– David de Vaus, Matthew Gray, Lixia Qu and David Stanton
Australian Institute of Family Studies, February 2007, 27 pp. ISBN 978 0 642 39546 7. ISSN 1446-9863 (Print); ISSN 1446-9871 (Online)
- 1. Contents
List of tables
List of figures
About the authors
2 International trends in divorce rates
3 The Australian system of provision for retirement incomes
4 Household, Income and Labour Dynamics in Australia (HILDA) Survey
5 Socio-economic and demographic characteristics by marital history
6 The effects of divorce on financial circumstances in older age
7 Concluding comments
List of tables
Payment rates of the Australian age pension system, September 2006
Selected characteristics by marital status and gender, aged 55Â–74 years, Australia, 2002
Financial circumstances by marital history and gender, aged 55Â–74 years, Australia, 2002
Financial situation by detailed marital history and gender, aged 55Â–74 years, Australia, 2002
Financial hardship and perceived prosperity by detailed marital history and gender, aged 55Â–74 years, Australia, 2002
Effects of divorce on financial assets, and marginal effects on various financial measures by gender, aged 55Â–74 years, Australia, 2002
A1. Determinants of probability of owning house outright, logit model
A2. Determinants of receiving an income support payment, logit model
A3. Determinants of having superannuation, logit model
A4. Determinants of equivalent household income ($Â‘000), ordinary least squares
A5. Determinants of per capita net household wealth ($Â‘000), ordinary least squares
A6. Determinants of amount of government pension/benefits per week, ordinary least squares
A7. Determinants of amount of superannuation ($Â‘000), ordinary least squares
List of figures
Annual divorce rates for six English-speaking OECD countries, 1951Â–1998
A1. Age-specific divorce rates, Australian males, 1981Â–2001
About the authors
David de Vaus is Professor of Sociology and Dean of the Faculty of Humanities and Social Sciences, La Trobe University. Matthew Gray is Deputy Director at the Australian Institute of Family Studies (AIFS). Lixia Qu is a Research Fellow at AIFS and PhD student at La Trobe University. David Stanton is a Visiting Fellow, Crawford School of Economics and Government, The Australian National University.
A previous version of this report was presented to the 13th International Research Seminar on Issues in Social Security, Â“Social Protection in an Ageing WorldÂ”, organised by the Foundation for International Studies on Social Security, Sigtuna, Sweden, 16Â–18 June 2006. Valuable comments on the earlier version of this report were provided by Alan Hayes, Boyd Hunter, Prem Thapa and Peter Whiteford.
As the first generation that experienced high rates of divorce reaches retirement age, the number of older Australians who have experienced divorce at some point in their lives will increase dramatically in coming decades. There is very little empirical evidence in Australia on the financial consequences of divorce for older people. This report begins to fill this gap by providing some of the first estimates of the financial consequences of divorce for Australians aged 55 to 74 years. Using data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey, it was found that, on average, having been divorced had negative consequences for income in older age for both men and women. However, the negative financial impacts of divorce were substantially reduced by remarriage. For some measures of financial circumstances, those who had remarried following divorce were very similar to the married and never-divorced. Older divorced single Australians were much more likely to experience material hardships and to report having a lower level of prosperity than the married and never-divorced. The divorced and single were more reliant on the public pension than those who had not divorced. This will have important implications for the financing of retirement incomes in Australia in coming decades and the extent to which the taxpayer will have to bear the costs of providing for retirement incomes.
As the first generation that experienced high rates of divorce reaches retirement age, the number of older Australians who have experienced divorce at some point in their lives will increase dramatically in coming decades. The effect of the increase in divorce rates in the mid-1970s is compounded by the structural ageing of the Australian population, with the first of the Â“baby boomersÂ” (those born between 1946 and 1965) turning 60 in 2006. This means that not only will larger numbers of people be entering older age, but also that a much greater proportion of these people will have experienced divorce at some point in their lives.
Divorce is relatively unlikely to occur in older age, with the most common age for divorce being in the late twenties to early forties. Although most divorces occur before people enter later life, this does not mean that these earlier divorces are without consequences for people later in life. One potential impact is financial. There is very little empirical evidence in Australia on the financial consequences of divorce for older people. This report begins to fill this gap by providing some of the first estimates of the financial consequences of divorce for Australians aged 55 to 74 years.
A number of measures of financial living standards are examined, including:
annual household income;
receipt of income support payments;
per capita net household assets;
perceived prosperity; and
experience of financial hardships.
The analysis reveals that, on average, having been divorced has negative consequences for income and financial circumstances in older age. However, the negative financial impacts of divorce are substantially reduced by remarriage. For some measures of financial circumstances, the divorced and remarried are very similar to the married never-divorced.
Home ownership rates of both the divorced and remarried and the divorced singles were lower than the rate of those who were married and had never-divorced.
Three-quarters of the married never-divorced men owned a home outright, compared to just 40.9 per cent of the divorced single men and 57.8 per cent of the divorced and remarried men.
The pattern was similar for women. The main difference between men and women following divorce is that older divorced single women had higher rates of outright home ownership (49.4 per cent) than older divorced single men (40.9 per cent).
Divorced and remarried people, especially men, were the most likely to be purchasing a home.
Those who were divorced and single were substantially more likely to be renting than the married (ever- and never-divorced). Of the divorced single men, 49.4 per cent were renting, compared to just 20.9 per cent of the divorced and remarried men and 15.0 per cent of the married never-divorced men.
Divorced single men and women had lower median levels of per capita household assets than those who were married and never divorced ($199,900 and $178,300 respectively).
For both men and women, per capita assets of the divorced and remarried and the married never-divorced were similar. Taken overall, it appears that remarriage following divorce returned men and women to a similar net asset position as the married never-divorced.
Married never-divorced older women were substantially less likely to have had superannuation (37.8 per cent) than divorced and single women (49.3 per cent) and divorced and remarried women (46.4 per cent). However, the divorce history of women makes relatively little difference to their average amount of superannuation.
Older divorced single men were less likely than either divorced remarried men or married never-divorced men to have superannuation (43.6 per cent, 57.5 per cent and 57.9 per cent respectively). Divorced singles also had much lower levels of superannuation assets ($44,600) than the other groups. Remarried divorced men had the highest amount of superannuation ($128,300), with the married never-divorced having an average amount of superannuation of $100,000.
For older men, the median household equivalent income (household income adjusted for household size and composition) was lowest for divorced single men ($15,500), followed by married never-divorced men ($24,500), and was highest for those who had divorced and remarried ($28,900).
For women, there was no relationship between marital history and median incomes (ranging from $22,000 for divorced singles to $22,900 for married never-divorced women).
Divorced and single men and women received higher levels of income support payments (including the age pension) than either the divorced and remarried or the married never-divorced.
Perceived prosperity and material hardships
Older divorced single Australians are much more likely to experience material hardships than the married never-divorced.
For men, the divorced and remarried are more likely to report having experienced financial hardship than the married never-divorced, but less likely than the divorced and single. For women, no difference in the rates of experiencing financial hardship were found between the divorced and remarried and the married never-divorced.
For both men and women, the divorced and single reported having a lower level of prosperity than the married never-divorced. The self-reported prosperity of the divorced and remarried and the married never-divorced were similar.
Differences in the educational, employment and other characteristics between the groups did not appear to explain these relationships between marital history and financial circumstances.
The finding that divorced singles were more reliant on the public pension than those who had not divorced has important implications for the financing of retirement incomes in Australia in coming decades and the extent to which the taxpayer will have to bear the costs of providing for retirement incomes.
While the Australian age pension system offsets some of the financial disadvantages faced by divorced older people who are single, the lower level of financial living standards experienced by those who have been divorced could be reduced by encouraging greater levels of labour force participation among divorced women prior to retirement age. Other strategies could involve assisting the divorced to obtain further education or retraining, and delaying retirement. However, increased labour market earnings alone will almost certainly not completely offset the negative financial consequences of divorce in older age.
As the first generation that experienced high rates of divorce reaches retirement age, the number of older Australians who have experienced divorce at some point in their lives will increase dramatically in coming decades. Due to the sharp increase in divorce in the mid-1970s, there is a Â“divorce bulgeÂ”, where an increasing number of divorced people are now reaching later life.1 The effect of this divorce bulge is compounded by the structural ageing of the Australian population, with the first of the Â“baby boomersÂ” turning 60 in 2006.2 This means that not only will larger numbers of people be entering older age, but also that a much greater proportion of these people will have experienced divorce at some point in their lives. Other countries have also experienced very substantial increases in divorce rates (see section 2) and will also have an increasing proportion of their older age population who have been divorced at some point in their lives.
While divorce can occur at any age, it is relatively unlikely to occur in older age, with the most common age for divorce being in the late twenties to early forties (Appendix, Figure A1). Although most divorces occur before people enter later life, this does not mean that these earlier divorces are without consequences for people later in life. This report focuses on the implications of divorce, occurring at any point in adulthood, for peopleÂ’s financial wellbeing in later life.
There is an extensive literature on the impact of divorce on the specific financial circumstances of women. Studies from a number of countries have found that women experience a decline in financial circumstances following divorce.3 What has received less attention is the lifetime financial consequences of divorce.4 There is little Australian research on the financial consequences of divorce in older age (retirement age). Whiteford and Stanton (2002) reported that divorced, separated and never-married women were less likely to have assets than widows, who, in turn, were less likely to have assets than married women. Overall, around 45 per cent of divorced or separated age pensioners had less than $5,000 worth of assets (Whiteford & Stanton, 2002).
This report extends our understanding of the consequences of divorce in older age using a new longitudinal Australian survey, the Household, Income and Labour Dynamics in Australia (HILDA) Survey.5 HILDA is AustraliaÂ’s first nationally representative large-scale household-based panel. It contains detailed information on income and the type and value of both assets and liabilities. In this report, income levels, home ownership rates, value of superannuation and net household assets, reliance on government income support and the experience of financial stress of older Australians are examined. Using these measures, we compare the financial situation of older people who have ever divorced to that of older Australians who are married and have never divorced. Comparisons are also made between older Australians who have been divorced and are currently remarried and those who have divorced and are currently single.6 The results in the paper provide some preliminary estimates of the likely impact of divorce on financial circumstances in older age. This is an area in which the authors intend to conduct further research, including using the longitudinal nature of the HILDA to extend the analysis presented in this report.
The focus on assets is important, given that asset ownership can confer a number of advantages on older people. Assets can be invested to produce an income or, in the case of home ownership, can reduce the need for income to pay rent. Assets can also be sold to meet consumption needs, including care (Whiteford & Stanton, 2002). Analysis of the range of financial variables available from HILDA provides a comprehensive picture of the financial position of older Australians who have been divorced, compared to older Australians who have never divorced.
The potential effects of divorce on retirement incomes and assets are complex and depend on a range of factors, including the effects on employment and retirement decisions post-divorce, the proportion of pre-divorce assets received by each partner, the stage of life at which the divorce occurred (including whether the couple had children), whether or not the divorced person remarried, the amount of child support received or paid, whether the couple had entered the housing market, and whether additional government benefits were received.
There are good reasons for expecting divorce to have a negative effect on retirement incomes. These include the effects of divorce on asset accumulation, the impact of legal fees incurred in negotiating property settlements, and the increases in living costs when a family separates (related to the loss of economies of scale). It is generally argued (supported by the limited data available) that the negative impact is greatest for women who have had children, since these women are the least likely to have a strong labour market position to enable them to recover financially. It is expected that the negative financial impact of divorce on women would be ameliorated if they repartnered with someone who had similar financial assets and earnings as their first husband. On the other hand, divorce may prompt an increase in labour force participation for financial reasons (in the economic literature, termed an Â“income effectÂ”).
The remainder of this report is structured as follows. Section 2 describes divorce rates across a number of countries. In the third section the Australian system of provision for retirement incomes is outlined, and in the fourth the data used to analyse the impact of divorce on financial living standards in older age are described. In section 5 the socio-economic and demographic characteristics of older Australians who have been divorced are compared to those who are married and never divorced, and in section 6 the effects of divorce on financial circumstances are analysed. In the final section concluding comments are made.
2 International trends in divorce rates
Australia is not alone in confronting the dual impact of the baby boom and the rapid increase in divorce rates. In English-speaking OECD countries, at least, there was a divorce boom between the late 1960s and the mid-1980s. Although the precise timing of the divorce boom varied between countries, the pattern of a rapid and substantial increase in divorce was common among most English-speaking OECD countries. Figure 1 shows the pattern in six such countries.
Figure 1: Annual divorce rates for six English-speaking OECD countries, 1951Â–1998
In England and Wales, rates started to climb slowly in the early 1960s, but rose sharply in 1971 following the 1969 Divorce Reform Act, which included Â“no-faultÂ” grounds for divorce. Similarly, in Australia, the divorce rates started to climb from the mid-1960s and rose very sharply following the introduction of the Family Law Act 1975 that introduced no-fault divorce. The sharpness of the divorce peak in 1976 was partly due to a backlog of long-term marital separations for which the end of the marriage was formalised as soon as the new Act came into effect. Following the 1976 peak, the divorce rate subsequently declined and has since been stable, but at a much higher level than prior to the introduction of the new Act.
New Zealand exhibited a very similar pattern a decade later, again with a pattern of gradual increase in divorce rates, and a sharp rise in 1981 following the introduction of no-fault divorce in the same year.
The pattern in the US is also similar to these countries. The rise in divorce began in the mid-1960s, but rose strongly from 1969 through to 1979. The first US no-fault divorce was introduced in 1969 in California under Governor Ronald Reagan. The rate of the US increase in divorce was less steep and more prolonged than in the countries discussed above, as over the decade no-fault divorce was only introduced gradually in its different state jurisdictions.
The Canadian pattern is a little different. The increase in divorce came in two waves. There was an initial rise in the early 1960s, followed by the passing of the Divorce Act in 1968, and there was another rise in 1986 due to further changes to the Divorce Act in the same year.
All these countries, and no doubt many other OECD countries, will now be experiencing a substantial increase of older people who at some earlier point in their lives have been divorced. High annual rates of divorce translate eventually into large numbers and high percentages of older people having experienced divorce. In Australia it is estimated (depending on the method of estimation) that between a third and 45 per cent of people who marry will divorce. The US has higher rates of people who ever divorce. We can therefore expect that, in many countries, up to half of the people who are entering later life, either now or in the future, will have been divorced at some point.
As governments explore ways of managing the financial consequences of structural ageing and seek to encourage financial self-sufficiency of older people, they will need to be aware of any financial consequences of divorce earlier in life on financial wellbeing and self-sufficiency in later life.
3 The Australian system of provision for retirement incomes
In order to understand the potential impact of divorce in Australia on assets and income in older age, it is necessary to understand the Australian system of provision for retirement incomes. This system differs in some important respects from that of many other OECD countries.7
The Australian system of provision for retirement incomes can be characterised as a three-pillar model.8 The three pillars are:
a flat-rate, means-tested pension financed from general government revenue (the age pension);
compulsory saving through an employment-based system, known as the Superannuation Guarantee; and
voluntary superannuation, assisted by tax concessions or other private saving, particularly housing.
The first pillar was introduced in Australia at the beginning of the 20th century. The second pillar of compulsory saving through the Superannuation Guarantee was introduced much more recently, during the 1990s, and now covers the majority of the employed population. However, it will take many years before those reaching retirement age will have had paid sufficient compulsory superannuation throughout their working lives to make a significant difference to their retirement incomes. The policy intention is that the mandatory second pillar and voluntary third pillar will reduce the pressure on the government of providing direct financial support for retirement incomes.
3.1 The age pension
A feature of the Australian system is that public spending on age pensions is low compared to most other OECD countries. Nevertheless, coverage is comprehensive, and the system appears to be highly redistributive to groups often poorly served by social insurance systems, such as women, those with long-term disabilities, low-wage earners and others with marginal or incomplete attachment to the labour force.9 Indeed, the Australian pension system has been described as Â“radically redistributiveÂ” (Aaron, 1992). Males aged over 65 are eligible to receive the age pension subject to income and asset tests. The age at which females are eligible to receive the age pension depends on when they were born and ranges from 60.5 years for those born before 31 December 1936 to 65 years for those born after 1 January 1949.
The Australian age pension is income-tested. Income above a specified amount (the Â“free areaÂ” or income disregard) reduces the amount of age pension received. For single age pensioners, each dollar of income above $128 per fortnight decreases the amount of pension received by 40 cents. For married (partnered) age pensioners, each dollar above $228 per fortnight reduces the amount of pension received by 20 cents (see Table 1 for details). The amount of private income at which the age pension cuts out is $1,423 per fortnight for single age pensioners and $2,381 per fortnight for (married) couples.
Table 1: Payment rates of the Australian age pension system, September 2006
Value in $A at September 2006
Standard (single) pension rate
$512.10 per fortnight (plus $18.20 pension supplement)
Married pension rate (each)
$427.70 per fortnight (plus $15.20 pension supplement)
Supplementary Rent Assistance
Up to $103.20 per fortnight for singles, $97.40 per fortnight for couples
$5.80 per fortnight (single and combined)
Income allowed before pension is reduced (free area or income disregards)
Single $128.00 per fortnight
Combined Married $228.00 per fortnight
40% for single and 20% for each of a couple
Single $1,422.75 per fortnight
Combined Married $2,381.00 per fortnight
No rate paid above:
Single home owners $161,500
Single non-home owners $278,500
Married home owners $229,000
Married non-home owners $346,500
No rate paid above:
Single home owners $334,250
Single non-home owners $451,250
Married home owners $516,500
Married non-home owners $633,500
Notes: * Assets over this amount reduce the pension received by $3 per fortnight for every $1,000 above the limit.
Source: A guide to Australian Government payments, 20 SeptemberÂ–31 December 2006.
In addition to the income test, there is an assets test applied to the receipt of the age pension. The rate of pension is calculated under both the income and assets tests, with the test that results in the lower rate being the one applied. The age pension is indexed twice each year, in line with the Consumer Price Index (CPI). The Australian Government also maintains the single rate of pension at a minimum of 25 per cent of male total average weekly earnings (MTAWE), and this flows on to the married pension rate.
The assets test thresholds are very high and exclude the value of the family home. Rather than being a residual system targeted only to the very poor, the Australian pension system is designed to exclude the relatively well-off. The extent to which the income test targets only the well-off is illustrated by the fact that nearly 80 per cent of older Australians receive at least a part pension. Indeed, a couple could still receive some pension even when their private income in retirement exceeds average male earnings.
In addition to the basic rates of payment set out in Table 1, pensioners may qualify for additional assistance, depending on their circumstances. This includes Rent Assistance, Pharmaceutical Allowance, Telephone Allowance, Remote Area Allowance and pension concession cards. Concession cards entitle the holder to Commonwealth health concessions, such as low-cost pharmaceuticals. State-based concessions may include reductions in property and water rates, reductions in energy bills, reduced fares on public transport, reductions on motor vehicle registration, and other health, household, educational and recreational concessions. Concession cardholders are also more likely to receive free medical care under Â“bulk-billingÂ” arrangements, although this is at the discretion of private doctors. In addition, many private sector companies provide discounts to concession card holders.
The proportion of employees with superannuation increased from 55 per cent in 1988 to 91 per cent in 2000 (Australian Bureau of Statistics [ABS], 2002). The amount of superannuation held is closely linked to the value of contributions made to superannuation over time. In 2000, the median total superannuation assets for employees aged 15Â–64 years was $10,200 and the median balance for male employees was more than double that of female employees. Data from the Australian Bureau of Statistics for 2000 reveal that Â“the difference between the superannuation balances of male and female employees increased with age to the point where male employees of retirement age (55Â–64 years) had more than twice the amount of superannuation of female employees ($44,700 compared with $19,800)Â” (ABS, 2002, p. 178).
A number of papers have discussed differences between men and women in amounts of superannuation accumulated over their working lives (Jefferson & Preston, 2005; Kelly, Percival, & Harding, 2002). In particular, there has been discussion on the impact on superannuation of broken work patterns and low earnings (both because of low hourly wages and because of part-time employment), which are much more common for women than men. The assumption that is often made is that the much lower levels of superannuation that will be accumulated by women who have had children will be compensated for by higher levels of superannuation accumulated by their partner, and that this superannuation is shared. While this is probably true while a couple remains married, it usually breaks down following divorce.
Superannuation fund assets increased from only 3 per cent of GDP in 1972 to 19 per cent in the early 1980s, and in 1999 were equivalent to 70 per cent of GDP (Whiteford & Stanton, 2002). Government estimates suggest that by 2020, fund assets could approach 120 per cent of GDP (Tinnion & Rothman, 1999). While the new compulsory superannuation system has grown rapidly in significance, it will be many years before the system is mature. As a result, the living standards of current pensioners are largely influenced by the public pension system and its interaction with private sources of income and wealth. The largest part of superannuation payouts is in the form of lump sums, which people then use in retirement as they see fit.
Divorce may have an impact upon financial circumstances (current financial living standards and net assets) in a number of ways. These include the following:
Divorce will normally result in the creation of two households rather than one. This may lead to a decline in living standards and to the loss of economies of scale, which in turn will make it more difficult to save and accumulate assets.
Divorce may affect labour force participation. In some cases, it may lead to withdrawal from the labour force (for example, to enable a lone parent to care for children or qualify for financial assistance or reduce the non-resident parentÂ’s liability for child support). In other cases, divorce may require a person to re-enter the workforce or to stay in the workforce longer than they may have intended so that they can accumulate sufficient funds for retirement.
The amount of age pension (and most other government income support payments) depends upon relationship status.
Following marriage breakdown, a coupleÂ’s assets need to be split. If the couple is unable to agree on how to do this, the Family Court (or Federal Magistrates Court) decides.10 The general approach used by the Family Court is to identify what each spouse owns and owes, the contributions made by the respective members of the couple during the course of the marriage and future requirements. There is no assumption that the starting position should be a 50Â–50 split. The Court considers both the financial and non-financial contributions made by each spouse.
In 2002, amendments to the Family Law Act gave courts the capacity to treat superannuation assets in the same way as any other form of property. The 2002 amendments require that superannuation assets be disclosed to the Court and that their value be taken into account when determining each member of the coupleÂ’s entitlements to assets.
Research by Sheehan and Hughes (2001), using data from the Australian Divorce Transitions Project collected by the Australian Institute of Family Studies in 1998, reported that, on average, 55 per cent of property and financial resources at the time of distributing assets following divorce was received by the wife and 45 per cent by the husband. In 42 per cent of cases, respondents reported that the wife received 60 per cent or more of the property. Twenty-nine per cent of respondents reported receiving the middle range of 40Â–59 per cent each, and 29 per cent reported the wife receiving less than 40 per cent of the property. Focusing on property only, Sheehan and Hughes (2001) found that 59 per cent of women received 60 per cent or more of the property.
4 Household, Income and Labour Dynamics in Australia (HILDA) Survey
4.1 Overview of HILDA
This analysis of the effects of divorce on the incomes, assets and living standards of older Australians is based upon the HILDA Survey. The HILDA Survey has a number of advantages for studying this issue. These advantages relate to its detailed information on current marital status and marital history, the size and national representativeness of the sample, the household nature of the survey and its detailed information on household wealth.
The annual survey, which commenced in 2001, tracks all members of an initial sample of 7,682 households across Australia. The HILDA Survey collects data in three main areas: economic and subjective wellbeing, labour market dynamics, and family dynamics. This report uses data from the second wave of the survey that were collected in 2002Â—the only wave to date that collected detailed information on household wealth.
Within each household, interviews were sought from all members aged 15 and older. In wave 1, face-to-face interviews were conducted with nearly 14,000 household members and further information was obtained from self-completed questionnaires from 13,055 of these household members. Watson and Wooden (2004) provide a detailed discussion of the design of the survey. The rate of attrition between wave 1 and 2 was 13.2 per cent, with 86.8 per cent of respondents in wave 1 successfully re-interviewed in wave 2 (Watson & Wooden, 2004).
4.2 HILDA measures of income, wealth, financial hardship and prosperity
Wave 2 of the survey contained detailed information on income and wealth. The following financial variables are analysed for this report:
annual household income;
annual household income adjusted for household size and composition (equivalised household income11);
housing tenure (own house without debt, purchasing house, other housing tenure (mainly rental));
whether a person had superannuation, and the value of that superannuation;12
whether a person was in receipt of a government income support payment (including the age pension); and
per capita net household assets13 (net assets in household divided by number of adults in the household).
A significant proportion of respondents were unable to quantify the value of one or more of the types of assets or liabilities. Taken overall, total net household wealth could not be directly determined for just over 39 per cent of all wave 2 responding households (Watson & Wooden, 2004). For individual respondents with missing information on assets or liabilities, a set of variables that imputed the missing information was included with the data set (see Watson and Wooden (2004) for a detailed description of the imputation process used). The analysis in this report uses measures of wealth that include imputed data; for individuals with missing information on income, the imputed income variable is used.
Although the proportion of respondents with imputed values for one or more types of assets or liabilities was relatively high, in many cases the missing information was restricted to only one or two of the asset and liability items.14 After a comprehensive assessment of the quality of the assets and liabilities data in wave 2 of HILDA, Watson and Wooden (2004) concluded that Â“overall, it is our assessment that the HILDA Survey has done a reasonable job in measuring total household wealth. Nevertheless, net worth is almost certainly overstated. This reflects both under-reporting of debts and, though more speculative, over-reporting of assetsÂ” (p. 24). They also emphasised that the main weakness associated with imputing missing data was that Â“even if imputation does generate unbiased cross-section estimates, estimates of change are unlikely to be so well behavedÂ” (p. 24).
While income and assets provide measures of financial circumstances, the welfare or poverty consequences of a lack of income depend upon a range of factors, including individual consumption preferences and budget constraints, and can only be drawn with reference to a set of prices (Mayer & Jencks, 1989; Travers & Richardson, 1993). Consequently, we include subjective measures of financial stress and wellbeing. In the HILDA Survey, respondents were asked the question: Â“Given your current needs and financial responsibilities, would you say that you and your family are prosperous, very comfortable, reasonably comfortable, just getting along, poor, or very poor?Â” We term this measure Â“perceived prosperityÂ”.
The HILDA Survey also contains questions on financial difficulties. Respondents were asked whether any of the following things had occurred in the previous 12 months due to a shortage of money: could not pay electricity bills, gas or telephone bills on time; could not pay the mortgage or rent on time; pawned or sold things; went without meals; was unable to heat their home; asked for financial help from friends or family; or asked for help from welfare/community organisations. Given the relatively low incidence of these financial hardships for older Australians, the measure used in this report is whether or not the respondent had experienced one or more of these hardships in the previous 12 months.
4.3 Characteristics of the sample for analysis
Our analysis is restricted to males and females aged 55 to 74 years who were married at the time of the interview or who had previously been married. This age grouping was chosen in order to obtain a sufficient sample size of older Australians to allow an analysis by gender.15 The age of 55 is also the age beyond which labour force participation rates drop substantially.
According to the HILDA Survey, 17 per cent of males aged 55 to 74 years who had ever been married had divorced in the past but were remarried at the time of interview, and 9 per cent were divorced and single at the time of interview. Overall, almost 26 per cent had ever divorced, 72 per cent were married and had never divorced and 3 per cent were widowed. For females, 11 per cent had been divorced but were remarried at the time of the interview, and 11 per cent were divorced and single at the time of interview. Overall, 22 per cent had ever divorced, 62 per cent were married and had never divorced and 16 per cent were widowed.
David de Vaus, Matthew Gray, Lixia Qu and David Stanton. Australian Institute of Family Studies, February 2007, 27 pp. ISBN 978 0 642 39546 7. ISSN 1446-9863 (Print); ISSN 1446-9871 (Online)
Read the entire paper here: