulate assets over the working life. The distribution of working-life income has been notionally split into three in terms of those with "low", "middle" and "high" working-life income levels.

Table 1 Working-Life Income and Likely Capacity to Accumulate Assets for Retirement

Income over working life	Asset accumulation over working life
	Low	Moderate	High
Low
Middle
High

The group defined as "low" are characterised as having low overall earnings over their working lives and very minimal private capacity to accumulate assets.

The group defined as "middle" are characterised as having moderate earnings over their working lives and having a capacity to accumulate moderate assets by making consumption choices that result in increased assets (for example, home ownership and investment). The extent to which consumption opportunities to accumulate assets are not taken results in low asset accumulation.

The group defined as having "high" earnings over their working lives are characterised as having high asset accumulation tendencies (for example, property and financial investments) with a reduction in asset accumulation resulting in moderate asset accumulation in retirement. Low asset accumulation at retirement is unlikely for this group due to their high earning capacity over their working lives and the tendency for the basis for some of this income to be asset related.

A framework for discussing the two thresholds of minimum adequacy and maintenance of pre-retirement living standards is depicted in the living standards (LS) outcomes grid (Figure 1). The four possible outcome combinations are shown in Boxes A to D.

Public provision of retirement income (New Zealand Superannuation) and government subsidies and social services are designed to ensure that people avoid hardship and are positioned in the right-hand side of the grid (boxes A and B). Asset accumulation that is commensurate with income (and thus low when income has been low) may enable people to maintain their pre-retirement living standards and be positioned in the top half of the grid (Boxes C and A). Box A is the most desirable outcome where hardship is avoided and pre-retirement living standards are maintained. Box D characterises the least desirable outcome such that hardship and a drop in living standards are experienced in retirement.

Figure 1 Living Standards (LS) Outcomes Grid: Achieving Minimum Adequate Living Standards and Maintaining Pre-Retirement Living Standards in Retirement

A reduction in New Zealand Superannuation or other government-provided subsidies and social services would result in a shift to the left, from avoidance of hardship towards hardship and a shift downwards from continuity in living standards towards a drop in post-retirement living standards.

In the next section, the living standards of the current cohort of older New Zealanders and contributing factors are presented. In the subsequent section, future cohorts of older people are considered in terms of emerging demographic trends, and possible levels of asset accumulation and potential retirement living-standards outcomes. The situation for current older people and scenarios relating to future cohorts of older people are discussed in terms of "low", "middle" and "high" working-life income groups and the LS outcomes grid.

The Current Cohort of Older New Zealanders

This section describes the current situation of older New Zealanders. Background demographics of older New Zealanders are presented and the living standards of older people are examined in relation to the achievement of a minimum adequate level of living standards and continuity of pre-retirement living standards in retirement. Following this, the factors that contribute to this group's living standards are examined. This information is brought together and discussed in terms of the asset accumulation and retirement outcomes framework in the final part of this section.

Demographics of Older New Zealanders

Older New Zealanders comprised 12% of the general population in 2001. There were 18 people aged 65 and over for every 100 aged 15 to 64 years. Other information based on the Census 2001 (Statistics New Zealand 2002b) and estimated resident population as at 31 March 2002 provides the following information for older New Zealanders.

Ethnic affiliation and gender

•  Most designated their ethnicity as European (93%), with a small proportion designating themselves as Maori (4%), Pacific (2%) or Asian (2%).
• About half (56%) were women, with women comprising 70% of those aged 85 and over.

Living arrangements and location

• Over half (58%) of older people were partnered, with a small percentage (2%) in de facto relationships.
• Most lived in urban areas and 11% lived in rural areas.
• Most (93%) lived in private dwellings, with 7% living in non-private dwellings such as retirement homes, residential care facilities, hospitals or retirement villages.
• One-third were living on their own, with the majority (three-quarters) being women.
• Thirteen per cent usually lived with their children.

Qualifications, employment and income

• Over half (54%) had a formal qualification.
• Eleven per cent were in paid employment in the week preceding the 2001 Census.
• The median annual income (from all sources, before tax) in the year to March 2001 was $13,100, with one-third receiving $20,000 or more and only 5% receiving more than $40,000 in income.
• More than half received income from other sources, including income from interest, dividends and rent (40%), from other superannuation and annuities (15%), and from wages and salaries (6%); about two in five older New Zealanders received income from just New Zealand Superannuation.

Information about the net worth of older New Zealanders is available from the Household Savings Survey (HSS) conducted by Statistics New Zealand in 2001 and commissioned by the Office of the Retirement Commissioner (Statistics New Zealand 2002c). This first nation-wide net worth survey involved 5,374 face-to-face interviews with New Zealanders living in private dwellings, and included a Maori booster sample. Results are presented for economic units (that is, non-partnered individuals and couples). Net worth was calculated by subtracting an economic unit's total debts from their total assets. The survey found the following information for older New Zealanders.

Assets

• The most common asset type was bank deposits (bank accounts in credit and bonus bonds).
• Two-thirds of non-partnered individuals and about three-quarters of couples owned residential properties (such as the house they lived in, rental properties and other residential properties, including empty sections), with 95% of non-partnered individuals and 90% of couples owning these properties mortgage free.

Debts

• Credit card debt was the most common debt type.

Net worth

• Net worth is accumulated over a lifetime and its distribution is closely related to age. In general, net worth increases with age until later retirement, where it begins to decrease as assets are used (Scobie and Gibson 2003).
• The median net worth for individuals (partnered and non-partnered) was $112,800.
• The top 20% (in terms of net worth) held about 60% of the net worth of non-partnered individuals and couples.
• Two per cent had a negative net worth.

The current overall picture of older New Zealanders is of a group that is predominantly European, with modest incomes but significant assets, and living in urban areas in their own homes, often mortgage free. Women comprise the majority of New Zealanders aged 85 and over due to their increased longevity.

The Living Standards of Older New Zealanders

Achieving minimum adequate living standards

The living standards of older New Zealanders were reported by Fergusson et al. (2001), who found that the majority of older New Zealanders were doing quite well, with most (79%) having no serious financial problems and unlikely to have restrictions or to economise a lot. A minority (around 5%) of older people had quite marked material hardship, and a further 5-10% had some restrictions and hardship. A different analysis of the same data by Krishnan et al. (2002) produced essentially the same conclusion. The research reinforced the importance of income from state superannuation, with over 60% of the income received by single respondents and couples being from that source.

Fergusson et al. (2001) concluded that New Zealand Superannuation and other associated payments such as Accommodation Allowances were able to protect the great majority of older people from hardship and material deprivation. Although this was the case, they noted that the avoidance of hardship in retirement is based on a combination of public and private provision. More specifically, therefore, these findings show that given the private provisions for retirement that the current cohort of older people had made, New Zealand Superannuation and associated payments were sufficient to enable the majority of older people to avoid hardship.

Maintenance of pre-retirement living standards in retirement

The living standards research did not specifically investigate whether older people had been able to maintain their previous working-age living standard in retirement. However, it is possible to draw some tentative conclusions based on what the research on the living standards of older people (Fergusson et al. 2001) and comparison of the living standards of older New Zealanders and the rest of the population have shown (Krishnan et al. 2002).

Research on the living standards of older people shows that the current cohort of older people generally have a comfortable or good standard of living. If a large number of this older cohort had substantially greater living standards prior to retirement, this would imply that the pre-retirement living standards of this group had been extremely favourable, and in particular much more favourable than the current group of 45 to 65 year olds. This seems unlikely, which tentatively suggests that the substantial proportion of currently retired people have probably retained their pre-retirement living standards.

Factors that Contributed to the Living Standards of Older New Zealanders

The research on the living standards of older New Zealanders (Fergusson et al. 2001) investigated the factors that may have led to variation in older people's living standards. Income was found to have only a modest relationship with living standards, accounting for up to 16% of variation in people's living standards. The research findings suggested that it was an accumulation of factors reflecting a person's current circumstances and previous life history, rather than one single factor such as annual net income, that determine a person's living standards in older age. The following factors were identified as contributing to diminished living standards:

• lower levels of net annual income
• lower levels of savings and investment (total value)
• high accommodation costs (mortgage and rental payments)
• increased exposure to economic stresses in the last year (such as replacement of household appliances, major house maintenance, large car repair bills, and funeral costs)
• increased exposure to adverse economic life events during the period from age 50-59 years (such as death of a partner, major illness or injury, separation or divorce, redundancy and prolonged unemployment)
• Maori ethnicity ⁵ 
• Pacific ethnicity
• lack of formal educational achievement
• low socio-economic status occupation when aged 50-59 years.

There were also other factors investigated in the research, which were found not to make a significant contribution to explaining variation in living standards in older age once the above factors were taken into account. Other factors considered in the research were: area of residence, partnership status, living with other household occupants, country of origin, value of own home, home production activities, number of children in family, currently having dependent children, gender (single people only), frequency of family contact and provision of family support.

An association between health and living standards was found for the older people in the living standards research, with diminished health relating to diminished living standards. However, it was beyond the scope of the research to clarify the causal direction of this association in terms of whether poor living standards led to poor health and disability or poor health and disability led to poor living standards. The relationship may be recursive, with various mechanisms operating to produce influences in both directions.

The research highlighted the importance of pre-retirement policies, including the encouragement of savings and investment during working life to meet economic needs in older age and developing social policies to ensure high levels of employment and adequate income levels over the life course before retiring age. These findings also emphasise the need to consider any major cohort differences in terms of people's capacity to accumulate assets over their working lives.

The government housing policies of the 1930s to 1970s, which supported home ownership through subsidised loans (Ferguson 1994), are of particular relevance to the cohort of New Zealanders aged 65 years and over in 2000. Ferguson (1994) provides a detailed account of New Zealand housing policy from 1840 to 1992. She notes that government lending, especially during the 1950s and early 1960s:

… kept the cost of existing housing low. It encouraged mortgage interest rates to move downwards and was effective in giving access to home-ownership to many people who might not otherwise have achieved it. (p.233)

Government promotion of home ownership through government lending declined from the 1970s onwards.

The research found that of the 5% who had quite marked material hardship, at least half reported having no savings or assets, paid rent or mortgage, reported exposure to economic stresses when aged 50 to 59 years, and had no formal qualifications.

The Contribution of Private Provision to Preventing Hardship

For the current population of older people, it is possible to demonstrate the importance of private provision in preventing hardship by considering the level of hardship among those who lack such provision. For the present purposes, lacking private provision is defined as (a) having no income in addition to New Zealand Superannuation, (b) living in accommodation that is rented in the private market, and (c) not having financial assets that exceed $1,000 in total value.

For the group of current older people who do not have any of these types of private provision, 31% are in the hardship range of the ELSI scale (Level One to Level Three), compared with less than 7% of older people generally. In other words, not having private provision of any of the types specified is associated with a four-fold increase in the likelihood of hardship. The more types of private provision that an older person has, the lower the likelihood of hardship. Those who lack any type of private provision have 13 times the likelihood of hardship compared to those who have all three types of provision.

A greater relative risk is found to apply when a more severe hardship criterion is adopted. Those who lack any type of private provision have 28 times the likelihood of being at ELSI Levels One or Two compared to those who have all four types of provision, and have 67 times the likelihood of being at Level One (severe hardship).

A large difference is similarly indicated when comparisons are made on the basis of ELSI mean scores. The mean score for those with no private provision is 34, compared with a mean score of 47 for older people generally. This is a substantial difference (being greater than one standard deviation).

The Overall Picture

The majority of older New Zealanders currently have good living standards. This is due to a combination of factors, including the availability of government social support and health services, high debt-free home-ownership rates (and low accommodation costs), income-generating asset and other asset accumulation over their working lives, and relatively lower occurrence of adverse economic life events later in their working lives.

Figure 2 presents a stylised picture of the asset accumulation and retirement outcomes for the current cohort of older New Zealanders. This depiction suggests the general accumulation patterns and outcomes that occurred for the majority of this population. Accumulation of assets mirrored the working-life income levels of those with middle and high working-life incomes. However, a proportion of those with low working-life incomes were able to achieve greater-than-expected asset accumulation. It is suggested that one of the specific initiatives that has contributed to this increase in asset accumulation was the government lending subsidies to promote home ownership in the 1930s to 1970s, which have resulted in people with low working-life incomes being able to purchase their own homes. This situation has led to the distribution of older New Zealanders being positioned mainly in the most desirable retirement outcome, where they have both avoided hardship and maintained their pre-retirement living standards.

Figure 2 Current Older New Zealanders: Retirement Asset Accumulation and Living Standard (LS) Outcomes

A minority of older New Zealanders with low working-life incomes were not able to make increased asset accumulation and had no or little assets or savings. The inability to augment NZS with some additional private provision of income resulted in a continuity of the hardship experienced in their working lives.

The Outlook: Emerging Trends and Implications for Future Generations of Older New Zealanders

Emerging trends for future cohorts of older people are described below (Office for Senior Citizens 2002). These trends show a very different context and working-life pattern for future cohorts of older New Zealanders than those experienced by the current generation of older people. Trends that indicate a potential reduction in living standards are described, followed by those that may enhance living standards for future cohorts of older people. These trends are also discussed in relation to the changing profile of older people likely to be in hardship.

Emerging Trends Indicating a Potential Reduction of Living Standards

An ageing population

New Zealand has an ageing population. The number of older people is expected to increase from 463,000 in 2002 to around 490,000 by 2005, and to 551,000 in 2010. The rate of growth in the older population is expected to increase from just over 1% currently to over 3% per year after 2011 as the oldest members of the post-war baby boom generation begin turning 65. In the longer term, the proportion of older New Zealanders is estimated to rise from 12% in 2001 to 18% in 2021, to about 26% in 2050.

Over the next few years, the number of people aged 85 years and over is expected to increase at a faster rate than the younger age groups. By 2005 the number of people aged 85 years and over is expected to have grown by about 13% to reach 59,000.

The current population of older New Zealanders are supported by government provision of services such as social and health services. As the population ages, it will be important that the ability for such services to cater for the growing population of older people is maintained. In particular, as people age their need for and usage of health services increases (Office for Senior Citizens 2002), and the increasing number of people aged 85 years and over will place additional pressure on health service provision.

The trend towards increasing numbers of individuals living longer extends the period of time that living standards need to be maintained in the context of having fixed retirement resources. "Inheritances" may be passed on later and/or be considerably reduced for the current working-age population due to the increased life expectancy among the current older population.

Reduced home-ownership rates

The emerging trend for future cohorts of older people is for reduced home ownership. This may, in part, be a reflection of the shift away from promotion of home ownership through government lending from the 1970s.

One of the main contributing factors associated with high living standards for older people in 2000 was the high levels of home ownership and low accommodation costs in retirement. This suggests that current working-age cohorts will need to have increased alternative asset accumulation or income available to offset the higher accommodation costs that may result in retirement. There will be less opportunity for low-income working-age people to own their homes as the previous housing policy's state-loan assistance is not available.

In this context, it is relevant to note that the Government has recently announced a modest initiative to make access to home ownership easier for low-to-modest income earners through a "home ownership mortgage insurance programme" provided by Kiwibank. This represents one of a number of possible mechanisms for seeking to reverse the declining rate of home ownership.

Increased changes in partnership status

Future cohorts of older New Zealanders are more likely to have experienced separation and divorce during their lives and to cohabit (rather than marry) than the current cohort of older people. This suggests a more disrupted pattern of asset accumulation than for the current cohort of older people, and indicates that a higher level of overall asset accumulation will be required to offset the additional transaction costs of relationship breakdown.

Later child bearing

The trend for later child bearing combined with the increased longevity of older people may result in increasing numbers of people who have both dependent elderly parents and children to care for at the same time. This will place expenditure demands on people later in their working lives during a period when consolidation and key asset accumulation may have occurred for the current older population.

Emerging Trends Indicating a Potential Enhancement of Living Standards

Employment and income variation

Future cohorts of older people are likely to have increased variation in income levels, relative to the current cohort of older people, which will increase the inequalities across this population and is likely to be associated with a broader range of living standards than is currently observed.

There is also a trend towards increasing levels of employment by women in their 50s and 60s and an increase in the proportion of people in employment after the age of 64 years. An increasing proportion of older people will have attained formal educational qualifications.

These trends suggest an increased earning capacity in the future, which provides greater opportunity for asset accumulation than for previous cohorts of older people through double-income families and extended working lives beyond the age of 64 years.

Fewer children

Future cohorts of older people are likely to have had fewer children. This suggests a reduction in expenditure demands and an increased capacity for asset accumulation during their working lives.

Emerging Trends Relating to the Profile of Older People in Hardship

Older Māori

The number and proportion of older Māori are projected to increase over the next three decades. Research on the living standards of older Māori show that they are materially disadvantaged relative to older non-Māori and many in the current cohort of older Māori face marked material hardship. The factors contributing to variation in living standards for older Māori were similar to those found for older non-Māori, with the research showing that pre-retirement policies were especially important for Māori. This included a focus on having sources of income additional to New Zealand Superannuation, the accumulation of savings and assets, reducing accommodation costs through home ownership, and avoiding redundancy and unemployment in one's 50s. The findings suggested that the disadvantage among older Māori was to some extent a consequence and continuation of disadvantage at a younger age (Cunningham et al. 2002). The current living standards of the population of working-aged Māori are also low (Krishnan et al. 2002).

These findings suggest that if effective methods to reduce this disparity are not found, a greater proportion of older people with low living standards will be Māori.

Older Pacific People

The number and proportion of older Pacific people are projected to increase rapidly over the next two decades. Research on living standards indicates that Pacific people (both younger and older) experience lower living standards than the population as a whole. Pacific peoples are also the most disadvantaged in terms of living standards of all the major ethnic groups in New Zealand. These findings suggest that if effective methods to reduce this disparity are not found, a greater proportion of older people with low living standards will be Pacific.

The Overall Picture of Emerging Trends

Overall, the emerging trends indicate several factors that may contribute towards a reduced distribution of living standards for future cohorts of older people relative to the current cohort of older New Zealanders.

To maintain the distribution of living standards of older people in the future, current working-age people need to accumulate more assets than the previous cohort to potentially offset several factors. These factors include an increased life expectancy and longer retirement period to resource, delay and reduction in inheritances received due to increased life expectancies of older generations, an increased likelihood of events such as separations and divorce, and an increased likelihood of needing to care for dependent elderly and dependent children during the same period.

The rearing of fewer children could potentially enhance the distribution of living standards for future cohorts of older people. However, this would only occur if the additional income for those in the high and middle working-life income groups and which would otherwise have been consumed on children was not required to offset the factors listed above. A level of asset accumulation would also be required that mirrored income levels if pre-retirement living standards were to be maintained. In particular, for those who chose not to purchase their own home, alternative asset accumulation options would need to be adopted to offset increased accommodation costs in retirement.

An increase in family earning capacity also suggests that relative to the current cohort of older people, future cohorts will need to have greater net worth accumulation at retirement to maintain pre-retirement living standards after they retire.

A substantial change to the profile of the living standards distribution of older people is the higher proportion of older Maori and Pacific peoples, the majority of whom will have low living standards should the current disparity evident for these groups continue. In addition, the assistance provided to older people with low working-life incomes and incentives to purchase their own homes (accumulate assets) through state loan subsidies in the 1930s to 1970s may have provided a strong boost to the asset accumulation of the low working-life income group. However, this is not an opportunity generally available to the current cohort of working-aged people.

These trends suggest that the pattern of asset accumulation that has been successful over the working lives of the current older population may not necessarily transfer successfully to future cohorts and may make it more difficult for people to make adequate private provision for retirement.

The next section of the report further discusses these emerging trends and implications using the stylised asset accumulation and LS outcomes grid approach.

Asset Accumulation and Retirement Outcome Scenarios

The framework relating income during the working life to asset accumulation by retirement offers a convenient means of formulating particular scenarios concerning future cohorts of older New Zealanders. Three particular scenarios, labelled X, Y and Z, are presented in this section depicting different asset accumulation and retirement living standards distribution outcomes for future cohorts of older New Zealanders. These scenarios are discussed, and suggestions for areas that could be further investigated to assist in enhancing the distribution of living standards of future cohorts of older age people are presented.

Scenario X presents a situation where each working-life income group (low, middle and high) has an asset accumulation pattern that mirrors their income levels. This results in an outcome where living standards are generally maintained, although a substantial proportion of the distribution are in hardship (minimum adequate living standards not achieved). This outcome is a result of an increased proportion of low working-life income people in future cohorts, who do not have opportunities to increase their accumulation of assets through government assistance as the current cohort did, and have no or negligible assets accumulated at retirement.

The arrows on the outside of the grid show that a reduction of New Zealand Superannuation and/or state subsidies or services would lead to more people being in hardship and more people experiencing a drop in living standards in retirement as additional private retirement provisions are required to replace the loss from reduced public provision of income or services.

Figure 3 Scenario X: Retirement Asset Accumulation and Living Standard (LS) Outcomes

Scenario Y depicts a situation where the retirement outcomes result in further reduction of the living standards distribution compared with scenario X. This adds the possibility that people in the middle and high working-life income groups chose consumption patterns that result in accumulated assets in retirement that are less than commensurate with their working-life income levels. As depicted in the LS outcomes grid side of the diagram this would result in a substantial increase in levels of hardship with both low-income and middle-income working-life groups being in hardship. However, for the middle-income group this would also represent a drop in their living standards, positioning them in the least desirable quadrant of neither avoiding hardship nor maintaining their pre-retirement living standards. Some of the cohort would maintain their living standards, but only those who had been in hardship during their working life.

The outcome depicted in this scenario highlights the importance of ensuring that there is ongoing promotion of savings behaviour and asset accumulation to working-age people. The tendency towards reduced home ownership, increased employment by women late in their working lives, and later child bearing with fewer children suggests that a different pattern of asset accumulation may need to be adopted for the distribution of living standards of current older people to be maintained in the future. More research needs to be undertaken to investigate how these life-stage changes may affect saving behaviour and the ability to accumulate assets for retirement. The longitudinal survey of family income and employment may be able to provide useful information to assist in understanding the life-cycle nature of asset accumulation, and this may help to provide milestone indications to working-age people about their saving behaviour and how they could best approach making private provisions for their retirement.

Figure 4 Scenario Y: Retirement Asset Accumulation and Living Standard (LS) Outcomes

Scenario Z presents the most desirable outcome for future cohorts of older people, where hardship is avoided and pre-retirement living standards are maintained. To achieve this outcome, a booster to the asset accumulation of the lower working-life income group would be needed, which would increase the proportion of low-income people who entered retirement with sufficient private assets accumulated to avoid hardship.

Figure 5 Scenario Z: Retirement Asset Accumulation and Living Standard (LS) Outcomes

Although it is not suggested that the housing policies of the previous generation could be replicated, it may be useful to further explore initiatives such as the trialling of the "home ownership mortgage insurance programme" provided by Kiwibank, and the possibility of asset-based welfare approaches of the type discussed in the United States, United Kingdom and Canada (Brynner and Paxton 2001, Friedman and Sherraden 2001, Nares et al. 2001).

The outcome also requires that middle-income and high-income people accumulate assets during their working lives that enable them to maintain their pre-retirement living standards in retirement.

Conclusions

This report has taken a living standards perspective to examine the adequacy of private provision for retirement. It has highlighted the importance of considering private provision, not only in terms of income (and income-generating assets), but also in terms of the assets (and in particular home ownership) that will reduce the level of income required to achieve at least a minimum level of living standards and to maintain living standards in retirement.

An examination of the situation of the current cohort of older people and emerging trends for the future shows the necessity of taking into consideration the differing social and economic conditions experienced by different cohorts as they move through the life stages and make choices through their working lives. Although the majority of older people are currently able to avoid hardship and have probably maintained their pre-retirement living standards through a combination of both public and private provision, the outcome could well be less favourable for future cohorts of older New Zealanders.

Indications for the future show that the pattern of asset accumulation that has been successful over the working lives of the current older population may not necessarily transfer successfully to future cohorts. Asset accumulation for the current cohort of working-age people may be more difficult than for the previous generation. Impediments to asset accumulation arise from more changeable life styles and later child bearing, leading to a shorter period of asset consolidation prior to retirement.

Other impediments arise from increased income inequality and a greater likelihood of spells of unemployment, with the likelihood that these are affecting the savings capacity of significant numbers of people. As a result, there is probably a rise in the proportion of people who experience difficulty in accumulating a substantial level of assets, and the proportion is likely to be higher for the current working-age population than it was for the present cohort of older people. In this connection, the trend towards reduced home ownership is of concern, as the good living standards of most current older people is partly a reflection of that group's high level of mortgage-free home ownership, aided by the home ownership promotion policies of the 1930s to 1970s. Alternative asset accumulation options will need to be adopted by lower-income working people who do not purchase their own homes if they are to have sufficient income after retirement to offset their accommodation costs (which will generally be higher than the accommodation costs of retired people owning their own homes). However, for the reasons indicated, asset accumulation is likely to be an uphill struggle for many of these people, even with high motivation.

An examination of potential asset accumulation patterns and retirement outcomes emphasised two key areas of focus for ensuring adequate private provision for retirement. These were:

• the ongoing promotion of savings behaviour and asset accumulation by working-age people
• the need to investigate ways of assisting low-income people to accumulate assets for retirement (whether in the form of home ownership or some other form).

Future research that will aid understanding of the adequacy of private provision for retirement by future cohorts of older New Zealanders includes the longitudinal research currently underway (SoFIE) and modelling of the asset accumulation scenarios and potential retirement outcomes described in this report.

References

Brynner, J. and W. Paxton (2001) The Asset-Effect, Institute for Public Policy Research, London.

Cunningham, C., M. Durie, D. Fergusson, E. Fitzgerald, B. Hong, J. Horwood, J. Jensen, M. Rochford and B. Stevenson (2002) Nga Ahuatanga Noho o te Hunga Pakeke Maori: Living standards of older Maori, Ministry of Social Development, Wellington.

Ferguson, G. (1994) Building the New Zealand Dream, Dunmore Press, Palmerston North.

Fergusson, D., B. Hong, J. Horwood, J. Jensen and P. Travers (2001) Living Standards of Older New Zealanders: A Technical Account, Ministry of Social Policy, Wellington.

Friedman, R.E. and M. Sherraden (2001) "Asset-based policy in the United States" in S. Regan and W. Paxton (eds.) Asset-Based Welfare: International Experiences, Institute for Public Policy Research, London.

Jensen, J., M. Spittal, S. Crichton, S. Sathiyandra and V. Krishnan (2002) Direct Measurement of Living Standards: The New Zealand ELSI Scale, Ministry of Social Development, Wellington.

Krishnan, V., J. Jensen and S. Ballantyne (2002) New Zealand Living Standards 2000, Ministry of Social Development, Wellington.

Nares, P., J. Robson Hadow and B. Gosse (2001) "The case for an asset-based approach to social policy in Canada" in S. Regan and W. Paxton (eds.) Asset-Based Welfare: International Experiences, Institute for Public Policy Research, London.

Office for Senior Citizens (2002) Briefing to the Incoming Minister for Senior Citizens: Towards Lifelong Participation and Independence, Ministry of Social Development, Wellington.

Randall, T., S. A'Court, K. Henshaw and D. Preston (2003) The Net Worth of New Zealanders: A Report on Their Assets and Debts, Statistics New Zealand, Wellington.

Scobie, G.M. and J.K. Gibson (2003) "Retirement wealth of New Zealand households: An initial analysis based on the household savings survey" presented to a Symposium on Wealth and Retirement, Retirement Commission, www.sorted.org.nz

Statistics New Zealand (2002a) 2001 Census Snapshot 9: Older People, Statistics New Zealand, Wellington.

Statistics New Zealand (2002b) Survey of Family Income and Employment: October Newsletter, www.stats.govt.nz.

Statistics New Zealand (2002c) The Net Worth of New Zealanders: Data from the 2001 Household Savings Survey - Standards Tables and Technical Notes, Statistics New Zealand, Wellington.

¹ Acknowledgements
We would like to thank the following Ministry of Social Development officials: Louise Dooley from the Office for Senior Citizens for her contribution; and Vasantha Krishnan and Alicia Wright from the Centre for Social Research and Evaluation for their feedback and comments in the development of this report.

² The Super 2000 Taskforce was disbanded in March 2000.

³ Previously the Ministry of Social Policy.

⁴ Also known as the Longitudinal Survey of Income, Employment and Family Dynamics (LSIEFD).

5 Ethnicity data were coded to one prioritised ethnicity code such that if a peson identified with more than one ethnic category the following priority order was used: Maori, Pacific peoples, other, European.