Introduction
To minimise the projected economic effects of an ageing population, government policy in many western countries now restrains early retirement (which was formerly encouraged) and promotes raised and extended labour-force participation (Auer and Fortuny Reference Auer and Fortuny2000/2; Organisation for Economic Co-operation and Development 1998, 2000; Taylor Reference Taylor2002; van Dalen and Henkens Reference van Dalen and Henkens2002; Walter, Jackson and Felmingham Reference Walter, Jackson and Felmingham2008). In New Zealand, the age of eligibility for the government pension (New Zealand Superannuation) was progressively increased from 60 to 65 years between 1992 and 2001, and compulsory retirement was abolished in 1999 to reduce discrimination against those who wish to work beyond 65 years of age (Hurnard Reference Hurnard2005). Aligned with the Organisation for Economic Co-operation and Development's recommendations, and as in other countries such as Australia and the United Kingdom, New Zealand has developed a ‘Positive Ageing Strategy’ (Dalziel Reference Dalziel2001). This emphasises a positive view of ageing, self-reliance, independence and ‘active’ and ‘productive’ participation in society particularly through economic activity (Davey and Glasgow Reference Davey and Glasgow2006). With these and other efforts, the government hopes that more people will work either full-time or part-time up to and beyond 65 years of age. Critical gerontologists argue that changing ideologies about society, ageing, work and retirement have implications for how older people think about themselves and experience the lifecourse (Biggs Reference Biggs2001; Biggs et al. Reference Biggs, Phillipson, Money and Leach2006; Gilleard and Higgs Reference Gilleard and Higgs2005; Kemp and Denton Reference Kemp and Denton2003; Rudman Reference Rudman2006).
Not surprisingly, the predictions of dire implications from population ageing have fuelled a boom in research on the issues of ageing, health, work and retirement. Some has examined the factors that lead late middle-aged people to stay in or to retire from the workforce, and there is now much research evidence that poor health encourages work exits among older workers (e.g. Alavinia and Burdorf Reference Alavinia and Burdorf2008; Cai and Kalb Reference Cai and Kalb2006; Humphrey et al. Reference Humphrey, Costigan, Pickering, Stratford and Barnes2003; McGarry Reference McGarry2004; Stephens and Noone Reference Stephens and Noone2008). This is generally understood to occur because poor health or a disability limit one's ability to continue working, but other influential factors include job satisfaction, attitudes to retirement, financial security, superannuation and pension eligibility and personal tax rules (e.g. French Reference French2005; Hansson et al. Reference Hansson, DeKoekkoek, Neece and Patterson1997; Humphrey et al. Reference Humphrey, Costigan, Pickering, Stratford and Barnes2003; Mein et al. Reference Mein, Martikainen, Stansfeld, Brunner, Fuhrer and Marmot2000). Given these confounding factors, the relative importance of health is disputed. Some claim that a ‘justification bias’ occurs, by which people cite ill-health for leaving the workforce to justify early retirement using a socially acceptable construct (Anderson and Burkhauser Reference Anderson and Burkhauser1985; Bound Reference Bound1991; Chirikos and Nestel Reference Chirikos and Nestel1984). Several studies have found no evidence of the justification bias (Au, Crossley and Schellhorn Reference Au, Crossley and Schellhorn2005; Dwyer and Mitchell Reference Dwyer and Mitchell1999; McGarry Reference McGarry2004). McGarry suggested that it may be less relevant now because of changing attitudes towards retirement. This debate brings to the forefront the social and moral norms and pressures in which work and retirement decisions are made and rationalised (Rudman Reference Rudman2006).
Despite the profusion of evidence that (poor) health impacts on work participation, hardly any research has examined the relationship closely. Agreeing with Magee (Reference Magee2004), we believe that there is a need to probe more deeply into the processes that link health and illness to work exits and retirement. McDonough and Amick (Reference McDonough and Amick2001) suggested that the social meanings that people attach to health, and the influence that these meanings have on work participation, deserve greater attention. These calls point to the need to study individual experiences, understandings and decisions in relation to the socio-cultural contexts of health and retirement. To this end, this paper describes a qualitative research project that has examined the relationships between health, work and retirement. More specifically, it has examined the different ways that health and illness impact on work participation and retirement using older workers' and retirees' accounts and explanations. It pays particular attention to the social meanings that older people attach to health in work and retirement decisions.
Methods
A qualitative, longitudinal design was used. Sixty New Zealanders (32 women, 28 men) aged 55–70 years were interviewed during October and November 2006. The participants had formerly participated in a large nationwide postal survey about health, work and retirement, and during that survey had indicated that they were willing to be interviewed face-to-face (Alpass et al. Reference Alpass, Towers, Stephens, Davey, Fitzgerald and Stevenson2007). The participants were exclusively from the lower North Island and selected to represent a range of part-time and full-time workforce and retirement positions; approximately one-half had the level of workforce participation that they desired. The sample had diverse occupations, educational levels, ethnicity, socio-economic status, and self-rated health. More specifically, the participants' current or most recent jobs (either as employees or self-employed) were as follows: 16 (27%) professionals, 15 (25%) managers, 12 (20%) clerical and administrative workers, six (10%) community and personal service workers, two (3%) labourers, and three each (5%) in the following occupational groups – technical and trades, sales, and drivers and machine operators. Thirteen participants (22%) had no secondary school education, 13 (22%) had a secondary school qualification but no tertiary education, and 21 (52%) had both a secondary and tertiary education qualification. Five (8%) participants self-identified as Maori, 19 (32%) as both Maori and New Zealand European, 34 (57%) as New Zealand European only, and two as having other ethnicity, so the Maori were over-represented. The participants' economic living standard, as measured by the short form Economic Living Standard Index (ELSISF) (Jensen, Spittal and Krishnan Reference Jensen, Spittal and Krishnan2005), varied considerably: 29 (49%) had a good or very good standard of living, 24 (40%) were comfortable or very comfortable, and seven (11%) were experiencing some, significant or severe hardship. The sample's ELSI mean score (23.1) was similar to the norm for older New Zealander's (24.8), although there were higher proportions in the hardship category (11% versus 6%) and in the comfortable category (40% versus 28%), and a lower proportion with a ‘good’ living standard (49% versus 67%). The participants' health varied also: 32 (53%) reported ‘excellent’ or ‘very good’ health, 20 (33%) reported ‘good’ health, and eight (13%) reported ‘fair’ or ‘poor’ health.
The first set of semi-structured interviews used a lifecourse approach to enquire about each participant's childhood, paid work, education and training, voluntary and caring roles, financial assets and housing, family and relationships, and health over the life span. The participants were also asked about retirement and the factors that influenced their workforce participation and retirement choices in middle-later life. Fifty of these participants (26 women, 24 men) were re-interviewed approximately 18 months later (in May and June 2008). Ten of the 60 participants were not re-interviewed for practical reasons, e.g. deceased, in hospital, or out of the country. The composition of the sample remained much the same. The second set of interviews asked about the participants' work, retirement, health and wellbeing over the 18 months, and again what influenced their employment and retirement choices. Together, the interviews produced rich, in-depth qualitative data that has elucidated the relationship between work, retirement, health and other factors.
Data coding and analysis
The first interviews were digitally recorded and transcribed. The first author then listened to the audio files and read through the transcripts several times so that the data became familiar. She then coded the data using ATLAS.ti software, first to identify parts of the transcripts that related specifically to work, retirement and health, and then to categorise the different ways that the constructs of interest were inter-related and articulated. During this process, it became apparent that health impacted on retirement in various ways. We focused on the similarities and differences in the participants' accounts of how health impacted on work withdrawal and retirement, the objective being to identify coherent, consistent and distinct themes and pathways. As the first author progressed through the data set transcript by transcript, she organised and collated relevant data extracts according to their common and distinguishing features. The specification of the pathways was repeatedly refined and reviewed, which involved moving back and forth between the newly categorised and the already collated extracts, to determine the coherence and underlying features of the health pathway that increasingly appeared to overarch each set of collated extracts. This on occasion led to changes, such as forfeiting a formerly perceived pathway or integrating related pathways into a broader coherent pathway.
Once available, the second interview transcripts were read and coded to identify extracts in which participants explained work withdrawal and retirement through health and other factors. They were then scrutinised methodically to find further examples of, and contradictory evidence about, the previously identified health pathways into retirement. In this way, the second set of interviews enabled verification of the health-related pathways. Importantly, it also enabled examination of changes in work and retirement over time, for the entire sample and for those who cited health as a reason for retiring. Writing an interpretative account of the identified pathways was an integral accompaniment of the analysis and continued throughout. The findings were discussed and critically reviewed by all three authors, during which the main themes were refined until there was agreement that the identified pathways were faithful representations of the data.
The findings
The participants discussed in detail their views, plans and experiences of work and retirement. One-half worked or intended to work part-time or full-time beyond the age of 65 years for reasons of preference or necessity. Also sizeable was the proportion of participants who had retired (17%) or intended to retire (33%) at or before age 65 years. They provided diverse reasons for staying in or leaving the workforce. In this analysis, we focus on health-related explanations of work reduction or retirement. Three health-retirement pathways were identified: the ‘impaired pathway’, in which health problems and disability affect people's ability to continue working; a ‘maximisation of life pathway’, in which people decide to retire specifically whilst they are healthy so that they can fulfil other goals; and a ‘protective pathway’, in which work and retirement decisions are motivated by the individual's concern to promote and protect their health. Each pathway is discussed in turn.
The impaired pathway: poor health affects the ability to work
Several participants mentioned that health problems had impaired their performance, comfort or ability to function at work, and consequently they had retired, reduced their hours or found alternative employment more suited to their health condition. There were cases of work participation being affected by an impairment caused by accidents and injuries (some were work-related), and others of musculoskeletal, cardiovascular, respiratory and multiple health conditions. Although conceivable, no one reported poor mental health as leading to retirement or work reductions from late middle age, although a couple mentioned that poor mental health had affected their earlier working life.
One participant, Edward, exemplified the participants who became functionally impaired through illness. (All the participants' names are fictitious.) He had worked for a government department as a systems analyst. At 52 years-of-age, he had a heart attack and stroke that caused neural damage and made him ‘commercially unviable’, something he had never anticipated. In response to the question, ‘what effect did that have on your life, the major heart attack and the stroke?’ he said:
I tried going back to work, part-time. … The neural damage wasn't immediately apparent. … That following year I struggled. … All the work I had previously known, the technical work, I would be referring to a book or my colleagues [thinking] oh, I can't remember how to do this. … Believe me, it took me nearly a year to come to terms with the fact that it wasn't going to happen. … The limitation is whether I wanted to or not, the practical reality is I do have health limitations. I'm not commercially viable in my old profession and I couldn't sustain full-time work anyway, purely for health reasons now.
Edward made a concerted effort to continue working, and attempted to re-train after-hours but with much frustration. He eventually accepted that he had irreversible neurological damage and ‘medically retired’. He has since built up his voluntary work to about 20 hours each week. He helps elders learns how to use information technology, and finds this unpaid work a satisfying alternative to paid work, better suited to his health condition, and so worthwhile and fulfilling that he no longer considers paid work an option.
Other participants stopped work for reasons related to illness-induced debilitation. Awhina reluctantly retired at 59 years of age because of repetitive strain injury and bronchitis before she had paid off her mortgage. Joe, when aged 60 years, fell and hurt his back so badly that he could not return to his job as a warehouse manager or do other activities he enjoyed. Ruth had to give up work at 62 years of age because of work-related poisoning and a stroke. Carol worked many years as a travelling salesperson until she was aged 57 years, when persistent pain from a car accident became so excruciating that she stopped. For some participants, the occurrence or possibility of health-induced retirement contradicted their wish to continue working because of the satisfaction derived or for financial goals.
Many who were working at the time of their first or second interview anticipated that declining health would lead them to retire. Ngaire, a formal care-giver, said in her first interview that health would probably make her stop work. When asked at what age she expected to retire, she replied, ‘never, I hope’. She was asked how she would arrive at a retirement decision, to which she responded, ‘oh, probably my health or maybe, I don't know … to be honest, I'm absolutely terrified of going potty, getting Alzheimer's’. As it transpired, Ngaire retired at 67 years of age through physical ill-health. During her second interview, she mentioned how lifting clients had damaged her back, leaving her temporarily unable to walk; she reluctantly followed her doctor's advice to give up work. The impaired pathway to retirement confirms that disability and illness have significant implications for many older people's working lives. Because bodily health is inextricably linked with interpretative processes (McDonough and Amick Reference McDonough and Amick2001), the impaired pathway is actually the process by which people's bodily ill-health and their interpretation of this interact to limit their ability to continue working as before.
The pathway that maximises a finite, precious life
Living one's life to the full is a prevalent aspiration in western societies, not least among older people. This is an important aspect of the ‘successful’ or ‘positive ageing’ discourse that advocates that one's older years, especially those before frailty and death, should be celebrated and used well; that during this life stage one should pursue developmental and productive goals and be active (cf. Baltes and Baltes Reference Baltes and Baltes1990; Rowe and Kahn Reference Rowe and Kahn1997). Fuelled in part by increasing longevity, these years of enjoyment are commonly referred to as the third age (cf. Laslett Reference Laslett1989). The participants articulated two distinct variants of the maximisation pathway: ‘optimising present good health’ and ‘maximising life following a health scare’.
Optimising good health whilst you have it
Several participants retired or planned to retire explicitly because they were healthy, with the intention of maximising their enjoyment of their remaining years of good health. Thomas, a (periodically) retired education consultant, said at his first interview that he retired at 59 years of age, and explained his decision:
Probably, there were probably three things: one was that at that stage, and we still are, we were both fit, healthy, and well, and felt that we probably had five or six really good years ahead of us and our preference would be to spend them doing the things we want to do rather than doing the things for other people and working and so on and so forth. So, I mean, that's the first decision you make is whether or not, what is it you want to do? And we really felt there were a lot of things we wanted to do, and we still had time to do it. If we continued to work, you know we ran the risk of getting to the point where we couldn't do them.
In his second interview, Thomas reiterated these views and explained that his brother-in-law's stroke had reinforced his fear of wasting the last years of health on work. Jerry also chose to retire from his full-time job when aged 67 years while he was healthy enough to continue working. He wanted to enjoy time with his grandchildren while he was mobile. As he explained:
I found myself on holiday one Christmas and I was lying on a beach and my daughter and granddaughter came up. … I thought I never do this. I was working six days a week at the golf [club job]. … And I thought in a few years time I will be 70 and I've got to do that. So I came back to work and told the boss, ‘It's a nice cruisy job … and I can do it for a few more years but in a few years time I will be 70, and when I turn 70 I am not going to be able to climb trees with the grand kids’. I just wanted to enjoy a few more years there with them so. It meant a drop in wages [income] but I thought, what the hell, I will do it.
In his second interview, Jerry, then 70, said his retirement situation had changed: he had taken up temporary part-time work, but explained that he prioritised family life, sport, and tramping (meaning hiking) over full-time work. He emphasised that uncertainty about the course of his health meant he wanted to enjoy his healthy years by spending them on fulfilling activities. Physically, Jerry and Thomas both enjoyed good health which, with the discourse of ‘maximising life’, informed their retirement decisions. Their decisions were also influenced by their understanding of mortality, ageing and health: both assumed an inevitable decline in health as one ages; they also talked of the precariousness and unpredictability of health and existence.
Several other participants referred to inevitable decline, the unpredictability of health and death, and the need to maximise non-work facets of life in the years that remained. Donna, aged 64 years, stated clearly that she would retire soon because she would like to do something other than work before she dies. When asked what factors had led to that decision, she said: ‘Probably the chief one is I don't know how long I'm going to live and therefore I would like some time to do something other than work before I die. Probably that sums it up quite well for me’. Several participants who had busy jobs referred to the unpredictability of death and their concern about leaving it too long before retiring. Karen (aged 61), who was prioritising current employment opportunities, had recently attended a good friend's funeral. This had made her wonder whether she should retire sooner and move to her beach house. Arthur (aged 59), who worked full-time and planned to work part-time after 65, referred to the risks of delaying retirement indefinitely by drawing on what happened to his cousin. Like Karen, his concern had not (yet) affected his work participation. He explained:
One of the things that I'm quite conscious of, I have a cousin who worked whatever hours were available. … Never owed anybody anything so he had the assets, he had a house and property and things like that … and about my age he was struck down with ME [chronic fatigue syndrome] or something like that. He can't enjoy what he's now got so I'm just making sure that I can.
The impact of witnessing others die or be afflicted with serious illness was raised repeatedly during the interviews. Death and serious or unexpected illnesses experienced by partners, friends, family members and acquaintances instilled awareness and anxiety about one's own vulnerability. Such reminders of mortality and that time is diminishing have been found to be significant for those in their middle-older years (Karp Reference Karp, Hess and Markson1991). The health histories and deaths of close relatives were also used as markers to predict one's own morbidity and longevity, to create either a sense of urgency or of sufficient time. Although some participants found work fulfilling and intended to continue, it was striking how many participants believed that work competed with other things that they thought were optimal in their life. This belief, interacting with the sense of mortality, decline and the unpredictability of health, led some participants to reduce or relinquish their work whilst healthy. Notably, this decision was dependent on financial security and being able to meet one's costs or sustain their desired living standard. A wish to relinquish work and maximise healthy remaining years could not be achieved by those who were experiencing socio-economic hardship and serious financial pressure.
Maximising life after a health scare
A few participants decided to prioritise life outside work after a serious health scare. Such scares were experienced as a biographical disruption, and they challenged taken-for-granted assumptions and demanded a fundamental rethinking of the individual's life project and self-concept (Bury Reference Bury1982; Charmaz Reference Charmaz1994). Those who experienced health scares were suddenly reminded of their mortality, and consequently they reappraised their life, priorities and aspirations for the future. For Douglas, this biographical disruption occurred after contracting cancer at 35 years-of-age. He realised that other goals were more important than a high-salary job. As he explained: ‘When I got cancer myself, that was the first wakeup call; that, you know, there has to be more to life than working your guts out to earn the high salary, you know, and have the best job’. When employed he could not achieve the work–life balance to which he aspired, and he retired from his high-salary career job at 49 years-of-age. Very high work stress (which his wife thought would ‘kill him’), and his post-cancer perspective on life and work, meant he was unwilling to continue in a situation that dissatisfied him. Since then, he has taken on various casual jobs ‘only to cover the bills’, and he and his wife moved to a ‘rural lifestyle’ property. As he has aged, he has retained a heightened awareness of mortality and of the importance of maximising life, which he lucidly explained:
It gave me that wakeup call that life is perhaps a bit more precious than, well it got me thinking about life, I suppose, and the fact that, really, you don't know what the future holds, and if you don't do some of the things that you've always been promising yourself you will do now, you may not actually get the opportunity to do them again. … Once you've had cancer, statistics show that you are more likely to get it again so you do need to think about the possibility that as you get older, you could have a problem and could be gone in a few months. So if there's things in life you want to do, you need to get on with them.
Martha also changed her priorities after being diagnosed with cancer when aged 60 years. She left an unhealthy, stressful city environment and her job to resettle in a small community where her son lived and which she found conducive to healing; she has since taken on voluntary work. Other research has provided evidence of the tendency to reappraise life and to retire after a health shock (Charmaz Reference Charmaz1994; Jimenez-Martin, Labeaga and Prieto Reference Jimenez-Martin, Labeaga and Prieto2006; Riphahn Reference Riphahn1999).
In summary, as other research on middle-older age adults has found (Karp Reference Karp, Hess and Markson1991), life's finitude was pertinent for some participants. Several referred to the importance of maximising the remainder of their life, typically whilst they were healthy enough to engage in fulfilling activities. Witnessing family members and friends succumb to serious illnesses, or experiencing a health shock oneself, clearly reminded these participants of their own vulnerability, and therefore of the importance of taking advantage of their remaining time. As many participants saw work as competing with the time available for fulfilling activities, some retired or reduced their work hours, but only if they could afford to do so. Those who understood health and work in this way but continued to work were conscious of the risks of delaying retirement. The maximisation of life pathway is consistent with notions of positive ageing and the third age, during which older people remain active and enjoy their remaining healthy years before frailty and dependency.
The retirement pathway that protects health and body
Protecting one's health by engaging in behaviour that promotes it, such as regular exercise and a healthy diet, and avoiding behaviour that compromises it, such as smoking and eating fatty foods, were clear concerns for most of the participants. Many engaged in physical activity and many monitored and controlled their diet, with the intention of maintaining or improving their health. Those who did not follow a healthy lifestyle referred to their ambitions or failings in this regard. This is not surprising considering the ubiquity of the ‘health promotion’ discourse, particularly in its individualised form (that people should take individual responsibility for their health: seeCrawford Reference Crawford2007 and Rudman Reference Rudman2006). Many participants conceptualised work as inimical to good health, particularly when time pressures, heavy workloads, conflicts with colleagues or employers, and organisational restructuring made it stressful. This reckoning led some participants to choose less stressful jobs, reduce their hours or retire. Here we focus on the people who strategically retired or relinquished stressful jobs to protect their health, and also examine the influence of healthy behaviours such as exercise.
As Daniel reasoned, ‘it was just ticking over in the back of my mind that I've got to retire sometime. I don't want to die in the job and somebody commented to me, “for every year you continue working after age 60, you take six months off your longevity or some figure like that”’. Another participant, Jerry, explained that he left work for two health reasons: to take advantage of his good health and as a protective measure. He explained that the mental stress of his job could increase the workload on his heart. In fact, he described the detrimental effects of mental stress on the heart in considerable detail:
It was one of the reasons I retired. I was starting to get that with a mental thing, and I know that just the other day, for instance, when I was on this treadmill, I looked across at the screen and it said my heart beat was 36. … I stood there and I was gripping the thing and I thought to myself, ‘I'm really, really annoyed that any minute now this damn treadmill is going to start, and I can be completely wasted’, and I was getting so annoyed with the fact that I had to do that. I looked across there and [the count] was already up to 46. I hadn't moved a muscle and it reminded me of the fact that when I did have my heart attack, I was in hospital and they said, ‘Listen, your mental health contributes so much towards it. You create these stresses for yourself and hey, forget it’. And this is one reason when I left the golf club [job], I was creating these mental stresses for myself and I had to remind myself that, hey, there's better things to do out there, I should retire.
As described earlier, Edward retired because a stroke impaired his work performance, but protecting his health and body was also important to him. Passing up paid employment was a component of his individual health-promotion strategy. As he reasoned: ‘even part-time paid employment is not a priority for me. … If it was part-time paid at the level I was paid before, it would be tempting but I wouldn't risk my health for it’. Several other participants changed their job or retired to protect themselves from work stress that was causing or exacerbating health problems (commonly high blood pressure and heart conditions), and some had taken into account their doctor's or partner's advice. Several participants who were still working made it clear that they would no longer tolerate stressful working conditions. Several expressed the idea that if their job became stressful, ‘that will be the day I'll leave’. Following contemporary health promotion messages, many people are very aware of the harmful effects of stress on their health and longevity. Having time to exercise was also raised during the interviews. Arthur wanted to reduce his hours to have more time to exercise and so to protect his health. He planned to leave his job if his request was not granted. He elaborated:
Well it's one of the reasons why I want to cut down work to protect it and increase my physical activity, play golf and keep fit. … One of the things I did negotiate with the deal that didn't come through was that I would work four-and-a-half days-a-week. [On] Wednesday afternoon I'd play golf, because I need to for my health, to get the exercise. But anyway I am working. Probably this time next year, if I'm not able to do that I won't be at [this company]. … I will say, ‘Hey, treat me right otherwise I will just go’.
Similarly, James, a semi-retired business advisor, believed his increasing interest in fitness and health would influence how long he continued to work. He also drew on the health promotion discourse in comments about the ‘toll’ of work pressure. When asked what influenced his decision to retire, he said:
I think, firstly, recognition that I didn't need to do that financially, [and] secondly, probably recognising that it does take a toll that I didn't need to impose on myself. … I'm being selective, recognising that stress is not a good thing to have at my age and something I don't really enjoy. … I think keeping fit is probably more of a priority for me than it's been in the past and that will have a bearing on time I put into employment or work. … I think the priority is to stay as fit and healthy as you can.
Many working participants expected that retirement would improve their health, mainly because they would have more time to exercise. Those who had retired spoke of the health benefits of less stress and more exercise, and some thought that retirement-related health benefits compensated for age-related health decline. There were, however, contradictory opinions about whether retirement negatively or positively influenced health. Some believed that people experience health benefits from work (particularly through the mental stimulation). Some thought that people are susceptible to decline or death shortly after they retire, and a few said this had deterred them from retirement. Notably, however, seeing retirement as a phase lacking purpose and stimulating activities that can contribute to decline or death, did not in all cases associate with delaying retirement. Instead, some participants explained, they had developed other interests to keep them invigorated or they had transitioned into retirement gradually.
In summary, as the health promotion discourse is now ubiquitous in New Zealand, it is not surprising that the participants were health conscious and took personal responsibility for maximising their own health. Surveillance and discipline of the body were a normal part of everyday life (Fullagar Reference Fullagar2002; Lupton Reference Lupton1995; Petersen Reference Petersen1996). People drew on the health promotion discourse to explain retirement or reduced working hours, typically in relation to stress. People were also mindful that retirement offers more time to exercise, and although none had retired solely for that reason, many had chosen work contracts that gave sufficient time to exercise, one was trying to negotiate reduced work hours so he could get fitter, and another expected he might soon work less for the same reason. Notably, the wish to protect one's health was also used to explain continued work participation, for the mental stimulation and to stave off (presumed) retirement-related decline and death. Following this health pathway to retirement seemed dependent on financial security.
It should be noted that the participants did not necessarily adopt one of the health pathways described when the opportunity arose. For example, Jack continued working even when he learned he had early signs of asbestosis. Although he engaged in protective health behaviours to retard the condition's progression, such as regular scans, losing weight and healthy eating, he had not given up work nor intended to retire to maximise the duration of his good health before the asbestosis became problematic. The pathways are ‘interpretative resources’ and were not necessarily acted upon, demonstrating the complexity of health-related work decisions. In Jack's case, he said that he continued working because of the effort he and his wife had invested over the decades in their business. Similarly, although several participants had experienced cancer and heart attacks, not all framed these incidents as health scares or experienced a biographical disruption which led to work changes or retirement. For example, Jerry experienced anxiety after a heart attack at 62 years of age, but he did not retire until five years later (to enjoy healthy years). It seems that he recovered physically and psychologically, and he described himself as healthy when he retired. Biographical continuity following a serious illness does occur (Faircloth et al. Reference Faircloth, Boylstein, Rittman, Young and Gubrium2004; Leveälahti, Tishelman and Öhlén Reference Leveälahti, Tishelman and Öhlén2007) and some who are initially shaken by a crisis gradually resume their normal lives (Charmaz Reference Charmaz1994). Finally, not all those in good health were motivated to leave work to maximise its advantages or to protect it: some found work fulfilling and some prioritised financial security and a higher standard of living.
Discussion
This paper has examined the complex ways in which health affects retirement decisions and work changes for middle-aged and older adults. While impairment through illness or accident commonly reduces work capacity, it has been shown that poor health is not the only health-related pathway into retirement. Others include optimising good health and protecting good health. In practice, these pathways are not independent, as several of the case examples have shown. Multiple, socially-constructed and shared meanings were used to interpret health-related strategies and to explain work and retirement decisions. Individuals do not retire or change their work hours in a social vacuum, of course, but in the context of their partnerships, families, communities and nations. In particular, the decision to retire is affected not only by the individual's beliefs about their future or their own activity priorities, but also by the needs of other family members and the felt obligations to other family roles, such as that of grandparent. Another evinced influence is close acquaintance with the health experiences of others, particularly relatives and friends who become critically ill or seriously impaired. Some participants, mostly women, reduced their work hours or retired earlier than they had intended to care for seriously-ill husbands or to support frail parents. Some participants were advised by doctors or partners to end stressful jobs as a protective measure.
In practice, these personal and family decisions were made within a broad socio-cultural context and alongside the moral discourse about what it is normatively ‘right’ to do with respect to personal ageing, work, health and retirement. When contemplating and describing life plans and justifying their decisions, people draw on discourses and employ constructs from many sources, including medicine, macro-social policy and the media. In this study, whilst physical illness significantly impacted on people's ability to work comfortably and effectively, two social meanings associated with health and ageing were also influential. First was a ‘positive ageing’ discourse, which included an imperative to be active and to enjoy a fulfilling later life. It is interesting to note, however, that when used in the context of retiring, ‘positive ageing’ was not necessarily equated with ‘productive ageing’ and economic usefulness. Second was an ‘individual health promotion’ or ‘responsible retiree’ discourse, which directs individuals to nurture their health through exercise and protect themselves from hazards such as work stress. In this light, unhealthy, stress-invoking working conditions are clearly inconsistent with human resources or labour-supply policies to extend working lives. In contradiction with these policies, the widespread acceptance of the ‘positive ageing’ and ‘individual health promotion’ discourses and the ways in which they informed decisions about work reductions and exits among the participants indicates a societal consensus that it is morally acceptable to choose early retirement. McGarry (Reference McGarry2004) argued that there has been another moral or attitudinal shift, namely that early withdrawal from work has increasingly been justified by pride in one's financial independence, while shame about one's ‘obsolescence’ is disappearing. Such changes in the moral climate call into question the prevalent ‘sickness justification bias’ for early retirement. This is not to deny the continuing force of the moral imperative to contribute to society through work: moral dilemmas were expressed by older people no longer in work and in more general terms about being old. Several retired participants constructed contributions to family, neighbourhoods, community and voluntary work as work – albeit unpaid – likely because of the high moral value placed on employment, contribution and busyness (Ekerdt Reference Ekerdt1986). The participants also valued independence and feared dependence, and they managed aspects of ageing, ill-health and disability in terms of self-presentation and social judgement. Future research could usefully examine retirees' and older persons' accounts for the moral imperatives which influence practices of ageing, ill-health, and leaving the workforce.
There are a few caveats to consider at this point. First, whilst the present study identified three distinct ways that health reportedly influences retirement decisions, the selected sample and qualitative methodology do not enable estimates of the proportion of older people (or of particular groups such as those of lower socio-economic status) that used or constructed these pathways, even within New Zealand. Instead, the study has added to previous understanding of the ways in which health impacts on retirement decisions, has identified and elucidated possible pathways, and highlighted the relevance of social meanings and discourse.
Second, although health was a significant factor contributing to retirement, other reasons included: organisational restructuring; dissatisfaction with employers, posts held and organisational philosophies; business disillusionment during an economic downturn; and family devotion and commitments including caring for grandchildren. Other research has shown that multiple factors work together to influence retirement decisions (Davey Reference Davey2008; Hansson et al. Reference Hansson, DeKoekkoek, Neece and Patterson1997; Henretta, Chan and O'Rand Reference Henretta, Chan and O'Rand1992). In the present study, non-health-related factors (e.g. dissatisfaction with work, wanting to spend more time with family) sometimes worked together with the health-related pathways to push or pull persons into retirement. Furthermore, personal health-work pathways often depended on financial security. Only those without debt and with sufficient savings or income to meet current or anticipated living costs could retire to protect their health or maximise their healthy remaining years. In some cases, being eligible for the government pension or being close to the ‘normal’ retirement age of 65 also intertwined with the health-related decisions.
Third, the work changes or withdrawals that occurred through the different health-related pathways were not necessarily endpoints. As noted by others (Davey Reference Davey2008; Gustman and Steinmeier Reference Gustman, Steinmeier, Smelser and Baltes2001; Hansson et al. Reference Hansson, DeKoekkoek, Neece and Patterson1997; Mutchler et al. Reference Mutchler, Burr, Pienta and Massagli1997), retirement is commonly a fluid and dynamic process, and people's intentions change over time. Changing circumstances and aspirations were plentiful, as this paper has illustrated. The qualitative and longitudinal methodology employed was useful in capturing changing work and retirement situations and preferences.
Fourth, the notion of health itself is complex. Maximising one's life in fulfilling ways whilst one has good health did not always mean that the person had perfect health. Many participants did indeed have very good health, with no signs of illness and excellent fitness, but some had had serious ailments from which they had recovered or with which they identified little. There were also instances of people with chronic illnesses such as diabetes who considered themselves healthy because they were rigorously engaging in behaviours that are considered good for health. Consequently, any participant's health status must be understood as the interplay between the body, interpretation and social meanings.
Embedded in current social policy and the positive ageing discourse is the notion that we can to an extent deny ageing and its physical realities and expect a long, healthy life (Biggs et al. Reference Biggs, Phillipson, Money and Leach2006), but not all the participants accepted this idea. In a rather practical way, they cautiously accounted for potential health hazards, decline and death, and consequently some engaged in (or wished to engage in) other paths alongside or instead of work. Some had already experienced poor health that had forced them out of work. On the basis of the present research findings, we suggest that macro-economic policy proposals in response to the ageing population (including those that encourage people to participate in the workforce until or beyond 65 years of age) need to understand the ways in which actual physical health impairments, societal explanations for retirement, and the imperative to maintain, enjoy and maximise good health interact to influence retirement decisions. Considering the diverse aspirations of people at this stage of life, we concur with statements in the United Nations Organization's (1991)Principles for Older Persons that older people should be able to participate in determining when and at what pace they withdraw from the labour force, and pursue the full development of their potential.
Acknowledgements
The data analysed for this paper were collected as part of a study funded by a grant to Fiona Alpass, Christine Stephens, Brendan Stevenson, Eljon Fitzgerald and Judith Davies by the Health Research Council of New Zealand. We also express warmly our gratitude to the participants for their generous and vital contributions to the study.
