Fatherhood by shock: does
marital separation bring about a more equal time allocation among parents?
University of Antwerp - Universiteit
Antwerpen - Centre for Social Policy Herman Deleeck - Authors: Joris Ghysels and Lieve Cornelis - Research papers : 2006
Paper presented at the 28th Annual Conference on Time Use Research (Copenhagen,
16-18 August, 2006); Version 1: July 2006
Abstract
Despite
government campaigning, a changing public perception of the contribution of
fathers to childrearing and rising female employment rates, time use data
continue to show a large gap between mothers' and fathers' childcare time. In
this paper, we investigate if a (largely) exogenous demographic evolution, the
rising rate of divorce, is likely to alter expectations on this issue. More
specifically, we want to know if the fathers' share of active childcare raises
following a separation. This is likely to be the case if they share custody
with their ex-partner.
Yet, it
is also possible that fathers call in other relatives, rely on childcare
services to a large extent or transfer their responsibilities to a new partner
as soon as possible. All the latter would result in a status quo, as far as the
gendered division of childcare time is concerned.
With
data from the 2004-2005 Flemish Families and Care Survey, we show that there
exists indeed a difference in the gender time balance, depending on the family
type of the male partner involved. More specifically, fathers who live in a
reconstituted family with some of their biological children, are significantly
more likely to share care tasks equally with their female partner than other
fathers. Hence, the rise in marital separations may coincide with a more equal
time allocation among parents.
JEL-codes: D13, J12, J13, J16
Themes: divorce, childcare time, gender, time
allocation
1. Introduction
By the
end of the twentieth century, female employment is widely considered to be self-evident.
As a result, more families face the everyday challenge of combining both labour
market duties, household chores and child care. This evolution has given way to
a (somewhat slowly) changing public perception of the contribution of fathers to
household work and childrearing. A ‘modern’ dad is expected to equally share
the burden of all domestic responsibilities which were previously taken up by women
according to traditional gender roles.
Furthermore,
governments have responded to this new social constellation by introducing a set
of measures to facilitate the combination of work and family (European Commission,
2005). Some of these actions (e.g. paternity leave and parental leave targeted at
fathers) try to induce fathers to participate more in child raising.
However,
despite government campaigning, a changing public perception of the contribution
of fathers to childrearing and rising female employment rates, time use data continue
to show a large gap between mothers’ and fathers’ childcare time (Stancanelli, 2003;
Glorieux, Koelet, Moens, 2001; Robinson, 2003; Craig, 2006; Kalenkoski, Ribar, Stratton,
2005). Furthermore, over the past decades the rate of change of the gap was so small
that it makes scholars remark that its closure is virtually out of sight.
Yet, another
societal change of the past decades, the rising divorce rate, may offer prospects
for faster change. In this paper, we investigate whether this (largely) exogenous
demographic evolution is likely to alter expectations. More specifically, we want
to know if the fathers’ share of active childcare raises following a separation.
This is likely to be the case if they share custody with their ex-partner. Still,
it is also possible that fathers call in other relatives, rely on childcare
services to a large extent or transfer their responsibilities to a new partner
as soon as possible. All the latter would result in a status quo, as far as the
gendered division of childcare time is concerned.
Data for
this analysis stem from the 2004-2005 Flemish Families and Care Survey, a survey
with representative data on families with children below 16 in the Flemish region
of Belgium. Before tackling the data, however, we go over results from previous
research (section 2) and present some basic data about fathers’ experiences with
divorce in Flanders (section 3). In section 4 we discuss the dataset in detail
and elaborate the empirical analysis. We show the differences in care time
according to a typology of fathers and confront this typology with other explanatory
factors in a multivariate analysis. Section 5 concludes.
2. Previous research about parental care time
In the following
paragraphs, we synthesize the most important findings of the relatively scarce literature
on childcare time. Anxo, Flood and Kocoglu (2002) compare the time allocation
of parents in France and Sweden. Their ‘double hurdle’ analyses of time diary
data indicate that parents tend to spend less time on childcare when their
children grow older, that lowly skilled parents tend to spend less time on childcare
and that women tend to be more sensitive to all kinds of explanatory factors. Their
analysis suggests, moreover, that institutions matter. Swedish fathers spend more
time with their children than the French, but the French tend to compensate for
the working time of the mothers while the Swedish do not. According to the
authors this may be attributed to the Swedish combination of a wide offer of childcare
services and extensive career flexibility regulations, which reinforces women’s
relative power in marriage and allows for a considerable externalisation of the
care burden.
Hallberg
and Klevmarken (2003) focus on the childcare time of Swedish dual-earner couples.
Their analysis of time diary data confirms the effect of the age of children. Furthermore,
it shows that Swedish parents treat their partner’s childcare time as a complement
to their own time rather than as a substitute.
Neuwirth
(2004) analyses primary childcare activities from the 1992 Austrian time use survey.
He finds that parents’ childcare time is mutually positively correlated, while
being inversely related to the own job time. Moreover, Austrian women tend to compensate
for their partner spending much time on his job, while men do not. This gender
differential and the overall observation of the smaller responsiveness of men
to observed variation in their household situation suggests that the male chauvinist
model is still important in Austria. Furthermore, Neuwirth’s ‘two stage least
squares’ analyses show clear links between the various time allocation decisions,
which indicates Austrian parents’ need to accommodate privately for, say, an increase
of their job time.
Deding and
Lausten (2004) analyse time diary data of Danish couples, which they enriched with
official register data. They compare market work with non-market work and differentiate
between housework and childcare within the latter category. The results
indicate that there exists a trade-off between market work and non-market work,
but that childcare is relatively ‘untouchable’. If the time for non-market work
is squeezed it is other categories of activities that shrink, not childcare. Moreover,
the authors confirm the previously found positive association between male and female
childcare time.
Kalenkoski,
Ribar and Stratton (2005) differentiate between market work, primary childcare
time and secondary childcare time and analyse British time diary data. As in
the French and Swedish case, highly skilled parents tend to spend more time
with their children and childcare time levels off as the children grow older. Interestingly,
primary childcare time is concentrated among the youngest and secondary
childcare time tends to rise somewhat with the age of the children before both categories
disappear altogether. They also examine parents’ child care time among three
family types: married, cohabiting and single-parent families. They find no differences
between married and cohabiting partners with regard to time devoted to
childcare and to market work, but single parents appear to spend more time on
child care and less time on market work than other parents.
Paley (2005)
considers the timing of parental childcare and includes both ‘active’ childcare
time and ‘passive’ childcare time, conditional on the other parent not being actively
caring at the same moment. She uses data from the 1997 time diary supplement of
the US Panel Study on Income Dynamics. Paley observes how the job schedules of parents
can effectively ‘force’ men to enhance their childcare time relative to other fathers.
However, she also finds that mothers do not treat their partner’s increase as a
substitute, but rather exhibit compensatory behaviour. In other words, when arriving
home later than the father and the children, the mother will spend relatively more
time with her children than other mothers. Among men, though, there is no such
compensatory behaviour.
Ghysels
(2004, 2005) analyses a 1997 ECHP-dataset of two-partner households in a simultaneous
equations framework incorporating both childcare time and job time. His
analyses show that childcare time is positively correlated among Danish,
Belgian and Spanish parents. However, the simultaneous equations framework revealed
no direct impact of a respondent’s job time on the own childcare time, nor his or
her partner’s, except in Belgium. In Denmark and Spain the link between
childcare and employment proved restricted to a one-way relationship with the
amount of personal childcare determining one’s job choice.
Besides
studies about parental childcare time in general, some researchers investigate childcare
time spent by one of the parents. Kimmel and Connelly (2006) focus on time mothers
spend with children because women tend to experience conflicts between market
work and family responsibilities more intensely than men do. They estimate a simultaneous
four-equation system using data from the 2003 American Time Use Survey. The
results of their analysis show that mothers’ time with children does not
respond to price or demographic changes as pure home production or leisure do.
As a result, maternal childcare time should be considered a distinct time
spending category, next to paid labour, leisure and pure household work production.
Furthermore, Kimmel and Connelly (2006) notice important differences between
time allocation on weekdays and during weekends. Therefore, a separate analysis
of timebetween these types of days seems appropriate.
Stancanelli
(2003) elaborates several regression analyses on ECHP-data [1] to investigate
fathers’ care time. Her results show that time allocated by fathers to child care
is responsive to their own hours of paid market work [2] (-) and to their
spouses’ paid working hours (+). In addition, caring time by fathers is found
to be positively related with employment in the public sector and with a high
level of education of the spouse. Self-employed men appear to spend substantially
less time caring for their children.
Apart from
childcare time, several researchers examine time people spend on domestic work
such as washing, cleaning, cooking meals, keeping the house in good order, etc.
One of the main conclusions from this strand of the literature is that women still
dedicate considerably more time to household chores than men, although the division
of household work has become slightly more gender equal over the past decades –
principally due to the fact that women do less now than they did before (Craig,
2006; Baxter, 2002; Stancanelli, 2003; Robinson, 2003).
A main
societal evolution of the twentieth century, the increased divorce rate, might alter
expectations on the balancing issue. During the various stages between a first and
second marriage, men are likely to acquire some experience running a household on
their own. This experience, as little as it can be, can be expected to lead
divorced men to carry out more household duties in their second marriage (or cohabitation)
than in their first marriage. This theoretical expectation is not immediately corroborated
in the literature. Most studies find that remarried woman continue to do the
bulk of housework (Demo & Acock, 1993; Pyke & Coltrane, 1996). Yet,
Ishii-Kuntz and Coltrane (1992) observe that remarried husbands in the US bear slightly
more household chores than husbands in their first marriage (National Survey of
Families and Households (NSFH) 1987-88). This result suggests that the time men
dedicate to household tasks has changed due to their experience of a divorce. In
our data, we will investigate whether this finding holds for childcare time as
well.
3. Fathers and divorce in Flanders
Recently,
Belgium has become one of the European countries with the highest rate of divorce.
[3] Statistics for Belgium show that two out of three couples who split up,
have children. Furthermore, many of these children are found to be very young.
Of all the children whose parents divorced in 2003 (about 40 000), nearly half
was 12 years old or younger at the time of divorce.
From a
social perspective, it is very interesting to examine what transformations with
regard to family structure take place after a divorce. Lodewijckx (2005)
investigated the living situation of all children in Flanders and found that
25% of Flemish children do not live with their biological parents. About 7% of Flemish
children stay with one biological parent and his or her new partner. The vast
majority of them live with their mother and a stepfather. When children stay
with only one parent, it almost always concerns their mother (11% of all children
have a single mother versus 1,6% who have single fathers). Moreover, the evolution
of children’s living conditions differs depending on whether they stay with their
mother or their father. One out of five divorced men [4] are part of a new family
within four years after their divorce, as opposed to just one out of ten women
(Corijn, 2005b). Therefore, children of single mothers are proportionally more likely
to stay in a single-parent family for a long while.
Corijn (2005a
and 2005b) investigated the marital situation of Flemish men and women four
years after their divorce. She observed that after this time interval 12% of young
men are single fathers. Other men have found a new partner and live together
with children, who can either be their own children from a previous marriage, the
children of their new partner or new children they join with their new partner.
Yet, in practice
the children involved are generally not the father’s own children, because
children of divorced couples usually stay with their mother. In Flanders, the mother
is very often (nearly automatically) given custody over the children, even though
the Belgian government has recently taken action to install residential co-parenting
arrangements as the norm. Given the current practice, most divorced fathers have
little opportunity to spend time with their own children. Moreover, there is no
systematic registration of co-parenting arrangements, which renders it a
difficult topic for empirical research.
4. An empirical analysis of paternal care time
The description
of the current divorce situation in Flanders - as provided in the previous
section - warrants a more precise definition of our research question. Most divorced
men in Flanders do not experience a transitional period of single parenthood. At
most they experience a type of partial single parenthood if they join custody
with their ex-partner. For most fathers this is not even the case and their
contacts with their children remain limited to (weekend) visits. Moreover, most
divorced fathers quickly form a new family and, hence, have every opportunity to
return to previous habits of unequal sharing of care time. Nevertheless, even
without the actual experience of the full sole responsibility for a household
with children, the divorce experience may lead to an increased involvement in
care time and household chores. Divorced fathers may have come to value the relationship
with their children more than they did before. Alternatively, they may also have
come to recognise the importance of an equal sharing of household responsibilities
for a lastingly successful relationship. Consequently, we will investigate
whether fathers living together with a partner show a different involvement in care
time, depending on the type of family they live in. For this analysis, we will
use a new Flemish dataset, which we describe below.
4.1 Data from the FFCS
The
Flemish Families and Care Survey (FFCS) was elaborated to provide data for a micro-distribution
analysis and contains information on income sources, expenses, care needs, time
use and the use of care services of families with children. It is coordinated
by the Centre for Social Policy Herman Deleeck (University of Antwerp) in
cooperation with the Flemish governmental organization Kind en Gezin [5] (Child
and Family) and the Flemish Fund for the Social Integration of Disabled Persons
(VLAFO) [6].
The
research population consists of families with a least one child between 0 and
15 years old who reside in the Flemish region [7]. The sample is divided into
four subsamples, representing four types of families. The largest group consists
of young families with a least one child under three years. A second group is
formed by ‘older’ families having at least one child between 3 and 15 years
old. Apart from these two groups, who were randomly selected from the National
Population Register, the sample is completed with two smaller subsamples, consisting
of poor families and families having a disabled child. The subsamples were randomly
selected from the client databases of Kind en Gezin and VLAFO respectively. For
this paper, only the samples from the National Population Register are used.
The time
use part of the FFCS refers to activities schedules to be filled by every parent
in the household for two days (one weekday and one day of the weekend, activities
on a quarterly basis), an employment matrix spanning a full week for every parent
and a care schedule for every child of the household (for a full week, on the basis
of a half hour). For the purpose of this paper, we could rely on 920 father’s activities
schedules of a weekday and 937 father’s activities schedules of a Saturday or Sunday.
These data were weighted to produce results representative of the full population
of families with at least one child younger than sixteen years old.
4.2 Fathers and care time in the FFCS
As
stated in the introduction to this section, we want to distinguish fathers on
the basis of their marital history. In our dataset there is no retrospective
data on marital history. However, detailed information is collected on the relationship
between the various family members. Consequently, we can differentiate between
three types of fathers.
The
first type groups those fathers living with children who are biologically
linked to both themselves and their spouses. These form the large majority of
the Flemish population and account for 92% of our sample. The second largest
group is constituted by fathers who live in a reconstituted family, but have a
biological link with at least one child in the household (5% of the sample).
Note that this can relate to the rare occasion of a father enjoying the custody
of his child(ren) from a former relationship, but is more likely to have to do
with offspring of the new relationship.
Moreover,
it should be noted that we cannot distinguish between men who became father for
the first time in the current relationship and fathers whose children from the previous
relationship live with their ex-partner. Consequently, the relationship data allow
no perfect delineation of the marital history of the fathers, but rather
provide a cross-sectional view of the biological and other links between the
parents and the children they take care of. Our final type of fathers refers to
those who live in a reconstituted family without any children who are
biologically linked to themselves (3%). Recall from section 3 that Flemish
fathers tend to enter this type of situation quite quickly after their divorce.
[8] Some, but not all, of them become fathers of the second type after a while
Before
describing the dependent variable, it should finally be noted that our dataset
is restricted to fathers living with a partner. There were only an insufficient
number of observations of single fathers, not surprising given their minority
position in Flemish society.
The
dependent variable of our analysis is primary care time. Parents were invited
to indicate any of six care activities (childcare, medical care, watching over
the children, play and related activities, transport for the child, other care for
own children) and other activities they combined with these care activities. Care
time is counted as primary care if in a given quarter only care activities are
reported or if care time is the only active act presented in the quarter (e.g.
playing with the children and resting are jointly noted). In the case of active
care activities marked jointly with active non-care activities, the care
activities are counted proportional to the total number of activities mentioned
(mostly one out of two, thus 0.5).
While
primary care time is in the focus of our attention, it is not just the number
of hours spent that is of interest for our analysis. More precisely we want to
know the balance between the mother and the father as far as care time is
concerned. Therefore, we will investigate two dependent variables: the amount
of time spent by the father and the proportion of this time in the total
parental care time (mother plus father).
Following
the suggestion of Kimmel and Connelly (2006) and given the predominance of full
time jobs among Flemish fathers, we additionally split our analysis between weekdays
and days of the weekend. In Table 2 and Table 3, the average number of quarters
of an hour fathers spend on primary care and fathers’ time as a proportion of
total parental care time on weekdays and weekends are presented. First, we
observe in the tables considerable differences between weekdays and days of the
weekend. An average Flemish father spends 1h 56’ on primary care for his children
on a typical weekday, while his care time rises to 3h 23’ on a Saturday or Sunday.
Furthermore,
ANOVA-analyses indicate significant differences for both dependent variables
between traditional fathers, fathers living in a reconstituted family with own children
and fathers living in reconstituted families with stepchildren only.
On weekdays
(see Table 2), fathers living in a reconstituted family with own children
devote on average two hours and three quarters to care activities. Their
contribution in total care time (44%) is higher than that of traditional
fathers and men who are just stepfathers. Fathers living in traditional families
spend about two hours daily on childcare activities, while men who are just stepfathers
spend even half an hour less.
In
weekends (see Table 3), traditional fathers appear to compensate for their
lower effort during the week by nearly doubling their daily childcare time.
Fathers who live in reconstituted families with own children spend a similar
amount of childcare time (three and a half hours). The share of parental care
time taken on by fathers increases, except for fathers of reconstituted
families with stepchildren only. Reconstituted families where fathers have own
children enjoy an equal distribution of childcare time between man and woman.
These
bivariate results provide the basis for our empirical venture. They suggest
that the hypothesized relationship between family type and paternal childcare
may indeed be true. Yet, it remains to be seen whether this association hold
true when controlled for the characteristics of the fathers involved. If not, it
will prove to be nothing but spurious correlation.
4.3 An empirical assessment of the impact of
the father typology
To validate
the previous bivariate analysis we submitted the association to a multivariate test:
an OLS estimate of fathers’ primary care time controlled for explanatory
variables commonly used in time estimates (see section 2). We include fathers’ characteristics
(age, educational level, family type, time spent in the labour market, whether
self-employed or not), child characteristics (number and age) and the job type
of the mother [9]
Table 4 lists
the results for the information on care time on a typical day of the working
week (Monday till Friday). Left in the table are the estimates for the amount of
time of the father, on the right hand side we show estimates for the
proportional contribution of the father, as compared to the total effort made by
the father and mother together. [10]
For the total
amount of time a Flemish father spends on primary care in weekdays (see Table
4), the marital history of himself or his spouse does not seem to matter. Crucial
variables are the number and age of the children and the number of hours the
father puts in his job. The latter has the obvious negative sign. If the father
has a time-demanding job (more than 40 hours a week), he spends about one hour
less on his children than fathers with a standard full-time job (3.6 quarters
of an hour = 54’). Furthermore, the estimates show that fathers with more than one
child spend relatively more time on their children, but there is no strong
difference between two, three or more children. The age of the children, on the
contrary, reveals a clear distinction. The younger children are, the more time
fathers spend on their care. Interestingly, this effect also exists for school
age children. On a typical week-day, a Flemish father invests three quarters of
an hour more in primary care when his youngest child is in nursery or primary
school than when his youngest is in secondary school (aged 13 to 15). [11]
Somewhat
unexpectedly, the job time of the mother does not enter the equation. Apparently,
fathers do not compensate for the time mothers spend in the labour market, at least
not in a directly observable way. [12] At first sight, this can be interpreted as
a sign of male chauvinism, but the proportions data we will discuss below show
that the explanation is rather more complex.
Indeed,
with the equal sharing of care tasks as the primary interest of our analysis,
we cannot limit ourselves to an analysis of paternal care time only. The right
hand pane of Table 4 is at least as important. It shows what makes fathers
diverge from their average weekday contribution of 32 %. In the balance with
their partner, some of the determinants of the total amount of time continue to
bear weight. A time-demanding job reduces the proportional contribution of a
father, much as a family of two or more children increases his proportional
contribution.
Different
from our earlier analysis of the total amount of time, the type of job of the female
partner does explain much of the variance in the proportional contribution of the
father. Compared with a wife without labour market responsibilities, a partner with
a part-time job raises the father’s contribution with 12 % and a full-time job
adds 23 %.
Another difference
with the previous analysis, derives from the family type of the father. Compared
with the large majority of fathers who live with their first spouse, a father who
lives in a reconstituted family with a new spouse and his own children, caters for
a larger proportion of the primary care time (+10%). [13] Moreover, the specific
nature of the relationship with the children proves important, because fathers who
join a mother and her children from a previous relationship, do not exhibit a
rise in their care contribution and behave as ‘standard’ fathers.
The
story of week-end care is slightly different, as shown in Table 5. First, job
time is less important than during the working week, as could be expected. The
type of job of the mother is not significant, neither for the total amount of
time, nor for the proportions data. Nevertheless, a time-demanding job does
have repercussions for the father on Saturdays and Sundays too. Yet, the effect
is smaller than during the week and, moreover, has no proportions parallel,
meaning that the partner of a father with a time-demanding job does not compensate
for his reduction in care time during the weekend. The latter contrasts with
her behaviour on weekdays, when the reduction in the care time of the father is
(potentially [14]) compensated for by an increase of maternal care time (see
Table 4).
On
Saturdays and Sundays, also the number and age of the children have a different
effect. In most families, the number of children is no longer important.
Fathers spend an equal amount of time on their children irrespective of whether
they live with one, two or three children. Only the minority of fathers with
four or more children (about 6% of our sample) top the daily average with an
additional 4.6 quarters of an hour (+ 1h 8’). More important than the number of
children, however, is their age. The primary care that babies and toddlers
require, is clearly different from the time needed by youngster in secondary
school (+ 2h 15’). A similar, though smaller, difference can be observed
between the latter and children in nursery and primary school (+ 1 h 31’). As noted
in previous research, fathers tend to focus their attention, relatively
speaking, on school-age children. This is the only age group that combines an
increase in the amount with a rise in the proportion of paternal care time. This
does not necessarily mean that over the weekend mothers spend less care time
when having children in nursery and primary school, but in any case they do not
parallel the increase in care time of their partners.
Finally,
we return to the impact of the family type of the father. For our analysis of weekend
care time, the latter provide the largest part of the explained variance. [16]
As for weekdays, the family type does not alter the amount of care time of
fathers, but it does influence the proportions. A father who forms a
reconstituted family with own children caters for a considerably larger
proportion of weekend care than a ‘standard’ father (+ 14%), while the contrary
is true for a father with stepchildren only, to whom this characteristic means
an average reduction of his proportional involvement by 12%. Thus, the exact
nature of the bond between the father and the children in the household again
proves crucial to the involvement of the father.
5. Summary and discussion
With
this paper, we wanted to investigate whether a divorce or separation experience
would work as a shock to fathers as far as their contribution to parental care
time is concerned. Theoretically, several explanations could be conceived for
such a shock. First, fathers may experience a period of lone parenthood
following a separation and this experience may incite them to participate more in
parenting than before. This reasoning follows a parallel line of thinking than the
one that is developed when policy makers restrict part of parental leave to
fathers only: the mere experience will develop the taste. Yet, in Flanders most
fathers do not gain custody of their children and, hence, do not experience
anything more than shared custody. Nevertheless, even part-time lone parenthood
may work. Furthermore, divorced fathers may come to value the relationship with
their children more than they did before, because of the changes in their
family relationships. Alternatively, they may also come to recognise the importance
of an equal sharing of household responsibilities for a lastingly successful
relationship.
Our data
cannot offer conclusive evidence on whatever drives the difference in care time.
Yet, it does confirm the existence of a difference. Flemish fathers in reconstituted
families do spend more time with their children than fathers who live with
their first spouse and joint biological children. More specifically, we
observed a sizeable difference in the gender balance of primary child care time
over the weekend. With full-time work as the norm for men, the weekend is the
time of the week when fathers have most degrees of freedom regarding their time
schedule. It is exactly at these moments that fathers who live in a
reconstituted family bear 14% more of the childcare burden than ‘standard’
fathers do. However, the biological link between the father and the children he
co-resides with, proves crucial. The former result applies only to fathers who
live a mixed situation: with some children they are biologically linked to and
some they are not. If fathers are pure stepfathers, their contribution to
parental childcare goes exactly the other way round. They reduce their
contribution over the weekend.
These results
offer promising prospects for future research. We may, for example, explore the
child part of the FFCS dataset to investigate to what extent the reaction of fathers
is driven by part-time co-residence. Do they enhance their contribution on days
the children are present and reduce it on other days? Is the total care time responsive
to these variations in residence? Furthermore, we may go into the actual care
activities and detail the type of care time fathers share. Our current total
primary care concept may well conceal differences in contributions, both
regarding the type of the activity and regarding the sole or joint nature of
the activity.
In any
case, our research results suggest that the answer to the question in the title
of this paper is: yes, marital separation may bring about more equality in
parental time allocation.
Footnotes
[1] Data
from the European Community Household Panel
[2] This
is also shown by Gray (2004).
[3] This
section is largely based on research of the Centre for Population and Family
Studies (CBGS): Corijn (2005a and 2005b) and Lodewijckx (2005).
[4] This
result regards men who divorced before the age of 40.
[5] Kind
en Gezin is –among other things- responsible for the recognition,
subsidy-giving and inspection of childcare facilities.
[6]
VLAFO is responsible for the non income related aid to disabled persons (care services,
special equipment,...). For the demanders of care it organises both the intake
and the assignment of services. Furthermore it is responsible for the
recognition of care providers.
[7]
Families living in the bilingual region of Brussels are not included in the
research population.
[8]
Again, this type of families is likely to contain in part men who became
stepfathers from a previous state as bachelor, hence without a personal divorce
experience.
[9] Descriptive
information on the explanatory variables is included in the Appendix. Table A1 also
shows the reference categories for the dummy indicators.
[10] The
total effort is a crude measure summing the number of quarters of an hour of fathers
and mothers. No compensation is made for overlapping time and hence joint care
activities count for two.
[11] In Flanders
the typical age interval for secondary school children is 12 to 18 (end of compulsory
schooling), but the FFCS sample is restricted to families whose youngest child
is below 16.
[12] Previous
research shows that mothers’ time allocation tends to depend on the size of her
family.Interaction effects between the mother’s type of job and the family size
indicators might reveal an indirect effect of the type of job, but this is left
for later research.
[13] The
fact that this proportional increase coincides with a lack of change in the
amount of time spent by fathers, suggests that mothers living in reconstituted
families spend less time on primary care than mothers living with their first
spouse and children. Future research should reveal, however, whether this has
to do with the relationship between the mother and the children in the new
relationship (e.g. all children of the father) or with time these children
spend outside the household (e.g. with the ex-spouse of the father, their
natural mother).
[14]
Mathematically, it is sufficient for mothers to reduce their care time to a
lesser extent than fathers to obtain this result. Hence, a status quo of
mothers’ care effort is also perfectly feasible.
[16] Yet,
it should be noted that the multivariate regression provides weak results on this
matter, with only 5% of the total variance of the fathers’ proportion of care
time explained by the variables in the equation.
6. Appendix
7. References
- · Anxo,
D., Flood, L., Kocoglu, Y., 2002, Offre de travail et répartition des activités
domestiques et parentales au sein du couple: une comparaison entre la France et
la Suède, Economie et Statistique,
No. 352, pp. 127-150.
- · Baxter,
J., 2002, Patterns of change and stability in the gender division of household labour
in Australia, 1996-1997, Journal of
Sociology, Vol. 38 (4), pp. 399-424.
- · Corijn, M., 2005a, Uit de echt gescheiden en dan…? De leefvorm enkele jaren na een echtscheiding
in het Vlaamse Gewest, ‘Uit het onderzoek’, 13 October 2005, online on: www.cbgs.be.
- · Corijn, M., 2005b, Echtscheidingen in België: met of zonder kinderen. ‘Uit het onderzoek’, 17 October
2005, online: www.cbgs.be.
- · Craig,
L., Children and the revolution. A time-diary analysis of the impact of motherhood
on daily workload, Journal of Sociology,
Vol. 42 (2), pp. 125-143.
- · Demo,
D.H., Acock, A.C., 1993, Family diversity and the division of domestic labor: How
much have things really changed?, Family
Relations, Vol. 42 (3), pp. 323-331.
- · European
Commission, 2005, Reconciliation of work and
private life: a comparative review of thirty European countries, Directorate-General
for Employment, Social Affairs & Equal Opportunities, Luxemburg: Office for
Official Publications of the European Communities. online: http://ec.europa.eu/employment_social/publications/2005/ke6905828_en.pdf
- · Eurostat,
2006, Crude divorce rate, online: http://epp.eurostat.ec.europa.eu.
- · Deding,
M. and Lausten, M., 2004, Choosing
between his and her time? The market work gap and housework gap of Danish couples,
28th IARIW Conference, Cork:IARIW/CSO, 30 p.
- · Ghysels,
J., 2004, Work, family and childcare: an empirical
analysis of European households, Cheltenham:Edward Elgar, 287 p.
- · Ghysels,
J., 2005, Holding the pencil or buying paint: parental childcare versus children’s
need for extra income, the case of Belgium, Denmark and Spain, Review of Economics of the Household,
vol. 3, nr. 3, pp. 269-289
- · Glorieux, I., Koelet, S., Moens, M., 2001, Een kwestie
van tijd. Tijdsbesteding in Vlaanderen anno 1999, Belgisch Tijdschrift voor Sociale Zekerheid, nr. 3, pp. 533-555.
- · Gray,
A., 2004, Towards a time economy of parenting,
Families and Social Capital Research Group, London South Bank University.
- · Hallberg,
D., Klevmarken, A., 2003, Time for children: A study of parent’s time allocation,
Journal of Population Economics, Vol.
16, pp. 205-226.
- · Ishii-Kuntz,
M., Coltrane, S., 1992, Remarriage, stepparenting, and household labor, Journal of Family Issues, Vol. 13 (2),
pp. 215-233.
- · Kalenkoski,
C.M., Ribar, D.C., Stratton, L.S., 2005, American
Economic review, Vol. 95 (2), pp. 194-198.
- · Kimmel,
J., Connelly, R., 2006, Is mothers’ time
with their children home production or leisure?, Discussion Paper no. 2058,
Bonn, Institute for the Study of Labor.
- · Lodewijckx, E., 2005, Kinderen en scheiding bij hun ouders in het Vlaamse Gewest. Een analyse
op basis van de Rijksregistergegevens, CBGS-werkdocument 2005/7, online: www.cbgs.be.
- · Neuwirth,
N., 2004, Parents' time, allocated for
child care? An estimation system on parents' caring activities, ÖIF Papers 2004-46,
Wien, Austrian Institute for Family Studies.
- · Paley,
I., 2005, Right place, right time: parents'
employment schedules and the allocation of time to children, ESPE 2005 Conference,
Paris, ESPE/Brown University.
- · Pyke,
K., Coltrane, S., 1996, Entitlement, obligation, and gratitude in family work, Journal of Family Issues, Vol. 17 (1),
pp. 60-82.
- · Robinson,
J.P., 2003, Changes in American parents’ use
of time, University of Maryland (draft version).
- · Stancanelli,
E., 2003, Do fathers care?, Paris,
Observatoire Francais de Conjonctures Economiques, Sciences-Po (preliminary
version).
==============================
Download a PDF version of the document:
Contact the author of the document:
* Corresponding author: Joris Ghysels
University of Antwerp
Room M-481
Sint-Jacobstraat 2
B-2000 Antwerpen
