‘Live to Work’ or ‘Work to Live’? A Qualitative Study of Gender and Work–life Balance among Men and Women in Mid-life

Authors


*Carol Emslie, Medical Research Council Social and Public Health Sciences Unit, 4 Lilybank Gardens, Glasgow, G12 8RZ, UK, e-mail: C.Emslie@sphsu.mrc.ac.uk

Abstract

Many contemporary studies of ‘work–life balance’ either ignore gender or take it for granted. We conducted semi-structured interviews with men and women in mid-life (aged 50 to 52 years) in order to compare their experiences of work–life balance. Our data suggest that gender remains embedded in the ways that respondents negotiate home and work life. The women discussed their current problems juggling a variety of roles (despite having no young children at home), while men confined their discussion of such conflicts to the past, when their children were young. However, diversity among men (some of whom ‘worked to live’ while others ‘lived to work’) and women (some of whom constructed themselves in relation to their families, while others positioned themselves as ‘independent women’) was apparent, as were some commonalities between men and women (both men and women constructed themselves as ‘pragmatic workers’). We suggest ways in which gender-neutral theories of work–life balance may be extended.

Introduction

This article explores the ways in which men and women in mid-life negotiate the intersections between paid work and other areas of life. As Gregory and Milner (2009) outline in the editorial of this special issue, research in this field has focused on ‘role conflict’, ‘role strain’, ‘work–home conflict’, ‘work–family conflict’ and, most recently, ‘work–life balance’ (defined as ‘satisfaction and good functioning at work and at home, with a minimum of role conflict’ by Clark 2000 p. 751). Rather than ‘work’ and ‘home’ being conceptualized as separate domains that have no bearing on each other, it is now recognized that domestic identities and responsibilities sometimes spill over into the workplace and that organizational identities and responsibilities often cross into home life (Halford et al., 1997; Kanter, 1977).

Many contemporary studies of work and home life either ignore gender or take it for granted (Gerson, 2004). One example is Clark's (2000) work–family border theory that aims to explain how people ‘manage and negotiate the work and family spheres and the borders between them in order to attain balance’ (p. 750). She compares the domains of work and family to different countries with contrasting cultures, and describes people as border-crossers who move between these worlds. Borders are an appropriate metaphor for this process, given that they vary in permeability (the extent to which elements from other domains may enter: this includes psychological permeations such as worrying about work when at home, as well as physical permeations such as a partner or child entering one's home office) and flexibility (the extent to which borders may expand or contract: this may apply to the flexibility of hours or to the location in which work takes place). Borders may also operate more strongly in one direction than another. For example, some employees are expected to work extra hours at short notice, whatever the consequences for their domestic life, while others may have flexibility in their working hours but not in the time they must collect their children from school. This is a useful theory for conceptualizing work–life balance. However, it is largely gender-blind. One of the aims of this article is to start to integrate gender into Clark's work–family border theory.

Here, gender is conceptualized as a dynamic set of socially constructed relationships, rather than as a fixed and binary category. Following West and Zimmerman (1987), gender is best understood as a verb (‘an ongoing activity embedded in everyday interaction’ p. 130), rather than as a property of individuals, and is continually constructed over the life course. Conceptualizing gender in this way disrupts the notion that masculine and feminine identities are the stable characteristics of individuals. Instead, gender identities are constantly renegotiated. Paechter (2003) integrates Butler's (1990) theories about the performative nature of gender with Wenger's (1998) ideas of communities of practice to try to discern which masculinities and femininities we perform, when we perform them and how this comes about. She suggests that: ‘the learning of what it means to be male or female … results in shared practices in pursuit of the common goal of sustaining particular localized masculine and feminine identities’ (p. 71).

Paechter suggests that masculine and feminine identities are related to the communities of practice of masculinity and femininity in five key ways: identity as the negotiated experience of self (the ways that we ‘do’ masculinity or femininity contribute to how we understand who we are), identity as community membership (competent and convincing performances of masculinity or femininity that conform to group norms are central to identity), identity as a learning trajectory (identity is always work in progress, as meanings of masculinities and femininities constantly change over time, place and institution), identity as a nexus of multi-membership (identity has to be constructed to encompass the intersections of participation in many different communities), and identity as an intersection between the local and the global (local masculinities and femininities are influenced by the mass media, popular culture, the law and so on). Here, we explore links between different gendered practices (that is, different ways of doing work–life balance) and constructions of masculine and feminine identities.

Gender is integral to any discussion about intersections between paid work and family life. The spatial separation of home and work brought about by industrialization in the 19th century was closely bound up with the ideology of separate spheres — public life for men and domesticity for middle-class women (Connell, 2005; Smith and Winchester, 1998) — and gender roles continue to be constructed around these masculine/public and feminine/private dualities. By the mid 1990s the ‘traditional family’ (with one full-time male breadwinner) ceased to be the most common family type in Britain, and dual earner families became the norm (Brannen et al., 1997). While there appear to be signs of ‘growing gender convergence, but not equity, in parents’ contribution to childcare time' (O'Brien, 2005, p. 4), women are still mainly responsible for domestic labour and childcare (Tang and Cousins, 2005), and also perform most of the emotional work in families (for example, listening to and comforting children, giving emotional support to partners, doing things to improve or maintain relationships) (Strazdins and Broom, 2004).

Given these responsibilities, most research has focused on how women (particularly women in high status jobs or in dual career families) have reconciled the worlds of family and employment (Guest, 2002). A small body of research has investigated how men perceive the connections between work and home life. Much of this work has taken place in the last decade (for example, Collinson and Hearn, 1996; Halford, 2006; Hatten et al., 2002; Smith and Winchester, 1998; Speakman and Marchington, 2004), although research began earlier in the USA (for example, Cohen, 1988; Pleck, 1985; Weiss, 1987). Recently, UK government policy documents also seem to be including men by noting that parents (not just mothers) need more choice about how to balance work and family life (HM Treasury, 2004). However, in practice, the large disparity in maternity and paternity leave entrenches traditional gender roles, and the assumption remains that flexible working and work–life balance are predominately issues for the working mothers of young children (Smithson and Stokoe, 2005).

Research on gender and ‘work–life balance’

Research on work–life balance is complicated by the gendered structure of the labour market. In many countries, women work shorter hours and occupy lower status jobs than men. Quantitative studies have attempted to control for this difference by comparing men and women working in similar occupations. The results are mixed: some studies have found that women report more conflict between work and home life than men (Frankenhaeuser et al., 1989; Lundberg et al., 1994); others have found that men and women report similar levels of conflict (Eagle et al., 1997; Emslie et al., 2004a; Hughes and Galinsky, 1994; Swanson et al., 1998; Triplett et al., 1999; Winslow, 2005), while one study (Chandola et al., 2004) found different results for different countries. This evidence is hard to interpret and does not tell us anything about how men and women understand and negotiate the intersections between work and home life. In order to do this it is necessary to take a qualitative approach.

Relatively few qualitative studies have set out to compare how men and women perceive the intersection of work and home life. Backett's (1982) study of parental negotiation is unusual in sampling middle-class couples in the UK at a particular stage in the life course (early family formation), rather than employees across a range of ages. Her findings underline the importance of gender. Mothers who were not in paid employment found it difficult to reconcile the demands of domestic work and childcare, while fathers managed to leave domestic problems behind them physically and mentally as they travelled to work each day. Parents used coping mechanisms to help maintain a belief in the fairness of division of labour in the household, despite evidence that fathers remained peripheral to family life: for example, a belief that fathers were willing, and available, to help at home sustained most families.

Organizational studies have also found gender differences. Loscocco's (1997) study of small business owners in the USA found that men saw the flexibility of their working hours as a symbol of the control they had through being their own boss, but tended to use it only occasionally, while women used flexibility as a key resource in trying to achieve a work–life balance. She concluded that women fulfil gender norms when they accommodate work to family life, while men fulfil them when they put their business first. Halford et al. (1997) also noted the gendered dimensions of home and work in their study of UK local government employees. They found a disjuncture between the emphasis employees placed on a shared commitment to work and home life in partnerships, and their descriptions of daily life in which women continued to bear the brunt of domestic work. This made the separation of home and work life particularly difficult for women. Similarly, Connell's (2005) study of public sector workers in Australia found that women were accountable both for running the household and for managing the relationship between the home and the workplace. Connell suggested that work–life problems for men and women may be quite different:

Dropping dead from career-driven stress, or shrivelling emotionally from never seeing one's children, is a different issue from exhaustion because of the double shift, or not getting promotion because of career interruptions. (Connell, 2005, p. 378)

In contrast, Hochschild (1997) found in her well-known study of an American corporation that there were increasing similarities in the way in which men and women regarded work and home life. She suggested that, in the past, the home was seen as a haven from which (male) workers could escape from the unpleasant world of paid work to relax and be appreciated. Now, both men and women regard home as an additional place of work, while the workplace is often seen as a haven (indicated by her subtitle ‘when work becomes home and home becomes work’):

Nowadays, men and women both may leave unwashed dishes, unresolved quarrels, crying tots, testy teenagers, and unresponsive mates behind to arrive at work early and call out, ‘hi, fellas, I'm here!’ (1997, p. 39)

Aims

Our study aims to make a contribution to the field by taking gender as a central theme. Gerson (2004) argues that we should understand work and family through a gender lens. Rather than assuming homogeneity within gender groups and gender differences between them, she suggests we should analyse diversity among women and among men, and look for convergences (rather than assuming divergences) between men and women. Therefore, in this qualitative study we compare the experiences of work–life balance among men and women in mid-life (aged between 50 and 52 years) in order to explore whether there were gender differences or similarities among the respondents. We also explored whether different gendered practices (that is, different ways of doing work–life balance) were linked to different gender identities, and thus to diversity among men and women. Through this empirical work, we hope to begin to integrate gender into Clark's work–family border theory.

We chose to take a life course approach as it helps us to appreciate the ‘larger social contexts in which personal choices and strategies are crafted’ (Gerson, 2004, p. 164). We focused on respondents in mid-life partly because most studies concentrate on younger respondents (usually the mothers of young children) and partly because of the rapid changes in gender relations over the lifetime of this cohort. The respondents were born in the early 1950s and so entered adolescence and adult life in the late 1960s and early 1970s, when widespread challenges to social and sexual mores, precipitated by second wave feminism and other political movements, were changing gender relations (Walby, 1997). They also entered their working and reproductive lives as significant pieces of legislation were enacted in the UK (for example, the Abortion Act of 1967, the Equal Pay Act of 1970 and the Sexual Discrimination Act of 1975), making them one of the first generations to have more control over their fertility throughout their reproductive lives (Hunt, 1990). However, the fact that a large gender pay gap (Perfect and Hurrell, 2003) remains over three decades after such legislative changes took place is one illustration of the power and persistence of gender divisions.

Methodology

This article is based on semi-structured interviews with 11 men and 12 women born in the early 1950s (and so aged between 50 and 52 at the time of interview in 2002–2004) who were sampled from the middle cohort of the West of Scotland Twenty-07 study (Macintyre et al., 1989). This longitudinal survey of the social patterning of health has collected quantitative data from three age cohorts in the west of Scotland, aged 15, 35 and 55 when first contacted in 1987–1988. The main focus of the Twenty-07 study is on health, but it also has a large number of questions relating to gender and work, and so provides an excellent sampling frame for this qualitative study.

The relationship between gender, work (both paid and unpaid) and health has been a major area of interest since the inception of the Twenty-07 study (Hunt, 2002). At a time when few studies systematically compared role conflict in women and men, a quantitative analysis of data from the middle cohort found that reports of work conflict were associated with higher levels of psychological malaise in both men and women (Hunt and Annandale, 1993). The Twenty-07 study also includes a measure of gender role orientation (the Bem Sex Role Inventory; Bem, 1981), which asks respondents to rate their personality according to qualities which have been judged to be stereotypically masculine (for example, dominant, defend my own beliefs, willing to take risks) or feminine (for example, affectionate, sympathetic) (Annandale and Hunt, 1990; Hunt, 2002; Hunt et al., 2004). While these instruments may be a fairly crude way of representing how people perceive themselves, they are useful in selecting respondents with a range of gender role orientations (see also Emslie et al., 2004b).

The aim of this study was to explore experiences of work–life balance among men and women in mid-life, and also to explore the ways in which respondents with different ways of ‘doing’ gender perceived and experienced work–life balance. We therefore wished to include in the sample some respondents who had conventional biographies and gender role orientations for their generation, and others who had less conventional biographies and gender role orientations. We used self-rated health as another way of sampling respondents, given that there is evidence that poor health can pose a challenge to identity and cause people to reflect on taken for granted gendered beliefs and behaviour (Charmaz, 1995; O'Brien et al., 2007). This method of purposive sampling is used when the aim is to select people who possess specific characteristics in order to illuminate the phenomena being studied, rather than to select a representative sample drawn from a population (Mays and Pope, 1995).

From the quantitative data in the Twenty-07 study, we knew that the typical pattern for this age-group was to get married and have children in their twenties. We therefore randomly selected 12 ‘conventional’ interview respondents who married and had children before the age of 30 (see Table 1). They also had average masculinity scores (according to the Bem Sex Role Inventory) for their sex and perceived themselves to be in reasonable health (that is, their self-assessed health was excellent, good or fair). In addition, 11 respondents were identified who had biographies which were less conventional (for example, never married, never had children, perceived their health to be poor, or had extremely high or low Bem Sex Role Inventory masculinity scores). Extreme masculinity scores were defined as those in the top or bottom 10 per cent of the sample, calculated separately for men and women using the quantitative data from the Twenty-07 survey. Although our main focus was on gender rather than on social class, we wanted to include men and women from a range of socioeconomic positions. We therefore included both working-class respondents (defined as those working in manual occupations) and middle-class respondents (those in non-manual occupations). All of our respondents, with one exception, were from the ethnic majority (white) population, which reflected the relatively homogeneous ethnic composition of this age group in this area. Ethical permission for the qualitative study was granted by the University of Glasgow.

Table 1. Selected characteristics of interview respondents
Name* (age)Current or most recent jobCurrent marital statusCurrent household composition (in addition to respondent)Reason for sampling
  1. Note:*All names are pseudonyms.

‘Conventional’ early adult biographies for their age and genderMarried and had children before they were 30 years old, average masculinity scores and in reasonable health
Rona (52)NurseMarriedHusband (children left home)
Maureen (51)TeacherDivorcedPartner and young son
Heather (51)SecretaryDivorcedTeenager
Michelle (51)Auxiliary nurseMarriedHusband and teenagers
Diane (51)Information technologyMarriedHusband and teenager
May (51)TechnicianMarriedHusband (children left home)
Alec (51)ArchitectMarriedWife (children left home)
Trevor (51)EngineerMarriedWife and teenager
Malcolm (52)ElectricianMarriedWife and teenager
Steve (51)PostmanMarried (2nd time)Wife and teenager
Gary (51)EngineerMarriedWife and teenager
Will (51)Own business (joiner)MarriedWife (children left home)
Less ‘conventional’ early adult biographies for their age and gender 
Penny (51)SecretaryMarriedHusbandNo children, very low masculinity score
Jackie (51)Social workDivorcedLives alone (children left home)Poor health
Marilyn (52)Social workSingleLives aloneNever married, no children, very low masculinity score
Gloria (52)Media (freelance)DivorcedLives alone (children left home)Very high masculinity score
Shona (52)Social careSinglePartnerNever married, no children
Ruth (51)Hospice support workMarriedHusband (children left home)Poor health
Ronald (51)Shift managerMarriedWife (children left home)Poor health, very low masculinity score
Douglas (50)SurveyorSingleBrotherNever married, no children, poor health
George (52)Housing managerMarried (2nd)Wife and stepdaughterNo children, very high masculinity score
Jimmy (51)Council cleansingDivorcedSon (21)Very low masculinity score
Kenny (52)Own business (training)DivorcedPartnerNo children, very high masculinity score

After an explanation of the study and assurances about confidentiality, all respondents were asked to give their informed consent and were asked whether their accounts could be tape-recorded. The object was to access the relationships and assumptions that made up the respondents' world-view so the interview began with a request for respondents to give a brief overview of their life. Using this overview as a guide, the interviewer (CE) then concentrated on particular stages in their biography: childhood; becoming a young adult; and work and domestic life as an adult. The respondents were also asked how they combined work with other aspects of their life and if there had been any occasions when this had been problematic. Therefore, respondents were free to decide what aspects, if any, of their work–life balance had been problematic (for example, combining childcare and paid-work, or elder-care and other responsibilities) and the period in their life when this had been most difficult.

The interviews were transcribed, and the accuracy of the transcripts was checked by listening to the tapes. Preliminary analysis began during fieldwork, with interviews being conducted in batches and then discussed by both researchers before further interviews were set up. Some questions were modified in the light of these discussions. The software package QSR Nvivo was used to facilitate the analysis of interviews and field notes. Following McCracken (1988), the analysis moved from the particular (a detailed analysis of language in each transcript) to the general (a comparison of patterns and themes across all the transcripts). Hypotheses were formulated, tested against the transcripts and, where necessary, reformulated in a cyclical process.

Findings

Looking back over their lives, the respondents who were parents described how they had occupied traditional gender roles when their children were young. The men had been the main breadwinners while the women had taken time out of the labour market or arranged their paid work around their family life. Apart from brief periods of unemployment, all but one of the men had been in continuous full-time employment. In contrast, only two women (neither of whom had children) had been in continuous full-time employment. At the time of interview all of the men were in paid work (with one working part-time), and 11 of the 12 women were in paid work (with three working part-time). Only one woman still had a young child at home; the other respondents' children were either teenagers or young adults, many of whom had left home. More men than women still had children living at home, while more women than men lived alone (see Table 1).

Experiences of work–life balance: revealing gender differences

Almost all the women had experienced difficulties in co-ordinating different areas of their lives and most of their accounts of these difficulties related to the present or the very recent past. Penny was an exception. She explained that she did not experience problems because she had few commitments, no children and worked traditional office hours. The other women found it difficult to combine roles such as employee, partner, mother, friend and daughter, and spontaneously used metaphors of juggling and balance to express these difficulties (for example, ‘can't keep all the balls in the air’, ‘juggling match’, need ‘more of a balance’). Many discussed caring for grandchildren or assisting elderly parents with household tasks. Maureen and Heather spoke about making difficult choices between spending time with ageing parents or with their children:

Maureen (51 years): When my father was in hospital … I [used to] come home from work, stay with my son for a couple of hours … and then go straight to the hospital and stay there overnight…. Well that was interfering with my family life … but it was just what you had to do.

Heather (51 years): After my dad died, I maybe spent too much time with my mother when I should have been spending time with the children…. you sit back and you think ‘Oh, maybe I should have done it differently’. But it's too late you know?

Many accounts centred on the impact of paid work on women's lives. Women described how they worried about work tasks while at home (for example, waking up at night to worry about phone calls they had not made and dreaming about work situations) and how their work sometimes left them too exhausted for other activities. Some women felt that these stressful situations at work had negatively influenced their health. Indeed, two women had recently been signed off sick and prescribed anti-depressants due to work stresses. However, paid work was also described in positive terms. Some women suggested that their workplaces had been a place of refuge when they had difficulties in their personal life. Comments that work was ‘therapeutic’ or prevented them from sinking into depression suggest that paid work can help to buffer stresses from other spheres of life.

Socioeconomic circumstances influenced the accounts women gave of how they managed different roles in their lives. Middle-class women readily described their busy lives and then reflected upon their desire to have more ‘time for themselves’ at this point in the life course. This idea was expressed by women both with and without children. Heather gave an eloquent account of this problem, including a description of the emotional work she still does for her teenage son:

Heather (51 years): I'm doing a course at the moment which takes up two nights a week, I go to my mother's one night a week…. My son still stays with me…. I have to spend a lot of time listening to him … I thought by this time I would have more time for me, to do really what I wanted.

In contrast, working-class women played down the difficulty of reconciling different roles and emphasized that they were coping or managing. Their suggestions that ‘you just got on with it’ served to normalize competing demands. For example, May said that even when she had to combine caring for her sick parents with her paid work ‘that didn't stop me doing anything’:

May (51 years): I never, ever found it difficult … anything I have wanted to do I have planned ahead and got it sorted but I could do what I wanted to do really…. There was a wee while when my mum and dad were both ill…. I was running up and down there, but again that didn't stop me doing anything.

Men's accounts differed from the women's in two important ways. First, most of the men focused on the way in which paid work dominated their lives, while the women had discussed the difficulties of balancing a range of roles at home, as well as balancing home and work life. Secondly, most men thought they currently had a good balance between paid work and other activities, although those who continued to work shifts reported that this made it difficult to organize their social lives. While women discussed recent or current problems, men's accounts of problems tended to be located firmly in the past. Many men described working longer hours in the past than they did now. For example, Kenny was a senior manager and felt he had more control over his career now than he had in the past:

Kenny (52 years): I've got quite a healthy balance. When I worked for the Council … I was going home and working at weekends, working at night … and no, I don't do that (now).

Many of the men who were fathers identified the time when their children were young as the period when they had experienced most conflict between work and home life: working long hours, weekends, overtime and working away from home were all problematic. Their accounts suggested they perceived this conflict as an individual problem for their family to solve. Alec was the only respondent (male or female) who gave a structural analysis of the situation, emphasizing that there has ‘got to be better ways of organizing family life’.

Middle-class men often described the need to work long hours in order to establish themselves in their profession, or to finish studying for professional qualifications. Alec, an architect, commented that in order to find time for the aspects of his job that he found interesting (design work), he had previously worked extended hours. Working-class men reported that they had often needed to work extra shifts or overtime to bring enough money into the household. Ronald commented that the timing of his shifts had restricted his ability to participate in his children's activities (see Speakman and Marchington, 2004 for similar findings):

Alec (51 years): Any kind of design work … would have to be done in the evenings and at weekends. Latterly, I kind of regret that, because in the early days when my children were growing up, I hardly saw them out of their pyjamas because I was leaving early in the morning and coming home late at night.

Ronald (51 years): The wages weren't brilliant then and to get the extra money you were doing extra shifts … your shifts restrict you for certain things, likes of the kids are in school plays, in pantos and you're working, you don't see a lot of these things.

These accounts suggest that men regretted missing important parts of their children's early years (see also Hochschild [1997]). However, middle-class men, in particular, balanced this regret with the recognition that the financial gain had been important. Some of the fathers' accounts suggested that time spent at work fulfilling their role as breadwinners conflicted with time their wives expected them to spend at home as fathers. Trevor and Malcolm alluded to past tensions over their working hours:

Trevor (51 years): At that time, the kids were young, so she [his wife] had to look after them all herself at night. I was working till 10 or 11 o'clock at night. She'd obviously had a problem doing it. I didn't realize the full extent at the time.

Malcolm (52 years): My wife and I were falling out, especially when he [his son] was younger, because I was never here, I was working.

A few men did not follow this general trend. Steve was the only man who had not had any problems co-ordinating different areas of his life. He worked for the Royal Mail, and so started and finished work by lunch time. He felt this arrangement had given him lots of free time to see his family and enjoy a social life. In addition, two men discussed current or recent problems they were having. Will was unusual as he described how he managed to spend time at home with his children when they were young. He also discussed currently trying to juggle work and family life. Jimmy was the only man who currently lived with his (adult) child but without a partner. Like some of the women, he described trying to juggle a number of tasks (running the household, looking after his 21-year old son and working):

Jimmy (51 years): It's only me and my son stay in the house…. I feel as if I'm coming in here now, and all I'm doing is ironing and tidying up, trying to decorate, oh it's hellish so it is! [laughs]…. And you've never got enough time.

Work–life balance and the construction of gender identities

After analysis of the interviews, we grouped respondents into the new categories that emerged from the rich qualitative data. (Interestingly, these new categories did not correspond to our sampling categories: for example, respondents who expressed less traditional attitudes in the qualitative interviews did not all come from the less conventional group sampled from the Twenty-07 study, using previously collected quantitative data). Some respondents constructed fairly traditional gender identities. These men noted the centrality of paid work in their lives (‘living to work’) while these women emphasized their caring responsibilities (‘female carers’). Others constructed less traditional gender identities (‘independent women’ and men who ‘worked to live’, who underscored the importance of their lives outside paid employment). A final, mixed-sex group of respondents who had similar class backgrounds and perceived their jobs as a means to an end (bringing in income), were categorized as ‘pragmatic workers’. We describe each in turn.

Constructing traditional gender identities: ‘female carers’ and men who ‘live to work’.  Ruth, Rona and Michelle (female carers) remarked on their caring responsibilities in both their home and work lives. All three had caring jobs which they enjoyed in hospital environments. Indeed, Ruth (a hospice support worker) stated that she would continue to do voluntary work in the hospice after she retired. All intended to continue working full-time until around the age of 60.

All had been married to their original partner for at least 30 years and had children. They positioned themselves as women whose identities were closely bound up with their families. This was despite the fact that their domestic situation was not always particularly traditional. Both Rona and Michelle were currently working full-time while their husbands had been made redundant. The following quotations illustrate the way this group of women describe the importance of their families to them:

Rona (52 years): We've always been an old-fashioned couple and we still are. My husband had his career and I had a job, which actually suited me very well. I have this lovely boss at work who's 30, and she is just climbing the ladder. All of a sudden she'll say when she's 36 or something, ‘Oh, I wanted a family!’— you know, you can't have it all.

Ruth (51 years): You could say that I just live for my family.

The accounts of two men (Trevor and Douglas) suggested that paid work had been crucial throughout their lives and that they ‘lived to work’. They enjoyed their work and did not intend to retire until at least the age of 65. Their largely unquestioning dedication to work made them more likely to accept that employment conflicted with other areas of life. Trevor stressed that he was ‘old fashioned’ because he accepted that paid work conflicted with family life. Douglas attributed his ability to work exceedingly long hours to his marital status (single), but admitted he no longer had a social life:

Trevor (51 years): Work did interfere (with family life) but I'm old fashioned: I tend to accept that as part of the job. I wouldn't say that the job comes first, but … you've got to give your best to your employer.

Douglas (50 years): I don't have a social life anymore…. Basically the invites become less and less because you're too busy working and the like…. But as I say I'm single … so I can get away with it.

Both of these men had a traditional division of labour at home, despite their different situations. Trevor was married with children and felt that it was a ‘mother's place to be at home and look after the children’. Douglas had lived with his parents until recently and his mother had done all the domestic work. After her death, he paid his niece to take over these chores.

Constructing less traditional gender identities: ‘independent women’ and men who ‘work to live’.  Unlike the female carers described above, the independent women did not position themselves as being dedicated to their families. Almost all these independent women (Marilyn, Maureen, Gloria, Jackie, Heather and Shona) had reduced, or were seeking to reduce, the hours they spent in paid work in order to have more time for themselves or to develop their own businesses. They had all worked in non-manual jobs and had sufficient resources to reduce their hours, either through reasonably well-paid jobs or redundancy money from previous work. Tiredness and worries about health had also acted as triggers prompting these women to reconsider the balance in their lives. For example, Marilyn, a social worker, decided to work part-time as she wanted more time and energy for other projects and to safeguard her health:

Marilyn (52 years): I was fed up finding that when I came home on a Friday night I was absolutely exhausted…. I'm diabetic and it's diet-controlled and I have high blood pressure…. so I kind of weighed up the pros and cons and decided that I wanted to go part-time and it's great: I love it!

Gloria was self-employed and made it clear that she had a large amount of control over which projects she worked on. Like Marilyn, health issues (in this case, mental health) influenced her decisions about work. In addition, she linked her independence at work with her independence in relationships:

Gloria (52 years): We were brought up to be very independent … and in order for you to keep being independent, you have to choose properly, so we just choose properly, and if you make a mistake then you just walk away from it and that's it. Same in relationships. If you make a mistake you just walk away from it …’cos at the end of the day, you need to keep your mental health.

Gloria's extract illustrates the importance that all the women in this group placed upon their independence. All were divorced or had never married. Those with current partners underscored their independence from these men and saw their future plans as an opportunity to fulfil their individual potential and to do what they wanted, rather than what others wanted. Extracts from Maureen and Jackie's interviews illustrate this emphasis on independence:-

Maureen (51 years): We're two independent people who have children who live together.

Jackie (51 years): I'm not just a mother or the job I do. First and foremost I'm a woman…. I've got a partner. We don't live together — that's something we decided. I like living on my own.

The group of men who constructed less traditional gender identities (Will, Malcolm, Alec, George and Kenny) were generally in high status professional jobs or ran their own businesses. Most wanted to retire in their mid-fifties, but this depended upon their finances. All of the men in this group either used the phrase ‘work to live’ or emphasized that they thought there was more to life than work. Some, like Will, attributed this to the influence of their father, while others, like Alec, referred to the context in which they grew up:

Will (51 years): My dad worked and worked and worked all his life…. There's no way I live to work. We work to live and we live well.

Alec (51 years): I don't consider work as an end in itself. I'm always one for life…. Growing up in the sixties … I suppose you are carried along by social movements and that does have an influence on you.

All these men had partners, and most stressed the egalitarian nature of their relationship. Men with children in this group discussed how they had sought to keep time free at weekends to spend with their families. For example, both Will and Malcolm worked long hours during the week, but had a caravan and a boat, respectively, where they spent time with their families at weekends. This also allowed them to get away — physically and mentally — from the demands of work.

Pragmatic workers.  These respondents (Diane, Penny, May, Gary, Ronald, Jimmy and Steve) had either always been pragmatic about their work (seeing it as a means to an end), or felt that it played a less central role in their lives now they were older. Male pragmatic workers differed from the men who lived to work because they had fewer resources to draw upon, and so their income was of greater importance for their families. While paid work did not take precedence over everything else in their lives, male pragmatic workers were aware that unemployment was infinitely worse than the everyday problems of paid work. The women noted the practical benefits of their jobs. For example, Penny remarked on the proximity of her work to her home and her holiday entitlement:

Penny (51 years): I've no bus fares, I'm 20 minutes from the house, I've got lots of holidays through longer service et cetera, so I would have to think seriously to move [jobs].

The pragmatic workers were all looking forward to retirement, although the men expressed some concern about being stuck at home and described plans to join clubs, or to continue working a few days a week to ensure they did not go ‘stir crazy’. This concern has been interpreted by others as an escape from the female-dominated space of the home (Wight, 1993).

These respondents occupied similar class positions: almost all the men were manual workers and all had physically demanding jobs, while the women were all married to men in physical, manual jobs. Their future plans for retirement were influenced by financial resources and anxieties about health. Penny commented that her husband would like to retire immediately because his job was so physically tiring, while Diane and May had health concerns of their own. Similarly, the men in this group either had concerns about their own health or indicated that they were aware of the possibility of early death amongst working [class] men:

Jimmy (51 years): If I reach 65 and I'm due my retiral [retirement], I'll look forward to it…. ’Cos I've seen I think four or five guys all retired in there and they've lasted a year, they're dying.

Ron (51 years): My theory [is] that in six or seven years time I'm off, I am out of there, and that's it…. Hopefully: if I live that long! If I live that long!

Conclusions

Most research on work–life balance concentrates on the experiences of the mothers of young children. However, lack of work–life balance is a problem for men as well as women, and for the parents of older, as well as preschool, children (Chandola et al., 2004; Emslie et al., 2004a). Our research contributes to the literature in exploring the experiences of men as well as women in mid-life, most of whom had teenage (or older) children. Despite the similarities in the current work and family circumstances of these men and women, our data suggest that gender remains interwoven in the business of negotiating home and work life. Our methodology meant that respondents were free to choose the situations and times in their lives when they felt it had been most difficult to reconcile paid work and other areas of life. It was notable that the female respondents discussed their current and varied concerns about juggling paid work, adult children and ageing parents, while the men tended to locate problems of work–life balance in the past; often when paid work conflicted with the demands of raising young children. Thus, while the presence of children in the household was associated with a lack of work–life balance for both men and women, these difficulties lasted longer, and took more complicated forms, for women. Our data suggest that, across the life course, women are seen as being responsible for maintaining smooth, or preferably imperceptible, transitions between the worlds of home and work life (see also Connell, 2005). As Loscocco (1997) argues, ‘changes in gender consciousness are the “final frontier” in the quest for greater gender equality in work–family linkages’ (p. 223).

Our research also contributes to the body of work which notes the variety of ways in which men and women do gender (Connell, 1985; Paechter, 2003). Exploring the boundary between the ‘male’ domain of work and the ‘female’ domain of home life is a fruitful way to explore how gender identities are continually (re)constructed (Smithson and Stokoe, 2005). The range of masculinities constructed by our sample is illustrated by the contrast, on the one hand, between men who said that they ‘worked to live’, noting the importance of life outside paid work, and on the other, the few men who ‘lived to work’, demonstrating a more traditional version of masculinity (also see Emslie, et al., 2004b for similar findings for an older cohort of men). Some men discussed how they had moved from a ‘live to work’ to a ‘work to live’ mentality over time, illustrating the way in which gender identities are continually reassessed and reconfigured (Paechter, 2003). Similarly, women constructed a range of femininities through their narratives; some emphasized their close family ties, while others underscored their independence from partners.

Our research also points to the importance of socioeconomic position when considering issues of work–life balance. Most research on work–life balance concentrates on non-manual (usually professional) employees (although see Collinson and Hearn, 1996; Speakman and Marchington, 2004). Through the analysis of our qualitative interviews, we identified a group of pragmatic workers (either male manual workers, or women married to manual workers) who discussed the practical and economic benefits of work. While health concerns often pushed them to consider early retirement, lack of resources prevented them from doing this. Our study suggests that, for some groups of employees, securing sufficient income may be a more pressing concern than balancing work and home life. However, our main focus in this study was on gender rather than social class. Future work should take a more nuanced approach to social class, rather than merely distinguishing, as we did, between manual and non-manual employees.

In the introduction, we referred to Clark's (2000) theory of border-crossing as a useful way to conceptualize work–life balance. Our empirical data provide some support for this theory and also suggest some ways to extend it. First, Clark's exclusive focus on the border between the worlds of work and family may have more resonance for men than for women: for women the family may comprise many worlds with conflicting demands (for example, children and elderly parents) and so women may cross more borders (more often) than men. Secondly, the temporal borders between work and family may vary for men and women. For some men in this sample, their role as a father was associated with attending key family events such as school plays or associated with spending time with their children outside the working week as ‘weekend dads’ (Hatten et al., 2002). The women did not discuss motherhood in this way. Finally, the ways that people position themselves in relation to social structures such as gender has consequences for the ways they shape the worlds of work and family (and vice versa). For example, female carers had relatively weak boundaries between home and work life as they integrated these worlds through the identification of similar (caring) aspects in each. In contrast, pragmatic workers — who were manual workers, or married to manual workers — had stronger boundaries between work and home life. Collinson and Hearn (1996) refer to the way that working-class men seek to ‘maintain an impenetrable psychological wall between “public” and “private” life’ as an attempt to distance themselves from organizational cultures that treat them as second-class citizens (p. 69). Respondents with greater resources were more able to renegotiate the boundaries between work and home life, through cutting down their hours in paid work (independent women) or by physically removing themselves to a place where they could not be contacted out of working hours (men who worked to live going to their boat or caravan for the weekend).

This project raises two methodological issues. Firstly, it is interesting that our findings cannot be explained simply by our sampling strategy. The respondents who expressed less traditional attitudes to work and gender roles in the semi-structured interviews (men who worked to live and the independent women) were not all from the less conventional group of respondents (for example, those who never married or never had children, or had very high or very low masculinity scores). Thus changes in marital status, socioeconomic status and resources, and parenting experiences seem as important, if not more important, in influencing gender identities than similarities or differences between people in the same age cohort.

Secondly, our study raises some issues about trying to access respondents' accounts of work–life balance. Integrating work and family life is intimately tied to constructions of identities and therefore questions about this topic may be experienced as threatening. Given that it is women who are expected to balance work and family life (Moen and Yu, 2000), we might expect women to find these questions more threatening than men. Our data provide some evidence that the working-class women respondents experienced these questions in this way. Their narratives suggest that they initially tried to minimize problems with work–life balance, instead noting that they ‘coped’ and ‘just got on with it’. Pill and Stott's (1982) study of perceptions of illness amongst working-class mothers came to similar conclusions. They found that their respondents defined a good mother as ‘one who “keeps going” and copes with the multifarious demands that her family make of her’ (p. 50).

In conclusion, while we have reported respondents' own accounts of work–life balance it is important to bear in mind that their individual choices were constrained by their socioeconomic resources and cultural norms about family, work and gender (Moen and Yu, 2000). For example, while the middle-class independent women had the freedom to reduce their working hours in order to enhance their work–life balance, this option was not open to those with more limited resources. In addition, it is striking that work–life balance was perceived as a personal issue to be dealt with using individual strategies and not as a structural problem caused by a lack of flexibility in the workplace and a lack of affordable childcare and elder-care in the UK. Indeed, Caproni (2004) suggests that balance may be an unachievable goal because it is built on an individualistic, achievement-orientated model that assumes that people have choice and control over their lives.

To the extent that the work/life discourse remains focused on the individual, power relations will remain beyond the scope of the discourse … who gets to define what work/life balance is? Who has access to available benefits, and who does not? (2004, p. 215).

Future research should concentrate on work–life balance among men and women in less privileged class positions in order to further illuminate the relationship between individual choice and structural constraints.

Acknowledgements

We would like to thank all the respondents who participated in this study. We would also like to thank Sally Macintyre, the editors and the anonymous referees for helpful comments on the manuscript. The authors are funded by the UK Medical Research Council (WBS U.1300.00.004).

Ancillary