Content uploaded by Isaure Delaporte
Author content
All content in this area was uploaded by Isaure Delaporte on Oct 19, 2022
Content may be subject to copyright.
Family Formation and Employment Changes among
Descendants of Immigrants and Natives in France: A
Multiprocess Analysis∗
Isaure Delaporte†Hill Kulu‡
Abstract
This paper investigates the association between family formation and the labour
market trajectories of immigrants’ descendants over the life course. Using rich data
from the Trajectories and Origins survey from France, we apply multilevel event his-
tory models to analyse the transitions in and out of employment for both men and
women by parity. We account for unobserved co-determinants of childbearing and em-
ployment by applying a simultaneous-equations modelling. Our analysis shows that
women’s professional careers are negatively associated with childbirth. There are dif-
ferences across descendant groups. The descendants of Turkish immigrants are more
likely to exit employment and less likely to re-enter employment following childbirth
than women from other groups. The negative impact of childbearing on employment
is overestimated among women due to unobserved selection effects. Among men, the
descendants of European immigrants are less likely to exit employment after having
a child than other descendant groups. The study demonstrates the negative effect of
childbearing on women’s employment, which is pronounced for some minority groups
suggesting the need for further policies to help women reconcile work with family life.
Keywords: Fertility, Employment, Life-course events, Multilevel event history analysis,
Descendants of immigrants, France.
∗This paper is part of a project that has received funding from the European Research Council (ERC)
under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No.
834103).
†University of St Andrews, St Andrews, KY16 9AL, Fife, Scotland, United Kingdom. E-mail: icmd1@
st-andrews.ac.uk.ORCID ID: 0000-0003-0279-2032.
‡University of St Andrews, St Andrews, KY16 9AL, Fife, Scotland, United Kingdom. E-mail: hill.
kulu@st-andrews.ac.uk.
1
1 Introduction
European labour markets are characterised by gender disparities. Women continue to
have lower labour force participation rates than men (ILO 2018). Besides, women work
fewer hours, are concentrated in specific occupations, and earn less (OECD 2021). One
important root cause of gender inequality in the labour market is family formation. A
large body of literature has found that women’s professional careers are negatively af-
fected following childbirth, whereas men’s employment trajectories are not (Loughran and
Zissimopoulos 2009;Bertrand et al. 2010;Angelov et al. 2016;Wilner 2016;Kleven and
Landais 2017;Kleven et al. 2019a;Kreyenfeld 2015). Some groups are more affected than
others: immigrant women experience a greater motherhood penalty than native women
(Kil et al. 2018;Nieto 2021;Vidal-Coso 2019). Yet, little is known about the effect of
childbirth on the employment trajectories of the descendants of immigrants.
The effect of having a child on individuals’ labour market decisions is likely to differ
across population groups for a number of reasons. First, the descendants of immigrants
often differ from each other and from the native population (here defined as native-born
individuals with two native-born parents) in their cultural background, social norms, and
preferences. They hold different preferences about the timing of family formation (De-
laporte and Kulu 2022;Kulu et al. 2021) and have different expectations regarding the
division of paid and unpaid work (Fleischmann and H¨ohne 2013;Khoudja and Platt 2018).
The descendants of immigrants do not fare equally in the labour market (Algan et al. 2010;
Silberman et al. 2007) and individuals who have limited labour market opportunities
might be more inclined to exit the labour market following childbirth. Furthermore, the
descendants of immigrants may differ from each other in the number of family members
available for informal care, and in their attitudes/access to childcare (Seibel and Hede-
gaard 2017;Biegel et al. 2021). Thus, it is important to investigate potential differences
in the effect of childbirth on employment by gender and migration background.
This paper investigates the relationship between family formation and the labour mar-
ket trajectories of immigrants’ descendants and natives. We focus on France which pro-
vides a rich context for the study of differences in employment trajectories among diverse
population groups. The minority population comprises a number of sizeable groups, with
differentiated employment histories, and family patterns (Delaporte and Kulu 2022). We
use a French survey – Trajectories and Origins – which holds information on immigrants,
immigrants’ descendants, and French natives, and contains retrospective biographical in-
formation on individuals’ childbearing events and their labour market outcomes over the
life course. We examine the employment trajectories of both men and women and focus
on the descendants of immigrants (including the 1.5G of immigrants who have arrived in
France before the age of 15) who belong to six origin groups, namely the descendants of
North Africans, Sub-Saharan Africans, South East Asians, Turkish, Southern Europeans,
and other Europeans. We apply multilevel event history models to study repeated events
of employment changes as well as the birth of several children. We examine three sets of
2
transitions: i) the transition to first employment after leaving full-time education, ii) the
transitions out of employment and iii) the transitions to second and higher order employ-
ment. For each set of transitions, we investigate the relative risks of experiencing these
changes separately for men and women by parity. We also explore differences between
immigrants’ descendants and natives as well as across origin groups.
This study extends previous research in the following ways. First, although the link
between fertility and employment has been investigated extensively among the majority
population (Loughran and Zissimopoulos 2009;Bertrand et al. 2010;Angelov et al. 2016;
Wilner 2016;Kleven and Landais 2017;Kleven et al. 2019a;Kreyenfeld 2015), only a
very limited number of studies have focused on immigrants and their descendants (Kil et
al. 2018;Nieto 2021;Vidal-Coso 2019;Lacroix and Vidal-Coso 2019). In this study, we
analyse the interaction between gender and migration background. This allows us to shed
light on the effect of family formation on the labour market trajectories of immigrants’
descendants and native men and women. We also look at differences across origin groups.
Second, we contribute to the literature on the economic integration of the descendants
of immigrants (Meurs et al. 2006;Clark and Drinkwater 2010;Piton and Rycx 2020;
Zwysen and Demireva 2020;Clark and Ochmann 2022;Algan et al. 2010;Silberman et
al. 2007) by studying repeated events of employment changes. Most studies focus on
one single transition in the entire professional career of individuals (Ganault and Pailh´e
2022). This enables us to shed light on the extent to which patterns of entry or exit
explain variation in labour force participation across descendant groups. While higher
rates of employment exit might indicate issues around retention or instability, lower rates
of employment entry are more likely to be a signal of structural or cultural obstacles
(Khoudja and Platt 2018). Therefore, examining differences in both employment entry
and exit rates will allow us to better understand some of the obstacles encountered by
individuals from different groups.
Lastly, most research on fertility and employment has not accounted for possible unob-
served selection effects. Yet, when studying the effect of childbirth on employment, there
are potential unobserved selection effects as individuals who are more likely to change
their employment status may also be more (or less) likely to have a birth because of
unobserved characteristics. For example, some individuals may be more career oriented,
whereas others are more family oriented. This would lead to a biased estimation of the
effect of childbirth on employment. In this paper, we adopt a simultaneous-equations
modelling approach which allows us to detect and control for unobserved time-constant
co-determinants of these two processes. Although simultaneous-equations hazard models
have been used in research on interrelated event histories of individuals before (Matysiak
2009;Kulu and Steele 2013;Mikolai and Kulu 2018;Steele et al. 2005,2006), to the best
of our knowledge, no study has applied this method to study the interrelationship between
employment and childbearing dynamics among migrant and ethnic minority populations.
3
2 Previous Research
2.1 Work-family balance: a gendered perspective
A large body of research highlights the role of family formation in explaining gender in-
equality in the labour market (Loughran and Zissimopoulos 2009;Bertrand et al. 2010;
Angelov et al. 2016;Wilner 2016;Kleven and Landais 2017;Kleven et al. 2019a;Kreyen-
feld 2015). While men’s labour force participation tends to be stable across the life course,
women’s labour force participation varies at different life stages, with lower or no partici-
pation often corresponding to periods of childbirth (Angrist and Evans 1996;Jacobsen et
al. 1999;Kleven et al. 2019a;Sieppi and Pehkonen 2019;Herrarte et al. 2012).
For women, childbirth results in lower employment rates (Gutierrez-Domenech 2005a;
Cristia 2008;Michaud and Tatsiramos 2011;Fitzenberger et al. 2013), lower earnings
(Angelov et al. 2016;Kleven et al. 2019b,c), and a reduction in working hours (Lundberg
and Rose 2000;Miller 2011;Kleven et al. 2019a;Gutierrez-Domenech 2005b;Wood et
al. 2016;Begall and Grunow 2015). It reduces women’ performance (Azmat and Ferrer
2017), experience (Klepinger et al. 1999;Daniel et al. 2013), occupational status (Cools et
al. 2017;Kleven et al. 2019a), productivity (Krapf et al. 2017), as well as full-time (Paull
2008;Daniel et al. 2013) and high-paid private sector employment (Daniel et al. 2013;
Lundborg et al. 2017;Kleven et al. 2019a). Many women move to more family-friendly
workplaces after having a child (Hotz et al. 2018).
This negative effect of having children on women’s work careers is due to a number
of reasons (Matysiak and Cukrowska-Torzewska 2021;Fiori and Di Gessa 2022). First,
career breaks and reduced working hours lead to a deterioration of human capital. It
can also be taken as a signal of low labour market attachment by employers who may
be reluctant to promote women after long parental leaves (Evertsson and Duvander 2011;
Puhani and Sonderhof 2011;Evertsson 2016). Second, mothers may display worse labour
market outcomes because they choose jobs which are more compatible with childcare,
but these positions often pay lower wages and offer fewer promotion prospects. Finally,
having children may also affect mothers’ productivity and thereby affect their labour
market outcomes.
By contrast, research on the effect of parenthood on men’s labour market trajecto-
ries points to mixed results. Most previous studies have shown that fathers may earn
higher wages and occupy higher positions than childless men. This phenomenon, called
‘fatherhood premium’, is attributed to a selection of highly successful men into parent-
hood (Baranowska-Rataj and Matysiak 2022), increased work effort of new fathers who
see themselves as the primary breadwinner of the family, and discriminatory practices of
employers who perceive fathers as highly reliable and committed employees (Hodges et
Budig 2010). Yet, recent studies have also shown that an increase in men’s involvement
in the family may also affect their work careers. For instance, fathers experience wage
penalties for taking parental leave or using flexible work arrangements and these penalties
4
can even be higher than among mothers (Evertsson 2016;Rudman and Mescher 2013).
Other studies find that the effect differs across men’s wage distribution, and that there
are premia among higher earning men (Cooke 2014;Glauber 2018) and penalties for low
earning ones (Cooke 2014). Therefore, more research is needed to understand the effect
of family formation on men’s employment trajectories.
2.2 Work-family balance among immigrant and native populations
Most studies looking at the role of childbirth behind gender inequality in the labour mar-
ket have focused on majority populations. In comparison, a relatively low number of
studies have examined migrant populations. Some studies have adopted a cross-sectional
approach to compare the labour force participation of women with and without children
across different groups (Holland and de Valk 2017;Lacroix and Vidal-Coso 2019). For
instance, Holland and de Valk (2017) find that the gap in labour force participation be-
tween mothers and childless women was similar for native and second-generation Turkish
women in Germany and Sweden but was larger in the Netherlands and France. Similarly,
Lacroix and Vidal-Coso (2019) find a greater drop in the probability of being employed
for immigrant women in more affluent households compared to native women.
More recently, a number of studies have adopted a longitudinal approach; they show
that employment levels decrease to a larger extent following the transition to parenthood
among migrant women than among natives (Kil et al. 2018;Nieto 2021;Vidal-Coso
2019). This is especially the case for women of non-European origin. However, there is
also a strong path-dependency of employment trajectories around parenthood for migrant
women and natives (Maes et al. 2021) and the fact that second-generation migrant women
generally have a lower pre-birth labour market attachment than native women plays a role
in explaining the observed migrant-native differentials in maternal employment. Lastly,
few studies have explored differences across origin groups in the effect of family formation
on individuals’ employment trajectories. For instance, Khoudja and Platt (2018) find that
Pakistani and Bangladeshi women’s labour market entries and exits are less sensitive to
childbearing events compared to those of other women in the UK.
Despite the lack of information, there are a number of reasons why we could expect
differences between immigrants’ descendants and natives, as well as across origin groups.
First, there are large strands of literature indicating that descendants of immigrants differ
from each other and from the native population in their social norms, and preferences. For
instance, the descendants of immigrants often exhibit different partnership and fertility
patterns depending on their parents’ country of origin. They also hold different preferences
about the timing of family formation (Delaporte and Kulu 2022;Kulu et al. 2021). For
instance, in Europe, the descendants of immigrants from culturally similar countries such
as European and Western countries often have similar partnership patterns as the ones
of natives (Hannemann and Kulu 2015;Mikolai and Kulu 2021;Pailhe 2015;Ferrari and
Pailhe 2017;Hannemann et al. 2020;Andersson et al. 2015;Liu and Kulu 2021). Their
5
fertility levels are also closer to levels observed for the native population. By contrast,
immigrants’ descendants from countries with conservative patterns of family formation
exhibit higher marriage, lower cohabitation, and lower separation rates than the natives
(Kulu and Hannemann 2016a;Andersson et al. 2015;Kuhnt and Krapf 2020;Liu and
Kulu 2021;Hannemann and Kulu 2015;Mikolai and Kulu 2021).
Besides, their fertility levels tend to be higher than those of natives (Van Landschoot
et al. 2017;Milewski 2007;Krapf and Wolf 2016;Gonzalez-Ferrer et al. 2013;Kulu and
Hannemann 2016a). In France, these patterns have been observed for the descendants
of immigrants from Turkey and North Africa (Pailhe 2015,2017;Hannemann et al. 2020;
Delaporte and Kulu 2022). These differences in partnership and fertility patterns could be
indicative of different social and gender norms (Diehl et al. 2009;Roder and Muhlau 2014),
which in turn might influence differently individuals’ labour supply following childbirth.
Indeed, the children of immigrants coming from countries with more conservative family
patterns might be more influenced towards family responsibilities, e.g., women might
reduce their employment after becoming mothers while men might feel that they have the
responsibility to provide financial support to their family after becoming fathers.
Second, the descendants of immigrants from different origin groups do not fare equally
in the labour market. A large strand of literature documents the disadvantaged labour
market positions of some groups among the descendants of immigrants in Europe (Meurs et
al. 2006;Clark and Drinkwater 2010;Piton and Rycx 2020;Zwysen and Demireva 2020;
Clark and Ochmann 2022;Silberman et al. 2007). Algan et al. (2010) show that the
labour market performance of immigrants’ descendants is worse compared to the natives
in France, Germany, and the UK. Silberman et al. (2007) find in France that groups who
come from former French colonies and/or are dominated by Muslims are substantially,
if not severely, disadvantaged in the process of labour market entry. Furthermore, the
descendants of immigrants of sub-Saharan African, North African, and Turkish origin
are at risk of experiencing labour market discrimination in France (Meurs et al. 2006).
Therefore, we could expect that those that have limited labour market opportunities may
have fewer incentives to continue to work after a child is born.
Third, access to family policies such as formal childcare and parental leave that help
to reduce work-family conflict may vary across descendant groups. Previous research has
found that the uptake of (in)formal childcare is substantially lower among immigrants
– especially non-European migrants – compared to the native population (Biegel et al.
2021;Schober and Spiess 2013;Wall and Jose 2004). These differences in the uptake of
childcare might be observed as well among the descendants of immigrants. Similarly, the
availability of social and family networks may vary across descendant groups. This might
be important to explain differences in the effect of childbirth on individuals’ employment
trajectories given that family members can help to act as support networks by taking over
childcare responsibilities.
6
2.3 Hypotheses
In the light of previous findings, we develop the following hypotheses. First, we expect
women’s professional careers to be negatively impacted by childbirth contrary to men’s
careers for whom we do not except to find any significant effect (Hypothesis 1). This
negative effect for women might be illustrated by lower rates of employment entry and
higher rates of employment exit for women with children compared to those without
children (Hypothesis 1a). We also expect this negative effect to increase as parity increases
(Hypothesis 1b). Second, we expect the negative effect among women to be stronger for
the descendants of immigrants than for native women (Hypothesis 2).
With regard to differences among origin groups, we expect this negative effect to be
more pronounced among the descendants of immigrants who come from origin countries
with more conservative norms, e.g., children of non-European immigrants, compared to
the descendants of immigrants who come from origin countries that are culturally close
to France such as the children of other European immigrants (Hypothesis 3). Lastly,
we also expect to detect unobserved time-constant co-determinants of childbearing and
employment (Hypothesis 4). While we expect the effect of childbirth on employment to
remain negative, an interesting question is how and to what extent taking into account
selection changes the effect of childbirth on employment trajectories.
3 Data and Methodology
The analysis is based on data from a rich French survey named Trajectories and Origins
which was collected in 2009. It contains information on a nationally representative sam-
ple of more than 20,000 individuals, including immigrants, immigrants’ descendants, and
French natives. For the purpose of this study, we focus on the descendants of immigrants
(including the 1.5G of immigrants who arrived to France before the age of 15) and the
French natives. We examine both men and women without imposing any restrictions on
age and study period. The final sample is composed of 10,886 immigrants’ descendants
(including 2,365 1.5G) and 3,462 French natives.
The survey contains retrospective biographical data with information on the employ-
ment and childbearing histories of individuals. More specifically, we have information on
the month and the year of each childbirth. We decide to subtract 7 months from the time
of birth since at this time, the majority of women are aware that they are pregnant, and
this knowledge may influence their subsequent employment trajectory and the one of their
partners. We also have yearly information on the employment status of individuals across
their life course. We convert the employment histories to a monthly format by assuming
that each event happens at the end of each year. To ensure that our results are robust,
we also conducted additional analyses (not reported in this study) where months were
assigned randomly to each employment event. We find similar results in both analyses.
Lastly, the survey also contains detailed information on individuals’ sociodemographic
7
characteristics such as gender, birth cohort, origin group, and educational level.
To study changes in the employment status of individuals across their life courses, we
estimate multilevel event history models. These models are an extension of conventional
event history models: rather than analysing a single employment transition, individuals
can move among different states. Regarding their employment status, individuals can
either be salaried, self-employed, in education, unemployed, housewife or other. To ensure
a reasonable number of events in each category, we regroup salaried and self-employed
individuals under the category “employed”. Individuals who are “out of employment”
are either “unemployed” or “inactive”. “Inactive” individuals are “housewife” or “other”.
It is important to note also that, in our framework, maternity/paternity leaves (which
typically last 16 and 6 weeks respectively in France) are not counted as employment exits.
We start by observing all individuals from the time they leave full-time education.
At this point, they can move to their first employment. For all individuals, once in
employment, they can go out of employment as unemployed or inactive. Individuals are
censored if they move to education or switch directly to another employment. Finally,
individuals who are out of employment can return to employment. Individuals are censored
if they move to education. To study the risk of a change in the employment status of
individuals by parity among men and women, we estimate three sets of processes: i) the
transition into first employment after leaving full-time education, ii) the transitions out
of employment, and iii) the transitions into second and higher-order employment. Each
of these transitions was specified as a hazard function as follows:
lnµF EN
i(t) = lnµ0(t) + X
j
αjxij +X
l
βlwil(t) + E N
i(1)
lnµEX
im (t) = lnµ0(t) + X
j
αjxijm +X
l
βlwilm(t) + E X
i(2)
lnµLEN
im (t) = lnµ0(t) + X
j
αjxijm +X
l
βlwilm(t) + E N
i(3)
where µF EN
i(t) denotes the hazard of first employment entry, µEX
i(t) is the hazard of mth
employment exit and µLEN
i(t) is the hazard of mth employment entry for individual i.
Considering equation (1), lnµ0(t) denotes the baseline log-hazard, which is specified as
piecewise constant. The baseline is time (in months) since leaving full-time education.
xij and wil(t) represent time-constant and time-varying characteristics, respectively, that
influence individuals’ propensities to change their employment status. For equations (2)
and (3), we estimate multilevel models because each individual can experience several
employment changes. For the outcomes of individuals who are out of employment, the
baseline is time (in months) since leaving employment; while for the outcomes of employed
individuals, the baseline is time (in months) since starting employment. For the three sets
of outcomes, our main covariate of interest is parity. Lastly, we include a joint random
effect for all employment entries (equations 1 and 3) denoted by EN
iand a separate
8
random effect for employment exits (equation 2) denoted by E X
i.
3.1 Joint model of employment changes and childbearing
Our explanatory variable – childbearing – is likely to be jointly determined with the
outcome of interest - employment. To address such concerns, we apply simultaneous-
equations hazard models (Lillard et al. 1995;Lillard and Panis 1996;Matysiak 2009;Kulu
and Steele 2013;Mikolai and Kulu 2018;Steele et al. 2005,2006). More specifically, we
estimate a joint model of employment changes and childbearing to detect and control for
individual-level unobserved factors, which may simultaneously influence both processes.
The models are as follows:
lnµF EN
i(t) = lnµ0(t) + X
j
αjxij +X
l
βlwil(t) + E N
i(4)
lnµEX
im (t) = lnµ0(t) + X
j
αjxijm +X
l
βlwilm(t) + E X
i(5)
lnµLEN
im (t) = lnµ0(t) + X
j
αjxijm +X
l
βlwilm(t) + E N
i(6)
lnµC1
i(t) = lnµ0(t) + X
j
αjxij +X
l
βlwil(t) + C
i(7)
lnµC2
im (t) = lnµ0(t) + X
j
αjxij +X
l
βlwil(t) + C
i(8)
lnµC3
im (t) = lnµ0(t) + X
j
αjxij +X
l
βlwil(t) + C
i(9)
where we have three additional equations. µC1
i(t) denotes the hazard of a first birth, µC2
i(t)
the hazard of a second birth and µC3
i(t) the hazard of a third birth for individual i. Each of
the hazard equations include a baseline log-hazard. The baseline log-hazard of first birth
was specified as age and the baseline log-hazard of second and third births was specified
as time since the birth of the previous child. The hazard equations also include a set of
time-constant and time-varying variables. Lastly, each of the hazard equations include a
random heterogeneity component which is person-specific. These individual-level random
effects aim to control for unmeasured time-constant characteristics that may influence
individuals’ likelihood of having a conception or of changing their employment status. We
assume that the residuals of the equations follow a joint bivariate normal distribution:
EN
i
EX
i
C
i
N
0
0
0
,
σ2
EN
i
ρEN
iEX
iρEN
iuC
i
ρEX
iEN
iσ2
EX
i
ρEX
iuC
i
ρuC
iEN
iρuC
iEX
iσ2
uC
i
(10)
9
where σ2
EN
i
,σ2
EX
i
and σ2
uC
i
denote the variances of the person-specific residuals of the
different processes, and ρis the correlation between the residuals. All models are estimated
via maximum likelihood using the aML software (Lillard and Panis 2003).
We estimate two models stepwise. First, we focus on the relationship between parity
and employment changes (Model 1). The first model is estimated twice; first (Model 1a) we
estimate single-process hazard models for the risk of first employment entry, employment
exits, and higher order employment entries and then (Model 1b) we estimate the multi-
process hazard model where we account for unobserved time-constant co-determinants of
the risk of employment entries, employment exits, and the risk of childbirth. Second,
we additionally examine whether and the extent to which the effect differs across origin
groups (Model 2). In other words, Model 2 includes an interaction term between parity and
origin group. Lastly, we report the results of Model 2 when accounting for the unobserved
time-constant co-determinants of childbearing and employment risks. For all models, we
analyse men and women separately.1
3.2 Variables
We include a number of variables in the models. First, respondents’ parity status is
treated as a time-varying variable using retrospective information on the year and month
of each birth, and is categorised as “childless”, “1 child”, “2 children”, and “3 or more
children”. For the first employment transition after leaving full-time education, we modify
this variable to have only two categories: “childless individuals” and “individuals with
children”. Partnership status is time-varying and is categorised as “single”, “cohabiting”,
“married” and “separated”. The categories “cohabiting” and “married” include both first
and higher order unions. Birth cohorts include 4 cohorts: 1948-1959 (reference), 1960-
1969, 1970-1979, and 1980-1999. Respondents’ educational level is categorized as low
(reference), medium, and high. The origin group for immigrants’ descendants includes
North Africa, Sub-Saharan Africa, South East Asia, Turkey, Southern Europe, and other
Europe. Lastly, we control for order of employment change and origin state. Individuals
can move into and out of employment several times. Besides, they can move out of
employment when being salaried or self-employed.
4 Results
Women and men experienced a large number of employment changes and births events.
We report the number and proportions in Tables A.1 and A.2 in Appendix. We show the
results of two event history models (Models 1-2) of the risk of a change in employment
status. To facilitate interpretation, Tables 1-2 show the relative risks for the key variables
of interest whereas Tables A.3-A.5 in Appendix report log-relative hazards for all variables
1We report the results of an additional model in the Appendix (Table A.4) where we also look at the
effect of time since birth on the employment changes of individuals.
10
in the models separately by gender.
4.1 Selection
We first estimate single-process hazard models for the risk of first employment entry,
employment exits, and higher order employment entries. We then model jointly the risk
of a change in the employment status and the risk of having a birth. The results are
reported in Table 1. We start our discussion by focusing on the estimates of the standard
deviations of the unobserved heterogeneity terms and their pairwise correlations. The
standard deviations of the person-specific residuals are significant in all models. This
implies that there is a significant portion of individual-specific heterogeneity that is not
accounted for by our covariates. It represents an individual-specific propensity to have
children in the fertility equations and an individual-specific propensity to work in the
employment equations. These results provide a justification for the need to estimate the
multi-process hazard model.
The person-specific unobserved heterogeneity terms are correlated. This means that
the hazards of birth and employment changes have unobserved co-determinants. Our re-
sults show that, for women, the unobserved characteristics that increase the propensity
to have a child are negatively correlated with the unobserved characteristics that increase
the propensity to enter employment and positively correlated with the unobserved char-
acteristics that increase the propensity to exit employment. We do not find significant
correlations for men. These findings indicate that women who are more likely to have a
child are more likely to leave employment and less likely to re-enter the labour market.
Due to the unobserved selection, the estimates of the impacts of fertility on employment
obtained from the single-process hazard models are biased, especially on the impact of
children on employment exits.
4.2 Impact of childbearing on employment
Our findings (reported in Table 1, Models 1a and 1b) show first some heterogeneity in
employment patterns across origin groups. Among women, all descendant groups except
the female descendants of Southern European immigrants are less likely to enter their first
employment spell compared to natives. Some descendant groups such as the descendants
of North African and Turkish immigrants are also more likely to go out of employment.
Among men, the male descendants of Sub-Saharan African immigrants are less likely than
other groups to enter their first employment spell. We also find that the male descendants
of Sub-Saharan African, North African and South East Asian immigrants are less likely
to experience higher order employment entries. The male descendants of North African
immigrants are more likely to exit employment, as opposed to the male descendants of
Southern European immigrants who are less likely to go out of the labour market.
Regarding parity, children have a strong and clearly negative impact on women’s em-
ployment. Mothers are less likely to take up a job than childless women do. They are also
11
more likely to exit employment following childbirth. We also find that as parity increases,
women are less likely to re-enter employment. Yet, the likelihood of exiting employment
decreases as parity increases. Regarding men, we find mixed results: while men are less
likely to enter employment following a birth, they are also less likely to exit employment.
The effect also seems lower in magnitude compared to women.
Table 1. Impact of childbearing on employment, relative risks (Model 1)
Women Men
Model 1a Model 1b Model 1a Model 1b
Single process Multi-process Single process Multi-process
RR Sig RR Sig RR Sig RR Sig
First employment entry
Parity
Childless (ref.) 1 1 1 1
With children 0.527 ∗∗∗ 0.536 ∗∗∗ 0.967 0.951
Origin Group
Native (ref.) 1 1 1 1
North Africa 0.710 ∗∗∗ 0.713 ∗∗∗ 0.909 ∗0.922
Sub-Saharan Africa 0.722 ∗∗∗ 0.720 ∗∗∗ 0.787 ∗∗∗ 0.793 ∗∗∗
South East Asia 0.897 0.902 1.049 1.071
Turkey 0.655 ∗∗∗ 0.654 ∗∗∗ 1.560 ∗∗∗ 1.655 ∗∗∗
Southern Europe 1.074 1.078 1.327 ∗∗∗ 1.359 ∗∗∗
Other Europe 0.815 ∗∗∗ 0.810 ∗∗∗ 1.063 1.066
Employment exits
Parity
Childless (ref.) 1 1 1 1
1 child 1.941 ∗∗∗ 1.692 ∗∗∗ 0.704 ∗∗∗ 0.691 ∗∗∗
2 children 2.028 ∗∗∗ 1.557 ∗∗∗ 0.780 ∗∗ 0.739 ∗∗
3+ children 2.368 ∗∗∗ 1.441 ∗∗∗ 1.047 1.011
Origin Group
Native (ref.) 1 1 1 1
North Africa 1.235 ∗∗∗ 1.305 ∗∗∗ 1.219 ∗∗∗ 1.166 ∗
Sub-Saharan Africa 0.938 1.075 1.143 1.132
South East Asia 0.973 0.986 1.055 1.000
Turkey 1.570 ∗∗∗ 1.761 ∗∗∗ 1.030 0.919
Southern Europe 0.851 ∗∗ 0.810 ∗∗∗ 0.824 ∗∗ 0.747 ∗∗∗
Other Europe 1.108 1.158 1.177 1.082
Higher order employment entries
Parity
Childless (ref.) 1 1 1 1
1 child 0.595 ∗∗∗ 0.607 ∗∗∗ 0.792 ∗∗∗ 0.798 ∗∗∗
2 children 0.488 ∗∗∗ 0.495 ∗∗∗ 0.641 ∗∗∗ 0.652 ∗∗∗
3+ children 0.379 ∗∗∗ 0.377 ∗∗∗ 0.681 0.646 ∗
Origin Group
Native (ref.) 1 1 1 1
North Africa 0.894 0.890 0.772 ∗∗ 0.750 ∗∗
Sub-Saharan Africa 0.863 0.839 0.721 ∗∗ 0.682 ∗∗
South East Asia 1.002 1.023 0.689 ∗∗ 0.695 ∗∗
Turkey 0.783 0.739 ∗0.957 1.031
Southern Europe 1.050 1.084 0.864 0.890
Other Europe 0.957 0.945 0.846 1.194
Unobserved heterogeneity
Standard deviation of residuals
Fertility 0.726 ∗∗∗ 0.730 ∗∗∗ 0.729 ∗∗∗ 0.732 ∗∗∗
Employment entry 0.618 ∗∗∗ 0.697 ∗∗∗ 0.656 ∗∗∗ 0.795 ∗∗∗
Employment exit 1.130 ∗∗∗ 1.307 ∗∗∗ 0.885 ∗∗∗ 1.205 ∗∗∗
Correlation between residuals
Fertility and employment entry -0.114 ∗∗∗ 0.095 ∗
Fertility and employment exit 0.302 ∗∗∗ -0.070
Employment entry and exit -0.658 ∗∗∗ -0.962 ∗∗∗
Source: Trajectories and Origins, authors’ own calculations.
Notes: See Table A3 in Appendix for the results of the full equations. * p < 0.10, ** p < 0.05, *** p < 0.01.
These conclusions can be drawn from the estimates obtained from the single-process
models as well as those from the multi-process hazard model. However, comparing the
12
results of the two models reveals that controlling for unobserved co-determinants reduces
the negative impact of parity on women’s employment. In other words, the negative effect
of childbirth on the risk of exiting or re-entering employment is overestimated (although
slightly) for women if we do not account for unobserved co-determinants of these two
processes. For men, we find that the negative effect of parity on employment entries is
marginally overestimated, although caution is needed when interpreting the results for
men given that they change between models.
Next, we examine possible differences across origin groups in the effect of parity on
the risks of employment entry and exit for both men and women (Table 2).
Table 2. Impact of childbearing on employment by origin, relative risks (Model 2)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Effects of parity on employment exits by origin group
Childless x North Africa 1.297 ∗∗ 0.913
Childless x Sub-Saharan Africa 1.082 0.725 ∗
Childless x South East Asia 1.125 0.946
Childless x Turkey 1.342 0.824
Childless x Southern Europe 0.898 0.775 ∗∗∗
Childless x Other Europe 1.022 1.099
Childless x Natives (ref.) 1 1
With children x North Africa 1.861 ∗∗∗ 0.935
With children x Sub-Saharan Africa 1.408 ∗∗∗ 1.019
With children x South East Asia 1.384 ∗0.583
With children x Turkey 2.886 ∗∗∗ 1.036
With children x Southern Europe 1.334 ∗∗∗ 0.503 ∗∗∗
With children x Other Europe 1.857 ∗∗∗ 0.564 ∗∗
With children x Natives 1.726 ∗∗∗ 0.452 ∗∗∗
Effects of parity on employment entries by origin group
Childless x North Africa 1.220 0.845
Childless x Sub-Saharan Africa 0.886 0.759
Childless x South East Asia 1.370 ∗0.712 ∗∗
Childless x Turkey 1.214 0.836
Childless x Southern Europe 1.395 ∗∗∗ 0.852
Childless x Other Europe 1.274 0.951
Childless x Natives (ref.) 1 1
With children x North Africa 0.708 ∗∗∗ 0.559 ∗∗∗
With children x Sub-Saharan Africa 1.025 0.688
With children x South East Asia 0.817 0.598 ∗
With children x Turkey 0.526 ∗∗∗ 0.882
With children x Southern Europe 0.811 0.653 ∗∗∗
With children x Other Europe 0.776 0.531 ∗∗∗
With children x Natives 0.628 0.658
Unobserved heterogeneity
Standard deviation of residuals
Fertility 0.658 ∗∗∗ 0.666 ∗∗∗
Employment entry 0.681 ∗∗∗ 0.796 ∗∗∗
Employment exit 1.322 ∗∗∗ 1.250 ∗∗∗
Correlation between residuals
Fertility and employment entry -0.302 ∗∗∗ 0.043
Fertility and employment exit 0.336 ∗∗∗ 0.004
Employment entry and exit -0.664 ∗∗∗ -0.936 ∗∗∗
Source: Trajectories and Origins, authors’ own calculations.
Notes: See Table A5 in the Appendix for the results of the full equations.
*p < 0.10, ** p < 0.05, *** p < 0.01.
In order to ensure a reasonable number of observations, we regroup all individuals who
have children under one category. Furthermore, it is important to note that, due to
limited sample size, the transition to first employment does not include an interaction
13
term between parity and origin group. The results show that, among women, all groups
are negatively affected by childbirth; yet the female descendants of Turkish immigrants are
the least likely to enter employment after childbirth and the most likely to exit employment
following childbirth. Among men, there is more heterogeneity: while the French natives
as well as the descendants of European immigrants are less likely to exit employment
after having a child, we do not find any significant relationship between childbirth and
employment exits for other descendant groups.
5 Conclusion and Discussion
This paper investigated the association between family formation and the labour market
changes of immigrants’ descendants in France. We examined whether and how the as-
sociation between fertility and employment differs by gender and migration background.
We contributed to the literature by simultaneously investigating the effect of childbearing
on all employment entries and exits in family ages among the descendants of immigrants.
Another novelty is that we also addressed the issue of selection on unobserved charac-
teristics, i.e. individuals who are more likely to have a child (or children) are more/less
likely to enter or exit employment. We applied a multi-process hazard model that allows
a correlation of person-specific error terms of fertility and employment transitions.
In line with our first hypothesis, our analysis reveals a strong negative impact of chil-
dren on women’s work. The arrival of the first child reduces the propensity of employment
entry and increases the risk of employment exit. Furthermore, higher order births further
reduce women’s likelihood of re-entering the labour market. We do not find evidence of a
stronger negative impact of childbirth on the employment of female descendants of immi-
grants compared to native women (Hypothesis 2). However, the effect of family formation
differs among immigrants’ descendants, which supports our third hypothesis. Although all
groups are negatively affected, the female descendants of Turkish immigrants are the most
likely to exit employment and the least likely to re-enter employment following childbirth.
Lastly, we have found that not controlling for unobservables led to an overestimation of
the negative effect of childbearing on women’s work, confirming our fourth hypothesis.
We expected to find no significant association between childbirth and men’s work. Yet,
we find that having a child also affects the employment trajectories of men: fathers are less
likely to exit employment but also less likely to re-enter employment following childbirth.
However, the direction of the effect differs across origin groups. There seems to be more
heterogeneity among men: while the French natives as well as the male descendants of
European immigrants are less likely to exit employment after having a child, we do not find
any significant association between childbirth and employment exits for other descendant
groups. Therefore, our results suggest that the association between childbirth and men’s
work is more complicated.
There are a number of potential explanations for the negative effect of childbirth on
14
employment of women. First, one factor that could play a role is the weak public support
for working parents. Yet, in France, it has been argued that this is less of an issue given
the existence of family-friendly policies such as free day-care facilities open all day long
and accommodating for children as young as 3 months old (Cukrowska-Torzewska 2017;
Lucifora et al. 2017). The French leave policy is also considered quite generous.
Still, there might be differences in the access to or use of childcare among individu-
als with a migration background compared to the native population. For instance, the
descendants of immigrants may hold an aversion towards formal childcare due to institu-
tional distrust. This was reported by second generation Moroccan immigrant mothers in
Flanders due to negative experiences such as discrimination as a school pupil, or negative
experiences as a childcare worker (Wood 2022). The descendants of immigrants might
then rely more on informal childcare than natives do but there again, the availability of
social and family networks may vary across groups. These potential differences in the use
of or access to childcare among immigrants’ descendants and natives might explain why
they have different labour market trajectories upon childbirth.
Another reason that could explain our result is that there is also a high instability of
employment contracts especially among minority populations (Algan et al. 2010;Meurs et
al. 2006). A history of unstable employment prior to family formation is likely to influence
individuals’ decision to remain or not in the labour market upon childbirth (Maes et al.
2021). It also has implications for access to family policies. Among other things, it affects
for instance the allowance that parents who partially/totally stop working in the labour
market to look after their children are entitled to. Therefore, having an unstable position
or poor employment conditions/prospects might explain why immigrants’ descendants are
more prone to leave the labour market upon childbirth compared to natives.
On a similar note, the descendants of immigrants are more prone to experience dis-
crimination in the labour market compared to the native population. In France, research
has found that the descendants of immigrants from Sub-Saharan Africa, North Africa and
Turkey are more likely to be victims of labour market discrimination (Meurs et al. 2006).
As a result of this, the children of immigrants may become demotivated to continue to
work and may consider family formation as a suitable alternative career.
Lastly, social norms are likely to play an important role. Although the descendants
of immigrants have been socialised in an egalitarian family context in France, they might
also remain influenced by parental attitudes, family networks and the wider migrant com-
munity. Previous studies have shown that some groups among the second generation of
immigrants in France continue to exhibit similar family patterns than those of their par-
ents rather than those of natives (Pailhe 2015,2017;Delaporte and Kulu 2022;Kulu et
al. 2021). Therefore, individuals who come from countries with more conservative values
might hold more strongly the perception that women are the main homemakers and care
providers. This could explain why for instance we find that the female descendants of
Turkish immigrants are the least likely to enter employment and the most likely to exit
employment after a birth.
15
This study contributes to the existing body of literature on gender and immigration
by analysing the interaction effect of motherhood, and migration background on women’s
and men’s labour market trajectories in France. More precisely, it sheds light on how
differently immigrant descendant and native men and women reconcile work with family
life in France.
References
Algan, Y., Dustmann, C., Glitz, A., & Manning, A. (2010). The Economic Situation of
First and Second-Generation Immigrants in France, Germany and the United Kingdom.
The Economic Journal, 120(542), F4-F30. https://doi.org/10.1111/j.1468-0297.
2009.02338.x
Andersson, G., Obu´cina, O., & Scott, K. (2015). Marriage and divorce of immigrants
and descendants of immigrants in Sweden. Demographic Research, 33, 31–64. https:
//doi.org/10.4054/DemRes.2015.33.2
Angelov, N., Johansson, P., & Lindahl, E. (2016). Parenthood and the gender gap in pay.
Journal of labor economics, 34(3), 545-579. https://doi.org/10.1086/684851
Angrist, J., & Evans, W. N. (1996). Children and their parents’ labor supply: Evidence
from exogenous variation in family size. https://doi.org/10.3386/w5778
Azmat, G., & Ferrer, R. (2017). Gender gaps in performance: Evidence from young
lawyers. Journal of Political Economy, 125(5), 1306-1355. https://doi.org/10.1086/
693686
Baranowska-Rataj, A., & Matysiak, A. (2022). Family Size and Men’s Labor Market Out-
comes: Do Social Beliefs About Men’s Roles in the Family Matter?. Feminist Economics,
1-26. https://doi.org/10.1080/13545701.2021.2015076
Begall, K., & Grunow, D. (2015). Labour force transitions around first childbirth in
the Netherlands. European Sociological Review, 31(6), 697-712. https://doi.org/10.
1093/esr/jcv068
Bertrand, Marianne, Claudia Goldin, and Lawrence F. Katz. 2010. Dynamics of the gender
gap for young professionals in the financial and corporate sectors. American economic
journal: applied economics 2, no. 3: 228-55. https://doi.org/10.1257/app.2.3.228
Biegel, N., Wood, J., & Neels, K. (2021). Migrant-native differentials in the uptake of (in)
formal childcare in Belgium: The role of mothers’ employment opportunities and care
availability. Journal of Family Research, 33(2), 467-508. https://doi.org/10.20377/
jfr-463
Clark, K., & Drinkwater, S. (2010). Patterns of ethnic self-employment in time and space:
evidence from British Census microdata. Small Business Economics, 34(3), 323-338.
https://doi.org/10.1007/s11187-008-9122-1
Clark, K., & Ochmann, N. (2022). Good job, bad job, no job? Ethnicity and employment
quality for men in the UK.
16
Cooke, L. P. (2014). Gendered family penalties and premiums across the wage distribution
in Australia, United Kingdom and the US. European Sociological Review, 30, 360–372.
https://doi.org/10.1093/esr/jcu044
Cools, S., Markussen, S., & Strøm, M. (2017). Children and careers: How family size
affects parents’ labor market outcomes in the long run. Demography, 54(5), 1773-1793.
https://doi.org/10.1007/s13524-017-0612-0
Cristia, J. P. (2008). The effect of a first child on female labor supply evidence from women
seeking fertility services. Journal of Human Resources, 43(3), 487-510.
Cukrowska-Torzewska, E. (2017). Cross-country evidence on motherhood employment and
wage gaps: the role of work-family policies and their interaction. Soc. Polit. Int. Stud.
Gender State Soc. 24, 178–220. https://doi.org/10.1093/sp/jxx004
Daniel, F. K., Lacuesta, A., & Rodr´ıguez-Planas, N. (2013). The motherhood earnings dip:
Evidence from administrative records. Journal of Human Resources, 48(1), 169-197.
Delaporte, I., & Kulu, H. (2022). Interaction between childbearing and partnership tra-
jectories among Immigrants and their descendants in France: An application of multi-
channel sequence analysis. Population Studies. https://doi.org/10.1080/00324728.
2022.2049856
Diehl, C., Koenig, M., and Ruckdeschel, K. (2009). Religiosity and gender equality: com-
paring natives and Muslim migrants in Germany. Ethn. Racial Stud. 32, 278–301.
https://doi.org/10.1080/01419870802298454
Evertsson, M. (2016), Parental leave and careers: Women’s and men’s wages after parental
leave in Sweden, Advances in Life Course Research, 29, 26–40. https://doi.org/10.
1016/j.alcr.2016.02.002
Evertsson, M. and A.-Z. Duvander (2011), Parental leave. Possibility or trap? Does family
leave length effect Swedish women’s labour market opportunities?, European Sociolog-
ical Review, 27 (4), 435–50. https://doi.org/10.1093/esr/jcq018
Ferrari, G., & Pailh´e, A. (2017). Transition to adulthood in France: Do children of im-
migrants differ from natives? Advances in Life Course Research, 31, 34-56. https:
//doi.org/10.1016/j.alcr.2016.10.001
Fiori, F., & Di Gessa, G. (2022). Influences on Employment Transitions around the Birth
of the First Child: The Experience of Italian Mothers. Work, Employment and Society,
09500170221082479. https://doi.org/10.1177/09500170221082479
Fitzenberger, B., Sommerfeld, K., & Steffes, S. (2013). Causal effects on employment
after first birth - A dynamic treatment approach. Labour Economics, 25, 49-62. https:
//doi.org/10.1016/j.labeco.2013.05.003
Fleischmann, F., & H¨ohne, J. (2013). Gender and migration on the labour market: Ad-
ditive or interacting disadvantages in Germany?. Social Science Research, 42(5), 1325-
1345. https://doi.org/10.1016/j.ssresearch.2013.05.006
17
Ganault, Jeanne, and Ariane Pailh´e. ”Work-life balance through the life course.” Re-
search Handbook on Work–Life Balance (2022): 72-87. https://doi.org/10.4337/
9781788976053.00011
Glauber, R. (2018). Trends in the motherhood wage penalty and fatherhood wage premium
for low, middle, and high earners. Demography, 55, 1663. https://doi.org/10.1007/
s13524-018-0712-5
Gonz´alez-Ferrer, A., Castro-Mart´ın, T., Kraus, E. K., & Eremenko, T. (2017). Childbear-
ing patterns among immigrant women and their daughters in Spain. Over-adaptation
or structural constraints? Demographic Research, 37, 599–634. https://doi.org/10.
4054/DemRes.2017.37.19
Guti´errez-Dom`enech, M. (2005a). Employment transitions after motherhood in Spain.
Labour, 19, 123-148. https://doi.org/10.1111/j.1467-9914.2005.00313.x
Gutierrez-Domenech, M. (2005b). Employment after motherhood: a European compari-
son. Labour economics, 12(1), 99-123. https://doi.org/10.1016/j.labeco.2004.04.
004
Hannemann, T., & Kulu, H. (2015). Union formation and dissolution among immigrants
and their descendants in the United Kingdom. Demographic Research, 33, 273–312.
https://doi.org/10.4054/DemRes.2015.33.10
Hannemann, T., Kulu, H., Gonz´alez-Ferrer, A., Pailh´e, A., Rahnu, L., & Puur, A. (2020).
Partnership dynamics among immigrants and their descendants in four European coun-
tries. Population, Space and Place, 26(5), e2315. https://doi.org/10.1002/psp.2315
Herrarte, A., J. Moral-Carcedo, and F. S´aez. 2012. The Impact of Childbirth on Spanish
Women’s Decisions to Leave the Labor Market. Review of Economics of the Household
10: 441–468.
Hodges, M.J. and M.J. Budig (2010), Who gets the baby bonus? Organizational hegemonic
masculinity and the impact of fatherhood on earnings, Gender and Society, 24 (6),
717–45. https://doi.org/10.1177/0891243210386729
Holland, J. A., & de Valk, H. A. (2017). Differences in labour force participation by mother-
hood status among second-generation Turkish and majority women across Europe. Pop-
ulation studies, 71(3), 363-378. https://doi.org/10.1080/00324728.2017.1319495
Hotz, V. J., Johansson, P., & Karimi, A. (2018). Parenthood, family friendly workplaces,
and the gender gaps in early work careers (No. w24173). National Bureau of Economic
Research. https://doi.org/10.3386/w24173
International Labour Organization. (2018). World employment and social outlook: Trends
for women 2018–Global snapshot.
Jacobsen, J. P., Pearce III, J. W., & Rosenbloom, J. L. (1999). The effects of childbearing
on married women’s labor supply and earnings: using twin births as a natural experi-
ment. Journal of Human Resources, 449-474. https://doi.org/10.2307/146376
18
Khoudja, Y., & Platt, L. (2018). Labour market entries and exits of women from different
origin countries in the UK. Social science research, 69, 1-18. https://doi.org/10.
1016/j.ssresearch.2017.10.003
Kil, T., K. Neels, J. Wood, and H. de Valk. 2018. Employment After Parenthood: Women
of Migrant Origin and Natives Compared. European Journal of Population. https:
//doi.org/10.1007/s10680-017-9431-7
Klepinger, D., Lundberg, S., & Plotnick, R. (1999). How does adolescent fertility affect
the human capital and wages of young women?. Journal of human resources, 421-448.
https://doi.org/10.2307/146375
Kleven, H., & Landais, C. (2017). Gender inequality and economic development: fertility,
education and norms. Economica, 84(334), 180-209. https://doi.org/10.1111/ecca.
12230
Kleven, H., Landais, C., Søgaard, J.E., 2019a. Children and gender inequality: evidence
from Denmark. Am. Econ. J. Appl. Econ. 11 (4), 181–209. https://doi.org/10.1257/
app.20180010.
Kleven, H., Landais, C., Posch, J., Steinhauer, A., Zweim¨uller, J., 2019b. The impact of
family policies on the dynamics of gender inequality. Mimeo.
Kleven, H., Landais, C., Posch, J., Steinhauer, A., Zweim¨uller, J., 2019c. Child penal-
ties across countries: evidence and explanations. Proceedings of the AEA Papers 109,
122–126.
Kuhnt, A., & Krapf, S. (2020). Partnership living arrangements of immigrants and natives
in Germany. Frontiers in Sociology, 5(538977). https://doi.org/10.3389/fsoc.2020.
538977
Kulu, H., Steele, F. 2013. Interrelationships between childbearing and housing transitions
in the family life course. Demography, 50(5), 1687-1714. https://doi.org/10.1007/
s13524-013-0216-2
Kulu, H., & Hannemann, T. (2016a). Introduction to research on immigrant and ethnic
minority families in Europe. Demographic Research, 35, 31–46. https://doi.org/10.
4054/DemRes.2016.35.2
Kulu, H., & Hannemann, T. (2016b). Why does fertility remain high among certain UK-
born ethnic minority women? Demographic Research, 35, 1441–1488. https://doi.
org/10.4054/DemRes.2016.35.49
Kulu, H., Mikolai, J., Delaporte, I., Liu, C., & Andersson, G. 2021. Family trajectories
among immigrants and their descendants in three European countries.
Krapf, M., Ursprung, H.W., Zimmermann, C., 2017. Parenthood and productivity of
highly skilled labor: evidence from the groves of academe. Journal of Economic Behavior
& Organization https://doi.org/10.1016/j.jebo.2017.05.010
Krapf, S., & Wolf, K. (2016). Persisting differences or adaptation to German fertility
patterns? First and second birth behavior of the 1.5 and second generation Turkish mi-
grants in Germany. In Social Demography Forschung an der Schnittstelle von Soziologie
19
und Demografie (pp. 137-164). Springer VS, Wiesbaden. https://doi.org/10.1007/
s11577-015-0331-8
Kreyenfeld, M. (2015). Maternal and paternal employment across the life course. Emerg-
ing trends in the social and behavioral sciences: An interdisciplinary, searchable, and
linkable resource, 1-15. https://doi.org/10.1002/9781118900772.etrds0210
Lacroix, J., & Vidal-Coso, E. (2019). Differences in labor supply by birthplace and family
composition in Switzerland: The role of human capital and household income. Journal of
International Migration and Integration, 20(3), 659-684. https://doi.org/10.1007/
s12134-018-0623-8
Lillard, L. A., Brien, M. J., & Waite, L. J. (1995). Premarital cohabitation and subsequent
marital dissolution: A matter of self-selection. Demography, 32, 437–457. https://doi.
org/10.2307/2061690
Lillard, L. A., & Panis, C. W. (1996). Marital status and mortality: The role of health.
Demography, 33(3), 313-327. https://doi.org/10.2307/2061764
Lillard, L. A., & Panis, C. W. (2003). aML Version 2. User’s Guide and Reference Manual.
EconWare, Los Angeles.
Liu, C., & Kulu, H. (2021). First comes marriage or first comes carriage? Family trajec-
tories for immigrants in Germany.
Loughran, David S., and Julie M. Zissimopoulos. Why wait? The effect of marriage and
childbearing on the wages of men and women. Journal of Human resources 44, no. 2
(2009): 326-349. https://doi.org/10.1353/jhr.2009.0032
Lucifora, C., Meurs, D., and Villar, E. (2017). Children, Earnings and Careers in an
Internal Labor Market. Available online at: http://www.aiel.it/cms/cms-files/
submission/all20170831111248.pdf
Lundberg, S., & Rose, E. (2000). Parenthood and the earnings of married men and women.
Labour Economics, 7(6), 689-710. https://doi.org/10.1016/S0927-5371(00)00020-
8
Lundborg, P., Plug, E., & Rasmussen, A. W. (2017). Can women have children and a
career? IV evidence from IVF treatments. American Economic Review, 107(6), 1611-
37. https://doi.org/10.1257/aer.20141467
Maes, J., Wood, J., & Neels, K. (2021). Path-Dependencies in Employment Trajectories
Around Motherhood: Comparing Native Versus Second-Generation Migrant Women in
Belgium. Journal of International Migration and Integration, 1-64. https://doi.org/
10.1007/s12134-020-00801-1
Matysiak, A. (2009). Employment first, then childbearing: Women’s strategy in post-
socialist Poland. Population studies, 63(3), 253-276. https://doi.org/10.1080/
00324720903151100
Matysiak, A., & Cukrowska-Torzewska, E. (2021). Gender and labour market outcomes.
In Research Handbook on the Sociology of the Family. Edward Elgar Publishing.
20
Meurs, D., Pailh´e, A., & Simon, P. (2006). The persistence of intergenerational inequalities
linked to immigration: Labour market outcomes for immigrants and their descendants
in France. Population, 61(5), 645-682. https://doi.org/10.3917/popu.605.0763
Michaud, P. C., & Tatsiramos, K. (2011). Fertility and female employment dynamics in
Europe: the effect of using alternative econometric modeling assumptions. Journal of
Applied Econometrics, 26(4), 641-668. https://doi.org/10.1002/jae.1133
Mikolai, J., & Kulu, H. (2018). Divorce, separation, and housing changes: A multiprocess
analysis of longitudinal data from England and Wales. Demography, 55(1), 83-106.
https://doi.org/10.1007/s13524-017-0640-9
Mikolai, J., & Kulu, H. (2021). The intersection of partnership and fertility trajectories
of immigrants and their descendants in the United Kingdom: A multilevel multistate
event history approach.
Milewski, N. (2007). First child of immigrant workers and their descendants in West Ger-
many: Interrelation of events, disruption, or adaptation? Demographic Research, 17,
859–896. https://doi.org/10.4054/DemRes.2007.17.29
Miller, A. R. (2011). The effects of motherhood timing on career path. Journal of Popu-
lation Economics, 24(3), 1071-1100. https://doi.org/10.1007/s00148-009-0296-x
Nieto, A. (2021). Native-immigrant differences in the effect of children on the gender pay
gap. Journal of Economic Behavior & Organization, 183, 654-680. https://doi.org/
10.1016/j.jebo.2021.01.015
OECD/European Union (2021), Gender wage gap, https://www.oecd.org/gender/.
Pailh´e, A. (2015). Partnership dynamics across generations of immigration in France:
Structural vs. cultural factors. Demographic Research, 33, 451–498. https://doi.org/
10.4054/DemRes.2015.33.16
Pailh´e, A. (2017). The convergence of second-generation immigrants’ fertility patterns in
France: The role of sociocultural distance between parents’ and host country. Demo-
graphic Research, 36, 1361–1398. https://doi.org/10.4054/DemRes.2017.36.45
Paull, G. (2008). Children and women’s hours of work. The Economic Journal, 118(526),
F8-F27. https://doi.org/10.1111/j.1468-0297.2007.02114.x
Piton, C., & Rycx, F. (2020). A broken social elevator? Employment outcomes of first-
and second-generation immigrants in Belgium. Essen, Germany. https://doi.org/10.
1007/s10645-021-09385-2
Puhani, P.A. and K. Sonderhof (2011). The effects of parental leave extension on training
for young women. Journal of Population Economics, 24 (2), 731–60. https://doi.org/
10.1007/s00148\OT1\textendash009\OT1\textendash0295-y
R¨oder, A., and M¨uhlau, P. (2014). Are they acculturating?: Europe’s immigrants and
gender egalitarianism. Soc. Forces 92, 899–928. https://doi.org/10.1093/sf/sot126
21
Rudman, L.A. and K. Mescher (2013), Penalizing men who request a family leave: Is
flexibility stigma a femininity stigma?, Journal of Social Issues, 69 (2), 322–40. https:
//doi.org/10.1111/josi.12017
Schober, P. S., & Spiess, C. K. (2013). Early childhood education activities and care
arrangements of disadvantaged children in Germany. Child Indicators Research, 6(4),
709-735. https://doi.org/10.1007/s12187-013-9191-9
Seibel, V., & Hedegaard, T. F. (2017). Migrants’ and natives’ attitudes to formal childcare
in the Netherlands, Denmark and Germany. Children and Youth Services Review, 78,
112-121. https://doi.org/10.1016/j.childyouth.2017.05.017
Sieppi, A., & Pehkonen, J. (2019). Parenthood and gender inequality: Population-based
evidence on the child penalty in Finland. Economics letters, 182, 5-9. https://doi.
org/10.1016/j.econlet.2019.05.034
Silberman, R., Alba, R., & Fournier, I. (2007). Segmented assimilation in France? Discrim-
ination in the labour market against the second generation. Ethnic and Racial studies,
30(1), 1-27. https://doi.org/10.1080/01419870601006488
Steele, F., Kallis, C., Goldstein, H., & Joshi, H. (2005). The relationship between child-
bearing and transitions from marriage and cohabitation in Britain. Demography, 42(4),
647-673.
Steele, F., Joshi, H., Kallis, C., & Goldstein, H. (2006). Changing compatibility of cohabi-
tation and childbearing between young British women born in 1958 and 1970. Population
Studies, 60, 137–152. https://doi.org/10.1080/00324720600598009
Van Landschoot, L., de Valk, H. A. G., & Van Bavel, J. (2017). Fertility among descen-
dants of immigrants in Belgium: The role of the partner. Demographic Research, 36,
1827–1858. https://doi.org/10.4054/DemRes.2017.36.60
Vidal-Coso, E. (2019). Female employment following childbirth: Differences between na-
tive and immigrant women in Switzerland. Journal of Ethnic and Migration Studies,
45(9), 1667-1692. https://doi.org/10.1080/1369183X.2018.1444983
Wall, K., & Jose, J. S. (2004). Managing work and care: A difficult challenge for immigrant
families. Social policy & administration, 38(6), 591-621. https://doi.org/10.1111/j.
1467-9515.2004.00409.x
Wilner, Lionel. Worker-firm matching and the parenthood pay gap: Evidence from
linked employer-employee data. Journal of Population Economics 29.4 (2016): 991-1023.
https://doi.org/10.1007/s00148-016-0597-9
Wood, J. (2022). Childcare ideals among second generation Muslim Moroccan immigrants
in Flanders. Community, Work & Family, 1-18. https://doi.org/10.1080/13668803.
2022.2086036
Wood, J., Neels, K., De Wachter, D., & Kil, T. (2016). Family formation and labour force
participation. Population, 71(1), 53-81.
22
Appendix A.
Table A.1. Number and proportions of person-months and events by sets of outcomes
and categories of variables
Women Men
Person-Months Events Person-Months Events
Number % Number % Number % Number %
First employment entry
Time since in education
0-1 year 29411 17 4165 72 27609 22 3740 69
1-3 years 33260 19 789 14 25325 20 1014 19
3-5 years 22223 13 278 5 14379 12 240 4
5+ years 92209 52 515 9 56247 46 401 7
Birth ohort
1948-1959 36247 20 823 14 13407 11 757 14
1960-1969 66290 37 1689 29 49984 40 1486 28
1970-1979 49472 28 2005 35 41749 34 1811 34
1980-1999 25094 14 1230 21 18420 15 1341 25
Partnership status
Single 64951 37 3650 64 76688 62 4183 78
Cohabiting 23125 13 947 16 15609 13 638 12
Married 73354 41 861 15 23090 19 413 8
Separated 15674 9 289 5 8174 7 161 3
Parity
Childless 83826 47 4993 87 94846 77 5008 93
With children 93277 53 754 13 28715 23 387 7
Educational level
Low 80075 45 861 15 36561 30 946 18
Medium 78058 44 2964 52 62085 50 2979 55
High 18970 11 1922 33 24913 20 1470 27
Origin group
Native 35554 20 1592 28 33534 27 1386 26
North Africa 58930 33 1229 21 32612 26 1064 20
Sub-Saharan Africa 13693 8 360 6 8449 7 330 6
South East Asia 6994 4 320 6 6887 6 342 6
Turkey 10946 6 220 4 3906 3 272 5
Southern Europe 31313 18 1440 25 26006 21 1446 27
Other Europe 13889 8 388 7 7910 6 388 7
Total 177103 100 5747 100 123560 100 5395 100
Employment exits
Time since previous employment
0-1 year 90773 11 329 12 86965 10 223 12
1-3 years 145729 17 693 26 138185 16 618 34
3-5 years 113854 14 461 17 107987 13 326 18
5+ years 484327 58 1211 45 509486 60 626 35
Birth cohort
1948-1959 235795 28 559 21 259957 31 401 22
1960-1969 338041 40 938 35 321567 38 605 34
1970-1979 217752 26 914 34 206696 25 552 31
1980-1999 43095 5 283 11 54404 6 235 13
Partnership status
Single 196925 24 459 17 246979 29 900 50
Cohabiting 135400 16 473 18 139148 17 250 14
Married 387062 46 1413 52 372557 44 404 23
Separated 115296 14 349 13 83939 10 239 13
Parity
Childless 348665 42 861 32 405428 48 1230 69
1 child 197950 24 802 30 158201 19 199 11
2 children 202354 24 677 25 180194 21 207 12
3+ children 85714 10 354 13 98800 12 157 9
Educational level
Low 144390 17 640 24 186593 22 524 29
Medium 449340 54 1616 60 477981 57 1038 58
High 240953 29 438 16 178049 21 231 13
24
Table A.1. Number and proportions of person-months and events by sets of outcomes
and categories of variables (Continued)
Women Men
Person-Months Events Person-Months Events
Number % Number % Number % Number %
Origin group
Native 281032 34 783 29 260609 31 460 26
North Africa 141905 17 605 22 136815 16 413 23
Sub-Saharan Africa 31325 4 111 4 31645 4 88 5
South East Asia 34527 4 105 4 36604 4 88 5
Turkey 17759 2 136 5 27500 3 71 4
Southern Europe 244096 29 683 25 268906 32 478 27
Other Europe 61979 7 181 7 62600 7 138 8
Order
1 672962 81 1934 72 632158 75 1289 72
2 114258 14 558 21 139062 17 335 19
3+ 47463 6 202 7 71403 8 169 9
Type of employment
Salaried 799304 96 2609 97 772092 92 1708 95
Self-employed 35379 4 85 3 70531 8 85 5
Total 834683 100 2694 100 842623 100 1793 100
Higher order employment entries
Time since out of employment
0-1 year 59192 16 672 22 47244 20 822 29
1-3 years 76334 21 1017 33 45062 19 1018 36
3-5 years 48697 13 383 12 27235 12 218 8
5+ years 180544 49 993 32 117010 49 736 26
Birth cohort
1948-1959 109673 30 532 17 47283 20 385 14
1960-1969 140282 38 1076 35 101317 43 989 35
1970-1979 86609 24 1044 34 68418 29 1052 38
1980-1999 28203 8 413 13 19533 8 368 13
Partnership status
Single 70985 19 835 27 98007 41 1367 49
Cohabiting 42852 12 498 16 36146 15 522 19
Married 207177 57 1280 42 77944 33 654 23
Separated 43753 12 452 15 24454 10 251 9
Parity
Childless 97934 27 1309 43 139391 59 1970 71
1 child 74029 20 645 21 33990 14 305 11
2 children 96780 27 668 22 40770 17 329 12
3+ children 96024 26 443 14 22401 9 190 7
Educational level
Low 133274 37 659 22 70825 30 586 21
Medium 193143 53 1801 59 131732 56 1590 57
High 38350 11 605 20 33994 14 618 22
Origin group
Native 104789 29 793 26 69033 29 757 27
North Africa 86682 24 705 23 51746 22 588 21
Sub-Saharan Africa 16387 4 164 5 10507 4 146 5
South East Asia 12877 4 145 5 12116 5 150 5
Turkey 16432 5 132 4 6334 3 95 3
Southern Europe 89752 25 798 26 62479 26 726 26
Other Europe 27827 8 220 7 18795 8 235 8
Order
1 273396 75 2195 72 175814 74 2050 73
2 68643 19 926 20 48379 20 550 20
3+ 22728 6 244 8 12358 5 194 7
Type of out of employment
Unemployed 73818 20 1209 39 60107 25 1043 37
Housewife 202147 55 1102 36 2076 1 20 0.7
Other 88802 24 754 25 174368 74 1731 62
Total 364767 100 3065 100 236551 100 2794 100
Source: Social Protection Survey of Chile, authors’ own calculations.
Notes: the proportions by origin group do not equal to 100 when we sum up because there is around 3% of
the sample that belong to other smaller origin groups.
25
Table A.2. Number and proportions of person-months and events by sets of outcomes
and categories of variables
Women Men
Person-Months Events Person-Months Events
Number % Number % Number % Number %
First birth
Age
15-19 year 432585 44 599 14 401852 39 129 4
20-24 years 293506 30 1750 40 305411 30 923 28
25-29 years 137728 14 1440 33 172850 17 1384 43
30-34 years 54954 6 468 11 74510 7 625 19
35+ years 53750 6 115 3 63737 6 190 6
Birth cohort
1948-1959 132116 14 753 17 141519 14 669 21
1960-1969 273985 28 1545 35 291052 29 1226 38
1970-1979 319226 33 1587 36 329635 32 1147 35
1980-1999 247197 25 487 11 256122 25 209 6
Partnership status
Single 742545 76 721 16 805263 79 446 14
Cohabiting 101992 10 1200 27 98952 10 964 30
Married 80836 8 2285 52 62228 6 1705 52
Separated 47151 5 166 4 51885 5 136 4
Educational level
Low 440943 45 1028 24 462941 45 703 22
Medium 392441 40 2283 52 436094 43 1842 57
High 139140 14 1061 24 119292 12 706 22
Origin group
Native 252575 30 1254 29 259603 25 921 28
North Africa 219542 23 918 21 210187 21 591 18
Sub-Saharan Africa 71571 7 241 6 69024 7 159 5
South East Asia 62093 6 207 5 77759 8 150 5
Turkey 35070 4 180 4 36557 4 152 5
Southern Europe 226771 23 1119 26 255519 25 956 29
Other Europe 67563 7 320 7 73172 7 240 7
Employment status
Abroad 8187 0.8 38 0.9 8047 0.8 23 0.7
Salaried 318148 33 2812 64 367528 36 2234 69
Self-employed 8835 0.9 62 1 17585 2 132 4
Unemployed 24581 3 153 3 25415 2 57 2
Student 537326 55 599 14 483873 48 238 7
Housewife 14572 1 323 7 1059 0.1 2 0
Inactive 21718 2 136 3 47274 5 372 11
Other 972524 100 4372 100 1018328 100 3251 100
Total
Second birth
Time since previous birth
0-1 year 50314 19 312 10 37305 20 220 10
1-3 years 72883 27 1526 50 52913 29 1168 52
3-5 years 38824 15 722 24 26404 14 509 23
5+ years 103595 39 506 17 66516 36 354 16
Birth cohort
1948-1959 69046 26 599 20 58673 32 537 24
1960-1969 111378 42 1185 39 73457 40 965 43
1970-1979 70189 26 1102 36 45321 25 688 31
1980-1999 15003 6 180 6 5687 3 61 3
Age at birth 1
15-19 year 24115 9 368 12 5932 3 72 3
20-24 years 108131 41 1282 42 49181 27 649 29
25-29 years 87047 33 1059 35 77618 42 993 44
30-34 years 34067 13 292 10 37898 21 429 19
35+ years 10684 4 42 1 11856 6 102 5
Partnership status
Single 22435 8 132 4 8522 5 56 2
Cohabiting 51422 19 580 19 43157 24 483 21
Married 142247 54 2168 71 107648 59 1611 72
Separated 49513 19 186 6 23810 13 101 4
26
Table A.2. Number and proportions of person-months and events by sets of outcomes
and categories of variables (Continued)
Women Men
Person-Months Events Person-Months Events
Number % Number % Number % Number %
Educational level
Low 59441 22 671 22 42037 23 478 21
Medium 149062 56 1645 54 109083 60 1254 56
High 57113 22 750 24 32018 17 519 23
Origin group
Native 86799 33 902 29 59385 32 642 29
North Africa 47266 18 641 21 26163 14 406 18
Sub-Saharan Africa 12560 5 143 5 6473 4 104 5
South East Asia 11356 4 133 4 5692 3 100 4
Turkey 6282 2 129 4 5552 3 115 5
Southern Europe 72829 27 817 27 60396 33 676 30
Other Europe 21404 8 215 7 14930 8 161 7
Employment status
Abroad 1113 0.4 22 0.7 798 0.4 16 0.7
Salaried 171763 65 1882 61 128451 70 1604 71
Self-employed 5816 2 59 2 11000 6 152 7
Unemployed 9059 3 69 2 3458 2 31 1
Student 9759 4 81 3 4803 3 51 2
Housewife 32350 12 643 21 172 0 4 0
Inactive 12999 5 128 4 6779 4 70 3
Other 22757 9 182 6 27677 15 323 14
Total 265615 100 3066 100 183138 100 2251 100
Third birth
Time since previous birth
0-1 year 36336 12 129 11 26654 12 94 11
1-3 years 59297 20 520 43 43811 20 371 43
3-5 years 42638 14 308 25 31088 14 221 25
5+ years 161913 54 266 22 115987 53 182 21
Birth cohort
1948-1959 106320 35 286 23 89583 41 236 27
1960-1969 130934 44 492 40 94046 43 399 46
1970-1979 57098 19 411 34 32422 15 221 25
1980-1999 5833 2 34 3 1489 0.7 12 1
Age at birth 1
15-19 year 34154 11 236 19 5565 3 50 6
20-24 years 139669 40 588 48 76662 35 316 36
25-29 years 99872 33 324 26 95921 44 338 39
30-34 years 21763 7 52 4 31898 15 137 16
35+ years 3082 1 7 0.6 6433 3 23 3
Partnership status
Single 8179 3 33 3 2741 1 14 2
Cohabiting 32416 11 139 11 28298 13 129 15
Married 216455 72 956 78 165360 76 662 76
Separated 43135 14 95 8 21140 10 63 7
Educational level
Low 64919 22 376 31 45474 21 246 28
Medium 172683 58 613 50 128040 59 447 51
High 62583 21 234 19 44026 20 175 20
Origin group
Native 112153 37 330 27 77172 35 221 25
North Africa 40357 13 329 27 24578 11 207 24
Sub-Saharan Africa 7924 3 66 5 5746 3 54 6
South East Asia 8541 3 54 4 7494 3 42 5
Turkey 6540 2 67 5 5576 3 58 7
Southern Europe 96956 32 242 20 77841 36 207 24
Other Europe 20690 7 98 8 15333 7 65 7
27
Table A.2. Number and proportions of person-months and events by sets of outcomes
and categories of variables (Continued)
Women Men
Person-Months Events Person-Months Events
Number % Number % Number % Number %
Employment status
Abroad 935 0.3 13 1 596 0.3 6 0.7
Salaried 180233 60 542 44 146481 67 619 71
Self-employed 11170 4 24 2 20986 10 59 7
Unemployed 7080 2 30 2 3365 2 16 2
Student 50059 17 442 36 218 0.1 2 0.2
Housewife 50059 17 442 36 218 0.1 2 0.2
Inactive 11439 4 59 5 5592 3 33 4
Other 35872 12 97 8 38014 17 116 13
Total 300185 100 1223 100 217539 100 868 100
Source: Social Protection Survey of Chile, authors’ own calculations.
Notes: the proportions by origin group do not equal to 100 when we sum up because there is around 3% of
the sample that belong to other smaller origin groups.
28
Table A.3. Impact of childbearing on employment, log-relative risks (Model 1, full
specification)
Women Men
Model 1a Model 1b Model 1a Model 1b
Single process Multi-process Single process Multi-process
RR Sig RR Sig RR Sig RR Sig
First employment entry
Constant -1.293 ∗∗∗ -1.364 ∗∗∗ -1.108 ∗∗∗ -1.212 ∗∗∗
Time since in education
0-1 year (slope) -0.260 ∗∗∗ -0.252 ∗∗∗ -0.216 ∗∗∗ -0.203 ∗∗∗
1-3 years (slope) 0.032 ∗∗∗ 0.031 ∗∗∗ 0.025 ∗∗∗ 0.027 ∗∗∗
3-5 years (slope) -0.026 ∗∗∗ -0.025 ∗∗∗ -0.061 ∗∗∗ -0.058 ∗∗∗
5+ years (slope) 0.001 0.001 0.004 ∗∗∗ 0.005 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0 0 0
1960-1969 -0.246 ∗∗∗ -0.238 ∗∗∗ -0.455 ∗∗∗ -0.471 ∗∗∗
1970-1979 -0.217 ∗∗∗ -0.200 ∗∗∗ -0.434 ∗∗∗ -0.443 ∗∗∗
1980-1999 -0.357 ∗∗∗ -0.340 ∗∗∗ -0.296 ∗∗∗ -0.274 ∗∗∗
Partnership status
Single (ref.) 0 0 0 0
Cohabiting 0.095 ∗∗ 0.080 0.389 ∗∗∗ 0.413 ∗∗∗
Married -0.187 ∗∗∗ -0.223 ∗∗∗ 0.281 ∗∗∗ 0.308 ∗∗∗
Separated 0.350 ∗∗∗ 0.338 ∗∗∗ 0.136 0.189 ∗
Parity
Childless (ref.) 0 0 0 0
With children -0.641 ∗∗∗ -0.624 ∗∗∗ 0.034 -0.050
Educational level
Low (ref.) 0 0 0 0
Medium 0.746 ∗∗∗ 0.773 ∗∗∗ 0.358 ∗∗∗ 0.373 ∗∗∗
High 1.138 ∗∗∗ 1.179 ∗∗∗ 0.391 ∗∗∗ 0.430 ∗∗∗
Origin group
Native (ref.) 0 0 0 0
North Africa -0.342 ∗∗∗ -0.338 ∗∗∗ -0.095 ∗-0.081
Sub-Saharan Africa -0.326 ∗∗∗ -0.328 ∗∗∗ -0.239 ∗∗∗ -0.232 ∗∗∗
South East Asia -0.109 -0.103 0.048 0.069
Turkey -0.423 ∗∗∗ -0.425 ∗∗∗ 0.445 ∗∗∗ 0.504 ∗∗∗
Southern Europe 0.071 0.075 0.283 ∗∗∗ 0.307 ∗∗∗
Other Europe -0.205 ∗∗∗ -0.211 ∗∗∗ 0.061 0.064
Employment exits
Constant -9.977 ∗∗∗ -9.919 ∗∗∗ -8.961 ∗∗∗ -8.994 ∗∗∗
Time since previous employment
0-1 year (slope) 0.313 ∗∗∗ 0.317 ∗∗∗ 0.329 ∗∗∗ 0.332 ∗∗∗
1-3 years (slope) -0.021 ∗∗∗ -0.018 ∗∗∗ -0.044 ∗∗∗ -0.040 ∗∗∗
3+ years (slope) -0.001 ∗∗ 0.001 0.001 ∗∗∗ 0.002 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0 0 0
1960-1969 0.258 ∗∗∗ 0.297 ∗∗∗ 0.173 ∗0.335 ∗∗∗
1970-1979 0.749 ∗∗∗ 0.796 ∗∗∗ 0.472 ∗∗∗ 0.615 ∗∗∗
1980-1999 1.365 ∗∗∗ 1.487 ∗∗∗ 0.610 ∗∗∗ 0.885 ∗∗∗
Partnership status
Single (ref.) 0 0 0 0
Cohabiting 0.346 ∗∗∗ 0.403 ∗∗∗ -0.508 ∗∗∗ -0.536 ∗∗∗
Married 0.482 ∗∗∗ 0.643 ∗∗∗ -0.875 ∗∗∗ -0.901 ∗∗∗
Separated 0.275 ∗∗∗ 0.323 ∗∗∗ -0.004 -0.089
Parity
Childless (ref.) 0 0 0 0
1 child 0.663 ∗∗∗ 0.526 ∗∗∗ -0.508 ∗∗∗ -0.536 ∗∗∗
2 children 0.707 ∗∗∗ 0.443 ∗∗∗ -0.249 ∗∗ -0.303 ∗∗
3+ children 0.862 ∗∗∗ 0.365 ∗∗∗ -0.046 0.011
Educational level
Low (ref.) 0 0 0 0
Medium -0.298 ∗∗∗ -0.512 ∗∗∗ -0.326 ∗∗∗ -0.450 ∗∗∗
High -1.194 ∗∗∗ -1.495 ∗∗∗ -0.901 ∗∗∗ -1.084 ∗∗∗
29
Table A.3. Impact of childbearing on employment, log-relative risks (Model 1, full
specification) (Continued)
Women Men
Model 1a Model 1b Model 1a Model 1b
Single process Multi-process Single process Multi-process
RR Sig RR Sig RR Sig RR Sig
Origin group
Native (ref.) 0 0 0 0
North Africa 0.211 ∗∗∗ 0.266 ∗∗∗ 0.198 ∗∗ 0.154 ∗
Sub-Saharan Africa -0.064 0.072 0.134 0.124
South East Asia -0.027 -0.014 0.054 -0.001
Turkey 0.451 ∗∗∗ 0.566 ∗∗∗ 0.030 -0.084
Southern Europe -0.161 ∗∗ -0.211 ∗∗∗ -0.194 ∗∗ -0.292 ∗∗∗
Other Europe 0.103 0.147 0.163 0.079
Order
1 (ref.) 0 0 0 0
2 0.095 0.293 ∗∗∗ 0.181 ∗∗∗ 0.317 ∗∗∗
3+ -0.614 ∗∗∗ -0.272 ∗∗∗ -0.046 0.241 ∗∗
Type of employment
Salaried (ref.) 0 0 0 0
Self-employed -0.412 ∗∗∗ -0.432 ∗∗∗ -0.436 ∗∗∗ -0.447 ∗∗∗
Higher order employment entries
Constant -7.765 ∗∗∗ -7.718 ∗∗∗ -8.876 ∗∗∗ -8.792 ∗∗∗
Time since out of employment
0-1 year (slope) 0.312 ∗∗∗ 0.315 ∗∗∗ 0.488 ∗∗∗ 0.498 ∗∗∗
1-3 years (slope) -0.020 ∗∗∗ -0.019 ∗∗∗ -0.081 ∗∗∗ -0.076 ∗∗∗
3-5 years (slope) -0.012 ∗∗∗ -0.011 ∗∗∗ -0.018 ∗∗∗ -0.018 ∗∗∗
5+ years (slope) 0.005 ∗∗∗ 0.005 ∗∗∗ 0.008 ∗∗∗ 0.009 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0 0 0
1960-1969 0.290 ∗∗∗ 0.282 ∗∗∗ 0.176 ∗∗∗ 0.127 ∗
1970-1979 1.452 ∗∗∗ 1.469 ∗∗∗ -0.163 ∗∗ -0.006
1980-1999 0.075 0.080 -0.240 ∗∗∗ -0.113
Partnership status
Single (ref.) 0 0 0 0
Cohabiting 0.032 0.036 0.227 ∗∗∗ 0.337 ∗∗∗
Married -0.221 ∗∗∗ -0.224 ∗∗∗ -0.163 ∗∗ -0.006
Separated 0.079 0.088 -0.240 ∗∗∗ -0.113
Parity
Childless (ref.) 0 0 0 0
1 child -0.519 ∗∗∗ -0.499 ∗∗∗ -0.233 ∗∗∗ -0.225 ∗∗∗
2 children -0.717 ∗∗∗ -0.703 ∗∗∗ -0.445 ∗∗∗ -0.428 ∗∗∗
3+ children -0.969 ∗∗∗ -0.975 ∗∗∗ -0.384 ∗∗∗ -0.437 ∗∗∗
Educational level
Low (ref.) 0 0 0 0
Medium 0.567 ∗∗∗ 0.636 ∗∗∗ 0.517 ∗∗∗ 0.575 ∗∗∗
High 1.179 ∗∗∗ 1.325 ∗∗∗ 1.101 ∗∗∗ 1.158 ∗∗∗
Origin group
Native (ref.) 0 0 0 0
North Africa -0.112 -0.117 -0.259 ∗∗ -0.288 ∗∗
Sub-Saharan Africa -0.147 -0.176 -0.327 ∗∗ -0.382 ∗∗
South East Asia 0.002 0.023 -0.373 ∗∗ -0.364 ∗∗
Turkey -0.245 -0.303 ∗-0.044 0.031
Southern Europe 0.049 0.081 -0.146 -0.117
Other Europe -0.044 -0.057 -0.167 -0.177
Order
1 (ref.) 0 0 0 0
2 0.402 ∗∗∗ 0.463 ∗∗∗ 0.249 ∗∗∗ 0.299 ∗∗∗
3+ 0.527 ∗∗∗ 0.692 ∗∗∗ 0.293 ∗∗∗ 0.488 ∗∗∗
Type of out of employment
Unemployed (ref.) 0 0 0 0
Housewife -0.732 ∗∗∗ 0.463 ∗∗∗ 0.249 ∗∗∗ 0.299 ∗∗∗
Other -0.554 ∗∗∗ -0.626 ∗∗∗ -0.330 ∗∗∗ -0.361 ∗∗∗
30
Table A.3. Impact of childbearing on employment, log-relative risks (Model 1, full
specification) (Continued)
Women Men
Model 1a Model 1b Model 1a Model 1b
Single process Multi-process Single process Multi-process
RR Sig RR Sig RR Sig RR Sig
First birth
Constant -7.039 ∗∗∗ -7.05 ∗∗∗ -9.309 ∗∗∗ -9.318 ∗∗∗
Age
15-19 year (slope) 0.014 ∗∗∗ 0.015 ∗∗∗ 0.040 ∗∗∗ 0.040 ∗∗∗
20-24 years (slope) 0.001 0.001 0.001 0.001
25-29 years (slope) 0.005 ∗∗∗ 0.005 ∗∗∗ 0.007 ∗∗∗ 0.008 ∗∗∗
30-34 years (slope) -0.004 ∗∗ -0.004 ∗∗ -0.003 ∗∗ -0.003 ∗∗
35+ years (slope) -0.024 ∗∗∗ -0.024 ∗∗∗ -0.016 ∗∗∗ -0.016 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0 0 0
1960-1969 0.126 ∗∗ 0.136 ∗∗ -0.017 -0.027
1970-1979 -0.019 -0.005 -0.279 ∗∗∗ -0.288 ∗∗∗
1980-1999 -0.287 ∗∗∗ -0.266 ∗∗∗ -0.710 ∗∗∗ -0.723 ∗∗∗
Partnership status
Single (ref.) 0 0 0 0
Cohabiting 2.249 ∗∗∗ 2.259 ∗∗∗ 2.479 ∗∗∗ 2.478 ∗∗∗
Married 3.251 ∗∗∗ 3.261 ∗∗∗ 3.644 ∗∗∗ 3.642 ∗∗∗
Separated 1.166 ∗∗∗ 1.171 1.173 ∗∗∗ 1.174 ∗∗∗
Educational level
Low (ref.) 0 0 0 0
Medium -0.302 ∗∗∗ -0.313 ∗∗∗ -0.091 -0.084
High -0.592 ∗∗∗ -0.613 ∗∗∗ -0.362 ∗∗∗ -0.350 ∗∗∗
Origin group
Native (ref.) 0 0 0 0
North Africa -0.028 -0.018 0.117 ∗0.119 ∗
Sub-Saharan Africa 0.320 ∗∗∗ 0.322 ∗∗∗ 0.303 ∗∗∗ 0.303 ∗∗∗
South East Asia -0.073 -0.069 -0.040 -0.037
Turkey 0.001 0.008 0.472 ∗∗∗ 0.486 ∗∗∗
Southern Europe -0.101 ∗-0.105 ∗0.098 ∗0.103 ∗
Other Europe 0.017 0.022 -0.107 -0.105
Employment status
Salaried (ref.) 0 0 0 0
Self-employed 0.079 0.064 0.162 0.162
Unemployed -0.111 -0.216 ∗∗ -0.391 ∗∗∗ -0.347 ∗∗
Student -1.025 ∗∗∗ -1.057 ∗∗∗ -0.746 ∗∗∗ -0.730 ∗∗∗
Housewife 0.609 ∗∗∗ 0.511 ∗∗∗ -0.673 ∗∗∗ -0.636
Inactive -0.118 -0.208 ∗∗∗ -0.120 -0.055
Other -0.064 -0.163 0.095 0.168 ∗∗
Second birth
Constant -7.702 ∗∗∗ -7.674 ∗∗∗ -8.065 ∗∗∗ -8.081 ∗∗∗
Time since previous birth
0-1 year (slope) 0.172 ∗∗∗ 0.174 ∗∗∗ 0.183 ∗∗∗ 0.183 ∗∗∗
1-3 years (slope) 0.035 ∗∗∗ 0.036 ∗∗∗ 0.039 ∗∗∗ 0.040 ∗∗∗
3-5 years (slope) -0.011 ∗∗∗ -0.011 ∗∗∗ -0.019 ∗∗∗ -0.019 ∗∗∗
5+ years (slope) -0.018 ∗∗∗ -0.017 ∗∗∗ -0.016 ∗∗∗ -0.016 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0 0 0
1960-1969 0.049 0.054 0.134 ∗0.125 ∗
1970-1979 0.204 ∗∗∗ 0.216 ∗∗∗ -0.001 -0.010
1980-1999 -0.291 ∗∗∗ -0.275 -0.509 ∗∗∗ -0.524 ∗∗∗
Partnership status
Single (ref.) 0 0 0 0
Cohabiting 0.897 ∗∗∗ 0.901 ∗∗∗ 0.884 ∗∗∗ 0.880 ∗∗∗
Married 1.452 ∗∗∗ 1.469 ∗∗∗ 1.480 ∗∗∗ 1.473 ∗∗∗
Separated 0.075 0.080 0.096 ∗∗∗ 0.089
31
Table A.3. Impact of childbearing on employment, log-relative risks (Model 1, full
specification) (Continued)
Women Men
Model 1a Model 1b Model 1a Model 1b
Single process Multi-process Single process Multi-process
RR Sig RR Sig RR Sig RR Sig
Educational level
Low (ref.) 0 0 0 0
Medium -0.218 ∗∗∗ -0.238 ∗∗∗ -0.106 -0.096
High -0.102 -0.143 ∗∗ 0.125 0.141 ∗
Origin group
Native (ref.) 0 0 0 0
North Africa 0.086 0.094 0.344 ∗∗∗ 0.348 ∗∗∗
Sub-Saharan Africa 0.083 0.098 0.510 ∗∗ 0.510 ∗∗∗
South East Asia -0.103 -0.105 0.315 ∗∗ 0.318 ∗∗
Turkey 0.175 0.197 0.454 ∗∗∗ 0.463 ∗∗∗
Southern Europe -0.124 ∗∗ -0.125 ∗∗ 0.002 0.008
Other Europe -0.076 -0.069 0.029 0.032
Employment status
Salaried (ref.) 0 0 0 0
Self-employed -0.032 -0.051 0.258 ∗∗ 0.259 ∗∗
Unemployed -0.368 ∗∗∗ -0.501 ∗∗∗ -0.278 -0.239
Student -0.563 ∗∗∗ -0.604 ∗∗∗ -0.308 ∗-0.293 ∗
Housewife 0.330 ∗∗∗ 0.209 ∗∗∗ 0.632 0.707
Inactive -0.063 -0.184 ∗-0.186 -0.110
Other 0.005 -0.095 0.020 0.087
Third birth
Constant -8.071 ∗∗∗ -8.026 ∗∗∗ -7.904 ∗∗∗ -7.928 ∗∗∗
Time since previous birth
0-1 year (slope) 0.140 ∗∗∗ 0.141 ∗∗∗ 0.143 ∗∗∗ 0.143 ∗∗∗
1-3 years (slope) 0.019 ∗∗∗ 0.019 ∗∗∗ 0.024 ∗∗∗ 0.024 ∗∗∗
3-5 years (slope) -0.011 ∗∗ -0.012 ∗∗ -0.028 ∗∗∗ -0.028 ∗∗∗
5+ years (slope) -0.022 ∗∗∗ -0.022 ∗∗∗ -0.017 ∗∗∗ -0.017 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0 0 0
1960-1969 0.099 0.107 0.071 0.062
1970-1979 0.156 0.163 ∗-0.020 -0.031
1980-1999 -0.686 ∗∗∗ -0.684 ∗∗∗ -0.328 -0.349
Partnership status
Single (ref.) 0 0 0 0
Cohabiting 0.649 ∗∗∗ 0.655 ∗∗∗ 0.542 ∗0.546 ∗
Married 1.062 ∗∗∗ 1.084 ∗∗∗ 0.893 ∗∗∗ 0.898 ∗∗∗
Separated 0.377 0.376 ∗∗∗ 0.701 ∗∗ 0.705 ∗∗
Educational level
Low (ref.) 0 0 0 0
Medium -0.501 ∗∗∗ -0.523 ∗∗∗ -0.484 ∗∗∗ -0.476 ∗∗∗
High -0.565 ∗∗∗ -0.617 ∗∗∗ -0.535 ∗∗∗ -0.519 ∗∗∗
Origin group
Native (ref.) 0 0 0 0
North Africa 0.618 ∗∗∗ 0.628 ∗∗∗ 0.813 ∗∗∗ 0.816 ∗∗∗
Sub-Saharan Africa 0.730 ∗∗ 0.755 ∗∗ 1.014 ∗∗∗ 1.017 ∗∗∗
South East Asia 0.386 ∗∗ 0.383 ∗∗ 0.509 ∗∗∗ 0.509 ∗∗∗
Turkey 0.496 ∗∗∗ 0.526 ∗∗∗ 0.996 ∗∗∗ 1.004 ∗∗∗
Southern Europe -0.373 ∗∗∗ -0.377 ∗∗∗ -0.171 -0.166
Other Europe 0.311 ∗∗ 0.319 ∗∗ 0.370 ∗∗ 0.372 ∗∗
Employment status
Salaried (ref.) 0 0 0 0
Self-employed -0.265 -0.279 -0.063 -0.065
Unemployed 0.184 0.050 0.028 0.062
Student 0.026 -0.021 0.013 0.030
Housewife 0.652 ∗∗∗ 0.520 ∗∗∗ 0.772 0.815
Inactive 0.122 -0.006 0.053 0.123
Other 0.089 -0.010 -0.216 ∗-0.156
32
Table A.3. Impact of childbearing on employment, log-relative risks (Model 1, full
specification) (Continued)
Women Men
Model 1a Model 1b Model 1a Model 1b
Single process Multi-process Single process Multi-process
RR Sig RR Sig RR Sig RR Sig
Unobserved heterogeneity
Standard deviation of residuals
Fertility 0.726 ∗∗∗ 0.730 ∗∗∗ 0.729 ∗∗∗ 0.732 ∗∗∗
Employment entry 0.618 ∗∗∗ 0.697 ∗∗∗ 0.656 ∗∗∗ 0.795 ∗∗∗
Employment exit 1.130 ∗∗∗ 1.307 ∗∗∗ 0.885 ∗∗∗ 1.205 ∗∗∗
Correlation between residuals
Fertility and employment entry -0.114 ∗∗∗ 0.095 ∗
Fertility and employment exit 0.302 ∗∗∗ -0.070
Employment entry and exit -0.6508 ∗∗∗ -0.962 ∗∗∗
ln-L -96771.47 -96689.48 -76239.96 -76109.02
Source: Social Protection Survey of Chile, authors’ own calculations.
Notes: * p < 0.10, ** p < 0.05, *** p < 0.01.
33
Table A.4. Impact of childbearing on employment, log-relative risks (additional
specification)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
First employment entry
Constant -1.341 ∗∗∗ -1.212 ∗∗∗
Time since in education
0-1 year (slope) -0.256 ∗∗∗ -0.203 ∗∗∗
1-3 years (slope) 0.030 ∗∗∗ 0.027 ∗∗∗
3-5 years (slope) -0.026 ∗∗∗ -0.058 ∗∗∗
5+ years (slope) 0.001 ∗∗∗ 0.005 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 -0.235 ∗∗∗ -0.474 ∗∗∗
1970-1979 -0.194 ∗∗∗ -0.447 ∗∗∗
1980-1999 -0.329 ∗∗∗ -0.276 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.089 ∗0.411 ∗∗∗
Married -0.209 ∗∗∗ 0.307 ∗∗∗
Separated 0.317 ∗∗∗ 0.194 ∗
Time since birth
Childless (ref.) 0 0
0-1 year after birth -0.746 ∗∗∗ 0.063
1+ year after birth -0.446 ∗∗∗ -0.083
Educational level
Low (ref.) 0 0
Medium 0.759 ∗∗∗ 0.373 ∗∗∗
High 1.153 ∗∗∗ 0.431 ∗∗∗
Origin group
Native (ref.) 0 0
North Africa -0.336 ∗∗∗ -0.078
Sub-Saharan Africa -0.337 ∗∗∗ -0.230 ∗∗∗
South East Asia -0.105 0.069
Turkey -0.426 ∗∗∗ 0.503 ∗∗∗
Southern Europe 0.071 0.309 ∗∗∗
Other Europe -0.210 ∗∗∗ 0.065
Employment exits
Constant -9.684 ∗∗∗ -8.992 ∗∗∗
Time since previous employment
0-1 year (slope) 0.319 ∗∗∗ 0.331 ∗∗∗
1-3 years (slope) -0.024 ∗∗∗ -0.039 ∗∗∗
3+ years (slope) 0.003 ∗∗∗ 0.002 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.247 ∗∗ 0.351 ∗∗∗
1970-1979 0.642 ∗∗∗ 0.624 ∗∗∗
1980-1999 1.336 ∗∗∗ 0.886 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.307 ∗∗∗ -0.527 ∗∗∗
Married 0.570 ∗∗∗ -0.859 ∗∗∗
Separated 0.444 ∗∗∗ -0.081
Time since birth
Childless (ref.) 0 0
0-1 year after birth 0.857 ∗∗∗ -0.197
1-3 years after birth 0.793 ∗∗∗ -0.425 ∗∗∗
3-5 years after birth -0.207 ∗∗ -0.596 ∗∗∗
5+ years after birth -0.442 ∗∗∗ -0.184
34
Table A.4. Impact of childbearing on employment, log-relative risks (additional
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Educational level
Low (ref.) 0 0
Medium -0.544 ∗∗∗ -0.459 ∗∗∗
High -1.536 ∗∗∗ -1.088 ∗∗∗
Origin group
Native (ref.) 0 0
North Africa 0.249 ∗∗∗ 0.164 ∗
Sub-Saharan Africa 0.069 0.153
South East Asia -0.018 0.002
Turkey 0.524 ∗∗∗ -0.057
Southern Europe -0.197 ∗∗∗ -0.295 ∗∗∗
Other Europe 0.148 0.083
Order
1 (ref.) 0 0
2 0.595 ∗∗∗ 0.331 ∗∗∗
3+ 0.313 0.260 ∗∗
Type of employment
Salaried (ref.) 0 0
Self-employed -0.424 ∗∗∗ -0.446 ∗∗∗
Higher order employment entries
Constant -7.767 ∗∗∗ -8.769 ∗∗∗
Time since out of employment
0-1 year 0.317 ∗∗∗ 0.498 ∗∗∗
1-3 years -0.024 ∗∗∗ -0.077 ∗∗∗
3-5 years -0.016 ∗∗∗ -0.019 ∗∗∗
5+ years 0.004 ∗∗∗ 0.009 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.337 ∗∗∗ 0.108
1970-1979 0.736 ∗∗∗ 0.366 ∗∗∗
1980-1999 0.880 ∗∗∗ 0.527 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.081 0.334 ∗∗∗
Married -0.244 ∗∗∗ -0.025
Separated 0.031 -0.105
Time since birth
Childless (ref.) 0 0
0-1 year after birth -0.962 ∗∗∗ -0.267 ∗∗
1-3 years after birth -0.895 ∗∗∗ -0.205 ∗∗
3-5 years after birth -0.112 -0.346 ∗∗∗
5+ years after birth -0.306 ∗∗∗ -0.433 ∗∗∗
Educational level
Low (ref.) 0 0
Medium 0.625 ∗∗∗ 0.581 ∗∗∗
High 1.275 ∗∗∗ 1.161 ∗∗∗
Origin group
Native (ref.) 0 0
North Africa -0.119 -0.295 ∗∗
Sub-Saharan Africa -0.182 -0.402 ∗∗
South East Asia 0.025 -0.372 ∗∗
Turkey -0.313 ∗∗ 0.020
Southern Europe 0.070 -0.115
Other Europe -0.044 -0.179
35
Table A.4. Impact of childbearing on employment, log-relative risks (additional
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Order
1 (ref.) 0 0
2 0.404 ∗∗∗ 0.298 ∗∗∗
3+ 0.560 ∗∗∗ 0.486 ∗∗∗
Type of out of employment
Unemployed (ref.) 0 0
Housewife -0.737 ∗∗∗ -0.560 ∗
Other -0.609 ∗∗∗ -0.364 ∗∗∗
First birth
Constant -6.963 ∗∗∗ -9.295 ∗∗∗
Age
15-19 year (slope) 0.014 ∗∗∗ 0.040 ∗∗∗
20-24 years (slope) -0.0002 0.001
25-29 years (slope) 0.004 ∗∗∗ 0.007 ∗∗∗
30-34 years (slope) -0.005 ∗∗ -0.004 ∗∗
35+ years (slope) -0.024 ∗∗∗ -0.016 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.146 ∗∗∗ -0.018
1970-1979 0.013 -0.276 ∗∗∗
1980-1999 -0.254 ∗∗∗ -0.709 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 2.253 ∗∗∗ 2.474 ∗∗∗
Married 3.241 ∗∗∗ 3.631 ∗∗∗
Separated 1.178 ∗∗∗ 1.172 ∗∗∗
Educational level
Low (ref.) 0 0
Medium -0.287 ∗∗∗ -0.082
High -0.562 ∗∗∗ -0.337 ∗∗∗
Origin group
Native (ref.) 0 0
North Africa -0.005 0.126 ∗
Sub-Saharan Africa 0.314 ∗∗∗ 0.303 ∗∗∗
South East Asia -0.062 -0.031
Turkey 0.009 0.464 ∗∗∗
Southern Europe -0.102 ∗∗ 0.101 ∗
Other Europe 0.025 -0.103
Employment status
Salaried (ref.) 0 0
Self-employed 0.056 0.164
Unemployed -0.230 ∗∗ -0.375 ∗∗
Student -1.056 ∗∗∗ -0.735 ∗∗∗
Housewife 0.459 ∗∗∗ -0.648
Inactive -0.238 ∗∗∗ -0.086
Other -0.170 ∗0.128
Second birth
Constant -7.429 ∗∗∗ -7.955 ∗∗∗
Time since previous birth
0-1 year (slope) 0.174 ∗∗∗ 0.183 ∗∗∗
1-3 years (slope) 0.034 ∗∗∗ 0.039 ∗∗∗
3-5 years (slope) -0.012 ∗∗∗ -0.020 ∗∗∗
5+ years (slope) -0.018 ∗∗∗ -0.016 ∗∗∗
36
Table A.4. Impact of childbearing on employment, log-relative risks (additional
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.051 0.135 ∗∗
1970-1979 0.166 ∗∗ -0.008
1980-1999 -0.319 ∗∗∗ -0.538 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.865 ∗∗∗ 0.864 ∗∗∗
Married 1.388 ∗∗∗ 1.439 ∗∗∗
Separated 0.047 0.081
Educational level
Low (ref.) 0 0
Medium -0.200 ∗∗∗ -0.097
High -0.048 0.162 ∗
Origin group
Native (ref.) 0 0
North Africa 0.119 ∗0.354 ∗∗∗
Sub-Saharan Africa 0.114 0.522 ∗∗∗
South East Asia -0.096 0.332 ∗∗
Turkey 0.225 ∗0.435 ∗∗∗
Southern Europe -0.110 ∗0.005
Other Europe -0.058 0.029
Employment status
Salaried (ref.) 0 0
Self-employed -0.029 0.262 ∗∗
Unemployed -0.482 ∗∗∗ -0.270
Student -0.582 ∗∗∗ -0.291 ∗
Housewife 0.186 ∗∗∗ 0.691
Inactive -0.193 ∗-0.142
Other -0.075 0.048
Third birth
Constant -7.650 ∗∗∗ -7.387 ∗∗∗
Time since previous birth
0-1 year (slope) 0.141 ∗∗∗ 0.143 ∗∗∗
1-3 years (slope) 0.019 ∗∗∗ 0.024 ∗∗∗
3-5 years (slope) -0.012 ∗∗ -0.028 ∗∗∗
5+ years (slope) -0.023 ∗∗∗ -0.017 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.270 ∗∗ 0.067
1970-1979 0.428 ∗∗∗ -0.017
1980-1999 -0.319 -0.486
Partnership status
Single (ref.) 0 0
Cohabiting -0.194 ∗0.673 ∗∗
Married -0.248 ∗∗ 1.018 ∗∗∗
Separated -0.035 0.824 ∗∗∗
Educational level
Low (ref.) 0 0
Medium -0.378 ∗∗∗ -0.467 ∗∗∗
High -0.104 -0.451 ∗∗∗
37
Table A.4. Impact of childbearing on employment, log-relative risks (additional
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Origin group
Native (ref.) 0 0
North Africa 0.474 ∗∗∗ 0.862 ∗∗∗
Sub-Saharan Africa 0.641 ∗∗ 1.016 ∗∗∗
South East Asia 0.170 0.581 ∗∗∗
Turkey 0.140 0.895 ∗∗∗
Southern Europe -0.392 ∗∗∗ -0.155
Other Europe 0.206 0.371 ∗∗
Employment status
Salaried (ref.) 0 0
Self-employed -0.045 -0.054
Unemployed 0.435 ∗∗ 0.036
Student -0.129 0.011
Housewife 0.687 ∗∗∗ 0.739
Inactive 0.232 0.089
Other 0.021 -0.177
Unobserved heterogeneity
Standard deviation of residuals
Fertility 0.660 ∗∗∗ 0.698 ∗∗∗
Employment entry 0.661 ∗∗∗ 0.795 ∗∗∗
Employment exit 1.222 ∗∗∗ 1.205 ∗∗∗
Correlation between residuals
Fertility and employment entry -0.194 ∗∗∗ 0.057
Fertility and employment exit 0.313 ∗∗∗ -0.005
Employment entry and exit -0.781 ∗∗∗ -0.975 ∗∗∗
ln-L -96381.75 -76095.14
Source: Social Protection Survey of Chile, authors’ own calculations.
Notes: * p < 0.10, ** p < 0.05, *** p < 0.01.
38
Table A.5. Impact of childbearing on employment, log-relative risks (Model 2, full
specification)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
First employment entry
Constant -1.449 ∗∗∗ -1.121 ∗∗∗
Time since in education
0-1 year (slope) -0.261 ∗∗∗ -0.205 ∗∗∗
1-3 years (slope) 0.029 ∗∗∗ 0.027 ∗∗∗
3-5 years (slope) -0.030 ∗∗∗ -0.059 ∗∗∗
5+ years (slope) 0.0004 0.004 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 -0.266 ∗∗∗ -0.442 ∗∗∗
1970-1979 -0.290 ∗∗∗ -0.427 ∗∗∗
1980-1999 -0.456 ∗∗∗ -0.282 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.055 0.409 ∗∗∗
Married -0.490 ∗∗∗ 0.313 ∗∗∗
Separated 0.168 ∗∗ 0.163
Educational level
Low (ref.) 0 0
Medium 0.818 ∗∗∗ 0.364 ∗∗∗
High 1.249 ∗∗∗ 0.389 ∗∗∗
Employment exits
Constant -9.981 ∗∗∗ -8.967 ∗∗∗
Time since previous employment
0-1 year (slope) 0.318 ∗∗∗ 0.333 ∗∗∗
1-3 years (slope) -0.017 ∗∗∗ -0.039 ∗∗∗
3+ years (slope) 0.0004 0.003 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.328 ∗∗∗ 0.308 ∗∗∗
1970-1979 0.850 ∗∗∗ 0.587 ∗∗∗
1980-1999 1.556 ∗∗∗ 0.912 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.438 ∗∗∗ -0.558 ∗∗∗
Married 0.676 ∗∗∗ -0.871 ∗∗∗
Separated 0.359 ∗∗∗ -0.075
Parity x Origin group
Childless x Natives (ref.) 0 0
Childless x North Africa 0.260 ∗∗ -0.091
Childless x Sub-Saharan Africa 0.079 -0.322 ∗
Childless x South East Asia 0.118 -0.056
Childless x Turkey 0.294 -0.194
Childless x Southern Europe -0.108 -0.255 ∗∗∗
Childless x Other Europe 0.022 0.094
1+ children x Natives 0.546 ∗∗∗ -0.794 ∗∗∗
1+ children x North Africa 0.621 ∗∗∗ -0.067
1+ children x Sub-Saharan Africa 0.342 ∗∗ 0.019
1+ children x South East Asia 0.325 ∗-0.540
1+ children x Turkey 1.056 ∗∗∗ 0.035
1+ children x Southern Europe 0.288 ∗∗∗ -0.688 ∗∗∗
1+ children x Other Europe 0.619 ∗∗∗ -0.572 ∗∗
Educational level
Low (ref.) 0 0
Medium -0.517 ∗∗∗ -0.453 ∗∗∗
High -1.510 ∗∗∗ -1.075 ∗∗∗
39
Table A.5. Impact of childbearing on employment, log-relative risks (Model 2, full
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Order
1 (ref.) 0 0
2 0.270 ∗∗∗ 0.300 ∗∗∗
3+ -0.313 ∗0.247 ∗∗
Type of employment
Salaried (ref.) 0 0
Self-employed -0.435 ∗∗∗ -0.399 ∗∗∗
Higher order employment entries
Constant -8.022 ∗∗∗ -8.886 ∗∗∗
Time since out of employment
0-1 year (slope) 0.314 ∗∗∗ 0.499 ∗∗∗
1-3 years (slope) -0.020 ∗∗∗ -0.077 ∗∗∗
3-5 years (slope) -0.013 ∗∗∗ -0.018 ∗∗∗
5+ years (slope) 0.005 ∗∗∗ 0.009 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.296 ∗∗∗ 0.151 ∗∗
1970-1979 0.644 ∗∗∗ 0.419 ∗∗∗
1980-1999 0.804 ∗∗∗ 0.585 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.004 0.336 ∗∗∗
Married -0.314 ∗∗∗ -0.020
Separated 0.031 -0.122
Parity x Origin group
Childless x Natives (ref.) 0 0
Childless x North Africa 0.199 -0.168
Childless x Sub-Saharan Africa -0.121 -0.276
Childless x South East Asia 0.315 ∗-0.339 ∗∗
Childless x Turkey 0.194 -0.179
Childless x Southern Europe 0.333 ∗∗∗ -0.160
Childless x Other Europe 0.242 -0.050
1+ children x Natives -0.466 -0.419
1+ children x North Africa -0.345 ∗∗∗ -0.581 ∗∗∗
1+ children x Sub-Saharan Africa -0.025 -0.374
1+ children x South East Asia -0.202 -0.515 ∗
1+ children x Turkey -0.642 ∗∗∗ -0.125
1+ children x Southern Europe -0.209 -0.426 ∗∗∗
1+ children x Other Europe -0.254 -0.633 ∗∗∗
Educational level
Low (ref.) 0 0
Medium 0.639 ∗∗∗ 0.582 ∗∗∗
High 1.316 ∗∗∗ 1.143 ∗∗∗
Order
1 (ref.) 0 0
2 0.424 ∗∗∗ 0.300 ∗∗∗
3+ 0.616 ∗∗∗ 0.474 ∗∗∗
Type of out of employment
Unemployed (ref.) 0 0
Housewife -0.781 ∗∗∗ -0.525 ∗
Other -0.610 ∗∗∗ -0.630 ∗∗∗
40
Table A.5. Impact of childbearing on employment, log-relative risks (Model 2, full
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
First birth
Constant -6.954 ∗∗∗ -10.237 ∗∗∗
Age
15-19 year (slope) 0.014 ∗∗∗ 0.102
20-24 years (slope) -0.0003 0.058 ∗∗∗
25-29 years (slope) 0.003 ∗∗∗ 0.022 ∗∗
30-34 years (slope) -0.005 ∗∗ 0.007 ∗∗∗
35+ years (slope) -0.024 ∗∗∗ -0.003 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.155 ∗∗∗ 0.045
1970-1979 0.029 -0.178 ∗∗∗
1980-1999 -0.233 ∗∗∗ -0.722 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 2.259 ∗∗∗ 2.469 ∗∗∗
Married 3.249 ∗∗∗ 3.633 ∗∗∗
Separated 1.181 ∗∗∗ 1.219 ∗∗∗
Educational level
Low (ref.) 0 0
Medium -0.294 ∗∗∗ -0.092 ∗
High -0.570 ∗∗∗ -0.277 ∗∗∗
Origin group
Native (ref.) 0 0
North Africa -0.025 0.161 ∗∗
Sub-Saharan Africa 0.289 ∗∗∗ 0.342 ∗∗∗
South East Asia -0.068 0.0001
Turkey -0.019 0.433 ∗∗∗
Southern Europe -0.098 ∗0.094 ∗
Other Europe 0.011 -0.104
Employment status
Salaried (ref.) 0 0
Self-employed 0.052 0.205 ∗∗
Unemployed -0.265 ∗∗∗ -0.430 ∗∗∗
Student -1.070 ∗∗∗ -0.776 ∗∗∗
Housewife 0.412 ∗∗∗ -0.736
Inactive -0.290 ∗∗∗ -0.125
Other -0.213 ∗∗ 0.155 ∗
Second birth
Constant -7.408 ∗∗∗ -7.884 ∗∗∗
Time since previous birth
0-1 year (slope) 0.174 ∗∗∗ 0.183 ∗∗∗
1-3 years (slope) 0.034 ∗∗∗ 0.038 ∗∗∗
3-5 years (slope) -0.012 ∗∗∗ -0.020 ∗∗∗
5+ years (slope) -0.018 ∗∗∗ -0.016 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.056 0.150 ∗∗
1970-1979 0.177 ∗∗∗ 0.016
1980-1999 -0.309 ∗∗∗ -0.522 ∗∗∗
Partnership status
Single (ref.) 0 0
Cohabiting 0.867 ∗∗∗ 0.847 ∗∗∗
Married 1.395 ∗∗∗ 1.409 ∗∗∗
Separated 0.052 0.076
41
Table A.5. Impact of childbearing on employment, log-relative risks (Model 2, full
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Educational level
Low (ref.) 0 0
Medium -0.210 ∗∗∗ -0.096
High -0.065 0.169 ∗∗
Origin group
Native (ref.) 0 0
North Africa 0.095 0.356 ∗∗∗
Sub-Saharan Africa 0.082 0.524 ∗∗∗
South East Asia -0.109 0.335 ∗∗
Turkey 0.200 0.415 ∗∗∗
Southern Europe -0.105 ∗0.002
Other Europe -0.069 0.028
Employment status
Salaried (ref.) 0 0
Self-employed -0.038 0.265 ∗∗
Unemployed -0.516 ∗∗∗ -0.275
Student -0.596 ∗∗∗ -0.289 ∗
Housewife 0.152 ∗∗∗ 0.645
Inactive -0.248 ∗∗ -0.152
Other -0.111 0.037
Third birth
Constant -7.624 ∗∗∗ -7.306 ∗∗∗
Time since previous birth
0-1 year (slope) 0.141 ∗∗∗ 0.143 ∗∗∗
1-3 years (slope) 0.019 ∗∗∗ 0.024 ∗∗∗
3-5 years (slope) -0.012 ∗∗ -0.028 ∗∗∗
5+ years (slope) -0.023 ∗∗∗ -0.017 ∗∗∗
Birth cohort
1948-1959 (ref.) 0 0
1960-1969 0.151 ∗0.083
1970-1979 0.173 ∗0.012
1980-1999 -0.769 ∗∗∗ -0.457
Partnership status
Single (ref.) 0 0
Cohabiting 0.625 ∗∗∗ 0.644 ∗∗
Married 1.025 ∗∗∗ 0.978 ∗∗∗
Separated 0.306 0.796 ∗∗
Educational level
Low (ref.) 0 0
Medium -0.407 ∗∗∗ -0.466 ∗∗∗
High -0.329 ∗∗∗ -0.446 ∗∗∗
Origin group
Native (ref.) 0 0
North Africa 0.649 ∗∗∗ 0.859 ∗∗∗
Sub-Saharan Africa 0.715 ∗∗∗ 1.011 ∗∗∗
South East Asia 0.404 ∗∗∗ 0.580 ∗∗∗
Turkey 0.489 ∗∗∗ 0.868 ∗∗∗
Southern Europe -0.350 ∗∗∗ -0.161
Other Europe 0.332 ∗∗∗ 0.366 ∗∗
42
Table A.5. Impact of childbearing on employment, log-relative risks (Model 2, full
specification) (Continued)
Women Men
Model 2 Model 2
Multi-process Multi-process
RR Sig RR Sig
Employment status
Salaried (ref.) 0 0
Self-employed -0.264 -0.056
Unemployed 0.022 0.030
Student -0.019 0.010
Housewife 0.427 ∗∗∗ 0.736
Inactive -0.112 0.076
Other -0.020 -0.185
Unobserved Heterogeneity
Standard Deviation of Residuals
Fertility 0.658 ∗∗∗ 0.666 ∗∗∗
Employment entry 0.681 ∗∗∗ 0.796 ∗∗∗
Employment exit 1.322 ∗∗∗ 1.250 ∗∗∗
Correlation Between Residuals
Fertility and employment entry -0.302 ∗∗∗ 0.043
Fertility and employment exit 0.336 ∗∗∗ 0.004
Employment entry and exit -0.664 ∗∗∗ -0.936 ∗∗∗
ln-L -96765.09 -76211.10
Source: Social Protection Survey of Chile, authors’ own calculations.
Notes:
*p < 0.10, ** p < 0.05, *** p < 0.01.
43
Figure A.1. Relative risks of employment entry and exit by origin and parity for women
(Model 2)
(a) Employment exits
(b) Second and higher order employment entries
Source: Trajectories and Origins, authors’ own calculations.
44
Figure A.2. Relative risks of employment entry and exit by origin and parity for men
(Model 2)
(a) Employment exits
(b) Second and higher order employment entries
Source: Trajectories and Origins, authors’ own calculations.
45
