4.1. Summary of Findings
Findings in our study indicate that the working proportion decreases slightly since age 45 and then goes down rapidly after age 50, especially after age 55. The majority of mature age workers moved from working to voluntarily not working and only a few to involuntarily not working. Before age 53, the proportion of involuntary workforce transitions is slightly higher than that of voluntary workforce transitions. While after age 53, voluntary workforce transitions increase rapidly and start to overwhelm involuntary workforce transitions. The positive age effect on voluntary workforce transitions, especially after age 53 years, is likely to be associated with the rising opportunities to be eligible to use income from superannuation or receive a disability pension [
41]. Once other factors have been controlled for, the age effect is insignificant for involuntary workforce transitions, indicating that the observed slight increase in involuntary workforce transitions with age in
Figure 1 is unlikely to relate to age itself.
It is found that there are four common factors (long term health conditions, prefer to work less hours, having a partner working and proportion of paid years) which had significant impacts on both voluntary and involuntary work force transitions. In which, proportion of paid years played the most important role, followed by long term health conditions, having a partner who is working, and prefer to work less.
Both historical and current employment statuses have significant impacts on workforce transitions and this is consistent to the accumulated life-long advantage and disadvantage theory used for labour market [
27,
42]. The longer the tenure in current occupations, the less likelihood to exit paid work involuntarily while the higher probability to exit paid work voluntarily. The higher the proportion of paid years after graduation, the lower likelihood of not working either voluntarily or involuntarily at later life, while the higher the proportion of unemployed years, the higher probability of involuntary not working at later life.
Having long term health conditions has a similar and strong power in predicting both voluntary (not going back to work), and involuntary workforce transitions (no matter going back to work or not). This reflects the fact that long term health condition is a major reason for mature age workers to leave paid work, as well as a barrier for them to go back to work. This could be explained that mature age workers with long term health conditions might value their free time more and are more likely to be eligible for government disability pension hence have a higher probability to leave their paid work voluntarily. On the other hand, workers with long term health conditions are less demanded by their employers hence their probability of involuntarily not working is higher.
Family structure, partnership and paying mortgage also have significant impacts on workforce transitions at mature ages. Unsurprisingly, home buyers still paying mortgage are less likely to exit paid work voluntarily. Workers, currently without a partner, are less likely to exit paid work voluntarily and are more likely to go back to work after involuntary workforce transitions, reflecting their high independence in both time and finance. Workers with a partner who is working are less likely to voluntarily exit paid work, reflecting the complementarities of joint arrangement of work and leisure time between partners at later life. Partner’s income had a significant, positive but small impact on voluntary workforce transitions. Workers with dependent students are less likely to exit paid work voluntarily, but once they exit their paid work involuntarily, they are less likely to come back to work, reflecting that they have strong incentives to stay at work to support their dependent students but face strong barriers to go back to work at mature ages.
There were also strong incentives among mature age workers to adjust their working hours when age arises, and there is room to improve workforce participation by hours for those who were under employed. When compared to those who prefer to work same hours, prefer to work less hours significantly predicts both voluntary and involuntary not working; dissatisfaction on working hours increased voluntary not working, while prefer to work more decreased voluntary not working. Part time, casual work, and prefer to work less hours might be a signal for a pathway to retirement for those workers who were financially prepared [
43]. While for workers with an overwhelming workload, work hours need to be adjusted down according to their changing health, capacity and preference.
The estimated coefficients of other predictors are mostly under our expectations and in line with existing literature. For instance, workers who were dissatisfied with their job security or work itself, working under fixed term or casual contracts were more likely to leave paid work involuntarily while working in public sectors predicts a lower probability of involuntary workforce transitions. Non-professional staff (workers, sales/clericals, administrative staff, operators, divers, labours) had a lower likelihood to go back to work after voluntary not working.
The general insignificance of education levels on workforce transitions at mature ages might simply reflect the combination of income and substitution effects as mentioned by human capital theory. On the one hand, the income effect predicts that higher educated people with higher earnings are more likely to be able to afford to enjoy free time by exiting their paid work earlier before the age pension age. On the other hand, the substitution effect states that higher educated people might stay in paid work longer due to a higher opportunity cost driven by their higher wages or better employment conditions [
8,
44].
4.2. Policy Implications
In order to remove the barriers that many older workers are facing to carry on working, the OCED called on the Australian authorities to take further actions to enhance the public awareness and effectiveness of age discrimination legislation, to prevent social securities as incentives to early retirement and to strengthen older workers’ employability [
6]. It is found that in New Zealand from 1992 to 2001, increasing age pensions age from 60 to 65 for both men and women, and allowing people to stay at paid work while receiving age pensions have effectively increased labour force participation at mature and older ages [
6].
The OECD has concerns that in Australia, the possibilities to draw superannuation benefits unconditionally as a lump-sum at an early age, to use disability pension as a pathway to early retirement, and to reduce income from age pensions while receiving income from paid work might contribute to the decrease of labour force participation at mature ages, which have not been well examined.
The existing policies and current efforts to increase workforce participation at later life in Australia include improving education and training, assisting attachment to labour market, enhancing long term campaign with age stereotypes and age discrimination, financially encouraging employers to hire workers aged 50 and plus, providing substantial superannuation tax incentives after age 60, as well as to increase progressively the preservation age for superannuation from 55 to 60 and age pension eligible age from 65 now to 67 and further to 70 [
2,
21,
45]. The current policies aiming to increase age eligibility for superannuation and age pensions are expected to delay some voluntarily not working with financial consideration but not for others without financial consideration.
The implications of our findings suggest that in order to facilitate longer workforce participation and enhance productivity and well-being at later life, different government policies and employment practices should be engaged to address the major causes of voluntary and involuntary workforce transitions at mature ages. Promoting mature age workers’ health, employability, work flexibility and friendly work environment, as well as providing rational and secure pathways before full retirement could help older workers to meet their changing health and family needs, hence stay in paid work longer [
4,
46,
47,
48].
A central and new challenge for an ageing society is to enable continued workforce participation at later life by preventing or ameliorating chronic diseases or disabilities when age arises for workers in their 50s or 60s [
49]. The ‘health first’ and “fitness” approaches, and life-long accumulative disadvantage theory, should be taken in consideration in making further policies to tackle the health-related or vulnerable worklessness for mature aged workers in Australia.
For voluntarily not working at later life, the ‘health first’ and “fitness” approaches suggest that the fundamental policies and employment practices are health promotion over life span, age-friendly workplaces, work flexibility to meet changing preference etc. The “health first” approach targets the root cause of worklessness through preventing, improving and managing chronic diseases at mature ages hence refining workers’ health capacity and employability [
50]. Health promotion can reduce both local worklessness and health inequalities but need joint efforts among clinical groups, work program providers work organizations and local authorities [
4,
51].
In Europe where some of countries have already taken important steps to tackle the challenge of ageing population, health promotion activities in workplace for ageing workers have recently been promoted as a new approach to improve occupational and populational health [
51]. Besides the historic approach that takes into account occupational risks, technical and medical expertise, and ergonomic adaptions in the work environment, this new approach promotes healthy habits which may delay the onset of diseases or help to manage the chronic diseases. A great effort has been made to increase the motivation of older workers to move to healthier habits, as older workers are more reluctant to make changes than younger workers. The existing health interventions in Australia have mainly occurred in communities for promoting physical activities and in primary care centers for disease prevention, as well as in health and aged care institutions for managing disabilities and chronic diseases, the health promotion activities in workplace are still inadequate. Australia shares the same obstacle for health promotion programs as in Europe on how to ensure continuity of funding and effectiveness after the end of the intervention programs.
With the extended life expectancy, many people in their 50s and 60s are expected to work with some form of mild chronic diseases or disabilities. It has been found that working longer hours than what is feasible is harmful to health, indicating that the standard full time work might no longer be the best fit for many older people with long term health conditions [
49,
52,
53]. The “fitness” approach encourages the employers and employees to discuss and negotiate work hours, hourly wage, and other work arrangement after reaching certain ages according to the changes in both labour supply and demand sides hence to achieve a new agreement which could be beneficial to both employers and employees. Our analysis supports this by evidence that almost one third of mature age workers had indicated their willingness to reduce working hours at later life. The alternative options could be secured part time or casual employment as a pathway to full retirement when health and energy declined at mature ages.
For involuntarily not working at later life, the life course approach suggests the implementation of policies in equalizing lifelong opportunities for health, education, training, retraining, employment assistance, can help to promote lifelong health, work capacities and continuous labour force participation for those with vulnerabilities, such as those without non-school qualifications or trainings.
Job insecurity is the most emerging issue to be addressed for those involuntarily not working. In last two or three decades there has been continuous shift away from secure (permanent) to insecure (fixed term or casual) jobs and from standard full time contracts to non-standard arrangements (part time, casual, short term or irregular hours) in Australia and many other developed countries [
29,
49] These changes are mainly driven by the economic restructure and technical change resulting a shift of economy from manufacturing sectors to service sectors [
29,
49]. It has been found that during this shift, older workers are often forced into precarious employment with the consequent cycle of fewer job opportunities, little training and lack of income security, exposure to discrimination, harassment and workplace bullying, non-portability of leave entitlements, as well as a reduced capacity to exercise autonomy in how the work is done, resulting damage to health and wellbeing [
29,
49,
54]
The new policy making will also need to take into account the changes in role model and work culture during the new economy, as these changes are likely to increase the demand for secure part time jobs at mature ages. On the labour force supply side, our analysis shows that the current generation of mature age workers are baby boomers who are still working under the male breadwinner female caregiver model, pursuing the norm of “good jobs” as secure and full time jobs, working as long as possible except for married females who participated in labour force flexibly to balance work and care needs [
29,
49,
55,
56,
57]. While for the new generations, the working model is changing significantly where people spend more years on formal education and trainings in early life in order to better survive in the new knowledge-based economy, with males working mostly full time but spending more time with children and reducing work effort in response to health change in later life, while females reducing working hours instead of fully withdrawing from work force for caring responsibilities [
49]. On the labour force demand side, the employer and recruiter perspectives are also changing in terms of what they want from mature age workers. For instance, the notion of employability has changed from “reliability, punctuality and the ability to accept direction” to “resourcefulness, adaptability and flexibility”, the requirement for formal qualification has grown, the low-skilled jobs have been reduced and casualization of the workforce is increasing in an uncertain economy [
29].
Current efforts in Australia, to increase job security of part time or casual work, include extra loading in earnings, wage subsidy, and protection against unfair dismissal etc. [
49]. Further investigation needs to be done on how to increase the options of more secure part-time, fix-termed or casual employment and other forms of more tenuous engagement in a new economy of services. Retraining programs with focus on broader skills and employability for workers who have lower level of qualifications and fewer training opportunities in a declining industry or occupation could assist transition from one industry to another hence to ensure lifelong continuous labour force participation [
8].
As discussed, the strategy of using pooled data drawn from different waves of existing longitudinal data allows us to estimate the workforce transitions at mature ages that would not be possible by using any single year data with small sample size. Nonetheless, we estimate predictors and advise caution in any attempt to interpret our results in terms of causality [
4]. Other limitations of this study include the utilization of one-digit occupational group to identify whether the work is physically demanding, and the use of job dissatisfaction to identify potential psychosocial risk factors in the workplace. For future research, more detailed occupational categories in HILDA can be used to identify possible impacts of hazardous work and impairing work on work force transitions. In addition, work-related stress [
58] and workplace psychological harassment [
59] are recognized world-wide as major challenges to workers’ mental health problems and other stress-related disorders, which are known to be among the leading causes of early retirement from work, high absence rates, overall health impairment [
58,
59]. Larger cross-sectional and longitudinal survey for mature age and older workers are necessary for further estimations.