Manajemen | Fakultas Ekonomi Universitas Maritim Raja Ali Haji 2005 25
TIME, MONEY AND JOB SPILLOVER: HOW
PARENTS’ JOBS AFFECT YOUNG PEOPLE
BARBARA POCOCK AND JANE CLARKE∗
T
his paper examines the perspectives of young people about their parents’ paid and
unpaid work, their preferences for time or money through more parental work, and
their views about how their parents’ jobs affect them. It analyses qualitative empirical data
collected in Australia in late 2003, by means of focus groups among 10–12 and 16–18year-old males and females in urban and rural locations in two Australian states, in both
high and low socioeconomic areas. It finds that more Australian children are looking for
more time from parents than more money from more parental work, though this varies
by income level, location and parental hours. This preference for ‘time over more money’
is consistent in single- and dual-earner couple households as well as sole parent/earner
households. Children are acute observers of parents and their jobs. Both positive and
negative spillovers are widely observed. Negative spillovers from long or unsocial hours
are especially marked, reinforcing other findings in support of policy interventions to
contain long or unsocial hours.
INTRODUCTION
Changes in patterns of paid and unpaid work, workplaces, and household shape
have been the subject of much recent analysis in Australia (HILDA 2001;
Campbell 2002; Megalogenis 2003; Pocock 2003; Pusey 2003; Summers 2003;
Tanner 2003; Watson, Buchanan, Campbell & Briggs 2003). They are driving
a lively policy and political interest in the work and family ‘collision’ and government responses to it, though action is slower to follow. The perspective of
children is missing from these accounts, yet is relevant to industrial, workplace
and household perspectives. Parents, in particular, act on certain assumptions
about children’s welfare as they determine household patterns of participation in
paid and unpaid work.
Whereas there has been some international research on the specific issue of
parental work patterns and children’s views of them, there has been very little in
Australia with the exception of Lewis, Tudball and Hand (2001). Galinsky (1999)
argues, based on her large US study, that the ‘work/family’ debate has been misframed with too much focus on whether working mothers in particular are ‘bad’
for children, rather than how work affects parents, and through them, children.
In the USA, the real question about parental work is ‘how parent’s work’, and
how attentive or focused they are able to be towards their children when they
are with them. Galinsky found that older children longed for more parental time,
especially with their fathers. These older children looked for ‘hang around time’
∗ School
of Social Sciences, University of Adelaide, Adelaide, SA, 5005. Australia.
Email: [email protected]
THE JOURNAL OF INDUSTRIAL RELATIONS, VOL. 47, NO. 1, MARCH 2005, 62--76
TIME, MONEY
AND
JOB SPILLOVER
63
more than younger children. She argues that the nature of parents’ jobs is very important to children who are very alert to parental moods. This result is confirmed
by Nasman’s study of Swedish children about parental work which found that
work affects parents’ ‘state and physical conditions’. This ‘colouring’ spills over
to family life, overriding the idea of work and family as separate social spheres’
(2003: 51). This spillover was higher for parents with irregular or long hours and
for single mothers, and manifested itself most obviously through parental fatigue.
In a Scottish study McKee, Mauthner and Galilee (2003) found that children were
profoundly aware of their parents’ work and its effects: they could ‘competently
assess how work made their parents feel’ (2003: 39) and many wanted to avoid the
stress arising from external control of work in their own future lives.
In Australia, Lewis, Tudball and Hand (2001) undertook interviews with a nonrandom group of parents and children from 47 Melbourne families (71 children
over eight years old). The majority of children in the study felt that their parents
worked ‘about the right amount of time’ and that—in accord with Galinsky’s
work—‘it is not whether and how much parents work, but how they work and
how they parent, that matters’ (2001: 23). Responses were divided ‘roughly evenly
between those saying that they wished their parents spent more time with them
and those who said their parents currently spent enough time with them’ (2001:
24). Only two parents in Lewis et al.’s study worked more than 50 hours a week.
This Australian and international literature mostly addresses the question of
spillover from parental work onto children. This paper addresses several questions. First, how do young people perceive the time/work patterns and trade-offs
in their households? Second, how do young Australians perceive spillover from
their parents’ jobs? Third, do these perceptions vary in relation to household
income, locations, household types, and different patterns of parental working
hours? Do young people with a parent at home also experience spillover from
that domestic form of work? Finally, how do these experiences and perceptions
affect their own plans for the future?
The analysis considers the perceptions of a group of young Australians at school
in Year 6 (10–12 years old) and Year 11 (16–18 years old) (Pocock and Clarke 2004,
and Pocock 2004 report more fully). We conducted 21 focus groups in two cities
(eight in Sydney, nine in Adelaide and four in rural South Australia) in late 2003.
The city schools were selected from a stratified sub-group of schools selected from
the two states, in high and low socio-economic groups based on their score on
the Index of Relative Socio-Economic Disadvantage (one of the five measures of
disadvantage published by the ABS). We call the higher-income schools ‘Leafy’
(northern Sydney) and ‘Comfort’ (southern Adelaide), the lower-income schools
‘Strive’ (western Sydney) and ‘Struggle’ (northern and western Adelaide), and
the country school ‘Country’. The average size of focus groups was four, and
most were separate-sex focus groups. All were age specific. The participants are
over-representative of males and higher-income areas, and children living with
employed parents. They are fairly representative by household type (breadwinner,
dual earner, single parent, blended).
We used the qualitative methodology of focus groups rather than interviews
because we were concerned that young people 10–17 years old may be shy in
64
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
one-on-one interviews with strangers. Further, we were interested in hearing
young people in conversation with each other about these issues, as well as hearing
their individual views in detail. Focus groups allowed in-depth consideration and
probing of issues and allowed new questions to surface. Each person was asked
their individual view on key issues, and the focus groups were attended by both
a note taker and facilitator and they were taped and transcribed. This allowed us
to identify each response by speaker, and to analyse their views by family type,
income and so on. The data were analysed by theme from the transcripts. Table 1
sets out the characteristics of interviewees.
Forty-three of the 93 lived-in households where children identify (without
prompting) long or unsocial hours as a significant issue affecting their parents (and
them). This may be an over-representation of children with a parent working long
or unsocial hours. However, it may not be, given that we are referring to both
long or unsocial hours, and given the rise in the proportion of Australian workers
who now work more than 45 hours a week (26 per cent in 2000 (Campbell 2002)).
Table 1 Focus group details
Location
Country Primary
Country High
Struggle Primary
Struggle High
Strive Primary
Strive High
Comfort Primary
Comfort High
Leafy Primary
Leafy High
Total
Higher income areas
Lower income areas
Country
Female only groups
Male only groups
Mixed sex groups
Females
Males
Family type
Two-parent, dual earner
Two-parent, single earner
Two-parent, no earner
Single-parent, earner
Single-parent, no earner
Total participants
No. groups
No. participants
Per cent
2
2
2
2
2
2
2
3
2
2
21
9
8
4
9
7
5
5
8
7
9
9
7
9
15
12
12
93
48
32
13
52
34
14
57
36
61
39
53
20
2
13
5
93
57
22
2
14
5
100
TIME, MONEY
AND
65
JOB SPILLOVER
Most children now live with two earners who each have a one in four chance of
working long hours. The study group includes a dozen children who lived with
at least one self-employed parent (ten of whom worked long or unsocial hours).
Children in the study described their parent’s hours with convincing precision,
even where they did not know their parent’s occupation. The impact of these
long or unsocial hours on the views of young people in this study is consistent and
strong.
The discussion below falls into three sections: firstly consideration of young
people’s preferences between time with parents and money from more parental
work; secondly their perceptions about how parents’ jobs affect them; and finally
some discussion of the implications for policy, action and theory.
TIME VERSUS MONEY: YOUNG PEOPLE’S PREFERENCES
In most households, parents’ jobs are valued by young people and understood
as a necessity. For some in lower-income areas in particular they are specifically
valued for the stability and security that they bring. Young people are well aware
of the need to earn and fully support it. Overall, however, more young people in
this study show a preference for more time with their parents, over more earnings
and less time with parents.
We asked participants: ‘If you could choose to have more time with your parents on the
one hand, or more money because they worked more, which would you choose?’ Almost half
of all 93 children would prefer more time with their parents, a fifth would prefer
more money, and many couldn’t decide between them, saying ‘both’ (see table 2).
Less than a tenth said that they liked things as they were, in contrast with the
results found by Lewis et al. (2001). It is interesting that only four young people
said they didn’t know. The preference for money was stronger in Sydney, and
weaker in Adelaide and the country. Those in lower income areas preferred more
money over more time. They were concerned about financial pressure and the
need to have money to pay bills and mortgages. However, even in lower income
areas, over a third of young people chose more time over more money, and only a
quarter unambiguously chose more money. More of them wanted to keep things
‘as is’ than in higher-income areas. Many in both low- and higher-income areas
wanted both, if only they could have them.
These results must be interpreted with caution. They are indicative of the views
of the 93 young people in our focus groups, rather than reliable indicators for
the larger population. However, they are suggestive of a significant preference
Table 2 Young people’s preferences for more time with parents or more money
Country
Lower income
Higher income
All
More time
More money
Both
As is
Don’t know
Total
92
38
44
48
8
25
23
20
0
22
21
19
0
16
4
8
0
0
8
4
100
100
100
100
66
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
for more time in the minds of many young people—even those located in less
financially comfortable areas.
Beneath this picture lies a much more complex story. Where children felt they
had enough time with parents, they often chose more money: Andre from Leafy
Primary put it like many others: ‘I’d probably choose more money because I can
see my parents basically whenever I want. My Mum [an architect] doesn’t do
that much work and my Dad [an artist] is at home’. A fifth wanted both: Charlie
at Leafy High wanted more time with his father (‘I don’t really see my Dad’),
and more money because he sees his parent’s employment in the entertainment
industry as unpredictable. Like many others, his ‘both’ answer reflects his view
that he sees enough of one parent (like most, his mother) and not enough of his
father. Brittany at the same school finds it hard to choose: she likes it ‘the way it
is. I would like to see my Dad more ‘cause I don’t see my Dad’.
In Adelaide households that are financially comfortable, and households in the
country, more young people preferred time with their parents, over more money.
In households feeling financial pressure, in Struggletown (northern or western
Adelaide) or Strivetown (western Sydney), views were pretty evenly split. Money
mattered more than in comfortable areas, but—despite financial pressure—a fair
number of young people wanted more time with their parents. Many mentioned
specific money pressures like meeting loan repayments and the bills. Some would
not choose: ‘I can’t really pick because we need the money, but I also need my
parents . . . so I don’t think I could choose’ (16 years old, Strive High).
There is surprisingly little difference in preferences between different household types, with around half of those in dual-earner couple, single-earner couple
and sole parent/earner households looking for more time, while about a fifth of
each would choose more money through more parental work. Although having
two parents in paid work might be expected to drive a preference for parental time
more than in single-earner couple households, it does not. Neither does living in
a sole parent/earner household.
Further conversation revealed that many young people in single-earner households miss their breadwinner parent and are looking for specific time with them.
They may see plenty of their mother, but this does not stop them for looking for
time with their father, especially a father with a demanding job. This may explain
the counter-intuitive result of little difference in preferences between children in
single-earner and dual-earner households.
The more striking differences in money/time preferences lie with location
and the demands of parents’ jobs, especially their hours. A sizeable difference
in time/money preferences exists between those who live in Sydney—whether in
low- or higher-income areas—and those who do not. Sydney-based young people
showed a stronger preference for more money from parents’ work. Three times
as many Sydney children wanted more money, than the proportion in Adelaide
and the country. Given that Sydney is a relatively high-cost city, with very high
housing costs, this is not surprising.
The second striking issue is the link to demanding jobs, especially those with
long or unsocial hours. We did not ask any direct questions about the impact of,
or views about, working hours; instead we asked about the ‘upsides/good’ and
TIME, MONEY
AND
JOB SPILLOVER
67
the ‘downsides/bad’ of mothers’ and then fathers’ jobs. Only a couple of the 43
children who mentioned that a parent worked long or unsocial hours did not
name this as a negative aspect of their parent’s job, and in at least one case this was
because they were ‘used to it’. Fifty-six per cent of those living with such a parent
wanted more time with parents (while only 19 per cent wanted more money).
This compares with 38 per cent of those who do not have a parent who is seen as
working long or unsocial hours who wanted more time.
A work/spend cycle (whereby more work drives more spending and more spending drives more work (Schor 1992)), was evident to many young people: ‘They
earn more money so they can buy you more things, but I don’t get to see them as
much if they’re working more’ as one young woman at Struggle High put it.
Young people in lower-income areas showed a high level of understanding
about how their parents’ work patterns fund necessities and loans. They wanted
the time, but they were understanding: ‘I’d prefer a bit more money because
we’ve got a lot of loans to pay off and we’re really tight for money recently, so
I’d prefer a bit more money . . . not too much that I don’t see them at all, but just
a bit more’ (Melinda, 16, Strive High). These children see their parents working
hard—and they feel that this is for them. They see more money as a means to
mitigate pressure on parents. Nearly all children in the country chose more time
with their parents over more time spent earning, even though each of these had
a mother at home or working part-time. They looked for more time from their
fathers in particular. Like young people in the city, they understood the need for
money. They were mostly sympathetic to their parents and their jobs.
The hyper-breadwinner
These discussions suggest that in single-earner couple households—the traditional ‘breadwinner’ home—a form of ‘hyper-breadwinner’ is evident: that is,
a breadwinner who is absent for long periods as he (they are mostly fathers)
takes on the whole task of household earning. For many breadwinners in this
study, this means overtime, longer hours at work, working on the phone or laptop after hours, or travelling long distances to work intensively or for extended
periods.
Having a parent who works long or unsocial hours drove a preference for time
over money. In some cases, children were clear that they wanted time with one
parent—always the one working longest or seen least. As Ali at Leafy Primary in
Sydney described his Dad’s situation: ‘I think it’s not good because I don’t see him
in the morning at all. He has to leave really early before I wake up and he comes
home really late so its annoying ‘cause I never really see him . . . I never really get
to interact with him. So it’s kind of lost time with him’ (Ali, 11, Leafy Primary).
The good thing about his father’s job is that he ‘gets lots of money’.
All four children of truck drivers in the study shared similar accounts about
their father’s hours. Eddie’s father worked six days a week driving a truck and
moving furniture but was now ‘really sick . . . Underneath his lungs it is all crappy.
And his lungs are just inflating’. He says he feels bad about his Dad’s illness
‘because I can’t play with him and, yeah, he can’t buy me stuff without the money’
(Eddie, 12, Struggle Primary). However, the issue is not confined to blue-collar
68
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
workers: children of managers, small business people and professionals shared this
perspective.
Parent-specific time hunger
At Comfort Primary, with relatively secure higher incomes, most children chose
more time with their parents over more money, especially more time with a parent
who is absent more. Nicky was an exception: she would choose more money, reflecting the fact that she lives with her single mother who works part-time while
her dad lives in another city. The general preference for time at Comfort Primary illustrates how breadwinner household structures—with a time-rich parent
at home, usually the mother—do not eliminate parental time hunger for children. Instead, they show a ‘parent-specific’ time hunger. In the hyper-breadwinner
household, where the breadwinner is working long or unsocial hours, that time
hunger is pronounced, focused upon the absent parent, most often their father.
For example, Bob, whose father runs a sporting range and sometimes works long
hours while his Mum works part-time, wants to spend more time with his parents
‘especially my Dad because he hasn’t spent a lot of time at home’, and would chose
that over more money.
A full-time job does not mean a lack of time with young people for some. Ellie,
for example, who lives with her mother who works full-time, says that her mother
‘really enjoys her job . . . but I still see her, and my Mum and I are pretty close and
we still can spend heaps of time together. So it’s good. She really likes it . . . She
says its really satisfying’ (Ellie, 16, Leafy High).
Long hours and money/time
Hours are an important determinant of children’s perspectives about their parents’
jobs. Long hours are consistently associated with negative views of parental work
patterns. A number of young people defined their own futures against their parent’s
working lives. They plan to make sure they spend enough time with their own
children, or want to avoid demanding jobs and make sure they have weekends off.
They name the kind of ‘work/eat/sleep’ cycle that long hours workers also name
for themselves (Pocock, van Wanrooy, Strazzari & Bridge 2001):
Dad earns money and gets out—but he hardly spends any time with you. He comes
home, eats tea and watches TV. We’d like time . . . He feels bad because he can’t spend
time with the family. (Kelsey, 12, Struggle Primary).
When it comes to choosing between time and money, Bonnie, daughter of
a country truck driver, did not hesitate: ‘more time . . . even if that means you
mightn’t get everything that you wanted’. Brittany, in Sydney at Leafy Primary
agreed: she wants more time. Some young people see a link between more time at
work and poorer quality relationships. Several were very conscious of long hours
and their impact on household relationships, seeing the need to balance money
with time. As Adam, son of a cabinet maker and aged carer put it:
It’s good to get money coming in and probably it’s good to work as hard as you can
when you’re younger so when you’re older you can retire with some money. But there
should probably be a limit to so much before your relationships with other people
start to strain because you are never there. (Adam, 16, Country High).
TIME, MONEY
AND
JOB SPILLOVER
69
Time with fathers
Young men do not stop wanting time with their fathers as they grow up based on
the consistent comments of 16–18 year old boys. While not all young men in our
study felt the same (indeed some spoke of feeling distant from their fathers because
of their absence over time), others were straight-forward about their sadness:
I reckon my Mum [who works part-time] is pretty much fine the way she is. Just leave
her like that. But I suppose a little more time with my Dad would be good, seeing
as I usually get to see him for about five minutes in the morning after I get up and
he usually gets home around the time I’m doing homework so I only get to see him
around tea time, and onwards at night. So not too much going on there. (Kyle, 16,
Country High).
Kevin (17, Country High), whose father worked very long hours in the country,
would like more time playing backyard cricket with his father, more ‘hang around
time’: ‘I wouldn’t mind if I just sat in the next room, but it would be better if he
wasn’t doing as much work’. Adam—from the same region—agreed. His father
works in an industry with peaks and troughs. These young men whose fathers
worked long or unsocial hours, missed time with their fathers, and were very alert
to paternal stress. It is not surprising that they do not miss their mothers, given
that most were at home or held part-time jobs, but their clear preference for time
with the fathers, especially with unstressed fathers, suggests that time with one
parent does not easily substitute for time with another: father-time has its own
function and is clearly desired by young men (and young women).
For many children, time is a key first requirement of a ‘good dad’. Kelsey, 12,
at Struggle Primary describes a ‘good Dad’ as someone who ‘spends time with
you’. Her friend Kelly agrees: ‘someone who mucks around with you’. For Zoe,
whose father left her life some time ago a ‘good dad’ is ‘some one who doesn’t just
come and leave’. In the same school, Harry whose father left when he was very
young because ‘he didn’t like kids’ agrees: ‘A good Dad would be if he was part of
my family again because it would be better if he still had a job and came back. If
he came back and lived with my Mum and still got paid $700, it would be great’
(Harry, 11, Struggle Primary).
When discussing a ‘good mother’, young women stressed the importance of
someone to talk to about issues of importance, particularly social relationships.
While some young men mentioned this, they stress the importance of time with
their fathers more. Young men were more likely to equate a good mother with a
good cook, and value their help with homework.
A number of young people were positive about their mothers being at home
after school and liked time with mothers at home. They enjoyed ‘hanging time’,
weekends, and time before and after school. Many also valued a clean house, and
young men in particular mention cooking and plenty of food in the house as a
benefit associated with a mother working at home. Young people tended to adopt
a different framework to the presence of a father at home: rather than carers they
were viewed as ‘workers in waiting’. In the four households with fathers at home,
they were viewed as ‘out of work’ (sick, retired, unemployed or retrenched) rather
than valued home-based carers.
70
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
Overwhelmingly young people spoke of the importance of their parents being
there for special events, such as sports, choir, public performances, when they
received an award, or for special events like birthdays. This confirms other findings
(Galinsky 1999; Lewis et al. 2001). Many sadly remember key events that their
parents have missed. They like to share their successes and public events, and they
also want parents around to help solve problems when they occur. Several didn’t
think that ‘make up’ time later really compensated: they wanted their parents to
witness their achievements and activities.
Many young men anticipate spending a considerable amount of time with their
own children—taking care not to miss weekends and evenings, and some will
parent very differently to their fathers, in terms of time and work allocations:
I’d love to be able to take a few years off and not work and spend it with my kids
and with my wife and just starting a family and being there for my kids for the first
few years of their life. But then I’d definitely go back to work like when they start
school, but I’d make sure I was there for them in the evenings, help them with their
homework and on the weekends do sporting activities and all that. And when they
are older and think I’m just boring and not cool, let them do their own thing, but I’ll
still try and sneak in some quality time. (Smithie, 17, Leafy High).
JOB SPILLOVER: HOW PARENTS’ JOBS AFFECT YOUNG PEOPLE
Alongside the issue of work and time trades, the issues of spillover (from work
to home and from home to work) have been widely discussed in the work and
family literature (Pleck 1977; Hertz and Marshall 2001). This study shows that,
from the children’s perspectives, not all jobs are the same. And not all children’s
perspectives are the same, even about jobs that might appear similar. The effects
of work on children are highly context-specific and diverse. Nonetheless, some
similarities are striking and reinforce existing research.
A great deal of public attention in Australia has focused on whether to work
when children are young—especially whether mothers should work. For young
people in this study, this is the wrong question. This reinforces Galinsky’s (1999)
US results, and those of Lewis et al. (2001) in Australia. Young people understand
that many of their parents need to work and they do not see parents’ work as
intrinsically bad or good. The important issue for the young people in this study
is not whether parents go to work, but the state in which they come home. Young
people comment upon the nature and effects of their parents’ jobs, rather than
about whether they work or not. They talk about the effects of their parents’
jobs on them and their brothers and sisters—rather than seeing a problem with
parents’ jobs per se. Many can see positive outcomes for their parents from their
paid jobs—outcomes that flow to children through material comfort and, beyond
this, to a happier parent and household. This is a consistent finding, evident in
most focus groups, age groups, and in all household types.
Young people’s alertness to how work affects their parents extends to domestic
work. Many young people with mothers doing domestic work at home, saw positives both for themselves and for their mothers. However, being at home was not
without spillover effects for mothers and children, confirming results that Lewis
TIME, MONEY
AND
JOB SPILLOVER
71
et al. (2001) report . These include physical and emotional effects. Fred sees that
‘my little brother was really hard work for my mum’ when she looked after him fulltime (Fred, 11, Comfort Primary). Others perceived tiredness, physical injuries,
social isolation or a heavy domestic load that mothers often carried alone. Young
people are also aware of the under-valuation of their mother’s work at home,
and several mention their mother’s depression. Other children commented upon
mothers who were happy doing domestic work. Once again, however, it is not the
job itself that causes noticeable spillover, but its fit with parental preferences.
With respect to paid work there are a clear spillovers onto children. These
include both positive and negative effects. Most obvious among the positive effects
is income. Alongside this, many young people see benefits for them arising from
their parents’ work-related skills. In addition, many children enjoy the things
parents bring home: pens, equipment, supplies. Young people also see that their
parents have fun and gain a sense of worth and contribution through their jobs.
Others feel that parents enjoy making friends, having laughs and social connection
through their jobs. As one put it ‘She tells funny stories from work, the people at
work. My mum likes her job’ (Susie, 16, Comfort High). Others speak positively
when their parents’ work is flexible, when they work hours that mean that they
are home at the same time as children, and when they are able to participate in a
meaningful way in their parents’ work.
On the negative side, many forms of spillover are obvious and common. The
daughter of two full-time working parents who said ‘My parents try not to bring
the work life home’ was unusual. She appreciated their containment: ‘I don’t really
need their stress as well as mine’ (Jade, 16, Comfort High). Most young people
could easily tell what kind of day their parents had had when they walk in the door.
Jobs ‘colour’ parental moods—to use Nasman’s expression (2003), affecting their
‘state and physical condition’. However, these effects go beyond mere ‘colouring’
of parental mood. They are directly transmitted to children and to others in the
household. Children not only observe their parent’s ‘colour’, but many are affected
by it themselves and feel its effects through yelling, arguments and household
tension. Young people are especially alert to these negative spillovers: it is these
they notice and respond too—often withdrawing from their parent to cope. In line
with studies in other countries, they notice when their parents are upset (Galinsky
1999; McKee et al. 2003).
Negative aspects of parent’s jobs include physical injury, emotional or mental injury, bad moods, stress, tiredness, sadness, uncertainty and fear. Negative
spillover was perceived in all socio-economic groups. However, it sometimes
takes different forms. Young people in lower-income areas spoke about the physical impact of their fathers’ and mothers’ jobs on their parents’ bodies—and they
worried about this. Changes in working hours or shifts sometimes created significant change for young people. However, more hours did not necessarily mean
more negative spillover: once again it was the nature of the job, and its fit with
parental preferences, which seemed more important. Very specific effects flowed
from bad events at work, to parents and to children. These effects were often
associated with the ongoing nature of their parent’s jobs: their hours and physical and emotional demands. However, there were other important spillovers
72
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
arising from more occasional or episodic events—like injuries, job loss, redundancy, demotion or reassignment.
The impact of long or unsocial hours: ‘he’s not violent, he just yells . . .’
Long or unsocial hours were frequently mentioned as a first effect of parents’
work, with almost universal negative effects. Like others, Ali noticed his father’s
stress and anger after a long shift: ‘sometime when he comes back in the afternoon
he gets angry at simple things cause he’s stressed’. Andre’s father, an artist, is also
‘stressed and tired’ when he has been working late on a painting.
Children identified direct negative effects of long hours upon their parents. The
word ‘grumpy’ was frequently used to describe parents—just as it is often used
by long hours workers (and their partners) to describe themselves (Pocock et al.
2001). Kyle (16, Country High) said of his farmer father ‘he’s got to work long
hours, go and do things if he finds something wrong in one of the paddocks just as
he is about to knock off. He has to go and fix it or whatever.’ As Kyle described it
‘You do something wrong and he gets up you’. ‘Yeah’ agreed Robert: ‘he usually
comes home in a pretty foul mood and you’ve got to tread lightly around him’
(17, Country High). James (15, Struggle High), whose dad drives a truck for long
hours, couldn’t see anything good about his father’s job. He described how its
effects spilled over into non-work time.
In a higher socio-economic setting, Mark’s father works two jobs. He has given
up a third which has made him ‘happier’. One of his jobs involves night shifts:
He’s incredibly tired, all the time, because he is just constantly going and he gets
aggravated. He’s not violent, he just yells and he can yell really loud. He doesn’t do
it all that often considering the stress he is under. (Mark, 17, Comfort High).
Like other young people, Mark is very understanding about his father’s work
and its effects. He sees his dad as ‘good fun’ and—for all his occasional yelling—
loves spending time with him. He keeps out of his father’s way when he is ‘incredibly tired’.
Young people respond to negative spillover in several ways: most commonly
they physically withdraw from a grumpy or shouting parent; they may turn to the
other parent and distance themselves from the absent or grumpy parent. Some
take steps to look after their tired parent. Others just worry about their parents,
while others dislike—or hate—the contamination of their emotional states by
their parents’ jobs.
Several children commented upon the fact that their parents’ long hours are
under-paid in their view: ‘He doesn’t get paid very well for the hours he puts in’
(Abraham, 16, Comfort High). Many specifically mentioned the loss of weekend
time with their parents, including time when all the family is together. For example, Sarah (11, Strive Primary) would like it if her father had a weekend day off
rather than Wednesday because she really enjoys family time on weekends. Others want their parents around on weekends for more utilitarian reasons: because
otherwise ‘we couldn’t get anywhere’. Several did not like their parents working
at night: ‘I wouldn’t like it if Mum worked at night’. ‘No I wouldn’t either.’ (16
year old girls at Struggle High). Shift work is also seen as a source of parental
TIME, MONEY
AND
JOB SPILLOVER
73
tiredness. These effects cross most boundaries: city and country, high and low
incomes, boys and girls, and they affected workers in blue, white and pink-collar
jobs, reaching into many occupations and industries. Both small business owners
and operators, and wage earners were affected in the perception of young people.
Where fathers worked long hours there was a considerable sense of loss, and
in some cases hurt and anger, particularly among boys and young men. Young
people spoke of how their parent’s anger and frustration with work, spilled over
onto them. With parents on a short fuse, young people knew when to ‘keep out
of the way’. They valued some explanation about their parents’ moods. They did
not like their parent being ‘shitty’, and it spilling over into interrogation of their
children, but an explanation for these spillovers mitigated their effect.
RE-FRAMING THE WORK AND FAMILY DEBATE: IMPLICATIONS FOR
POLICY, ACTION AND THEORY
This analysis affirms research that tells us that it is not whether parents go to work
or not, but the state in which they come home, that really affects children. This ‘state’
reflects objective characteristics of jobs (like hours and intensity) as well as the
extent to which parents’ preferences match their jobs. The debate about whether
to work or not needs to be re-framed in Australia (as Galinsky (1999) has argued
in the USA). In particular, less attention should be paid to the issue of whether
mothers should hold paid jobs, and more to the work of fathers. Most importantly,
a mother at home does not make up for a father who is absent a great deal.
For many children, their mothers’ and fathers’ jobs are associated with positive
spillover. Young people value the money and security that parents’ paid work
brings. Beyond this, they can see that many parents enjoy their jobs, or aspects
of them. They love their stories, sense their social connection through work, and
see that their parents have fun and feel good about doing a job well.
But negative spillover is also widespread. It is especially associated with disappointed parental preferences (for example, a parent who doesn’t want to work
part-time but has to, or a parent who cannot work the shifts they want). It is
also associated with some specific job characteristics: risk of physical harm, job
insecurity, work overload, or long or unsocial hours. These often send a parent
home from work angry or upset. Their moods are obvious to young people—from
physical, verbal and behavioural clues. Children ‘read’ their parents easily.
Job spillover, however, does not end with the parents. Both good and bad job
spillover is directly transmitted to children. It affects them—their moods, concerns
and behaviour. The fit between job preferences and job reality seems critical to
the level and nature (good or bad) of spillover. For example, a mother doing
domestic work who is happy with that, who is not physically exhausted or sick,
will have much less spillover than a mother who does such work but is resentful,
bored or feels isolated or under-valued. A parent who loves his job and works long
hours may bring home less negative spillover than one who works part-time but
hates his job. That said, long or unsocial hours are consistently associated with
perceived negative spillover and the majority of parents who work them appear
to children to be sometimes grumpy, tired and stressed.
74
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
How should parents, workplaces and public policy respond to these findings?
At the individual level, it is clear that ‘working mothers’ are not, in themselves, a
problem in the view of children. Many see benefits arising from parents’ jobs. The
nature of work itself, and its impact upon parents, emerge as the more important
considerations. Where jobs send parents home worried, sick, hurt, tired or bad
tempered, children notice and are negatively affected. Such spillovers are not
confined to paid jobs: they also arise from domestic work if it fits poorly with
parental preferences, or leaves parents isolated, depressed, tired or under-valued.
Too much worry about having a job at all—especially with respect to working
mothers—has diverted attention from more significant aspects of work that young
people notice—like how absent working fathers or their long hours affect children,
and how work affects parental bodies, tempers and emotions. It seems that negative
spillover is reduced when parents do work that they like, for hours that suit their
preferences, while minimising physical injury and long or unsocial hours that
leave them bad tempered or tired.
Of course, this is at once both obvious and more easily said than done. For many
parents their preferences do not—alone or even in significant part—determine
their labour market outcomes; these are instead overshadowed by larger structural, systemic and institutional forces. Changing these requires new public and
workplace policies and practices. The finding that long or unsocial hours affect
children negatively, in their perception, reinforces the well-documented concerns
of adults (Pocock et al. 2001; Peetz et al. 2004). This strengthens the argument
for policies that contain hours of work to social and reasonable levels.
Other policies and practices that assist parents to find a good fit between how
they want to parent and how they work, are also important to the perceptions of
children. These include measures that increase worker say over the allocation of
working time and its organisation (or greater ‘worker time sovereignty’), including
over hours of work, start and finish times, days off, holidays, shifts and rosters.
Greater worker capacity to move into and out of full-time and part-time work,
without risk to employment or earnings, and into and out of paid work itself
will also help by protecting home and children from fears about income and job
security. This includes a policy regime of tax and social security arrangements that
does not penalise people for changing jobs or for moving into and out of the labour
market to get the best possible fit of preferences with circumstances. Predictable
and secure job arrangements that do not destabilise and frustrate preferences, with
negative consequences for the security of children, are also likely to be significant.
An improved suite of paid and unpaid leave rights for working parents will also
assist parents to better meet the changing demands of their dependents so that
they can better parent around their work to meet both their own preferences and
those of their children.
In theoretical terms, this analysis confirms the close relationship of work to
household life: they cannot be understood as separate social or industrial spheres
(Nasman 2003). Of course, this close relationship is nothing new: it has characterised the work and households of adults and their children for centuries, with
children long involved in forms of labour, and often affected in various ways by
the waged and unwaged work of parents. However, this relationship is changing
TIME, MONEY
AND
JOB SPILLOVER
75
at present, as more carers (especially mothers) enter paid work, and experience
personal guilt and very variable institutional supports (Pocock 2003). This analysis of the perceptions of young people confirms that theoretical treatment of work
and family as separate social and policy spheres is unhelpful. They are intimately
connected. This connection requires analysis of work that recognises that paid
work is located within a larger ‘total labour’, and within diverse households and
communities. It also requires policy action that takes account of these connections.
Increasingly, producing bodies are also reproducing bodies, which affects how we
must think about workers, workplaces, dependency, childhood and households.
In the perception of these young people, the spheres of work and households
are not consistently in opposition to one another, the one draining the other.
Work does not always have negative effects upon young people: positive spillover
from work for some is also evident, encouraging us to consider ways and means
to increase positive spillover, and to reconsider the conventional binary and oppositional theorisation of ‘work’ and ‘home’.
ACKNOWLEDGEMENTS
This research was funded by The Australia Institute and the Australian Research Council through
the fellowship project ‘Theoretical and policy implications of changing work/life patterns and
preferences of Australian women, men and children, household and communities’.
REFERENCES
Campbell I (2002) Extended working hours in Australia. Labour & Industry 13(1), 73–90.
Galinsky E (1999) Ask the Children: What America’s Children Really Think About Working Parents.
New York: William Morrow and Company.
Hertz R, Marshall NL (2001) Working Families. The Transformation of the American Home. Berkeley
CA: University of California Press.
HILDA (2001) Annual Report. Melbourne: Melbourne Institute.
Jensen A, McKee L (Eds) (2003) Children and the Changing Family; Between Transforamtion and
Negotiation. London: RoutledgeFalmer.
Lewis V, Tudball J, Hand K (2001) Family and work: the family’s perspective. Family Matters, No
59, Winter, 2001. Melbourne: Australian Institute of Family Studies.
McKee L, Mauthner N, Galilee J (2003) Children’s perspectives on middle-class work-family arrangements. In Jensen An–Margritt, McKee Lorna (Eds), Children and the Changing Family;
Between Transformation and Negotiation. London: RoutledgeFalmer, pp. 27–45.
Megalogenis G (2003) Fault Lines. Race, Work and the Politics of Changing Australia. Scribe,
Melbourne.
Nasman E (2003) Employed or unemployed parents: A child perspective. In Jensen An–Margritt,
McKee Lorna (Eds.), Children and the Changing Family; Between Transformation and Negotiation.
London: RoutledgeFalmer, pp. 46–60.
Peetz D, Fox A, Townsend K, Allan C (2004) The big squeeze: Domestic dimensions of excessive work, time and pressure. New Economies: New Industrial Relations, Proceedings of the 18th
AIRAANZ Conference, 2–6 February 2004. pp. 381–390.
Pleck JH (1977) The work/family system. Social Problems, No. 24, 417–427.
Pocock B (2004) Work and Family Futures. How Young Australians Plan to Work and Care. Canberra:
The Australia Institute.
Pocock B (2003) The Work/Life Collision. What work is doing to Australians and what to do about it.
Sydney: Federation Press.
Pocock B, Clarke J (2004) Can’t buy me love? What young Australians views on parental work, time,
guilt and their own consumption. Canberra: The Australia Institute.
76
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
Pocock B et al (2001a) Fifty Families: What Unreasonable Hours are Doing to Australians, Their Families
and Their Communities. Melbourne, ACTU.
Pusey M (2003) The Experience of Middle Australia, The Dark Side of Economic Reform. Cambridge,
Cambridge University Press.
Schor J (1992) The Overworked American: The Unexpected Decline of Leisure. New York: Basic Books.
Summers A (2003) The End of Equality. Sydney: Random House.
Tanner L (2003) Crowded Lives. Sydney: Pluto Press.
Watson I, Buchanan J, Campbell I, Briggs C (2003) Fragmented Futures. Sydney: Federation Press.
PARENTS’ JOBS AFFECT YOUNG PEOPLE
BARBARA POCOCK AND JANE CLARKE∗
T
his paper examines the perspectives of young people about their parents’ paid and
unpaid work, their preferences for time or money through more parental work, and
their views about how their parents’ jobs affect them. It analyses qualitative empirical data
collected in Australia in late 2003, by means of focus groups among 10–12 and 16–18year-old males and females in urban and rural locations in two Australian states, in both
high and low socioeconomic areas. It finds that more Australian children are looking for
more time from parents than more money from more parental work, though this varies
by income level, location and parental hours. This preference for ‘time over more money’
is consistent in single- and dual-earner couple households as well as sole parent/earner
households. Children are acute observers of parents and their jobs. Both positive and
negative spillovers are widely observed. Negative spillovers from long or unsocial hours
are especially marked, reinforcing other findings in support of policy interventions to
contain long or unsocial hours.
INTRODUCTION
Changes in patterns of paid and unpaid work, workplaces, and household shape
have been the subject of much recent analysis in Australia (HILDA 2001;
Campbell 2002; Megalogenis 2003; Pocock 2003; Pusey 2003; Summers 2003;
Tanner 2003; Watson, Buchanan, Campbell & Briggs 2003). They are driving
a lively policy and political interest in the work and family ‘collision’ and government responses to it, though action is slower to follow. The perspective of
children is missing from these accounts, yet is relevant to industrial, workplace
and household perspectives. Parents, in particular, act on certain assumptions
about children’s welfare as they determine household patterns of participation in
paid and unpaid work.
Whereas there has been some international research on the specific issue of
parental work patterns and children’s views of them, there has been very little in
Australia with the exception of Lewis, Tudball and Hand (2001). Galinsky (1999)
argues, based on her large US study, that the ‘work/family’ debate has been misframed with too much focus on whether working mothers in particular are ‘bad’
for children, rather than how work affects parents, and through them, children.
In the USA, the real question about parental work is ‘how parent’s work’, and
how attentive or focused they are able to be towards their children when they
are with them. Galinsky found that older children longed for more parental time,
especially with their fathers. These older children looked for ‘hang around time’
∗ School
of Social Sciences, University of Adelaide, Adelaide, SA, 5005. Australia.
Email: [email protected]
THE JOURNAL OF INDUSTRIAL RELATIONS, VOL. 47, NO. 1, MARCH 2005, 62--76
TIME, MONEY
AND
JOB SPILLOVER
63
more than younger children. She argues that the nature of parents’ jobs is very important to children who are very alert to parental moods. This result is confirmed
by Nasman’s study of Swedish children about parental work which found that
work affects parents’ ‘state and physical conditions’. This ‘colouring’ spills over
to family life, overriding the idea of work and family as separate social spheres’
(2003: 51). This spillover was higher for parents with irregular or long hours and
for single mothers, and manifested itself most obviously through parental fatigue.
In a Scottish study McKee, Mauthner and Galilee (2003) found that children were
profoundly aware of their parents’ work and its effects: they could ‘competently
assess how work made their parents feel’ (2003: 39) and many wanted to avoid the
stress arising from external control of work in their own future lives.
In Australia, Lewis, Tudball and Hand (2001) undertook interviews with a nonrandom group of parents and children from 47 Melbourne families (71 children
over eight years old). The majority of children in the study felt that their parents
worked ‘about the right amount of time’ and that—in accord with Galinsky’s
work—‘it is not whether and how much parents work, but how they work and
how they parent, that matters’ (2001: 23). Responses were divided ‘roughly evenly
between those saying that they wished their parents spent more time with them
and those who said their parents currently spent enough time with them’ (2001:
24). Only two parents in Lewis et al.’s study worked more than 50 hours a week.
This Australian and international literature mostly addresses the question of
spillover from parental work onto children. This paper addresses several questions. First, how do young people perceive the time/work patterns and trade-offs
in their households? Second, how do young Australians perceive spillover from
their parents’ jobs? Third, do these perceptions vary in relation to household
income, locations, household types, and different patterns of parental working
hours? Do young people with a parent at home also experience spillover from
that domestic form of work? Finally, how do these experiences and perceptions
affect their own plans for the future?
The analysis considers the perceptions of a group of young Australians at school
in Year 6 (10–12 years old) and Year 11 (16–18 years old) (Pocock and Clarke 2004,
and Pocock 2004 report more fully). We conducted 21 focus groups in two cities
(eight in Sydney, nine in Adelaide and four in rural South Australia) in late 2003.
The city schools were selected from a stratified sub-group of schools selected from
the two states, in high and low socio-economic groups based on their score on
the Index of Relative Socio-Economic Disadvantage (one of the five measures of
disadvantage published by the ABS). We call the higher-income schools ‘Leafy’
(northern Sydney) and ‘Comfort’ (southern Adelaide), the lower-income schools
‘Strive’ (western Sydney) and ‘Struggle’ (northern and western Adelaide), and
the country school ‘Country’. The average size of focus groups was four, and
most were separate-sex focus groups. All were age specific. The participants are
over-representative of males and higher-income areas, and children living with
employed parents. They are fairly representative by household type (breadwinner,
dual earner, single parent, blended).
We used the qualitative methodology of focus groups rather than interviews
because we were concerned that young people 10–17 years old may be shy in
64
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
one-on-one interviews with strangers. Further, we were interested in hearing
young people in conversation with each other about these issues, as well as hearing
their individual views in detail. Focus groups allowed in-depth consideration and
probing of issues and allowed new questions to surface. Each person was asked
their individual view on key issues, and the focus groups were attended by both
a note taker and facilitator and they were taped and transcribed. This allowed us
to identify each response by speaker, and to analyse their views by family type,
income and so on. The data were analysed by theme from the transcripts. Table 1
sets out the characteristics of interviewees.
Forty-three of the 93 lived-in households where children identify (without
prompting) long or unsocial hours as a significant issue affecting their parents (and
them). This may be an over-representation of children with a parent working long
or unsocial hours. However, it may not be, given that we are referring to both
long or unsocial hours, and given the rise in the proportion of Australian workers
who now work more than 45 hours a week (26 per cent in 2000 (Campbell 2002)).
Table 1 Focus group details
Location
Country Primary
Country High
Struggle Primary
Struggle High
Strive Primary
Strive High
Comfort Primary
Comfort High
Leafy Primary
Leafy High
Total
Higher income areas
Lower income areas
Country
Female only groups
Male only groups
Mixed sex groups
Females
Males
Family type
Two-parent, dual earner
Two-parent, single earner
Two-parent, no earner
Single-parent, earner
Single-parent, no earner
Total participants
No. groups
No. participants
Per cent
2
2
2
2
2
2
2
3
2
2
21
9
8
4
9
7
5
5
8
7
9
9
7
9
15
12
12
93
48
32
13
52
34
14
57
36
61
39
53
20
2
13
5
93
57
22
2
14
5
100
TIME, MONEY
AND
65
JOB SPILLOVER
Most children now live with two earners who each have a one in four chance of
working long hours. The study group includes a dozen children who lived with
at least one self-employed parent (ten of whom worked long or unsocial hours).
Children in the study described their parent’s hours with convincing precision,
even where they did not know their parent’s occupation. The impact of these
long or unsocial hours on the views of young people in this study is consistent and
strong.
The discussion below falls into three sections: firstly consideration of young
people’s preferences between time with parents and money from more parental
work; secondly their perceptions about how parents’ jobs affect them; and finally
some discussion of the implications for policy, action and theory.
TIME VERSUS MONEY: YOUNG PEOPLE’S PREFERENCES
In most households, parents’ jobs are valued by young people and understood
as a necessity. For some in lower-income areas in particular they are specifically
valued for the stability and security that they bring. Young people are well aware
of the need to earn and fully support it. Overall, however, more young people in
this study show a preference for more time with their parents, over more earnings
and less time with parents.
We asked participants: ‘If you could choose to have more time with your parents on the
one hand, or more money because they worked more, which would you choose?’ Almost half
of all 93 children would prefer more time with their parents, a fifth would prefer
more money, and many couldn’t decide between them, saying ‘both’ (see table 2).
Less than a tenth said that they liked things as they were, in contrast with the
results found by Lewis et al. (2001). It is interesting that only four young people
said they didn’t know. The preference for money was stronger in Sydney, and
weaker in Adelaide and the country. Those in lower income areas preferred more
money over more time. They were concerned about financial pressure and the
need to have money to pay bills and mortgages. However, even in lower income
areas, over a third of young people chose more time over more money, and only a
quarter unambiguously chose more money. More of them wanted to keep things
‘as is’ than in higher-income areas. Many in both low- and higher-income areas
wanted both, if only they could have them.
These results must be interpreted with caution. They are indicative of the views
of the 93 young people in our focus groups, rather than reliable indicators for
the larger population. However, they are suggestive of a significant preference
Table 2 Young people’s preferences for more time with parents or more money
Country
Lower income
Higher income
All
More time
More money
Both
As is
Don’t know
Total
92
38
44
48
8
25
23
20
0
22
21
19
0
16
4
8
0
0
8
4
100
100
100
100
66
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
for more time in the minds of many young people—even those located in less
financially comfortable areas.
Beneath this picture lies a much more complex story. Where children felt they
had enough time with parents, they often chose more money: Andre from Leafy
Primary put it like many others: ‘I’d probably choose more money because I can
see my parents basically whenever I want. My Mum [an architect] doesn’t do
that much work and my Dad [an artist] is at home’. A fifth wanted both: Charlie
at Leafy High wanted more time with his father (‘I don’t really see my Dad’),
and more money because he sees his parent’s employment in the entertainment
industry as unpredictable. Like many others, his ‘both’ answer reflects his view
that he sees enough of one parent (like most, his mother) and not enough of his
father. Brittany at the same school finds it hard to choose: she likes it ‘the way it
is. I would like to see my Dad more ‘cause I don’t see my Dad’.
In Adelaide households that are financially comfortable, and households in the
country, more young people preferred time with their parents, over more money.
In households feeling financial pressure, in Struggletown (northern or western
Adelaide) or Strivetown (western Sydney), views were pretty evenly split. Money
mattered more than in comfortable areas, but—despite financial pressure—a fair
number of young people wanted more time with their parents. Many mentioned
specific money pressures like meeting loan repayments and the bills. Some would
not choose: ‘I can’t really pick because we need the money, but I also need my
parents . . . so I don’t think I could choose’ (16 years old, Strive High).
There is surprisingly little difference in preferences between different household types, with around half of those in dual-earner couple, single-earner couple
and sole parent/earner households looking for more time, while about a fifth of
each would choose more money through more parental work. Although having
two parents in paid work might be expected to drive a preference for parental time
more than in single-earner couple households, it does not. Neither does living in
a sole parent/earner household.
Further conversation revealed that many young people in single-earner households miss their breadwinner parent and are looking for specific time with them.
They may see plenty of their mother, but this does not stop them for looking for
time with their father, especially a father with a demanding job. This may explain
the counter-intuitive result of little difference in preferences between children in
single-earner and dual-earner households.
The more striking differences in money/time preferences lie with location
and the demands of parents’ jobs, especially their hours. A sizeable difference
in time/money preferences exists between those who live in Sydney—whether in
low- or higher-income areas—and those who do not. Sydney-based young people
showed a stronger preference for more money from parents’ work. Three times
as many Sydney children wanted more money, than the proportion in Adelaide
and the country. Given that Sydney is a relatively high-cost city, with very high
housing costs, this is not surprising.
The second striking issue is the link to demanding jobs, especially those with
long or unsocial hours. We did not ask any direct questions about the impact of,
or views about, working hours; instead we asked about the ‘upsides/good’ and
TIME, MONEY
AND
JOB SPILLOVER
67
the ‘downsides/bad’ of mothers’ and then fathers’ jobs. Only a couple of the 43
children who mentioned that a parent worked long or unsocial hours did not
name this as a negative aspect of their parent’s job, and in at least one case this was
because they were ‘used to it’. Fifty-six per cent of those living with such a parent
wanted more time with parents (while only 19 per cent wanted more money).
This compares with 38 per cent of those who do not have a parent who is seen as
working long or unsocial hours who wanted more time.
A work/spend cycle (whereby more work drives more spending and more spending drives more work (Schor 1992)), was evident to many young people: ‘They
earn more money so they can buy you more things, but I don’t get to see them as
much if they’re working more’ as one young woman at Struggle High put it.
Young people in lower-income areas showed a high level of understanding
about how their parents’ work patterns fund necessities and loans. They wanted
the time, but they were understanding: ‘I’d prefer a bit more money because
we’ve got a lot of loans to pay off and we’re really tight for money recently, so
I’d prefer a bit more money . . . not too much that I don’t see them at all, but just
a bit more’ (Melinda, 16, Strive High). These children see their parents working
hard—and they feel that this is for them. They see more money as a means to
mitigate pressure on parents. Nearly all children in the country chose more time
with their parents over more time spent earning, even though each of these had
a mother at home or working part-time. They looked for more time from their
fathers in particular. Like young people in the city, they understood the need for
money. They were mostly sympathetic to their parents and their jobs.
The hyper-breadwinner
These discussions suggest that in single-earner couple households—the traditional ‘breadwinner’ home—a form of ‘hyper-breadwinner’ is evident: that is,
a breadwinner who is absent for long periods as he (they are mostly fathers)
takes on the whole task of household earning. For many breadwinners in this
study, this means overtime, longer hours at work, working on the phone or laptop after hours, or travelling long distances to work intensively or for extended
periods.
Having a parent who works long or unsocial hours drove a preference for time
over money. In some cases, children were clear that they wanted time with one
parent—always the one working longest or seen least. As Ali at Leafy Primary in
Sydney described his Dad’s situation: ‘I think it’s not good because I don’t see him
in the morning at all. He has to leave really early before I wake up and he comes
home really late so its annoying ‘cause I never really see him . . . I never really get
to interact with him. So it’s kind of lost time with him’ (Ali, 11, Leafy Primary).
The good thing about his father’s job is that he ‘gets lots of money’.
All four children of truck drivers in the study shared similar accounts about
their father’s hours. Eddie’s father worked six days a week driving a truck and
moving furniture but was now ‘really sick . . . Underneath his lungs it is all crappy.
And his lungs are just inflating’. He says he feels bad about his Dad’s illness
‘because I can’t play with him and, yeah, he can’t buy me stuff without the money’
(Eddie, 12, Struggle Primary). However, the issue is not confined to blue-collar
68
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
workers: children of managers, small business people and professionals shared this
perspective.
Parent-specific time hunger
At Comfort Primary, with relatively secure higher incomes, most children chose
more time with their parents over more money, especially more time with a parent
who is absent more. Nicky was an exception: she would choose more money, reflecting the fact that she lives with her single mother who works part-time while
her dad lives in another city. The general preference for time at Comfort Primary illustrates how breadwinner household structures—with a time-rich parent
at home, usually the mother—do not eliminate parental time hunger for children. Instead, they show a ‘parent-specific’ time hunger. In the hyper-breadwinner
household, where the breadwinner is working long or unsocial hours, that time
hunger is pronounced, focused upon the absent parent, most often their father.
For example, Bob, whose father runs a sporting range and sometimes works long
hours while his Mum works part-time, wants to spend more time with his parents
‘especially my Dad because he hasn’t spent a lot of time at home’, and would chose
that over more money.
A full-time job does not mean a lack of time with young people for some. Ellie,
for example, who lives with her mother who works full-time, says that her mother
‘really enjoys her job . . . but I still see her, and my Mum and I are pretty close and
we still can spend heaps of time together. So it’s good. She really likes it . . . She
says its really satisfying’ (Ellie, 16, Leafy High).
Long hours and money/time
Hours are an important determinant of children’s perspectives about their parents’
jobs. Long hours are consistently associated with negative views of parental work
patterns. A number of young people defined their own futures against their parent’s
working lives. They plan to make sure they spend enough time with their own
children, or want to avoid demanding jobs and make sure they have weekends off.
They name the kind of ‘work/eat/sleep’ cycle that long hours workers also name
for themselves (Pocock, van Wanrooy, Strazzari & Bridge 2001):
Dad earns money and gets out—but he hardly spends any time with you. He comes
home, eats tea and watches TV. We’d like time . . . He feels bad because he can’t spend
time with the family. (Kelsey, 12, Struggle Primary).
When it comes to choosing between time and money, Bonnie, daughter of
a country truck driver, did not hesitate: ‘more time . . . even if that means you
mightn’t get everything that you wanted’. Brittany, in Sydney at Leafy Primary
agreed: she wants more time. Some young people see a link between more time at
work and poorer quality relationships. Several were very conscious of long hours
and their impact on household relationships, seeing the need to balance money
with time. As Adam, son of a cabinet maker and aged carer put it:
It’s good to get money coming in and probably it’s good to work as hard as you can
when you’re younger so when you’re older you can retire with some money. But there
should probably be a limit to so much before your relationships with other people
start to strain because you are never there. (Adam, 16, Country High).
TIME, MONEY
AND
JOB SPILLOVER
69
Time with fathers
Young men do not stop wanting time with their fathers as they grow up based on
the consistent comments of 16–18 year old boys. While not all young men in our
study felt the same (indeed some spoke of feeling distant from their fathers because
of their absence over time), others were straight-forward about their sadness:
I reckon my Mum [who works part-time] is pretty much fine the way she is. Just leave
her like that. But I suppose a little more time with my Dad would be good, seeing
as I usually get to see him for about five minutes in the morning after I get up and
he usually gets home around the time I’m doing homework so I only get to see him
around tea time, and onwards at night. So not too much going on there. (Kyle, 16,
Country High).
Kevin (17, Country High), whose father worked very long hours in the country,
would like more time playing backyard cricket with his father, more ‘hang around
time’: ‘I wouldn’t mind if I just sat in the next room, but it would be better if he
wasn’t doing as much work’. Adam—from the same region—agreed. His father
works in an industry with peaks and troughs. These young men whose fathers
worked long or unsocial hours, missed time with their fathers, and were very alert
to paternal stress. It is not surprising that they do not miss their mothers, given
that most were at home or held part-time jobs, but their clear preference for time
with the fathers, especially with unstressed fathers, suggests that time with one
parent does not easily substitute for time with another: father-time has its own
function and is clearly desired by young men (and young women).
For many children, time is a key first requirement of a ‘good dad’. Kelsey, 12,
at Struggle Primary describes a ‘good Dad’ as someone who ‘spends time with
you’. Her friend Kelly agrees: ‘someone who mucks around with you’. For Zoe,
whose father left her life some time ago a ‘good dad’ is ‘some one who doesn’t just
come and leave’. In the same school, Harry whose father left when he was very
young because ‘he didn’t like kids’ agrees: ‘A good Dad would be if he was part of
my family again because it would be better if he still had a job and came back. If
he came back and lived with my Mum and still got paid $700, it would be great’
(Harry, 11, Struggle Primary).
When discussing a ‘good mother’, young women stressed the importance of
someone to talk to about issues of importance, particularly social relationships.
While some young men mentioned this, they stress the importance of time with
their fathers more. Young men were more likely to equate a good mother with a
good cook, and value their help with homework.
A number of young people were positive about their mothers being at home
after school and liked time with mothers at home. They enjoyed ‘hanging time’,
weekends, and time before and after school. Many also valued a clean house, and
young men in particular mention cooking and plenty of food in the house as a
benefit associated with a mother working at home. Young people tended to adopt
a different framework to the presence of a father at home: rather than carers they
were viewed as ‘workers in waiting’. In the four households with fathers at home,
they were viewed as ‘out of work’ (sick, retired, unemployed or retrenched) rather
than valued home-based carers.
70
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
Overwhelmingly young people spoke of the importance of their parents being
there for special events, such as sports, choir, public performances, when they
received an award, or for special events like birthdays. This confirms other findings
(Galinsky 1999; Lewis et al. 2001). Many sadly remember key events that their
parents have missed. They like to share their successes and public events, and they
also want parents around to help solve problems when they occur. Several didn’t
think that ‘make up’ time later really compensated: they wanted their parents to
witness their achievements and activities.
Many young men anticipate spending a considerable amount of time with their
own children—taking care not to miss weekends and evenings, and some will
parent very differently to their fathers, in terms of time and work allocations:
I’d love to be able to take a few years off and not work and spend it with my kids
and with my wife and just starting a family and being there for my kids for the first
few years of their life. But then I’d definitely go back to work like when they start
school, but I’d make sure I was there for them in the evenings, help them with their
homework and on the weekends do sporting activities and all that. And when they
are older and think I’m just boring and not cool, let them do their own thing, but I’ll
still try and sneak in some quality time. (Smithie, 17, Leafy High).
JOB SPILLOVER: HOW PARENTS’ JOBS AFFECT YOUNG PEOPLE
Alongside the issue of work and time trades, the issues of spillover (from work
to home and from home to work) have been widely discussed in the work and
family literature (Pleck 1977; Hertz and Marshall 2001). This study shows that,
from the children’s perspectives, not all jobs are the same. And not all children’s
perspectives are the same, even about jobs that might appear similar. The effects
of work on children are highly context-specific and diverse. Nonetheless, some
similarities are striking and reinforce existing research.
A great deal of public attention in Australia has focused on whether to work
when children are young—especially whether mothers should work. For young
people in this study, this is the wrong question. This reinforces Galinsky’s (1999)
US results, and those of Lewis et al. (2001) in Australia. Young people understand
that many of their parents need to work and they do not see parents’ work as
intrinsically bad or good. The important issue for the young people in this study
is not whether parents go to work, but the state in which they come home. Young
people comment upon the nature and effects of their parents’ jobs, rather than
about whether they work or not. They talk about the effects of their parents’
jobs on them and their brothers and sisters—rather than seeing a problem with
parents’ jobs per se. Many can see positive outcomes for their parents from their
paid jobs—outcomes that flow to children through material comfort and, beyond
this, to a happier parent and household. This is a consistent finding, evident in
most focus groups, age groups, and in all household types.
Young people’s alertness to how work affects their parents extends to domestic
work. Many young people with mothers doing domestic work at home, saw positives both for themselves and for their mothers. However, being at home was not
without spillover effects for mothers and children, confirming results that Lewis
TIME, MONEY
AND
JOB SPILLOVER
71
et al. (2001) report . These include physical and emotional effects. Fred sees that
‘my little brother was really hard work for my mum’ when she looked after him fulltime (Fred, 11, Comfort Primary). Others perceived tiredness, physical injuries,
social isolation or a heavy domestic load that mothers often carried alone. Young
people are also aware of the under-valuation of their mother’s work at home,
and several mention their mother’s depression. Other children commented upon
mothers who were happy doing domestic work. Once again, however, it is not the
job itself that causes noticeable spillover, but its fit with parental preferences.
With respect to paid work there are a clear spillovers onto children. These
include both positive and negative effects. Most obvious among the positive effects
is income. Alongside this, many young people see benefits for them arising from
their parents’ work-related skills. In addition, many children enjoy the things
parents bring home: pens, equipment, supplies. Young people also see that their
parents have fun and gain a sense of worth and contribution through their jobs.
Others feel that parents enjoy making friends, having laughs and social connection
through their jobs. As one put it ‘She tells funny stories from work, the people at
work. My mum likes her job’ (Susie, 16, Comfort High). Others speak positively
when their parents’ work is flexible, when they work hours that mean that they
are home at the same time as children, and when they are able to participate in a
meaningful way in their parents’ work.
On the negative side, many forms of spillover are obvious and common. The
daughter of two full-time working parents who said ‘My parents try not to bring
the work life home’ was unusual. She appreciated their containment: ‘I don’t really
need their stress as well as mine’ (Jade, 16, Comfort High). Most young people
could easily tell what kind of day their parents had had when they walk in the door.
Jobs ‘colour’ parental moods—to use Nasman’s expression (2003), affecting their
‘state and physical condition’. However, these effects go beyond mere ‘colouring’
of parental mood. They are directly transmitted to children and to others in the
household. Children not only observe their parent’s ‘colour’, but many are affected
by it themselves and feel its effects through yelling, arguments and household
tension. Young people are especially alert to these negative spillovers: it is these
they notice and respond too—often withdrawing from their parent to cope. In line
with studies in other countries, they notice when their parents are upset (Galinsky
1999; McKee et al. 2003).
Negative aspects of parent’s jobs include physical injury, emotional or mental injury, bad moods, stress, tiredness, sadness, uncertainty and fear. Negative
spillover was perceived in all socio-economic groups. However, it sometimes
takes different forms. Young people in lower-income areas spoke about the physical impact of their fathers’ and mothers’ jobs on their parents’ bodies—and they
worried about this. Changes in working hours or shifts sometimes created significant change for young people. However, more hours did not necessarily mean
more negative spillover: once again it was the nature of the job, and its fit with
parental preferences, which seemed more important. Very specific effects flowed
from bad events at work, to parents and to children. These effects were often
associated with the ongoing nature of their parent’s jobs: their hours and physical and emotional demands. However, there were other important spillovers
72
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
arising from more occasional or episodic events—like injuries, job loss, redundancy, demotion or reassignment.
The impact of long or unsocial hours: ‘he’s not violent, he just yells . . .’
Long or unsocial hours were frequently mentioned as a first effect of parents’
work, with almost universal negative effects. Like others, Ali noticed his father’s
stress and anger after a long shift: ‘sometime when he comes back in the afternoon
he gets angry at simple things cause he’s stressed’. Andre’s father, an artist, is also
‘stressed and tired’ when he has been working late on a painting.
Children identified direct negative effects of long hours upon their parents. The
word ‘grumpy’ was frequently used to describe parents—just as it is often used
by long hours workers (and their partners) to describe themselves (Pocock et al.
2001). Kyle (16, Country High) said of his farmer father ‘he’s got to work long
hours, go and do things if he finds something wrong in one of the paddocks just as
he is about to knock off. He has to go and fix it or whatever.’ As Kyle described it
‘You do something wrong and he gets up you’. ‘Yeah’ agreed Robert: ‘he usually
comes home in a pretty foul mood and you’ve got to tread lightly around him’
(17, Country High). James (15, Struggle High), whose dad drives a truck for long
hours, couldn’t see anything good about his father’s job. He described how its
effects spilled over into non-work time.
In a higher socio-economic setting, Mark’s father works two jobs. He has given
up a third which has made him ‘happier’. One of his jobs involves night shifts:
He’s incredibly tired, all the time, because he is just constantly going and he gets
aggravated. He’s not violent, he just yells and he can yell really loud. He doesn’t do
it all that often considering the stress he is under. (Mark, 17, Comfort High).
Like other young people, Mark is very understanding about his father’s work
and its effects. He sees his dad as ‘good fun’ and—for all his occasional yelling—
loves spending time with him. He keeps out of his father’s way when he is ‘incredibly tired’.
Young people respond to negative spillover in several ways: most commonly
they physically withdraw from a grumpy or shouting parent; they may turn to the
other parent and distance themselves from the absent or grumpy parent. Some
take steps to look after their tired parent. Others just worry about their parents,
while others dislike—or hate—the contamination of their emotional states by
their parents’ jobs.
Several children commented upon the fact that their parents’ long hours are
under-paid in their view: ‘He doesn’t get paid very well for the hours he puts in’
(Abraham, 16, Comfort High). Many specifically mentioned the loss of weekend
time with their parents, including time when all the family is together. For example, Sarah (11, Strive Primary) would like it if her father had a weekend day off
rather than Wednesday because she really enjoys family time on weekends. Others want their parents around on weekends for more utilitarian reasons: because
otherwise ‘we couldn’t get anywhere’. Several did not like their parents working
at night: ‘I wouldn’t like it if Mum worked at night’. ‘No I wouldn’t either.’ (16
year old girls at Struggle High). Shift work is also seen as a source of parental
TIME, MONEY
AND
JOB SPILLOVER
73
tiredness. These effects cross most boundaries: city and country, high and low
incomes, boys and girls, and they affected workers in blue, white and pink-collar
jobs, reaching into many occupations and industries. Both small business owners
and operators, and wage earners were affected in the perception of young people.
Where fathers worked long hours there was a considerable sense of loss, and
in some cases hurt and anger, particularly among boys and young men. Young
people spoke of how their parent’s anger and frustration with work, spilled over
onto them. With parents on a short fuse, young people knew when to ‘keep out
of the way’. They valued some explanation about their parents’ moods. They did
not like their parent being ‘shitty’, and it spilling over into interrogation of their
children, but an explanation for these spillovers mitigated their effect.
RE-FRAMING THE WORK AND FAMILY DEBATE: IMPLICATIONS FOR
POLICY, ACTION AND THEORY
This analysis affirms research that tells us that it is not whether parents go to work
or not, but the state in which they come home, that really affects children. This ‘state’
reflects objective characteristics of jobs (like hours and intensity) as well as the
extent to which parents’ preferences match their jobs. The debate about whether
to work or not needs to be re-framed in Australia (as Galinsky (1999) has argued
in the USA). In particular, less attention should be paid to the issue of whether
mothers should hold paid jobs, and more to the work of fathers. Most importantly,
a mother at home does not make up for a father who is absent a great deal.
For many children, their mothers’ and fathers’ jobs are associated with positive
spillover. Young people value the money and security that parents’ paid work
brings. Beyond this, they can see that many parents enjoy their jobs, or aspects
of them. They love their stories, sense their social connection through work, and
see that their parents have fun and feel good about doing a job well.
But negative spillover is also widespread. It is especially associated with disappointed parental preferences (for example, a parent who doesn’t want to work
part-time but has to, or a parent who cannot work the shifts they want). It is
also associated with some specific job characteristics: risk of physical harm, job
insecurity, work overload, or long or unsocial hours. These often send a parent
home from work angry or upset. Their moods are obvious to young people—from
physical, verbal and behavioural clues. Children ‘read’ their parents easily.
Job spillover, however, does not end with the parents. Both good and bad job
spillover is directly transmitted to children. It affects them—their moods, concerns
and behaviour. The fit between job preferences and job reality seems critical to
the level and nature (good or bad) of spillover. For example, a mother doing
domestic work who is happy with that, who is not physically exhausted or sick,
will have much less spillover than a mother who does such work but is resentful,
bored or feels isolated or under-valued. A parent who loves his job and works long
hours may bring home less negative spillover than one who works part-time but
hates his job. That said, long or unsocial hours are consistently associated with
perceived negative spillover and the majority of parents who work them appear
to children to be sometimes grumpy, tired and stressed.
74
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
How should parents, workplaces and public policy respond to these findings?
At the individual level, it is clear that ‘working mothers’ are not, in themselves, a
problem in the view of children. Many see benefits arising from parents’ jobs. The
nature of work itself, and its impact upon parents, emerge as the more important
considerations. Where jobs send parents home worried, sick, hurt, tired or bad
tempered, children notice and are negatively affected. Such spillovers are not
confined to paid jobs: they also arise from domestic work if it fits poorly with
parental preferences, or leaves parents isolated, depressed, tired or under-valued.
Too much worry about having a job at all—especially with respect to working
mothers—has diverted attention from more significant aspects of work that young
people notice—like how absent working fathers or their long hours affect children,
and how work affects parental bodies, tempers and emotions. It seems that negative
spillover is reduced when parents do work that they like, for hours that suit their
preferences, while minimising physical injury and long or unsocial hours that
leave them bad tempered or tired.
Of course, this is at once both obvious and more easily said than done. For many
parents their preferences do not—alone or even in significant part—determine
their labour market outcomes; these are instead overshadowed by larger structural, systemic and institutional forces. Changing these requires new public and
workplace policies and practices. The finding that long or unsocial hours affect
children negatively, in their perception, reinforces the well-documented concerns
of adults (Pocock et al. 2001; Peetz et al. 2004). This strengthens the argument
for policies that contain hours of work to social and reasonable levels.
Other policies and practices that assist parents to find a good fit between how
they want to parent and how they work, are also important to the perceptions of
children. These include measures that increase worker say over the allocation of
working time and its organisation (or greater ‘worker time sovereignty’), including
over hours of work, start and finish times, days off, holidays, shifts and rosters.
Greater worker capacity to move into and out of full-time and part-time work,
without risk to employment or earnings, and into and out of paid work itself
will also help by protecting home and children from fears about income and job
security. This includes a policy regime of tax and social security arrangements that
does not penalise people for changing jobs or for moving into and out of the labour
market to get the best possible fit of preferences with circumstances. Predictable
and secure job arrangements that do not destabilise and frustrate preferences, with
negative consequences for the security of children, are also likely to be significant.
An improved suite of paid and unpaid leave rights for working parents will also
assist parents to better meet the changing demands of their dependents so that
they can better parent around their work to meet both their own preferences and
those of their children.
In theoretical terms, this analysis confirms the close relationship of work to
household life: they cannot be understood as separate social or industrial spheres
(Nasman 2003). Of course, this close relationship is nothing new: it has characterised the work and households of adults and their children for centuries, with
children long involved in forms of labour, and often affected in various ways by
the waged and unwaged work of parents. However, this relationship is changing
TIME, MONEY
AND
JOB SPILLOVER
75
at present, as more carers (especially mothers) enter paid work, and experience
personal guilt and very variable institutional supports (Pocock 2003). This analysis of the perceptions of young people confirms that theoretical treatment of work
and family as separate social and policy spheres is unhelpful. They are intimately
connected. This connection requires analysis of work that recognises that paid
work is located within a larger ‘total labour’, and within diverse households and
communities. It also requires policy action that takes account of these connections.
Increasingly, producing bodies are also reproducing bodies, which affects how we
must think about workers, workplaces, dependency, childhood and households.
In the perception of these young people, the spheres of work and households
are not consistently in opposition to one another, the one draining the other.
Work does not always have negative effects upon young people: positive spillover
from work for some is also evident, encouraging us to consider ways and means
to increase positive spillover, and to reconsider the conventional binary and oppositional theorisation of ‘work’ and ‘home’.
ACKNOWLEDGEMENTS
This research was funded by The Australia Institute and the Australian Research Council through
the fellowship project ‘Theoretical and policy implications of changing work/life patterns and
preferences of Australian women, men and children, household and communities’.
REFERENCES
Campbell I (2002) Extended working hours in Australia. Labour & Industry 13(1), 73–90.
Galinsky E (1999) Ask the Children: What America’s Children Really Think About Working Parents.
New York: William Morrow and Company.
Hertz R, Marshall NL (2001) Working Families. The Transformation of the American Home. Berkeley
CA: University of California Press.
HILDA (2001) Annual Report. Melbourne: Melbourne Institute.
Jensen A, McKee L (Eds) (2003) Children and the Changing Family; Between Transforamtion and
Negotiation. London: RoutledgeFalmer.
Lewis V, Tudball J, Hand K (2001) Family and work: the family’s perspective. Family Matters, No
59, Winter, 2001. Melbourne: Australian Institute of Family Studies.
McKee L, Mauthner N, Galilee J (2003) Children’s perspectives on middle-class work-family arrangements. In Jensen An–Margritt, McKee Lorna (Eds), Children and the Changing Family;
Between Transformation and Negotiation. London: RoutledgeFalmer, pp. 27–45.
Megalogenis G (2003) Fault Lines. Race, Work and the Politics of Changing Australia. Scribe,
Melbourne.
Nasman E (2003) Employed or unemployed parents: A child perspective. In Jensen An–Margritt,
McKee Lorna (Eds.), Children and the Changing Family; Between Transformation and Negotiation.
London: RoutledgeFalmer, pp. 46–60.
Peetz D, Fox A, Townsend K, Allan C (2004) The big squeeze: Domestic dimensions of excessive work, time and pressure. New Economies: New Industrial Relations, Proceedings of the 18th
AIRAANZ Conference, 2–6 February 2004. pp. 381–390.
Pleck JH (1977) The work/family system. Social Problems, No. 24, 417–427.
Pocock B (2004) Work and Family Futures. How Young Australians Plan to Work and Care. Canberra:
The Australia Institute.
Pocock B (2003) The Work/Life Collision. What work is doing to Australians and what to do about it.
Sydney: Federation Press.
Pocock B, Clarke J (2004) Can’t buy me love? What young Australians views on parental work, time,
guilt and their own consumption. Canberra: The Australia Institute.
76
THE JOURNAL
OF
INDUSTRIAL RELATIONS
March 2005
Pocock B et al (2001a) Fifty Families: What Unreasonable Hours are Doing to Australians, Their Families
and Their Communities. Melbourne, ACTU.
Pusey M (2003) The Experience of Middle Australia, The Dark Side of Economic Reform. Cambridge,
Cambridge University Press.
Schor J (1992) The Overworked American: The Unexpected Decline of Leisure. New York: Basic Books.
Summers A (2003) The End of Equality. Sydney: Random House.
Tanner L (2003) Crowded Lives. Sydney: Pluto Press.
Watson I, Buchanan J, Campbell I, Briggs C (2003) Fragmented Futures. Sydney: Federation Press.