Introduction

The Commonwealth Government's Intergenerational Reports (IGRs) highlight concerns that population ageing is increasing costs to the state and reducing workforce supply (Treasury 2010). Related research studies by the Australian Institute of Health and Welfare (AIHW; Jenkins et al. 2003) and the National Centre for Social and Economic Modelling (NATSEM; Nepal et al. 2011a) present the projected care needs of an ageing population and projected shortfalls of informal caregivers. Demographic ageing, increasing longevity and continuing economic uncertainty have also generated intense concern for the sustainability of retirement income systems and care systems in developed and developing countries around the world (OECD 2006, 2011).

The Government's third and most recent IGR (Treasury 2010) concluded that maintaining workforce participation and increasing productivity are priorities for Australia in responding to the future challenges of economic uncertainty and population ageing. Positive views on the role of older people have been presented by National Seniors Australia (NSA 2009) and the Treasury's Advisory Panel on the Economic Potential of Senior Australians (EPSA; Treasury 2011). The Government Response to the final EPSA report (Treasury 2012) accepted recommendations for action in key areas including active ageing, volunteering and age discrimination, outlining a number of important initiatives now underway that have prospects for constructively enabling more contributions by older people. According to the Council on the Ageing, the most important is the 10 year Plan on Positive Ageing, which 'provides an opportunity to develop a comprehensive approach across all aspects of economic and social life to ensure older people are able to participate fully as citizens and members of society' (COTA 2012: 2). A common theme in these reports is the importance of recognising, enabling and assisting Australians to continue productive contributions as they age.

However, there has been limited research outside these commissioned reports containing population trends and projections specifically for older age groups, and past research typically has not presented corresponding data on younger age groups. Providing comparative data for those in early, middle and later life is important for research on productive engagement over the life course. Therefore, the main objective of the present study is to present an independent descriptive analysis of recent data on the productive contributions of older Australians alongside comparative data for younger age groups.

This study aims to contribute to the literature by 1) providing a review of recent research and 2) presenting comprehensive data on engagement in both paid and unpaid productive activities by younger, middle and older age groups, as an indication of change over the life course, which may act as a useful reference for future research on productive engagement in Australia. The results are disaggregated by key demographic and human capital factors that have previously been associated with productive engagement, including age, gender, health, education, income, occupation and marital status. The paper concludes with a summary of research directions and policy actions that could enable productive engagement in mid and later life.

Theoretical framework

Human capital

Opportunities and necessities to make economic and social contributions can depend on an individual's physical and mental health capacities, their motivations and skills, their family and community context, competing pressures on time use and the structural incentives and constraints imposed by employment markets and public policies (WHO 2010; NSA 2012). The idea that personal and social factors, such as health and education, are resources that can either enable or constrain an individual's ability to make productive contributions, and that these resources are modifiable, is consistent with the concept of human capital (Mirowsky & Ross 1998). A recent working paper by the Productivity Commission (Shomos 2010: 1) linking literacy and numeracy skills to labour market outcomes defined human capital as 'the set of attributes that individuals possess, including knowledge, skills, work experience, health and intangible characteristics such as motivation'. According to human capital theory, such resources may be acquired or enhanced through education, training or other informal means including 'the experiences of undertaking daily activities at home or at the workplace' (Shomos 2010: 1). The social policy implications of this are that an investment in human capital can yield positive returns by increasing future productivity. Supporting this view and the Government's current interest in improving human capital, recent research in Australia has found that improvements in education, training and health are related to increased wages and labour market productivity (McDonald 2010; Shomos 2010; Nepal et al. 2011b).

Given that the return on investments in human capital is likely to appear in the medium to longer term, a life course approach is useful for examining productive activities in middle to later life. A key idea consistent with contemporary developmental life course perspectives (for example, Elder 1995) is that a person's life course or developmental trajectory is influenced by earlier life experiences and exposures, which have in part been shaped and reinforced by age-, gender- and other socially-graded norms, expectations and institutions within the particular cultural context (Moen 1996; Moen & Spencer 2006; Hershey et al. 2010). In addition, because education, income, health and productive engagement can vary throughout the life course, examining productive engagement not only in middle to later life, but also amongst earlier age groups will add to our understanding of how productive engagement varies across the life course by gender, age and human capital resources.

Marital status and occupation are not typically included in the human capital framework, apart from use as demographic control variables, but they may also be useful indicators of life-long personal resources and available kinship networks (Kendig et al. 2007; Kjellsson 2013). Marital status may be positively associated with productive engagement when one's personal resources are extended to include the resources of their partner (Glass et al. 1995; Hesketh et al. 2011). Being partnered can widen one's kinship network and increase the probability of providing care for children, grandchildren and adult relatives; it has also been associated with other benefits such as good health (Kendig et al. 2007), particularly for men, that may directly affect productive engagement. In addition to its association with education and income, occupation can provide further human capital relevant to engagement in productive activities. For example, individuals in managerial and professional jobs may have more opportunity to develop further work-related skills and knowledge, while those in more physical or manual jobs may experience limitations including poorer health and wage outcomes (Kjellsson 2013).

Productive ageing

The global significance of ageing well and ageing productively has been brought to attention by three major worldwide trends. First, the demographic trend of people living longer increases the potential for continued productive engagement, particularly if the prevalence and/or negative impacts of poor health and chronic disease are reduced (Mor 2005; Mathers 2007). Second, there has been a corresponding sociocultural trend in which people have been showing less inclination to retire from the paid workforce in the more traditional sense, with many choosing (if possible) to remain active and continue productive activities (Morrow-Howell et al. 2001a). Third, the global trend of continuing economic uncertainty highlights the increasing need for productive engagement across all age groups (Chomik & Piggott 2012). While the United States has experienced recent rises of unemployment among older people, the main effect of the recent global financial crisis (GFC) in Australia appears to have been to encourage some delays in planned retirement and adjustments to lower expenditure for those reliant on investment returns for retirement income (O'Loughlin et al. 2010; Kendig et al. in press).

The concept of productive ageing emerged in the United States in the 1980s, largely as a positive reaction against the dominant, ageist view of ageing as a process of declining physical and cognitive health and functioning, resulting in increased dependency and dwindling contributions (O'Reilly & Caro 1994). The problem with this dominant view is that it devalued older people, failing to acknowledge their ongoing contributions and potential for productive engagement in a range of activities, not only paid work. In contrast, the productive ageing perspective recognises that individuals can and often do continue their engagement in activities that have socioeconomic value within their social context as they age.

Although the meaning of productive ageing can vary greatly across different cultures, societies and individuals (Carr et al. 2013), productive ageing may be defined as the engagement of older people in activities that produce, or develop the capacity to produce, goods or services that are either paid for or unpaid, but otherwise would have to be paid for (O'Reilly & Caro 1994; Bass & Caro 2001). Productive ageing can be defined at the societal level or at the individual level. At the societal level, productive ageing refers to older people's engagement in activities that can be assigned some social or economic value such as contributions of money, time, effort and information. At the individual level, productive ageing includes activities that improve an individual's physical or mental health, which have potential positive carry-on effects of decreasing dependency and increasing capacity to contribute at the broader social or economic level (Birren 2001). Defined broadly, productive activities can include paid work, volunteering, caregiving, and other informal help to family and friends, as well as activities that have indirect value to the broader community and directly benefit the individual, such as self-care or self-maintenance (that is, looking after one's self), and further education and training (Morrow-Howell et al. 2001a; Kendig & Browning 2010).

A recent Australian review has warned that the concept of productive ageing can have pejorative, normative connotations, particularly when less privileged workers are expected to continue working at the expense of a period of personally-rewarding, meaningful activity including leisure after retirement (Carr et al. 2013). Thus, it is important to note that the alternative to productive ageing or activities is not necessarily unproductive ageing or activities (Morrow-Howell et al. 2001b; Walker 2008). Rather, Morrow-Howell and colleagues (2001b) suggest that productive ageing should be viewed as one of several goals that one may wish to pursue in later life, and different individuals may have different goals for later life, all of which have value. Thus, although productive ageing can have real and immediate economic value to both the individual and the broader community, it is not an ultimate goal that should be imposed on all people. In our view, public policy should aim to promote and support or enable productive ageing as an option for those who wish to continue productive activities in later life.

Prior research

With the exception of a few published studies (such as Dosman et al. 2006; Fernfindez-Ballesteros et al. 2011), the majority of prior research on productive ageing has been conducted within the United States. As Hinterlong (2008) noted, research on productive ageing has been limited in part by the lack of consensus about which activities should count as productive. Paid work has been the most extensively researched productive activity (for example, McDonald 2011), although there has been increasing research on less formalised, typically unpaid activities that are more difficult to measure, such as volunteering (for example, Tang et al. 2010), caregiving (for example, McMunn et al. 2009), childcare and domestic work (for example, de Vaus et al. 2003). Even less attention has been paid to individual-level productive activities such as self-care or self-maintenance, education and training (for example, Brandon & Temple 2006).

From his review of American research prior to 2006, Hinterlong (2008) reported that the majority of older individuals were involved in at least one productive activity, with a significant proportion involved in multiple activities. Following 1,644 Americans aged 60 and over in 1986 from the Americans' Changing Lives Study (ACL) for three waves (1986, 1989, 1994), Hinterlong (2008) also found that the majority (74 per cent) of respondents engaged in one or more productive activities at each wave, although individuals varied significantly in the amount of time spent on each activity and this individual variability increased over time while mean time decreased.

Much of the research on the contributions of individuals in Australia has typically focused on only one or two productive activities. For example, McDonald (2011) presented useful data on paid work participation for Australians aged 55 years and over, while Bittman and colleagues (2007) examined how caregiving impacts the paid work participation of Australians aged 15 to 64 years. Using data from the Australian Bureau of Statistics (ABS), National Seniors Australia (NSA 2009) reported that 71 per cent of Australians aged 55 to 59 years of age were engaged in paid work, which was the highest participation rate for this group since 1978. NSA (2009) also reported that, among Australians aged 55 and over, the percentage of those engaged in voluntary work increased steadily between 1995 and 2006. One of the few studies to examine multiple activities was conducted by Brandon and Temple (2006), but did not include paid work. Their time diary data for 1,350 individuals not in the workforce and aged 55 years and over from the 1997 Australian Time Use Survey (TUS), suggested that while all older Australians spent time on domestic work, many (42 per cent) also spent time in volunteering, caregiving, and childcare activities.

Gender

In an early overview of productive ageing in the United States, O'Reilly and Caro (1994) noted significant gender differences, with a greater proportion of women than men engaged in unpaid productive activities such as domestic work, childrearing, caregiving and volunteering. They also reported that more women experienced conflicts between paid work and these unpaid activities, which could lead to a reduction in or withdrawal from engagement in paid work earlier in life. Consistent with a life course perspective, they noted that having an interrupted or reduced engagement in paid work earlier in life could also limit one's access to opportunities for paid work in later life. Although gender was not a strong predictor of volunteering, women tended to volunteer more than men in later life, and also tended to give more time.

More recently, in their study of 458 healthy Spanish adults aged 55 to 75, Fernfindez-Ballesteros and colleagues (2011) reported that women spent significantly more time performing productive activities than men, but this was largely due to more time spent in domestic work. In Australia, McDonald (2011) presented data from 2000 to 2010 indicating a trend of increasing paid work engagement among people 55 years and older in more recent years, particularly for women, although their rates of engagement remained substantially lower than men in every year. Interestingly, while partnered men aged 55-59 and 60-64 were more likely than unpartnered men to be in paid work, there was little difference at older age groups or for women. Amongst older individuals not in paid work, Brandon and Temple (2006) also found that, controlling for age group, marital status, education level, English as a first language, disability status, income, vehicle ownership and geographic location, being female increased the likelihood of engagement in volunteering and caring activities in addition to domestic work.

Human capital and other personal resources

The literature also suggests that, health permitting, the majority of older individuals is likely to continue engagement in a range of productive activities as they age, although the extent of engagement can vary greatly between individuals, particularly at older ages (Li & Ferraro 2006; Hinterlong et al. 2007; Hinterlong 2008). Analysing data from the 2004 Survey of Health, Ageing and Retirement in Europe (SHARE), Erlinghagen and Hank (2006) found that being younger, having better health and higher levels of education increased the likelihood of engagement in volunteer work.

Similarly, Brandon and Temple (2006) found that having lower rates of disability, being younger, widowed and more highly educated increased the likelihood of engagement in volunteering and caring activities in addition to domestic work amongst older Australians not in paid work. Regarding paid work, McDonald (2011) reported higher rates of engagement amongst older Australians with no long-term health conditions or functional disabilities, higher education levels, greater English proficiency and those living in a capital city. These results provide early evidence that personal and human capital factors such as health or disability status and educational attainment can enable or constrain one's capacity to engage in multiple productive activities beyond paid work.

Time as a resource and constraint

Time is another personal resource that can affect engagement in productive activities. This is most evident in the literature providing evidence of the potential time trade-off or competition for time between different productive activities, usually between paid work and other productive activities. For example, Fernfindez-Ballesteros and colleagues (2011) found that retirees spent more time on adult caregiving and formal volunteering than non-retirees. Similarly, Erlinghagen and Hank (2006) found that workers were less likely than retirees to be engaged in volunteer work. In contrast, being an informal helper or carer increased the likelihood of engagement in volunteer work. In a smaller American study, Tang and colleagues (2010) found that the top two reasons for stopping volunteer work were declining health and commitments to other productive activities including paid work, caregiving or other volunteering programs.

Using a relatively young cross-sectional sample of 3,581 Canadians aged 45 and older from Statistics Canada's 1998 General Social Survey, Dosman and colleagues (2006) examined engagement in productive activities and daily time allocation by gender and work status. They found that compared to their still-employed counterparts, retirees spent more time in self-care and unpaid productive work (domestic work, caregiving and volunteering), and a small percentage (four to six per cent) of retirees still spent on average four hours per day in paid work. Retired men were more likely than employed men to be engaged in volunteering and domestic work, and were also likely to spend more time in domestic work than employed men. Surprisingly, fewer retired women than employed women were engaged in caregiving, but of those engaged, retired women spent more time caregiving than employed women. This mixed finding highlights the value of including time spent as a measure of productive engagement.

In their Australian study, Bittman and colleagues (2007) found that, among those initially in full-time work, the likelihood of moving to either part-time or no work was greater among carets than non-carets. In addition, among those initially not in the labour force, carers were much less likely than non-carers to begin either full-time or part-time work, suggesting that caring had both immediate and longer-term effects on engagement in paid work. These findings underscore the value of a life course approach that considers engagement not only in middle or later life, but also at earlier ages. Baxter and colleagues (2007) showed that both men and women with dependent children felt more time pressure than those without, and time spent with pre-school children on weekdays generally decreased with greater working hours. This suggests that younger age groups may also experience competition between engagement in paid work and childcare activities.

Current study

The present study adds to the existing literature by presenting relatively recent, nationally representative data on the extent of engagement in multiple productive activities including paid work, childcare, volunteering, caregiving and domestic work among Australians aged 15 years and over. We expect that consistent with earlier US and Australian research findings (Brandon & Temple 2006; Hinterlong 2008), the majority of individuals across all age groups will be engaged in one or more productive activities.

Engagement in each productive activity is expected to vary by age and gender, according to dominant social and cultural expectations of ageing and gender roles in Australia. Consistent with the literature (Cai & Kalb 2006; McDonald 2010, Nepal et al. 2011b), we expect that having more human capital and other personal resources such as better health, higher levels of education and income and being partnered, will be associated with higher engagement in paid work across all age groups. The relationship between human capital or personal resources and engagement in other productive activities is less clear. Prior research from overseas (Erlinghagen & Hank 2006; Li & Ferraro 2006; Hinterlong et al. 2007) suggests that the same relationship occurs for productive activities other than paid work. However, because of the potential for competition between paid work and other activities, and the positive association between paid work and human capital, it is possible that there might be a negative or no relationship between higher resources and other productive activities.

The current study takes a first step towards examining the relationship between personal resources and engagement in paid work and productive activities other than paid work within the Australian context. The research questions being addressed include 'what is the extent that Australians are engaged in single and multiple activities?', 'how are demographic and human capital variables related to engagement?' and 'how does engagement vary for different age groups?' Adding to prior research that has typically focused on one or two specific productive activities, this study presents data on the rates of engagement in a broader array of productive activities such as childcare, volunteering, caregiving and domestic work. These results are disaggregated across all working age groups by gender, health and education to examine how these factors are related to engagement over the life course. The study has a particular focus on individuals in the baby boomer cohort (born 1946-1965) who are currently under most pressure to continue productive engagement as they age. We present additional data on engagement for these older age groups disaggregated by household income, occupation and marital status, and examine which of these factors are significantly associated with their engagement in productive activities.

Method

Data and sample

The data were obtained from Wave 10 (2010) of the Household, Income and Labour Dynamics in Australia (HILDA) Survey. HILDA is a nationally representative household panel survey conducted annually since 2001 with an initial Wave 1 sample of 7,682 households (66 per cent response rate), comprising 19,914 individuals. Within households, only the 15,127 individuals aged 15 years and over were eligible for personal interviews, resulting in a Wave 1 sample of 13,969 eligible individuals (92 per cent response rate). At each wave, the sample is extended to include new household members following changes in household composition, household members turning 15 years of age and non-responding Wave 1 household members who decide to participate in a later wave. Attrition rates across waves are comparable to similar surveys internationally (Melbourne Institute of Applied Economic and Social Research 2011). Further details of the sample and survey methodology have been reported elsewhere (Watson & Wooden 2002).

Wave 10 included data from 13,526 individuals aged 15 to 93. Of these, 12,048 (89 per cent) returned their Self-Completion Questionnaire, which contained the time-use questions examined in this paper. The final sample comprised 10,131 individuals with complete age, gender, health, educational level and time-use data.

Measures and analysis

Demographic variables including age, gender, health, educational level, household income, occupation and marital status were obtained in the personal interview, while time spent on productive activities was obtained from the Self-Completion Questionnaire (SCQ). In the SCQ, respondents reported the number of hours they would spend in a typical week doing paid work, household errands, housework, outdoor tasks, looking after one's own children, looking after other people's children, volunteering and caring for a disabled or elderly relative.

Prior research (such as Van der Meet 2006) has often combined childcare with caregiving to adults or with other forms of instrumental support that assists others, including housework and household errands. However, given the possibility that caring for one's own children, caring for other people's children and caring for a disabled or elderly relative are likely to show different patterns of engagement over the lifespan, we kept these three types of caregiving separate.

Time spent on household errands, housework and outdoor tasks was summed to form a single domestic work variable. Engagement in each activity was then computed as a dichotomous measure with zero hours spent on an activity indicating non-involvement and more than zero hours indicating engagement. However, engagement in domestic work was very high (above 90 per cent) across all categories of age, gender, health, education, household income, occupation and marital status, with some subgroups containing fewer than five individuals not engaged in domestic work. To improve the reliability of both the domestic work engagement variable and the chi-square tests, domestic work engagement was divided into tertiles for a more continuous categorical measure comprising low (less than 10 hours), mid (10 to 21 hours) and high (more than 21 hours) domestic work engagement in a typical week.

To provide a relatively informative yet succinct snapshot of how gender, health and education relate to productive activities over the life span, respondents were divided into 10-year age groups from 15-24 years of age to 75 years and over. To examine the relationship between productive engagement and household income, occupation and marital status specifically for the baby boomer cohort, only respondents aged 45 to 64 were included, divided into 5-year age groups to maximise both sample size and information on age-graded trends.

Given that health is viewed as a potential enabler or resource for productive contributions over the life course, a measure of longer-term functional health was used instead of subjective health ratings, which can change more frequently following short-term fluctuations in one's current health status (Bailis et al. 2003; Han et al. 2005). Respondents who indicated they did not have any long-term health condition, impairment or disability that restricts them in their everyday activities were classified as having good health, while those who did were classified as having a long-term health condition. (Respondents were presented a list of examples including conditions such as uncorrected sight, hearing or speech problems, any condition that restricts physical activity or physical work and any mental illness which requires help or supervision.)

Consistent with prior research (for example, McDonald 2011), three levels of education were used: no post-school qualification, post-school qualification and university degree.

Separate chi-square tests were run for each productive activity with 10-year age group as the control variable and gender, health or education entered as the independent variable. In the baby boomer analysis, separate chi-square tests were run for each productive activity with 5-year age group as the control variable and income, occupation or marital status entered as the independent variable. Finally, to identify which of these factors distinguished between the baby boomers who engaged in one, multiple or zero activities, separate multinomial and binomial logistic regressions were run using data from the 45-64 year-olds.

Results

Sample descriptives for the whole sample and the baby boomer sub-sample are presented in Table 1. In the total sample, 19.5 per cent of respondents were aged between 15 and 24, 15.6 per cent aged between 25 and 34, 16.6 per cent aged between 35 and 44, 17.6 per cent aged between 45 and 54, 14.1 per cent aged between 55 and 64, 9.8 per cent aged between 65 and 74 and the remaining 6.8 per cent were aged 75 years and over. Amongst those who indicated engagement in an activity, the most time was spent in paid work, followed by domestic work, caring for one's own children, adult caregiving, caring for other people's children and volunteer work.

Table 2 shows engagement disaggregated by age and gender. The bottom third of Table 2 shows engagement disaggregated by age group only. With the exception of the youngest age group who had the highest engagement in part-time work (34.4 per cent), engagement in full-time work tended to be lower for older age groups, with a clear decrease for the 55-64 and 65-74 year-olds relative to the next youngest age group.

Not surprisingly, engagement in caring for one's own children was relatively high for those aged 25-54, and was highest among 35-44 year-olds (74.6 per cent), whereas engagement in caring for others' children was generally higher for older age groups, and was highest among 55-64 year-olds (17.8 per cent). Volunteering and caregiving show similar trends to engagement in caring for others' children, but volunteering was highest among 65-74 year-olds (31.9 per cent). Finally, engagement in domestic work tended to be lower in younger age groups and higher in the older age groups with the majority of 15-34 year-olds reporting low engagement in domestic work, while the majority of those aged 55 and over reported high engagement.

The mean number of activities across all respondents was 2.4 (SD = 1.0). Most respondents reported engaging in at least one activity (98.4 per cent), while 81.5 per cent reported engagement in multiple activities. Excluding domestic work, which most people engaged in, 82.2 per cent reported engaging in at least one activity, while 40.4 per cent engaged in multiple activities. Figure 1 displays the percentages of respondents engaged in zero, single and multiple activities, excluding domestic work and disaggregated by age and gender. Interestingly, the majority of younger and older age groups engaged in one or fewer activities (excluding domestic work), while the majority of those in the middle age groups (35-54 year-olds) engaged in multiple activities.

Age and gender

Across all ages, males and females showed different rates of engagement in all productive activities other than caring for one's own children (see Table 2 column headings). The most obvious gender difference in engagement across the life course was in paid work. With the exception of the oldest age groups, engagement in full-time work was generally lower among females than males, whilst engagement in part-time work was generally higher.

Females generally reported higher rates of engagement than males in caring for others' children and volunteering, and were also more engaged in caregiving for adult relatives from ages 35 to 64. However, engagement in caring for one's own children was higher for females than for males only in the 25-44 year age groups, whereas engagement among the 45-64 year-olds was higher for males than for females, possibly due to men having children at later ages than women.

To further examine the role of gender in engagement over the life course, we plotted the average number of hours spent by those engaged in each activity, disaggregated by age and gender (Figure 2). Consistent with age-graded norms and the above findings for rates of engagement, time spent on paid work and caring for one's own children tended to be higher among younger age groups and lower for older age groups. In contrast, time spent in other activities tended to be lower for younger age groups and higher for older age groups. In the age groups between 15 and 64, males tended to spend more time in paid work than in any other activity, whereas females aged 15-34 spent as much time caring for one's own children as in paid work. As time spent caring for one's own children decreases in older age groups for both males and females, time spent in domestic work and caregiving seems to increase and is highest in the two oldest age groups. These data highlight the potential competition for time between different activities and the variation in time spent in productive engagement across the life course.

Age and health

The data in Table 3 suggest different rates of engagement in all activities other than volunteering between those with a long-term health condition and those with good health across all ages. With the exception of caring for adult relatives and other people's children, engagement rates were generally higher for those with good health than those with a long-term health condition. Consistent with the human capital literature, the most notable difference is the generally higher engagement in paid work, particularly in full-time paid work, for those with good health compared to those with a long-term health condition across all age groups. The relatively higher engagement in volunteering of those over 65 years with good health compared to those with a long-term health condition highlights the importance of health for engagement, particularly at older ages.

Good health was associated with higher engagement in caring for one's own children among those aged 25-54, and caring for others' children only for the oldest age group. However, 15-24 year-olds with good health were less engaged in caring for others' children than those with a long-term health condition. In addition, among those aged 25-54, engagement in caregiving was higher for those with poorer health than those with good health.

Age and education

Consistent with the findings of previous research, Table 4 shows that higher levels of education tended to be associated with greater engagement in paid work (either full-time or part-time) and volunteering across the life course. Engagenaent in caring for one's own children was also significantly associated with education level. The data in Table 4 suggest that this is largely due to differences mostly between those with a degree and those without one. For example, in the two youngest age groups, only 1.0 per cent and 32.5 per cent of those with a degree were engaged in caring for one's own children, compared to 18.5 per cent and 52.2 per cent of those with a non-degree post-school qualification and 8.9 per cent and 51.1 per cent of those with no post-school qualification. However, the situation is reversed among those in middle to later life, with higher education generally being associated with higher engagement for those over 45 years of age. A similar pattern was observed for engagement in caring for others' children and caregiving across the age groups, although the associations with education were generally not significant.

Baby boomer analysis

Descriptive findings for the baby boomer sample are presented in Table 1, Column 3. In terms of age, 29.1 per cent of baby boomer respondents were aged between 45 and 49, 26.5 per cent aged between 50 and 54, 22.8 per cent aged between 55 and 59 and the remaining 21.6 per cent were aged between 60 and 64.

Age and income

Table 5 presents engagement rates disaggregated by household income tertiles for the baby boomer cohort aged 45-64 years in 2010. Having a higher household income was generally associated with higher engagement in full-time paid work, but lower engagement in part-time paid work, suggesting that personal income, which is expected to be higher for those in full-time work, is strongly related to household income. Higher incomes were also generally associated with higher engagement in caring for one's own children, although this relationship was not significant in the 60-64 year age group. While higher incomes were on average related to lower engagement in caring for others' children and adult caregiving across age groups, this relationship was significant only in the 45-49 year group. Finally, although those with incomes in the highest tertile have the highest engagement in volunteering, the relationship between income and volunteering was not significant either within or across age groups.

Age and occupation

The engagement rates for different occupational groups are displayed in Table 6. Across age groups, occupational group was related to engagement in paid work, caring for one's own children, volunteering and caregiving. Managers and professionals generally had the highest engagement in full-time paid work and caring for one's own children and the lowest engagement in part-time work. Conversely, clerical and sales workers generally had the lowest engagement in full-time paid work and caring for one's own children and the highest engagement in part-time work. The relationship between occupation and paid work was significant within all age groups except for the 60-64 year-olds, while for caring for one's own children, the relationship was only significant within the 45-49 and 55-59 year age groups.

The data also suggest that managers, professionals and clerical and sales workers generally had higher engagement in both volunteering and caregiving, while machinery operators and labourers generally had lower engagement across all ages. Trades and community workers in the 45-49 and 50-54 year age groups had similarly low engagement in volunteering and caregiving, but engagement tended to be higher in the two older age groups. However, only the relationships between occupation and volunteering in the 45-59 year age groups were significant. Occupational group was not related to caring for others' children either across or within age groups.

Age and partner status

As seen in Table 7, across all ages, those married or in a de facto relationship generally had higher engagement in paid work, caring for one's own and others' children and volunteering than those with no partner. However, these relationships were not significant within the 60-64 year age group. In addition, marital status was only significantly related to engagement in caring for one's own children within the two youngest age groups, and to caring for others' children in the 55-59 year age group. The results also suggest no significant relationship between marital status and adult caregiving either across or within age groups.

Multivariate analysis of baby boomer engagement

The results from the multinomial and logistic regressions examining which of the above factors distinguish between those engaged in single, multiple or no activities are shown in Table 8. Domestic work was again excluded from the activity count due to very high engagement rates, while occupation, which was only collected from a sub-set of the sample, was excluded to maximise sample size. Controlling for age, gender, partner status, education and household income, the odds of being engaged in one activity or multiple activities rather than no activity is about three times higher for those in good health compared to those with a long-term health condition. Other factors that increased the odds of engagement in a single activity rather than no activity included being younger, unpartnered and having a higher household income. Being younger and having a higher household income also increased the odds of being engaged in multiple activities rather than no activity, as did being partnered and having a degree.

Paid work engagement was included in the binomial regression to examine whether it distinguishes between those engaged in single and multiple activities. Controlling for the other factors in the model, not being in paid work compared to being in full-time work halved the odds of being engaged in multiple activities rather than in a single activity, while being in part-time work increased the odds by 1.48 times. Other significant factors that increased the odds of engagement in multiple activities were being younger and having a degree.

Discussion

The present study was based on a secondary analysis of recent data from working-age and older Australians in the nationally representative HILDA Survey, collected in 2010. The findings update and extend previous findings from Australia and elsewhere that individuals continue engagement in productive activities as they age over the life course. The extent and ways in which people contribute was found to vary with age and gender, as well as with their social position and personal resources that can either enable or constrain their capacities and opportunities.

Overall, the findings are generally consistent with prior research on age- and gender-related expectations and opportunities within the Australian context and provide support for the utility of the life course perspective (see, for example, Elder 1995). The life course trajectory suggested by the data is that young Australians are likely to begin their contributions to paid work part-time, when they are also still studying, before entering the workforce more fully for a period of time and then having their own children. Those in middle life have relatively high engagement in all productive activities, but the contributions shift from paid work and caring for one's own children towards caring for other people's children or adult relatives, and towards volunteering in later life.

Our findings also supported those of earlier research on the influence of gender expectations and opportunities on engagement in paid work, childrearing and caregiving. Men had relatively higher engagement than women in paid work, particularly in full-time paid work, and generally lower engagement in part-time paid work and other activities beyond paid work. Interestingly, the percentage engaged in caring for one's own children was higher for males than females in the 45-64 year age groups, suggesting that men have children at older ages than women. This is possibly related to their higher engagement in paid work, but is more likely a reflection of the common age gap found in relationships, with men generally being older, and the different reproductive timing for men and women. Even so, the mean time spent by those engaged in caring for one's own children was generally higher for females than males in all age groups. These mixed results highlight the value of including different measures of productive engagement and examining engagement over the life course.

As predicted by human capital theory, the results highlight the potential for health and education to enable or constrain the types of activities individuals engage in across the life course. Those with long-term health conditions and lower educational qualifications generally had lower engagement than those with good health in all activities other than caring for other's children and adult relatives, two activities that are more likely to occur within one's home or place of residence. These activities are also more likely to be unpaid, raising concerns about the potential for those with long-term health conditions and lower educational qualifications to experience social disadvantage and social exclusion through reduced opportunities to engage in paid work and other productive activities across the life course (McDonald 2010).

The additional analyses of the baby boomer cohort aged 45-64 in 2010 suggested that, in addition to the traditional human capital factors of health and education, personal resources such as having a higher household income and being partnered were also significantly related to productive contributions. Those with a partner generally reported higher engagement in all activities other than adult caregiving, which was unrelated to marital status. Higher household incomes were generally associated with higher engagement in full-time paid work and caring for one's own children, but lower engagement in part-time work and caring for other people's children or adult relatives. Being unpartnered and having a lower household income was also related to decreased odds of engagement, raising concerns about the likely social and financial disadvantages of those not engaged in productive activities outside domestic work.

The results for occupation further support the idea that personal resources can enable or constrain productive engagement. Occupations that are generally regarded as having higher social and economic status, such as managerial and professional jobs, were related to higher engagement in productive activities, while lower status jobs, such as machinery operators and labourers, tended to be associated with lower engagement in all activities. Overall, these findings support the utility of both the human capital and life course frameworks for research on productive ageing, but they also highlight the added value of examining other personal resources not typically included in the human capital framework, such as marital status, household income and time.

Limitations and future directions

In interpreting the findings it is important to recognise that age differences may reflect cohort influences as well as life course influences. For example, educational background and gender expectations, as well as employment markets, differ considerably for those currently in early adulthood from those experienced by people now in middle and later life. Longitudinal analyses are required to understand the impacts of earlier experiences on the life course and life transitions such as entries to and exits from paid work and other productive activities (see, for example, Baxter et al. 2008).

Our future research will extend these findings by examining transitions over time using several waves of the HILDA Survey. More in-depth, longitudinal analyses of time use data will add to our knowledge of the potential competitions or trade-offs in allocating scarce time between different activities. Future multivariate analyses will test the relative importance of, and interactions among, enabling and constraining influences on activities as well as the consequences for health and wellbeing when people have multiple, demanding responsibilities such as full-time work and caregiving. Future research may also examine other potential determinants such as household structure, number of dependent children, attitudes towards productive engagement and levels of social capital.

Conclusions and policy implications

Following concerns that governments will be unable to continue providing adequate pension and healthcare support for an ageing population as the proportion of those working decreases, governments have begun implementing strategies including raising pension eligibility ages, increasing mandatory retirement savings and addressing barriers such as age discrimination to further encourage and enable older individuals to remain in the paid workforce longer (OECD 2006; Griffin et al. 2012; Ryan 2012). The appointment of the first Age Discrimination Commissioner to the Australian Human Rights Commission in 2011 provided national leadership in identifying the structural barriers and pervasive ageism that undermine the recognition and contributions of older people in the wider community as well as in the workforce. However, our findings suggest that social and demographic factors such as gender, human capital and other personal resources may also significantly enable or constrain engagement in productive activities over the life course and as individuals age.

Despite the cross-sectional nature of the data limiting our ability to draw causal inferences, these findings can inform government and social policies designed to improve the well-being of individuals as they age and their ability to age productively. For example, current government actions aimed at improving access to human capital resources such as health and education, should not only target younger Australians, but also those in middle and later life, who increasingly expect and are expected to continue productive engagement as they age. The clear gender differences found in the data also suggest that there is great value in promoting policies that reduce gender differences in paid and unpaid productive contributions, while at the same time recognising and supporting the contributions of both sexes.

Our findings also underscore the importance of valuing people across all age groups in terms of their broader contributions beyond tax-generating economic activities. This is particularly pertinent for older people, females and those with fewer personal resources, who are more likely to have reduced engagement in paid work but increased engagement in unpaid work relative to younger people, males and those with more personal resources, respectively. However, policy actions aimed at improving the situation of females and those with fewer personal resources must not only recognise but also reward unpaid as well as paid contributions, which can benefit individuals as they age. The monetary value of unpaid volunteering, childcare and caregiving work by older Australians has been estimated at approximately $6.8 billion a year (NSA 2009), providing an important counterpoint to generational attacks that scapegoat older people for the costs of population ageing with adverse effects on their social standing and self-esteem (Kendig 2010).

In conclusion, developing policies aimed at reducing social inequality and increasing human capital and personal resources can help facilitate the government's productive ageing agenda. More fundamental action to enable productive economic and social participation requires a comprehensive view of ageing across the life course. Our findings suggest that improving education, health, employment opportunities and the removal of barriers in accessing these opportunities for those with limited resources can increase people's capacities for both paid and unpaid contributions across all ages. Ongoing investments enabling people across the life course to remain healthy and to have economic security are sound investments in the future and will be crucial for an increasingly active, engaged and vital older population.

Acknowledgements

This paper uses unit record data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey. The HILDA Project was initiated and is funded by the Australian Government Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA) and is managed by the Melbourne Institute of Applied Economic and Social Research (Melbourne Institute). The findings and views reported in this paper, however, are those of the authors and should not be attributed to either FaHCSIA or the Melbourne Institute.

This research was conducted by the Australian Research Council Centre of Excellence in Population Ageing Research (project number CE110001029). However, the views expressed herein are those of the authors and are not necessarily those of the Australian Research Council, who were not involved in decision-making in the research and publication process.

References

Bailis, D.S., Segall, A. & Chipperfield, J.G. (2003) 'Two views of self-rated general health status', Social Science and Medicine, 56 (2), 203-17.

Bass, S.A. & Caro, F.G. (2001) 'Productive aging: a conceptual framework'. In N. Morrow-Howell, J. Hinterlong & M. Sherraden (eds) Productive Aging: Concepts and Challenges, Baltimore, John Hopkins University Press.

Baxter, J., Gray, M. & Hayes, A. (2007) A snapshot of how Australian families spend their time, AIFS Facts Sheet, Melbourne, Australian Institute of Family Studies.

Baxter, J., Hewitt, B. & Haynes, M. (2008) 'Life course transitions and housework: marriage, parenthood, and time on housework', Journal of Marriage and Family, 70 (2), 259-72.

Bittman, M., Hill, T. & Thomson, C. (2007) 'The impact of caring on informal carers' employment, income and earnings: a longitudinal approach', Australian Journal of Social Issues, 42 (2), 255-72.

Brandon, P.D. & Temple, J.B. (2006) 'Determinants of time allocation combinations among non-employed older persons: evidence from Australian time use diaries', Australian Journal of Social Issues, 40 (1), 7-26.

Cai, L. & Kalb, G. (2006) 'Health status and labour force participation: evidence from Australia', Health Economics, 15(3), 241-61.

Carr, A., Biggs, S. & Kimberley, H. (2013) Meanings of a long life: cultural, social and historical perspectives, Social Policy Working Paper No. 17, Brotherhood of St Laurence and Centre for Public Policy, University of Melbourne.

Chomik, R. & Piggott, J. (2012) Mature-age labour force participation: trends, barriers, incentives, and future potential, CEPAR Briefing Paper 2012/01, Sydney, Centre of Excellence in Population Ageing Research.

COTA (Council on the Aging) (2012) Policy alert no.7--Advisory Panel on the Economic Potential of Senior Australians, COTA, April 2012.

de Vaus, D., Gray M. & Stanton D. (2003) Measuring the value of unpaid household, caring and voluntary work of older Australians, Research Paper No. 34, Melbourne, Australian Institute of Family Studies.

Dosman, D., Fast, J., Chapman, S.A. & Keating, N. (2006) 'Retirement and productive activity in later life', Journal of Family and Economic Issues, 27 (3), 401-19.

Elder, G.H. (1995) 'The life-course paradigm: social change and individual development'. In P. Moen, G.H. Elder & K. Lusher (eds) Examining Lives in Context: Perspectives on the Ecology of Human Development, Washington, American Psychological Press.

Erlinghagen, M. & Hank, K. (2006) 'The participation of older Europeans in volunteer work', Ageing and Society, 26 (4), 567-84.

Fernandez-Ballesteros, R., Zamarron, M.D., Diez-Nicolas, J., Lopez-Bravo, M.D., Molina, M.A. & Schettini, R. (2011) 'Productivity in old age', Research on Aging, 33 (2), 205-26.

Glass, T., Seeman, T.E., Herzog, A.R., Kahn, R. & Berkman, L.F. (1995) 'Change in productive activity in late adulthood: MacArthur studies of successful aging', Journal of Gerontology: Social Sciences, 50B (2), S65-76.

Griffin, B., Loh, V. & Hesketh, B. (2012) 'Age, gender, and the retirement process', In M. Wang (ed.), The Oxford Handbook of Retirement, New York, Oxford University Press.

Han, B., Phillips, C., Ferrucci, L., Bandeen-Roche, K., Jylha, M., Kasper, J. & Guralnik, J.M. (2005) 'Change in self-rated health and mortality among community-dwelling disabled older women', The Gerontologist, 45 (2), 216-21.

Hesketh, B., Griffin, B. & Lob, V. (2011) 'A future-oriented Retirement Transition Adjustment Framework', Journal of Vocational Behavior, 79 (2), 303-14.

Hinterlong, J.E. (2008) 'Productive engagement among older Americans: prevalence, patterns, and implications for public policy', Journal of Aging and Social Policy, 20 (2), 141-64.

Hinterlong, J., Morrow-Howell, N. & Rozario, P.A. (2007) 'Productive engagement and late life physical and mental health: findings from a nationally representative panel study, Research on Aging, 29 (4), 348-70.

Jenkins, A., Rowland, F., Angus, P. & Hales, C. (2003) The Future Supply of Informal Care 2003 to 2013: Alternative Scenarios, Australian Institute of Health and Welfare (AIHW) Cat. No. AGE 32, Canberra, AIHW.

Kendig, H. (2010) 'The Intergenerational Report 2010: a double-edged sword', Australasian Journal on Ageing, 29 (4), 145-46.

Kendig, H. & Browning, C. (2010) 'A social view on healthy ageing: multi-disciplinary perspectives and Australian evidence'. In D. Dannefer & C. Phillipson (eds) The SAGE Handbook of Social Gerontology, London, Sage Publications.

Kendig, H., Dykstra, P.A., van Gaalen, R.I. & Melkas, T. (2007) 'Health of aging parents and childless individuals', Journal of Family Issues, 28 (11), 1457-86.

Kendig, H., Wells, Y., O'Loughlin, K. & Heese, K. (in press) 'Australian baby boomers face retirement during the Global Financial Crisis', Journal of Aging and Social Policy.

Kjellsson, S. (2013) 'Accumulated occupational class and self-rated health. Can information on previous experience of class further our understanding of the social gradient in health?', Social Science and Medicine, 81, 26-33.

Li, Y. & Ferraro, K. F. (2006) 'Volunteering in middle and later life: is health a benefit, barrier or both?' Social Forces, 85 (1), 497-519.

Mathers, C. (2007) 'The health of older Australians', In A. Borowski, S. Encel & E. Ozanne, (eds) Longevity and Social Change in Australia, Sydney, UNSW Press.

McDonald, I. (2010) 'Economic perspective on social exclusion', Australian Journal of Social Issues, 45 (2), 213-25.

McDonald, P. (2011) 'Employment at older ages in Australia: determinants and trends'. In T. Griffin & F. Beddie (eds) Older Workers: Research Readings, Adelaide, Commonwealth of Australia.

McMunn, A, Nazroo, J., Wahrendorf, M., Breeze, E. & Zaninotto, P. (2009) 'Participation in socially-productive activities, reciprocity and wellbeing in later life: baseline results in England', Ageing and Society, 29 (5), 765-82.

Melbourne Institute of Applied Economic and Social Research (2011) HILDA Survey Annual Report 2010, Melbourne Institute of Applied Economic and Social Research, University of Melbourne.

Mirowsky, J. & Ross, C.E. (1998) 'Education, personal control, lifestyle and health: a human capital hypothesis', Research on Aging, 20 (4), 415-49.

Moen, P. (1996) 'A life course perspective on retirement, gender and well-being', Journal of Occupational Health Psychology, 1 (2), 131-44.

Moen, P., & Spencer, D. (2006) 'Converging divergences in age, gender, health, and well-being: strategic selection in the third age.' In R. H. Binstock & L. K. George (eds) Handbook of Aging and the Social Sciences, Amsterdam, Elsevier.

Mor, V. (2005), 'The compression of morbidity hypothesis: a review of research and prospects for the future', Journal of the American Geriatrics Society, 53 (9s), S308-S309.

Morrow-Howell, N., Hinterlong, J. & Sherraden, M. (eds) (2001a) Productive Aging: Concepts and Challenges. Baltimore, Johns Hopkins University Press.

Morrow-Howell, N., Hinterlong, J., Sherraden, M. & Rozario, P. (2001b) 'Advancing research on productivity in later life'. In N. Morrow-Howell, J. Hinterlong & M. Sherraden (eds) Productive Aging: Concepts and Challenges, Baltimore, John Hopkins University Press.

Nepal, B., Brown, L., Kelly, S., Percival, R., Anderson, P., Hancock, R. & Ranmuthugala, G. (2011a) Projecting the need for formal and informal aged care in Australia: a dynamic microsimulation approach, National Centre for Social and Economic Modelling (NATSEM) Working Paper 11/07, Canberra, NATSEM, University of Canberra.

Nepal, B., Brown, L., Ranmuthugala, G. & Percival, R. (2011b) 'A comparison of the lifetime economic prospects of women informal carers and non-carers, Australia, 2007', Australian Journal of Social Issues, 46 (1) 91-108.

NSA (National Seniors Australia) (2009) Still Putting In: Measuring the Economic and Social Contributions of Older Australians, Research Report, Canberra, National Seniors Australia Productive Ageing Centre, Department of Health and Ageing.

--(2012) Barriers to Mature Age Employment: Final Report of the Consultative Forum on Mature Age Participation, Canberra, National Seniors Australia Productive Ageing Centre, Department of Health and Ageing.

OECD (Organisation for Economic Co-operation and Development) (2006) Live Longer, Work Longer, Paris, Organisation for Economic Co-operation and Development.

--(2011) Pensions at a Glance 2011: Retirement-income Systems in OECD and G20 Countries, Paris, Organisation for Economic Co-operation and Development.

O'Loughlin, K., Humpel, N. & Kendig, H. (2010) 'Impact of the global financial crisis on employed Australian baby boomers: a national survey', Australasian Journal on Ageing, 29 (2), 89-91.

O'Reilly, P. & Caro, F.G. (1994) 'Productive ageing: an overview of the literature', Journal of Aging and Social Policy, 6 (3), 39-71.

Ryan, S. (2012) Working Past Our 60s: Reforming Laws and Policies for the Older Worker 2012, Sydney, Australian Human Rights Commission.

Shomos, A. (2010) Links Between Literacy and Numeracy Skills and Labour Market Outcomes, Productivity Commission Staff Working Paper, Melbourne, Productivity Commission.

Tang, F., Morrow-Howell, N. & Choi, E. (2010) 'Why do older adult volunteers stop volunteering?', Ageing and Society, 30 (5), 859-78.

Treasury (2010) Australia to 2050: Future Challenges, Intergenerational Report (IGR) 2010, Canberra, Commonwealth of Australia.

--(2011) Realising the Economic Potential of Senior Australians: Turning Grey into Gold, Advisory Panel on the Economic Potential of Senior Australians (EPSA) Report 3, Canberra, Commonwealth of Australia.

--(2012) Government Response to the Final Report of the Advisory Panel on the Economic Potential of Senior Australians, Canberra, Commonwealth of Australia.

Van der Meer, M. (2006) 'Productivity among older people in the Netherlands: variations by gender and the socio-spatial context in 2002-03', Aging and Society, 26 (6), 901-23.

Walker, A. (2008) 'Commentary: The emergence and application of active aging in Europe', Journal of Aging and Social Policy, 21 (1), 75-93.

Watson, N. & Wooden, M. (2002) The Household, Income and Labour Dynamics in Australia (HILDA) Survey: Wave 1 Survey Methodology, HILDA Project Technical Paper Series No. 1/02, Melbourne Institute of Applied Economic and Social Research, University of Melbourne.

WHO (World Health Organization) (2010) Mental Health and Development: Targeting People with Mental Health Conditions as a Vulnerable Group, Geneva, World Health Organization.

Table 1: Participant characteristics of the
whole sample and of the baby boomers only

                                                  Baby
                                     All         boomers

n                                  10,131         3,208
Female (%)                          52.7          51.9
Married or de facto (%)             62.8          74.6
Dependent child/children (%)        40.4          32.7
Education (%)
  Degree qualification              22.8          26.7
  Non-degree qualification          30.7          36.3
  No post-school qualification      46.5          37.1
Paid work (%)                       62.8          70.7
  Mean hours per week            35.9 (15.4)   38.0 (14.5)
Own childcare (%)                   33.3          35.1
  Mean hours per week            18.1 (19.9)   9.7 (10.9)
Other childcare (%)                 12.0          13.7
  Mean hours per week            7.4 (11.1)    9.3 (12.7)
Volunteer work (%)                  20.3          24.4
  Mean hours per week             5.0 (6.8)     5.2 (7.6)
Caregiving (%)                       9.3          16.5
  Mean hours per week            16.1 (26.0)   14.8 (25.7)
Domestic work (%)                   97.6          99.3
  Mean hours per week            19.4 (16.6)   21.9 (16.3)

Mean hours is for those engaged in each
activity. Standard deviations are in parentheses.

Table 2: Percentage engaged in each productive
activity by age and gender

                          Paid work            Childcare
                             ***

                         FT       PT       Own      Other ***

Females (n = 5,341)
15-24                 25.3 ***   40.0      11.0      11.9 **
25-34                 46.3 ***   27.4    48.6 ***    11.7 **
35-44                 35.4 ***   36.5    78.2 ***   16.8 ***
45-54                 391 ***    36.6    43.6 ***    12.3 **
55-64                 25.2 ***   22.9    10.2 ***   21.7 ***
65-74                  1.8 **     7.5      5.9       18.8 **
75 and over             0.0       2.6      2.1         5.5

Males (n = 4,790)
15-24                   37.6     28.0      8.7         7.7
25-34                   83.7      7.8      40.2        7.5
35-44                   85.9      4.9      70.4       11.1
45-54                   79.1      8.0      59.0        8.4
55-64                   51.3     16.7      20.4       13.4
65-74                   5.7       8.8      7.4        11.9
75 and over             0.3       3.3      4.0         5.3

All (n = 10,131)
15-24                   31.1     34.4      10.0        9.9
25-34                   63.7     18.3      44.7        9.7
35-44                   59.2     21.6      74.6       14.1
45-54                   58.5     22.7      51.1       10.4
55-64                   37.6     19.9      15.0       17.8
65-74                   3.7       8.2      6.6        15.4
75 and over             0.1       2.9      2.9         5.4

                      Volunteer    Caregiving         Domestic work
                                      ***                  ***

                       work ***                   Low      Mid    High

Females (n = 5,341)
15-24                    11.2         2.7       68.4 ***   24.2   7.4
25-34                  16.5 **        2.7       30.8 ***   34.3   34.9
35-44                  26.6 ***     9.8 ***     11.0 ***   32.2   56.8
45-54                   25.0 *      18.8 ***    13.8 ***   36.1   50.1
55-64                    26.8       21.3 ***    12.5 ***   32.8   54.7
65-74                   34.9 *        11.7      7.5 ***    22.6   69.9
75 and over              21.7         9.2         20.9     28.0   51.0

Males (n = 4,790)
15-24                    9.8          3.1         76.5     18.8   4.7
25-34                    11.0         3.5         51.2     35.4   13.4
35-44                    17.5         5.2         36.9     43.1   20.0
45-54                    20.8         12.3        32.3     45.3   22.4
55-64                    25.4         13.6        29.2     41.3   29.5
65-74                    28.9         10.5        19.3     33.8   46.9
75 and over              22.2         11.3        21.9     35.1   43.0

All (n = 10,131)
15-24                    10.6         2.9         72.2     21.6   6.2
25-34                    14.0         3.1         40.3     34.8   24.9
35-44                    22.3         7.6         23.2     37.4   39.4
45-54                    23.0         15.6        22.8     40.6   36.7
55-64                    26.1         17.6        20.4     36.9   42.7
65-74                    31.9         11.1        13.3     28.1   58.6
75 and over              21.9         10.1        21.3     31.1   47.5

N = 10,131. FT = Full-time (35 hours or more in a typical week);
PT = Part-time; Domestic work tertiles: Low = Less than 10 hours;
Mid = 10 to 21 hours; High = More than 21 hours.

* p < .05; ** p < 01; *** p < .001

Table 3: Percentage engaged in each
productive activity by age and health

                            Paid work          Childcare
                               ***

                           FT        PT     Own ***    Other *

Long-term health condition (n = 2,720)
15-24                   23.3 ***    29.6     10.9      16.7 ***
25-34                   44.3 ***    23.5    37.7 *       12.6
35-44                   42.9 ***    20.7   65.8 ***      14.4
45-54                   36.6 ***    20.9   42.3 ***      11.9
55-64                   24.8 ***    16.0     13.7        19.2
65-74                    2.1 ***    5.5       6.9        13.7
75 and over                0.0      2.1       2.5       3.2 **

Good health (n = 7,411)
15-24                     32.2      35.1      9.8        8.9
25-34                     66.2      17.6     45.6        9.4
35-44                     63.0      21.8     76.6        14.0
45-54                     66.5      23.4     54.3        9.9
55-64                     46.2      22.6     15.9        16.8
65-74                      5.2      10.6      6.4        17.0
75 and over                0.4      4.5       3.7        9.4

                        Volunteer   Caregiving        Domestic work
                                       ***                 ***

                          work                     Low      Mid    High

Long-term health condition (n = 2,720)
15-24                     12.8         3.9         70.4     21.4   8.2
25-34                     13.7       8.7 ***       45.4     31.1   23.5
35-44                     18.8       14.1 ***    27.6 **    30.1   42.3
45-54                    19.7 *       18.6 *     22.6 **    34.9   42.5
55-64                     24.8         18.8        19.2     37.1   43.8
65-74                   25.5 ***       11.2        15.2     26.5   58.3
75 and over             17.5 ***       9.6       25.1 ***   33.0   41.9

Good health (n = 7,411)
15-24                     10.2         2.7         72.5     21.7   5.9
25-34                     14.0         2.4         39.6     35.3   25.1
35-44                     23.1         6.1         22.2     39.1   38.8
45-54                     24.2         14.5        22.8     42.6   34.5
55-64                     27.0         16.8        21.3     36.7   42.0
65-74                     37.8         11.0        11.6     29.5   58.9
75 and over               29.8         11.0        14.7     27.8   57.6

N = 10,131. FT= Full-time (35 hours or more in a typical week);
PT = Part-time; Domestic work tertiles. Low = Less than 10 hours;
Mid = 10 to 21  hours; High = More than 21 hours.

* p < .05; ** p < .01; *** p < .001

Table 4: Percentage engaged in each
productive activity by age and education

                           Paid work         Childcare       Volunteer
                              ***

                          FT       PT    Own ***    Other    work ***

Degree (n = 2,311)
15-24                  72.3 ***   16.8   1.0 ***     5.0       13.9
25-34                  69.1 ***   18.9   32.5 ***    8.2      16.8 *
35-44                  62.8 ***   25.5     75.5      13.7    28.0 ***
45-54                  66.1 ***   22.1   59.4 ***    7.8     33.9 ***
55-64                  44.5 ***   258    21.6 ***    16.8    34.5 ***
65-74                  6.8 ***    14.9     6.2       17.4    47.8 ***
75 and over            1.4 ***    11.1   12.5 ***   11.1 *    33.3 *

Non-degree post-school
qualification (n = 3,106)
15-24                    56.9     24.9     18.5      9.4        7.1
25-34                    62.9     20.6     52.2      11.8      13.8
35-44                    64.6     18.5     75.5      13.3      20.0
45-54                    61.6     22.9     50.1      11.8      20.6
55-64                    43.5     18.5     13.8      16.3      25.5
65-74                    3.7      7.6      5.2       15.6      31.5
75 and over              0.0      2.2      1.7       3.4       23-0

No post-school
qualification (n = 4,714)
15-24                    23.6     37.3     8.9       10.3      11.0
25-34                    57.8     14.7     51.1      9.3       10.6
35-44                    49.6     21.1     72.6      15.4      19.1
45-54                    48.4     22.9     45.2      11.0      16.5
55-64                    28.9     17.6     12.1      19.5      21.6
65-74                    2.8      6.3      7.7       14.7      27.1
75 and over              0.0      1.8      1.8       5.3       19.6

                       Caregiving         Domestic work
                           *                   ***

                                      Low      Mid    High

Degree (n = 2,311)
15-24                     1.0       62.4 ***   31.7   5.9
25-34                     2.6       41.8 **    37.7   20.5
35-44                    4.8 **       21.1     40.8   38.1
45-54                     18.9       20.7 *    45.6   33.7
55-64                     19.3      22.4 ***   40.9   36.7
65-74                     12.4        8.7      33.5   57.8
75 and over               12.5        23.6     30.6   45.8

Non-degree post-school
qualification (n = 3,106)
15-24                     4.4         47.2     47.2   47.2
25-34                     3.3         47.2     47.2   47.2
35-44                     9.2         47.2     47.2   47.2
45-54                     14.3        47.2     47.2   47.2
55-64                     15.9        47.2     47.2   47.2
65-74                     13.5        47.2     47.2   47.2
75 and over               5.6         47.2     472    47.2

No post-school
qualification (n = 4,714)
15-24                     2.7         75.5     18.8   5.7
25-34                     3.5         42.0     28.6   29.4
35-44                     8.6         24.4     34.2   41.4
45-54                     14.5        22.8     36.8   40.5
55-64                     18.0        19.3     31.5   49.2
65-74                     9.1         14.5     25.9   59.6
75 and over               11.5        23.0     29.0   47.9

N = 10,131. FT = Full-time (35 hours or more in a typical week);
PT = Part-time. Domestic work tertiles: Low = Less than 10 hours;
Mid -10 to 21 hours; High = More than 21 hours.

* p < .05; ** p < .01; *** p < .001

Table 5: Percentage engaged in each productive
activity by age and household income

                           Paid work         Childcare
                              ***

                          FT       PT    Own ***    Other **

Under $60,000 (n = 1,018)
45-49                  35.9 ***   25.2   51.3 ***   12.0 **
50-54                  28.2 ***   24.1   27.7 ***     14.4
55-59                  17.9 ***   21.4   12.7 ***     18.3
60-64                  13.6 ***   19.2     10.0       18.6

$60,000 - $124,999 (n = 1,200)
45-49                    65.8     21.9     57.0       11.7
50-54                    59.7     24.6     42.9       11.8
55-59                    57.8     20.4     20.4       18.3
60-64                    34.5     24.6     7.9        17.7

$125,000 and over (n = 917)
45-49                    71.2     23.0     69.0       8.6
50-54                    75.0     18.9     51.1       5.0
55-59                    69.8     15.6     28.1       15.1
60-64                    51.8     17.9     12.5       20.5

                       Volunteer   Caregiving       Domestic work
                                       **                 ***

                         work                     Low     Mid    High

Under $60,000 (n = 1,018)
45-49                    23.9       22.2 ***     27.8     34.6   37.6
50-54                    15.9         19.5      21.5 *    37.4   41.0
55-59                    23.1         21.0      17.9 **   34.9   47.2
60-64                    28.1         17.5      14.4 **   31.7   53.9

$60,000 - $124,999 (n = 1,200)
45-49                    22.8         12.0       23.6     40.5   35.9
50-54                    20.2         14.8       26.1     38.4   35.6
55-59                    25.6         17.0       21.5     46.4   32.2
60-64                    24.6         16.3       23.2     37.9   38.9

$125,000 and over (n = 917)
45-49                    29.1         10.7       18.7     43.3   38.0
50-54                    23.2         17.9       18.9     49.6   31.4
55-59                    27.1         19.1       26.6     37.7   35.7
60-64                    29.5         12.5       24.1     31.3   44.6

N = 3,135. FT = Full-time (35 hours or more in a typical week);
PT = Part-time. Domestic work tertiles: Low = Less than 10 hours;
Mid = 10 to 21  hours; High = More than 21 hours. Household
income was divided into tertiles according to gross income band
for the last financial year.

* p < .05: ** p < .01; *** p < .001

Table 6: Percentage engaged in each
productive activity by age and occupation

                        Paid work ***          Childcare   Volunteer

                          FT       PT    Own ***   Other   work ***

Managers and professionals (n = 991)
45-49                  78.0 ***   20.1   64.4 *    10.8     31.9 **
50-54                  75.0 **    20.9    49.3      6.8    28.4 ***
55-59                  78.4 **    18.8   28.6 **   11.3     30.5 *
60-64                    48.4     39.6    11.3     14.5      25.8

Trades and community workers (n = 482)
45-49                    67.7     30.6    64.5      9.1      18.8
50-54                    70.4     28.2    40.1     12.7      10.6
55-59                    57.0     36.6    23.7     22.6      24.7
60-64                    57.4     31.1    11.5      115      29.5

Clerical and sales workers (n = 531)
45-49                    62.5     35.6    52.5      8.1      25.6
50-54                    58.1     35.6    38.8     10.0      21.9
55-59                    63.8     29.9    11.8     18.1      22.0
60-64                    51.2     40.5     4.8     17.9       250

Machinery operators and labourers (n = 347)
45-49                    63.5     27.8    57.4     12.2      17.4
50-54                    66.3     28.7    50.5     12.9      11.9
55-59                    63.8     28.8    22.5     15.0      13.8
60-64                    51.0     39.2    13.7      7.8       98

                       Caregiving       Domestic work
                           **

                                    Low    Mid    High

Managers and professionals (n = 991)
45-49                     13.6      25.4   44.3   30.3
50-54                     20.3      22.6   47.0   30.4
55-59                     21.1      26.8   47.9   25.4
60-64                     14.5      23.9   428    33.3

Trades and community workers (n = 482)
45-49                     15.1      24.2   37.6   38.2
50-54                     12.0      22.5   45.1   32.4
55-59                     20.4      23.7   37.6   38.7
60-64                     11.5      27.9   44.3   27.9

Clerical and sales workers (n = 531)
45-49                     10.6      19.4   46.9   33.8
50-54                     18.1      25.6   40.6   33.8
55-59                     18.1      22.8   45.7   31.5
60-64                     22.6      19.0   310    50.0

Machinery operators and labourers (n = 347)
45-49                     8.7       27.8   36.5   35.7
50-54                     10.9      25.7   38.6   35.6
55-59                     8.8       27.5   43.8   28.8
60-64                     15.7      31.4   35.3   33.3

N = 2,351. FT = Full-time (35 hours or more in a typical week);
PT = Part-time. Domestic work tertiles: Low = Less than 10 hours;
Mid = 10 to 21 hours; High = More than 21 hours.

* p < 05; ** p < .01, *** p < .001

Table 7: Percentage engaged in each productive
activity by age and marital status

                       Paid work ***     Childcare

                          FT       Pr    Own ***    Other **

Married or de facto (n = 2,391)
45-49                  60.8 ***   25.0   67.3 ***     10.4
50-54                  59.1 ***   24.0   46.8 ***     10.8
55-59                   49.9 *    20.4     21.3     19.8 ***
60-64                    27.3     20.1     9.1        19.9

No partner (n = 812)
45-49                    57.7     17.0     38.2       11.6
50-54                    50.0     18.2     27.3       8.2
55-59                    43.2     15.9     15.3       8.5
60-64                    23.4     22.3     12.6       14.3

                       Volunteer   Caregiving        Domestic work *

                       work ***                  Low     Mid    High

Married or de facto (n = 2,391)
45-49                   27.5 *        13.6      21.7 *   38.2   40.1
50-54                   22.3 *        16.7       21.5    41.9   36.6
55-59                   27.4 *        19.5       22.7    39.8   37.5
60-64                    26.9         16.8       18.6    31.9   49.5

No partner (n = 812)
45-49                    19.9         15.8       26.6    44.0   29.5
50-54                    14.5         18.6       25.5    40.9   33.6
55-59                    18.8         18.2       18.8    42.0   39.2
60-64                    27.4         13.7       20.6    36.6   42.9

N = 3,203. FT = Full-time (35 hours or more in atypical week);
PT = Part-time. Domestic work tertiles: Low = Less than 10
hours; Mid = 10 to 21 hours; High = More than 21 hours.


 * p < .05; ** p < .01; *** p < .001

Table 8: Multinomial (no activity vs. single/multiple activities)
and binomial (single vs. multiple activities) logistic regressions,
45-64 year olds

                                        No activity vs.
                                        single activity

                                  B      ExpB         95% CI

Age                             -.04     .96 *    .94       .98
Gender (0 = Males)              -.22      .80     .62       1.04
Good health                     1 09    2.98 *    2.30      3.86
Married or de facto             -.37      .69     .52       .91
Degree vs. No post-school       -.32      .73     .50       1.05
  qualification
Degree vs. Non-degree           -.06      .94     .64       1.39
  post-school qualification
High vs. low household income   -1.57   .96 ***   .94       .98
High vs. mid household income   -.72      .80     .62       1.04
Full-time vs. no paid work       --       --       --        --
Full-time vs. part-time work     --       --       --        --

                                N = 3,180; Nagelkerke R2 = 225;
                                 [chi square](16) = 677.03 ***

                                       No activity vs.
                                     multiple activities

                                  B       ExpB         95% CI

Age                             -.12    .89 ***    .87       .91
Gender (0 = Males)              -.36    .70 ***    .54       .91
Good health                     1.16    3.19 ***   2.45      4.15
Married or de facto              .39    1.48 **    1.10      1.98
Degree vs. No post-school       -1.00   .37 ***    .25       .53
  qualification
Degree vs. Non-degree           -.51     .60 **    .41       .88
  post-school qualification
High vs. low household income   -1 62   .20 ***    .12       .31
High vs. mid household income   -.85    .43 ***    .27       .68
Full-time vs. no paid work       --        --       --        --
Full-time vs. part-time work     --        --       --        --

                                          Single vs.
                                     multiple activities

                                  B       ExpB         95% CI

Age                             -.08    .93 ***    .91       .94
Gender (0 = Males)              -.14      .87      .73       1.03
Good health                     -.05      .95      .79       1.14
Married or de facto              .79    2.21 ***   1.80      2.70
Degree vs. No post-school       -.64    .53 ***    .43       .65
  qualification
Degree vs. Non-degree           -.47    .63 ***    .51       .77
  post-school qualification
High vs. low household income    .10      1.11     .36       1.42
High vs. mid household income   -.13      .88      .72       1.06
Full-time vs. no paid work      -.69    .50 ***    .40       .64
Full-time vs. part-time work     .39    1.48 ***   1.20      1.83

                                N = 2,816; Nagelkerke R2 = .152;
                                 [chi square](10) = 338.75 ***

ExpB refers to change in odds ratio with one unit change in the
predictor. Household income tertiles: Low = Under $60,000;
Mid = $60,000 to $124,999; High = $125,000 and over.

* p < .05; ** p < .01; *** p < .001

Figure 1: Percentage engaged in single, multiple and no activities
by age and gender

Males
                           Percentage (%) engaged
Age group
(years)      No Activities   One Activity   Multiple activities

15-24           26.6             55.7              17.7
25-34            4.2             48.9              46.9
35-44            3.4             23.4              73.2
45-54            4.6             30.3              65.1
55-64           15.2             45.6              39.2
65-74           44.5             41.0              14.5
75+             60.6             33.8               5.6

Females
                           Percentage (%) engaged
Age group
(years)      No Activities   One Activity   Multiple activities

15-24           20.2             61.1              18.7
25-34            3.1             52.9              44.0
35-44            3.1             25.9              70.9
45-54            9.2             35.3              55.5
55-64           19.6             44.6              35.7
65-74           40.0             42.8              17.2
75+             66.0             29.3               4.7

Note: Table made from bar graph.