Cover photo of Social Policy Journal

Overworked Families? Changes in the Paid Working Hours of Families with Young Children, 1986 to 2001

Paul Callister 1
School of Government
Victoria University


Internationally, there is much research interest in the potential challenges associated with overwork within households, particularly for parents raising children. New Zealand census data show that, when individuals are considered, average hours of paid work for employed women and men changed very little between 1986 and 2001. Yet, in this time period, there were significant changes in both employment rates for women and men and a polarisation of hours of work among these individuals. Further affecting household working hours have been changes in family structure. When total hours of work for both single parents and couple families with young children are considered, a polarisation of hours of work is also evident. However, the shift to long hours was larger than the growth in short hours, while the average hours worked by couples with young children also increased. These findings help explain why many sole parents and couples feel that paid working time has increased and, conversely, that family time has decreased. The paper concludes by examining policy options available to governments in order to curb long working hours.


International research indicates that over the past couple of centuries the average hours spent by individuals in paid work have reduced (Bosch and Lehndorff 2001). Yet, in the last decade, concerns about overwork in New Zealand have re-emerged (e.g. Department of Labour 2004, New Zealand Council of Trade Unions 2002). A comparison of the proportion of employees working 50 or more hours per week among a selection of OECD countries shows that New Zealand has one of the highest proportions of workers putting in long hours of paid work (Messenger 2004). Some of the concerns about long hours of paid work relate to workplace health and safety issues, while others focus on the possible negative effect on families and children (Dawson et al. 2001, Golden and Jorgensen 2002, Pocock 2001, White and Beswick 2003).

This paper has four aims. The first is to document changes between 1986 and 2001 in hours of paid work by New Zealand families with preschool children. While this includes employed sole parents and employed child-rearing couples, most of the attention is placed on couples given that there has been no significant research on the total working hours of couples in New Zealand.

Second, the paper canvasses possible explanations for the increase in working hours. This is primarily to assess whether shifts in hours worked reflect changing working-hour preferences or whether there are other drivers of change. Third, drawing on a variety of studies, some of the effects of increased working hours on unpaid work – including childcare and, ultimately, on child outcomes – are outlined. The final part of the paper highlights policy options available to governments wishing to curb long working hours, particularly among parents.

While this paper focuses on “overwork”, any research on balancing paid work and family responsibilities in employed families needs to be set against an overall polarisation in employment; that is, the division of child-rearing families into either “work-rich” (i.e. all adults are in paid work) or “work-poor” (i.e. all adults are jobless). Singley and Callister (2003), using Statistics New Zealand’s Household Labour Force Survey data, have shown that while by 2002 strong economic growth had reduced household joblessness back to near 1986 levels, other trends were of potential concern. In particular, between 1986 and 2002 joblessness rose substantially among households in which all working-aged members were Maori, and joblessness also became more concentrated in child-rearing and prime-aged (25–49 years) households. While jobless households, and involuntary underwork in general, are also an important policy concern, they are not addressed in this paper.

Data and Methods

Although a wider programme of research uses data from the 1986, 1991, 1996 and 2001 New Zealand Censuses of Population and Dwelling, this paper focuses on changes between 1986 and 2001. The censuses provide a record of usual weekly hours of paid work, with individual hours recorded. Total work hours are used in this research; that is, hours in all jobs are added together. No attempt was made to examine the relationship between changes in multiple job holding and changes in hours of work.

The individual data can be linked at the household level for sole parents and for child-rearing couples. In terms of couples, the paper firstly looks at changes in working hours for partnered parents as individuals. It then considers total hours for couples. The measure used in this research is a household-based family unit. For example, sole parents could be living in a household with other wider family or non-family members but these other individuals are not considered. Equally, couples could be part of a wider household unit. In addition, a child in a sole-parent family may be spending time with a non-custodial parent, but this cannot be determined using census data.

Age for children and parents is an important variable in this research. The analysis is restricted to parents with a youngest child under five. Although this is the age group in which mothers’ working patterns are most strongly affected, in recent decades growth in employment of mothers of preschool children has been particularly strong. At the 2001 Census, within the 89,286 families with a youngest child aged under five, 47% of the female parents were employed in the week preceding the date of the Census (Ministry of Women’s Affairs 2002). Yet international comparative data indicate employment rates of New Zealand mothers with a preschool child are still relatively low, particularly among sole mothers (Table 1).

Table 1 Employment Rates for Mothers with a Child under 6 Years of Age

Total Mothers
Sweden 75 81 65
Portugal 72 70 83
Finland 67 58 65
Austria 67 66 76
Netherlands 66 62 39
Canada 63 70 68
Belgium 62 72 49
United States 59 61 68
France 56 57 52
United Kingdom 55 61 37
Ireland 53 46 35
Germany 52 51 50
New Zealand 47 53 32
Greece 47 48 63
Italy 47 45 72
Poland 46 50 33
Australia 45 48 30
Spain 45 42 65

Source: Johnston and White (forthcoming)

In terms of parent’s age, the main focus of this paper is on the largest child-rearing age groups for children under five, that is the 25–34 and 35–44 years age groups. However, the age group 25–34 years is of particular interest as, in early 2005, the government stated that, while overall New Zealand’s labour force participation rates are high the rate for some groups of New Zealand women, particularly those aged 25–34, are below the OECD average (Clark 2005). This is not surprising given the overall employment rates shown in Table 1. However, for individuals, some data outside this age range are presented while, for couples, some initial data are presented based only on the child’s age. While addressing age among sole parents is straightforward, defining couples by age creates some conceptual challenges. In this study, couples are defined by the age of the female partner. In addition, in this study only opposite-sex couples are considered.

The hours reported are only for those in paid work (and only those who recorded actual hours). With regards to calculating average hours of couples only those couples that were linked into paid work were included. In other words, couples where neither partner was in paid work were excluded.

This means that the calculation of hours worked includes couples where one partner, usually the mother, was not in paid work. This analysis focuses on average hours and long hours of paid work. For this research, 50 or more hours per week is used as an indicator of long working hours for individuals and sole parents. This was the cut-off point used by the Ministry of Social Development in its 2003 Social Indicators Report (Ministry of Social Development 2003). For long hours for couples, two measures are used: combined 80 or more hours, and combined 100 or more hours per week. Some data on short hours are also included to illustrate a polarisation of hours. For individuals (this includes sole parents and partnered parents), under 20 hours is considered short hours, while for couples 30 hours is used as the benchmark.

Aside from age, the main variable considered for couples is education. Three educational groups are used:

  • both partners have a degree or higher qualification
  • either has a qualification
  • other combinations of qualifications.

Ethnicity is also briefly considered. However, due to major methodological challenges, this is based on the ethnicity of the child, not of the parents, and data are presented only for 2001.

Finally, a number of factors such as the changing age structure of the population; shifts in the number of sole parents versus couples and increases in parental educational achievements (particularly for mothers) all influence changes in working hours. In this summary paper it is not possible to present all these data. However, background data can be found in a number of recent reports, including the technical appendices of Callister (2004a) and the Ministry of Women’s Affairs (2002).


To set the scene for changes in employment for child-rearing families, some background data on changes in average hours of employment for individuals in the target age range are set out in Table 2. Although, when calculated across all age groups 15 years or older, average hours of paid work changed very little between 1986 and 2001, there were some significant changes in average hours worked among some age groups, notably a decline among those aged 15–24. The strongest increase in average working hours of employed individuals was among women aged 45–54, with an increase of just over two hours per week.

However, averages can disguise changes in the distribution of hours of paid work. For individuals in the target age groups, there was an increase in the proportion working either short hours or long hours between 1986 and 2001. Table 2 shows the proportion of men and women in the target age groups working 50 or more hours per week. It indicates increases in the proportion working long hours in most of the age groups shown, particularly among men and women in the 45–54 age group. This is an age group where, if the individual is a parent, the children will have already reached school age or, in some situations, will have left home. When 1991 and 1996 data are included in the analysis, there was a plateauing of this growth between 1996 and 2001 for most groups of women and men; the notable exceptions were women aged 25–34 and 45–54, where increases were seen over the whole time period. Table 2 also shows the proportion of individuals working under 20 hours per week. The most significant trend was the increase in the proportion of younger people working short hours. Some of this will relate to students working part time.

Table 2 Average Hours, Long Hours and Short Hours Worked per Week, by Sex and Age, 1986 and 2001

Men Women
15–24 25–34 35–44 45–54 15–24 25–34 35–44 45–54
? 86–01
% working
? 86–0
1 18.3
< 20
? 86–01

Sole Parents

A number of studies have identified increasing employment rates for sole parents in New Zealand (e.g. Goodger 2001, Singley and Callister 2003). These data also show that employment rates for sole fathers (a numerically much smaller group than sole mothers) are much higher than for sole mothers. In addition, these studies (and Table 1) demonstrate that employment rates for sole parents are much lower than for partnered parents.

Table 3 shows the percentage point change in the proportion of male and female sole parents with a preschool child working four main blocks of hours of paid work. There was some polarisation of hours over this period, with an increasing proportion working either short hours or long hours. For sole mothers, the strongest percentage point growth was among those working under 20 hours.

Table 3 Percentage Point Changes in Hours of Work for Employed Sole Parents with a Child Under 5 Years, by Age of Parent, 1986 and 2001

Hours Male Sole Parents Female Sole Parents
25–34 35–44 25–34 35–44
< 20 4.0 5.3 9.5 9.4
20 < 40 -0.5 -6.6 1.1 5.6
40 < 50 -5.2 -4.5 -9.9 -17.3
50+ 1.6 5.7 -0.6 2.3

Table 4 shows the proportion of sole parents working particular hours in 2001. It demonstrates that the largest proportion of sole fathers with a preschool child worked within the standard 40–49-hour group, but that the group working either short or long hours was still quite significant. Older fathers were more likely to work long hours. Table 4 shows that hours worked by sole mothers are also influenced by the age of the mother. For both mothers and fathers this will be in part influenced by the age of the child, with more children closer to turning five years old in older age groups of parents. In addition, a far higher proportion of sole mothers worked under 20 hours per week in 2001 than did sole fathers.

Table 4 Proportion of Employed Sole Parents with a Child Under 5 Years Working in Each Group of Hours, by Age of Parents, 2001 data only

Hours Male Sole Parents Female Sole Parents
25–34 35–44 25–34 35–44
< 20 8.3 8.6 38.7 34.8
20 < 40 10.9 9.9 30.9 33.1
40 < 50 53.9 50.5 24.3 22.8
50+ 26.9 31.0 6.1 9.3
Total 100.0 100.0 100.0 100.0

The data on sole mothers and ethnicity suggest that, on average, mothers of Asian children worked the longest hours per week (32 hours) and mothers of New Zealand European children the lowest (25 hours).2 It is likely that educational differences underlie some of the variations in hours between men and women and between the various ethnic groups. Further in-depth research would be needed to better understand the causes of these differences.

Employed Couples

Changes in Hours for Employed Mothers and Fathers

This first section considers working hours of partnered parents without considering partner’s working hours.3 This is so some insight can be gained into how the changes in working hours have been affecting mothers and fathers independently. Table 5 shows changes in average hours worked by partnered employed mothers and fathers in the main age groups for couples raising preschool children. While the changes in hours worked between 1986 and 2001 were not dramatic, there was some slight increase in average hours worked.

Table 5 also helps assess whether there has been a polarisation of working hours of partnered mothers and fathers with young children. While for mothers the data do not support a growing polarisation of hours of work as seen among the wider population, they do for fathers. There emerges a strong decline in the proportion of mothers working short hours (under 20) and a slight increase, from a very low base, in the proportion of fathers working under 20 hours. When long hours (50 or more) are considered, there was a strong increase in the proportion of partnered fathers working these hours. However, the increase was primarily for those recording exactly 50 hours rather than more than 50 hours. For partnered mothers, the increase in employment rates was not equally matched by an increase in the proportion working long hours. Nevertheless, due to both demographic and employment changes in the broad 25–44 age group, there were numerically slightly more partnered mothers with a preschool child working 50 or more hours per week in 2001 than in 1986. The increase was from nearly 3,978 in 1986 to just over 5,586 in 2001.

Table 5 Average Hours Worked per Week, % of Employed Partnered Mothers and Fathers Working Under 20 Hours, and % Working 50 or More Hours Per Week, by Age of Parent with a Child Under 5 Years, 1986 and 2001

Hours Mothers Fathers
25-34 35-44 25-34 35-44
Average hours 1986 25.3 26.6 47.1 47.5
2001 26.9 27.4 47.5 48.2
? 86–01 1.6 0.8 0.4 0.7
% working
> 20 hours
1986 45.8 41.9 1.6 1.4
2001 37.5 36.6 2.8 2.7
? 86–01 -8.3 -5.3 1.2 1.3
% working?
50 hours
1986 9.0 10.2 34.3 37.0
2001 7.9 9.8 38.6 43.0
? 86–01 -1.1 -0.4 4.3 6.0

The data also show that although there are strong gender-based differences in work patterns, when short hours are considered over time mothers’ patterns of paid work have been changing to become closer to those of fathers. Yet the proportion of fathers working long hours showed a strong rise, increasing the gap between mothers and fathers in this area.

These data also need to be interpreted against a backdrop of changes in the number of employed partnered mothers and fathers. In 1986, there were just under 45,000 employed partnered mothers aged 25–44 with a child under five years. This had increased to nearly 64,000 by 2001. In contrast, the number of employed partnered fathers aged 25–44 with a child under five declined from 116,000 to 100,000. These changes reflect both changing demographics and changing employment rates of child-rearing partnered mothers and fathers.

Tables 6 and 7 explore the change in working hours of partnered parents by level of highest education. To further simplify the analysis, only the 25–34 age group is considered. Table 6 shows that in 2001 the group of partnered fathers most likely to work short hours were those with no formal qualifications. In contrast, the fathers most likely to work long hours were those with degrees or higher educational qualifications. When a more detailed analysis was carried out, it was found that many of the well-educated fathers recorded exactly 50 hours of paid work, and it was poorly educated fathers who were slightly more likely to work more than 50 hours.

Table 6 Changes in Hours of Partnered Employed Fathers Aged 25–34 with a Child under 5 Years, by Highest Level of Education, 1986 and 2001

Highest qualifications % < 20 hours per week % = 50 hours per week
1986 2001 ? 86–01 1986 2001 ? 86–01
Degree or higher 0.9 2.5 1.6 36.3 40.2 3.9
Other tertiary 0.9 2.1 1.2 34.3 38.8 4.5
School 1.4 2.8 1.4 36.2 38.0 1.8
No qualifications 2.9 4.1 1.2 32.0 38.5 6.5

The changes in working hours by education were also strong for employed mothers (Table 7). Although well-educated mothers (any tertiary qualification) were the most likely to be working short hours in 1986, this reversed in 2001. Linked to this, the strongest decline in the proportion of mothers working short hours was among those with a degree or higher. This was also the only group of mothers where the proportion working 50 or more hours per week increased. In 2001, well-educated mothers with a preschool child were the most likely to work long hours. However, across educational groups, long hours of paid work were relatively uncommon among partnered mothers with a preschool child. If long hours are required for career advancement in most senior professional and managerial occupations, then compared with partnered fathers few mothers are working such hours and this places them at a relative disadvantage.

Table 7 Changes in Hours of Partnered Employed Mothers Aged 25–34 with a Child under 5 Years by Highest Level of Education, 1986 and 2001

Highest qualifications % < 20 hours per week % = 50 hours per week
1986 2001 ? 86–01 1986 2001 ? 86–01
Degree or higher 47.7 33.7 -14.0 8.8 11.5 2.7
Other tertiary 49.6 39.1 -10.5 9.2 7.7 -1.5
School 46.3 37.6 -8.7 9.7 7.0 -2.7
No qualifications 39.5 38.1 -1.4 7.9 7.7 -0.2

The trends for those older parents with a child under five were broadly similar. However, in this age group, a slightly higher proportion of both mothers and fathers were likely to work longer hours. There are many possible explanations for this. One contributing factor is that workers in this age group will be more likely to be in senior positions in workplaces and therefore expected to work longer hours.

Underlying the changes in hours are strong changes in the proportion of qualified working parents. For example, in 1986, 7.5% of partnered working mothers aged 25–34 with a child under five had a degree or higher qualification. By 2001, this had doubled to 15.3%. At the other extreme, the proportion of working mothers with no formal qualification more than halved from 25% to 10.8%. This reflects changes in educational participation, as well as in employment and marriage rates by education.

Changes in Total Hours of Employed Couples

The following data are again based around preschool children and, initially, do not take into account the age of parents. Table 8 shows the proportion of child-rearing couples working a total of under 30 hours per week or, in terms of long hours, between 80 and 100 hours and 100 or more hours per week in both 1986 and 2001. Again, it shows an increase at both ends of the weekly working hours’ spectrum, but with the strongest growth in longer hours. The increasing hours of work for couples primarily reflect three trends: an increase in the proportion of fathers working long hours, increasing employment rates for mothers and, at the same time, a decline in the number of employed mothers working short hours.

Table 8 Changes in Weekly Combined Hours of Paid Work for Employed Couples with a Child Under 5 Years, 1986 and 2001

% in each group of hours
< 30 30 < 40 40 < 50 50 < 80 80 < 100 100+
1986 2.3 5.5 40.7 36.4 10.1 5.0
2001 4.2 3.4 27.4 42.2 15.6 7.2
? 86–01 1.9 -2.1 -13.3 5.8 5.6 2.2

Changes by age and education levels of both parents of a New Zealand preschool child were then considered. At an aggregate level there was a rise in average combined hours worked by couples aged 25–34 and 35–44 between 1986 and 2001. For the former group, the rise was from 56 hours to 62 hours per week; for the latter it was from 58 to 63 hours.

Table 9 again restricts the analysis to couples where the mother was aged 24–34. It shows the smallest rise in average hours was for couples where both partners had no formal educational qualifications. Underlying these data, however, are again major changes in the qualification mix of couples in this age range. For instance, in 1986, 19% of these couples had no qualifications, but by 2001 this had dropped to just 7%. The decline was even stronger in the 35–44 age group, from 22% to just 6%. This reflects a number of trends, including increased education levels among both men and women and delayed childbearing, particularly among the well educated.

Table 9 Average Hours Worked per Week by Employed Couples Where the Female Was Aged 25 to 34 Years with a Child Under 5 Years, by Highest Qualification Gained, 1986 and 2001

1986 2001 ? 86–01
Both partners have degree or higher 56.5 60.9 4.5
Other combination of qualifications 56.6 61.9 5.3
Neither has a qualification 55.2 57.1 1.9
Total 56.3 61.5 5.2

In terms of longer hours (Table 10), while the differences were not great in 1986 it was poorly educated couples that had the highest proportion working 80 or more hours of combined work. By 2001 there was little difference by education level. This shows that long working hours can now be found across the educational spectrum.

Table 10 Proportion of Employed Couples with a Preschool Child Working 80 or More Hours of Combined Work Per Week Where the Female Was Aged 25 to 34 Years, by Highest Qualification of Both Partners, 1986 and 2001

1986 2001 ? 86–01
Both partners have degree or higher 14.0 22.5 8.5
Other combination of qualifications 13.9 22.2 8.3
Neither has a qualification 16.7 21.1 4.4
Total 14.4 22.2 7.8

The proportion of couples with a preschool child working 100 or more hours per week of combined work was then calculated. In total, 6.8% of couples aged 25 to 34 years worked these hours. Well-educated couples were again very marginally over-represented among those working long hours. At the other end of the hours spectrum, poorly educated couples were slightly over-represented.

While qualifications are of some importance with regard to hours of work, both at the overall individual and household levels (particularly when average hours are considered), the detailed analysis showed that it was partnered fathers with no formal qualifications who tended to work longer hours (more than 50) than other partnered fathers. Yet, overall, couples with no qualifications did not stand out as working the longest hours. This suggests that some poorly educated partnered fathers may be putting in long hours because their partner is less likely to be employed or, if employed, is both low paid and puts in low hours of work.

Finally, when ethnicity was considered, the data showed only a small variation in average hours worked by couples with a child under five. Couples with an Asian child put in the longest hours (as did Asian sole mothers) but (unlike sole mothers) parents of European children were not far behind. Couples with a Pacific child worked the lowest average hours. Again, the reasons for these differences are unclear.

Why Might Families be Working Longer Hours of Paid Work?

While not an ideal data source for understanding decision-making in families, census data provide some grounds for speculating why some sole parents and couples are working longer hours of paid work. Although the education data for couples suggest there will be a group of poorly educated parents working long hours to make ends meet, the data also indicate that some of the shift to longer hours for families is likely to have come about through choice rather than simply economic need. As demonstrated, most of the change in working hours has come through changes in women's employment. As a group, women are becoming better educated, with better-educated women tending to have higher levels of employment and to work longer hours than poorly educated women. A significant group of women are making positive choices to participate in education, and higher employment rates are one outcome of this. If gender equity both in a society and within individual opposite-sex couples is seen to be a societal goal then the increasing education and employment of women should be viewed as a positive trend.

However, as research in the United States shows, choice of working hours by both women and men is still conditioned by a range of societal factors. For example, Schor (1998) has argued that long hours have been driven by a culture of consumption. Competitive materialism constantly requires additional financial resources, which many Americans (and possibly New Zealanders) try to meet by increasing the number of hours they work for pay. In addition, Bell and Freeman (2000) suggest that there are particular incentives to work long hours in countries with a high level of wage inequality, such as the United States, Britain and New Zealand. Working hard is potentially rewarded through career advancement, while a lack of perceived effort can attract a major earnings' penalty. Even well-educated individuals may face wage penalties and career barriers if they challenge working-hour norms.

There are a variety of other reasons for long hours of work. For instance, working such hours may provide a higher level of job security. Hours of paid work may therefore increase in times of high unemployment or other periods of job insecurity. Equally, hours of work might increase in times of economic expansion and labour shortages. For example, Hays New Zealand (2004) reported in August 2004 that 52% of New Zealand companies had increased their overtime levels in the previous year. This was a period when unemployment levels were very low and there were reported skill shortages.

The housing market also potentially provides incentives to work longer hours. In a time of rapidly rising property prices, owning a house may become unaffordable for first homeowners working standard hours. Initial home ownership might become more dependent on longer hours of work for individuals and couples.

Other data show that single people and couples without children tend to work the longest hours (Callister 2004a). These also indicate that individual and family earnings are closely correlated with hours of work. This suggests that, when compared with non-parents, parents and even well-educated parents are being economically disadvantaged through working shorter hours of paid work. Based on both New Zealand and international fertility trends, it also appears that a significant number of well-educated women are making a trade-off between fertility and success in their careers (Callister 2002, Grant et al. 2004, Sceats 2003). Success in careers may not only require unbroken periods of paid work, but also long hours worked.

However, census data are cross-sectional. It also needs to be kept in mind that over a life-cycle many individuals and families, including the well educated, will have times of high hours of paid work and other times of fewer hours. The changing employment rates and hours of work (mainly for women) by age of child support this life-cycle view. Higher employment rates and longer hours among older women (Table 2), either with older dependent children or no dependent children in the home, suggest that while time out of work, or reduced working hours, could be important for a period in their working lives, there can also be long periods of relatively high employment over a life-cycle. As noted in the OECD (2004) report, Babies and Bosses, while New Zealand mothers tend to exit paid work when their children are very young, we have a stronger return to paid work, and from part time to full time work, as the youngest child moves into older age groups than in many other countries.

Census data provide little information on preferences for hours of work, aside from "revealed preferences". It would be useful to have a better idea of families' preferences regarding working hour, and particularly fathers' preferences, given they tend to put in the longest hours of paid work in both sole-parent and couple households. International research suggests a sizeable number of workers would prefer a different workload. However, some workers want longer hours, some shorter ones. While it is more likely that those working part time want longer hours and those working full time the opposite, there are also part-time workers who want to reduce their hours and full-time workers who want longer hours. This suggests that although flexibility in time schedules and work patterns is growing, a significant number of workers are still constrained by organisational schedules or economic necessities in their time allocation (e.g. Bielenski et al. 2002, Böheim and Taylor 2004, Clarkberg and Moen 2001, Drago and Yi-Ping 2003, Evans et al. 2001, Jacobs and Gerson 2001, Reynolds 2003, Stier and Lewin-Epstein 2003, Wooden and Loundes 2001).

Focusing specifically on fathers, research carried out in Australia on the working preferences of partnered fathers suggests that, overall, fathers' satisfaction with their work hours decreased as the number of hours worked increased (Weston et al. 2004). In addition, the proportion of fathers who would prefer to work fewer hours (taking into account the impact this would have on their income) increases with the number of hours worked. Yet, the research also found that a quarter of fathers working very long hours (60 or more) were satisfied with their working hours. In New Zealand, an on-line survey of fathers indicated that 80% of them wished they could spend more time with their children (EEO Trust 2003). Also in New Zealand, the Ministry of Social Development (forthcoming) has carried out some exploratory work on parents' working-hour preferences. This again shows some mismatch between actual and preferred working hours.

The Effects of Increased Working Hours on Unpaid Work, including Childcare

Consideration also needs to be given to the impact increasing working hours may have on children. Discussions about the effect of paid work on family life are not new. In the 1950s, when mothers began increasing their participation in paid work concerns about maternal deprivation were first raised (e.g. Bowlby 1952). However, research on the effects of mothers’ involvement in paid work has become far more sophisticated. Based on a review of international literature, the Ministry of Women’s Affairs (2004: 43) conclude, “Maternal employment in itself has no significant negative or positive effects on children, although some international evidence suggests small negative effects for very young children”. However, the Ministry of Women’s Affairs goes on to note that, “Early, extensive and poor quality non-maternal care combined with poor quality home care leads to negative effects.”

Like many studies of the effects of maternal employment, a range of factors influences outcomes, including the quality of maternal employment and quality of time spent with children. Where and when the work is carried out can also be important (Callister 2004c). However, while focusing primarily on maternal employment, the Ministry of Women’s Affairs noted that one of the important influences on child outcomes is paternal care, and certainly the census data show that it is primarily fathers who work long hours. While there are still few studies of the effects of paternal employment – and particularly long hours of work – on child outcomes, interviews with young Australians about work–life balance issues suggest that many young people wanted to spend more time with their fathers (Pocock 2004).

Knowing more about time spent with children requires information on total hours of work, both paid and unpaid. New Zealand and overseas data provide some guide as to what the effects on unpaid work might be. While total hours worked by sole parents is important, given that this paper focuses primarily on couples the following review data also only consider couples.

In New Zealand there has only been one time-use survey carried out, and so changes in total working hours over time cannot be determined. This time-use study was carried out much nearer to 2001 than to 1986 so will include families that have increased their working hours. Across the total population the data show that men’s and women’s total hours of work are very similar, but that men undertake more paid work and women more unpaid work. When the sample is restricted to partnered men and woman with a child under five, Stevens (2002) demonstrates that total hours of work are higher for parents of young children than for men and women without children.4 Steven’s data also show that the ratio of total hours of women’s to men’s work was 0.96; that is, on average partnered men with a child under five work longer total hours than partnered women. These data indicated that 19% of women’s total work was paid, while for men it was 64%. The total hours data suggest that, contrary to popular discourse, in New Zealand it is, on average, fathers who suffer more from the “double burden” than mothers. The OECD (2004) has also produced a ratio of total paid and unpaid time of women to men for couple families with a child under six years of age. This ratio again shows that, on average, men work longer total hours than women. The New Zealand ratio was 0.7, compared with ratios of 1.2 in Portugal (where the double burden falls more heavily on women), and 1.0 in Switzerland.5

However, these data are for all partnered men and women, not just couples where both are employed. They include couples where the father works full time and the mother stays at home. It may be that in New Zealand it is the potential for a “double burden” that prevents some mothers from entering the workforce or moving from part-time to full-time employment. International time-use data restricted to couples where both partners work full-time suggest that there is a significant double burden in some countries but not others. Table 11 shows the ratio of women’s to men’s total work time (paid and unpaid) in couple households with a child under five where both partners work full-time. These data do not include New Zealand.6 The total work time for couples in Sweden, a country where paid working hours are relatively short, is nearly equal. However, in the other countries, women working full-time have a higher total workload than men.

Table 11 Ratio of Women’s to Men’s Total Work Time (Paid and Unpaid) in Couple Households with a Child under 5 Years and Where Both Partners Work Full-Time

USA (1995) 1.05
UK (1999) 1.16
Sweden (1991) 0.99
Italy (1989) 1.26
Source: OECD (2001)

While the “double burden” may be a problem for some parents, is it a problem for children? In the United States, every five years, the Families and Work Institute conducts a National Study of the Changing Workforce (Bond et al. 2002). This study shows that the proportion of married wage and salaried employees who lived in dual-earner couples has increased substantially, from 66% in 1977 to 78% in 2002. In the same period, combined work hours for dual-earner couples with children rose 10 hours a week, from 81 hours a week in 1977 to 91 hours 25 years later. But in a somewhat surprising result, the combined time that parents spent caring for their children on workdays increased – from 5.2 hours in 1977 to 6.2 hours in 2002.

This is not the only study to find such a result.7 Also in the United States, Bianchi (2000) found that, despite the rapid rise in mothers’ labour-force participation, mothers’ time with children has been quite stable over time. She notes that in the past, non-employed mothers’ time with children was reduced by the demands of unpaid family work and domestic chores and by the use of mother substitutes for childcare, especially in large families. Bianchi comments that employed mothers now try to find new ways to maximise time with children. For example, in all the years studied employed mothers undertook less housework than non-employed mothers, although total hours of housework was also declining for both groups. The reduction in housework hours can come about in a variety of ways. Standards may be lowered or housework time may be “intensified”; that is, more work is carried out in less time. For example, dishwashers or clothes driers may speed up housework, or individuals may simply work harder. For those who can afford it, “professionals” are increasingly cleaning houses, while other forms of household work, such as food preparation, are increasingly being “outsourced”. Such strategies can lessen the potential “double burden”.

However, just as importantly, Bianchi also found that within couples, fathers are spending more time with their children than in the past, potentially increasing the total time children spend with parents even as mothers work longer hours outside the home. This trend of increasing paternal care has also been shown by other overseas time-use studies in a range of industrialised countries (e.g. Gershuny 2000, Yeung et al. 2001). For example, in the United States, Bond et al. found that in dual-earner couples, since 1977 fathers have increased the time they spend on workdays doing household chores, including childcare, by approximately 42 minutes, while mothers have reduced their time by the same amount, although still doing more than fathers.

New Zealand time-use data also show that some parents are undertaking paid work at home in the evenings and weekends, time that children will also be generally home (Callister 2004c). A further investigation of these data indicates that nearly 24% of working parents recorded undertaking simultaneous paid work and childcare with, not surprisingly, almost all of the simultaneous childcare passive rather than active childcare (Callister and Singley 2004). But given the popular idea that it is mainly women who multi-task, an unexpected result was that when both passive and active childcare are considered, there was little difference between women and men. The data showed that just over a third of employed New Zealand mothers and fathers undertook a spell of simultaneous work in weekends, while during the week the figures were a fifth for fathers and a quarter for mothers. However, these data provide no information on the quality of such childcare.

Finally, another reason why parents may have been able to increase their hours of work but not reduce time spent with their children is that they may have less time for themselves. Research by Bond et al. (2002) has supported this notion. In the United States, in 2002 fathers spent 1.3 hours on themselves on workdays, down from 2.1 hours in 1977. But the study found mothers have even less time for themselves – 0.9 hours versus 1.6 hours in 1977.

Overall, the various sources of data, including the data on why families are working longer hours and the effects on children, suggest a very complex picture. There appear to be both positive and negative outcomes associated with longer family working hours. However, preference literature from countries with long working hours suggests some parents working long hours consider themselves to be overworked and would like to find ways of reducing their paid work hours, but often face barriers to doing so.

Policies to Reduce Working Hours

Is the rise in working hours for some families likely to abate? In early 2005, the government raised the issue of lifting employment rates for some groups of New Zealand women (Clark 2005). A range of policies is to be developed to encourage more mothers to enter (or re-enter) paid employment. Policies designed to bring more sole parents into paid work may not increase average working hours for those employed and, in fact, if new entrants work part-time may even reduce them. However, simply through increased employment rates, unless this paid work only occurs in school time, more sole parents are likely to spend less time with their children. If greater support is given to working parents, it is possible that employment rates for partnered mothers will also continue to increase. Unless there are quite dramatic changes in employment rates and/or hours of work for fathers, this is likely to result in a further increase in the average hours of work of child-rearing couples.

A number of New Zealand studies have already provided reasons to expect some further increase in mothers’ employment (e.g. Bryant et al. 2004, OECD 2004). If gender equality is to be achieved in New Zealand through women’s patterns of work (both in employment rates and hours worked) becoming more similar to those of men rather than through changes in both men’s and women’s working hours, then inevitably family hours of work will continue to rise. If, for example, employment rates for partnered mothers reach those of Sweden (Table 1) and partnered New Zealand fathers’ hours of paid work stay above those of Swedish fathers, then New Zealand children will have parents working longer combined hours than their Swedish peers.8

Alternatively, can governments do anything to reduce excessive working hours for some parents? Initiatives to reduce working hours can be broadly divided into three groups.9 The first involves government encouraging voluntary working-hour reduction agreements between workers and employers. This is the approach taken by Australia. This model relies on increasing flexibility in the labour market so employers and employees can individually or collectively work out agreements that suit both parties. This has been supported by campaigns to encourage the provision of work–life balance policies in the workplace. This is, principally, also the model currently used in New Zealand.

There are two, not mutually exclusive, types of legislation that have been used internationally to try to reduce working hours. The first endeavours to place a cap on working hours. The French 35-hour working week legislation is perhaps the best-known example. The second type of legislation primarily gives support to those wanting to work part-time. Examples can be found in both the United Kingdom and the Netherlands. The latter policies are designed mainly to support workers fulfilling their preferences rather than placing restrictions on workers’ choices.

The British legislation, designed to support a reduction in working hours, falls into both broad categories of legislation.10 The first, which aims to place a “cap” on working hours, has a cut-off point of 48 hours. The second aims to support part-time work. For instance, the Employment Act 2002 introduced new employment legislation designed to help working parents. This legislation has many aims, but one of them is to reduce any discrimination against part-time workers and workers who wish to work non-standard hours. In this legislation, from April 2003 parents with young and disabled children have the right to request flexible work to facilitate childcare (Tiger 2004). Under the regulations, employees can request to: change the hours they work, change the times they are required to work, or work from home (whether for all or part of the week).

In a paper published in 2003, Barnard et al. argue that the various working-hours legislation has so far had little impact on an ingrained culture of long working hours in the United Kingdom. However, another paper puts forward evidence that suggests that working time has been reducing (United Kingdom Government 2004). According to the United Kingdom Labour Force Survey, the proportion of full-time employees usually working over 48 hours per week has fallen since 1998. In the first quarter of 2003, 20.4 per cent of full-time employees usually worked more than 48 hours, compared to 23.3 per cent in the same time period in 1998. This decline was driven by a reduction in the number of men reporting long working hours. Between 1998 and 2003, no decline was evident in the proportion of women working over 48 hours (United Kingdom Department of Trade and Industry 2004).

There is much ongoing debate about both the philosophy and the actual workings of the United Kingdom working-hour legislation. The United Kingdom government believes that having a highly flexible labour market has been a key factor behind its recent strong employment performance. It also recognises working-hour preferences are diverse of and the government does not want to overly restrict free choices made by individuals. Overall, the government “believes strongly that the UK’s competitiveness should not depend on people working long hours but equally that people should be free to determine their own working patterns” (United Kingdom Department of Trade and Industry 2004: 10).

In the context of New Zealand’s aim to support a relatively high level of labour market flexibility, the strategies adopted by the United Kingdom government appear a useful model to explore further if the New Zealand government wishes to develop policies to support a reduction in the proportion of New Zealanders working long hours of paid work. However, the legislation that puts a cap on working hours potentially diminishes preferences and there are a variety of ways of avoiding this particular United Kingdom legislation. These methods include the following (Wooden 2003).

  • Many of the individuals working long hours tend to be in occupations that are not much affected by regulation, such as managers or the self-employed.
  • Many of the additional hours are not directly paid for on an hourly basis and hence presumably not recorded, thus making the policing of restrictions on hours difficult.
  • Many of these extra hours are worked at home which, in most instances, will be beyond the reach of regulation.
  • Restrictions can always be circumvented, for example by shifting employees to self-employed status, or by workers by taking additional part-time jobs.

A further problem is that if such schemes are designed with the aim of ensuring children spend more time with their parents, both individuals and couples’ hours need to be considered. For example, in a couple with children, one partner may work 55 hours a week with their partner at home looking after the children full-time. If a cap was placed at 48 hours, as in the United Kingdom, then the main earner would have their hours reduced. This may result in this worker spending more time with their children, but is likely to reduce total family income and may force the other partner into part-time work. It may potentially cut across the preferences of both partners in this couple. In contrast, a couple where both partners worked 45 hours would not have their working hours capped, yet potentially would be spending 35 hours more hours per week away from their children.

In contrast, it seems that those United Kingdom policies that promote higher-quality part-time work may be more helpful in supporting the working-hour preferences of many workers, and there appear to be fewer incentives to circumvent such regulation.

However, given the worldwide problems of women’s over-representation in part-time work and the resulting gender equity implications, particularly in terms of earnings, a model developed in the Netherlands for supporting part-time work may merit further investigation (Callister 2004b). The proportion of men working part-time in the Netherlands is relatively high and it would be worth exploring the reasons for this. One possible factor is that achieving gender equity in both paid and unpaid work takes a higher priority in many continental European and Nordic countries than it does in the remainder of the OECD, including New Zealand, with many policies designed to support equal sharing of work in both spheres.

Finally, it needs to be kept in mind that government policies to reduce long hours of work can only go so far. A review of work–life balance issues in New Zealand concludes that organisational culture and businesss incentives are important considerations in the design of work–life balance policy (Varuhas et al. 2003). The authors argue that policy interventions that result in costs for the firm that are not offset by benefits can create perverse incentives. These may cause firms to be reluctant to hire staff who are likely to draw on family-friendly benefits. The researchers note that government policies that require firms to provide family-friendly measures, including working-time reduction policies, for employees are likely to be ineffective in improving work–life balance if the workplace culture is not family friendly. Varuhas et al. go on to suggest that while the government has limited ways of influencing workplace culture in private firms, as a major employer itself in New Zealand it has the opportunity to lead by example. This is a sensible suggestion that needs further investigation.


Barnard, C., S. Deakin and R. Hobbs (2003) “Opting out of the 48-hour week: Employer necessity or individual choice? An empirical study of the operation of article 18(1)(b) of the Working Time Directive in the UK” The Industrial Law Journal, 32(4):223-252.

Bell, L.A. and R.B. Freeman (2000) The Incentive for Working Hard: Explaining Hours Worked Differences in the U.S. and Germany, NBER Working Paper 8051, National Bureau of Economic Research, Inc., Cambridge, Massachusetts.

Bianchi, S.M. (2000) “Maternal employment and time with children: Dramatic change or surprising continuity?” Demography, 37(4):401-414

Bielenski, H., G. Bosch and A. Wagner (2002) Working Time Preferences in Sixteen European Countries, European Foundation for the Improvement of Living and Working Conditions, Ireland.

Böheim, R. and M.P. Taylor (2004) “Actual and preferred working hours” British Journal of Industrial Relations, 42(1):149-167.

Bond, J.T. with C. Thompson, E. Galinsky and D. Prottas (2002) “Highlights of the national study of the changing workforce: Executive summary” Families and Work Institute No. 3

Bosch, G. and S. Lehndorff (2001) “Working time reduction and employment: Experiences in Europe and economic policy recommendations” Cambridge Journal of Economics, 25:209-43.

Bowlby, J. (1952) Maternal Care and Mental Health, a report prepared on behalf of the World Health Organization as a contribution to the United Nations programme for the welfare of homeless children, World Health Organization, Geneva.

Bryant, J., V. Jacobsen, D. Garrett and M. Bell (2004) Labour Force Participation and GDP in New Zealand, Treasury Working Paper 04/07, The Treasury, Wellington.

Callister, P. (2002) “Aging population and social policy in New Zealand: Could ‘family-friendly’ policies increase both fertility and women’s employment?” New Zealand Population Review, 28(2):221-251.

Callister, P. (2004a) The Future of Work Within Households: Understanding Household-Level Changes in the Distribution of Hours of Paid Work, Department of Labour, Wellington.

Callister, P. (2004b) “Long hours of work and work-life balance: A scan of the literature” Unpublished report prepared for the Department of Labour, Wellington.

Callister, P. (2004c) “Time use data and work/life policy development” Social Policy Journal of New Zealand, 22:108-118. Callister, P. and S.G. Singley (2004) “A double burden or the best of both worlds? A research note on simultaneous paid work and childcare in New Zealand” working paper

Clark, H. (2005) “Moving towards a better NZ future” NZ Herald, 1 February

Clarkberg, M. and P. Moen (2001) “Understanding the time-squeeze: Married couples’ preferred and actual work-hour strategies” American Behavioral Scientist, 44 (7):1115-1136.

Craig, L. (2005) “How do they do it? A time-diary analysis of how working mothers find time for the kids” Social Policy Research Centre Discussion Paper No. 136, Social Policy Research Centre University of New South Wales.

Dawson, D., K. McCulloch and A. Baker (2001) Extended Working Hours in Australia: Counting the Costs, report commissioned by the Department of Industrial Relations, Australia.

Department of Labour (2004) Achieving Balanced Lives and Employment: What New Zealanders Are Saying About Work–Life Balance PDFs/wlb-consultation-summary.pdf.

Drago, R. and T. Yi-Ping (2003) “Family structure, usual and preferred working hours, and egalitarianism” paper for presentation at the HILDA Conference, March, University of Melbourne.

EEO Trust (2003) “What do Kiwi fathers want?” news/media.php?id=53.

Evans, J.M., D.C. Lippoldt, and P. Marianna (2001) “Trends in working hours in OECD countries” Labour Market and Social Policy Occasional Paper No 45, OECD, Paris.

Golden, L. and H. Jorgensen (2002) “Time after time: Mandatory overtime in the U.S. economy ” EPI Briefing paper

Goodger, K. (2001) “Recent trends in the employment rate of sole mothers in New Zealand” Social Policy Journal of New Zealand, 16:191-198.

Gershuny, J. (2000) Changing Times: Work and Leisure in Postindustrial Society, Oxford University Press, New York

Grant, J., S. Hoorens, S. Sivadasan, M. van het Loo, J. DaVanzo, L. Hale, S. Gibson and W. Butz (2004) Low Fertility and Population Ageing: Causes, Consequences, and Policy Options, Rand Europe

Hays New Zealand (2004) “Overtime tipping work/life balance

Jacobs, J. A. and K. Gerson (2001) “Overworked individuals or overworked families?: Explaining trends in work, leisure, and family time” Work and Occupations, 28(1):40-63.

Jacobs, J.A. and J.C. Gornick (2001) “Hours of paid work in dual-earner couples: The U.S. in cross-national perspective” Luxembourg Income Study Working Paper No. 253

Johnston, G. and A. White (forthcoming) Labour Force Participation of Women in New Zealand, New Zealand Treasury, Wellington.

Messenger, J. C. (2004) “Finding the balance: Working time and workers’ needs and preferences in industrialized countries: A summary of the report and its implications for working time policies” paper presented at the 9th International Symposium on Working Time Paris, 26–28 February

Ministry of Social Development (2003) The Social Report 2003: Te Purongo Oranga Tangata 2003, Ministry of Social Development, Wellington

Ministry of Social Development (forthcoming) Work, family and parenting study: Phase 2 research findings, Ministry of Social Development, Wellington.

Ministry of Women’s Affairs (2002) “Participation in paid work of mothers with a youngest child aged under five years”

Ministry of Women’s Affairs (2004) “Influences of maternal employment and early childhood education on young children’s cognitive and behavioural outcomes”

New Zealand Council of Trade Unions (2002) Thirty Families: Interim Report of the Thirty Families Project: The Impact of Work Hours on New Zealand Workers and Their Families, New Zealand Council of Trade Unions, Wellington.

OECD (2001) “Balancing work and family life: Helping parents into paid employment” in OECD Employment Outlook, June, OECD, Paris, pp.129-166.

OECD (2004) Babies and Bosses: Reconciling Work and Family Life: Volume 3: New Zealand, Portugal and Spain, OECD, Paris.

Pocock, B. (2001) The Effect of Long Hours on Family and Community Life: A Survey of Existing Literature, report for the Queensland Department of Industrial Relations, Brisbane.

Pocock, B. (2004) “Dad or a DVD – No contest

Reynolds, J. (2003) “You can’t always get the hours you want: Mismatches between actual and preferred work hours in the US” Social Forces, 81(4):1171-1199.

Sceats, J. (2003) “The impossible dream: Motherhood and a career?” New Zealand Population Review, 29(1):155-171.

Schor, J. B. (1998) The Overspent American: Upscaling, Downshifting, and the New Consumer, Basic Books, New York.

Singley, S.G. and P. Callister (2003) “Work poor or working poor? A comparative perspective on New Zealand's jobless households” Social Policy Journal of New Zealand, 20:134-155.

Stevens, D. (2002) “Equity or equality: A study of the time men and women spend in paid and unpaid work in New Zealand households” paper presented at the Time Pressure, Work–Family Interface, and Parent–Child Relationships International Time Use Conference, Kitchener-Waterloo, Ontario, Canada, March 21-23.

Stier, H. and Lewin-Epstein, N. (2003) “Time to work: A comparative analysis of preferences for working hours” Work & Occupations, 30 (3):302-327.

Tiger (2004) Taylored Interactive Guidance on Employment Rights

United Kingdom Department of Trade and Industry (2004) "Working Time - widening the debate: A preliminary consultation on long hours working in the UK and the application and operation of the working time opt out”

United Kingdom Government (2004) “Communication from the Commission concerning the re-examination of directive 93/104/ec concerning certain aspects of the organisation of working time: Response of the United Kingdom”

Varuhas, J., L. Fursman, L. and V. Jacobsen (2003) Work and Family Balance: An Economic View, Treasury Working Paper 03/26, Wellington

Weston, R., M. Gra, L. Qu and D. Stanton (2004) Long Work Hours and the Wellbeing of Fathers and Their Families, Research Paper No 35, Australian Institute of Family Studies, Melbourne.

White, J. and J. Beswick (2003) “Working long hours” Health and Safety Laboratory, WPS/02/10

Wooden, M. (2003) “Long-hours working and enterprise bargaining” Agenda, 10(3): 259-271

Wooden, M. and J. Loundes (2001) How Unreasonable are Long Working Hours? Melbourne Institute Working Paper No. 1/02, Melbourne Institute of Applied Economic and Social Research, University of Melbourne.

Yeung, W. J., J.F. Sandberg, P.E. Davis-Kean and S.L. Hofferth (2001) “Children's time with fathers in intact families” Journal of Marriage and Family, 63(1):136-154.


1 Acknowledgement
A research grant from the Department of Labour’s Future of Work Contestable Fund enabled this research to be carried out. This paper draws on two more detailed papers. The first is a technical working paper produced as part of a Labour Department Future of Work project (Callister 2004a). The second is a literature scan prepared for the Department of Labour’s work–life balance programme (Callister 2004b)

2 Data was not obtained on ethnicity and working hours of sole fathers.

3 Due to a coding error by the author, some incorrect data on the proportion of partnered mothers and fathers working 50 or more hours was contained in the “Final 4 November 2004” version of the main research report (Callister 2004a). A revised paper was placed on the Department of Labour’s website on 16 February, 2005.

4 These data are not calculated using couples as the unit of analysis but are calculated for individuals who live in couples.

5 It is not clear why the OECD ratio and the ratio calculated by Stevens are so different.

6 Statistics New Zealand has not deposited the time-use data at the international time-use study centre at Essex University, where data shown in Table 11 are drawn from. This calculation could be repeated using New Zealand Time Use data (Callister 2004c).

7 For an Australian example, see Craig (2005).

8 Data indicating weekly hours of paid work for Swedish parents can be found in Jacobs and Gornick (2001).

9 There is a fourth model, which appears to involve a consensus between employees, employers and the government, where the goal is high productivity but relatively short working hours. This is the Swedish model. However, given that the Swedish model is so different from New Zealand in many areas of economic and social policy it is not considered in this paper.

10 A particularly useful site for information on British working-hour legislation is:

Cover photo of Social Policy Journal


Social Policy Journal - Issue 24

Overworked Families? Changes in the Paid Working Hours of Families with Young Children, 1986 to 2001

Print this page.