Löwe, Paul Se e in
A icle — Published Ve sion
Does aCi il Se ice Job Ma e ? The In luence o Ci il Se ice
Employmen on he T ansi ion o he Fi s and he Second Child o
Women and Men in Ge many
KZ SS Kölne Zei sch i ü Soziologie und Sozialpsychologie
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Löwe, Paul Se e in (2025) : Does aCi il Se ice Job Ma e ? The In luence o Ci il
Se ice Employmen on he T ansi ion o he Fi s and he Second Child o Women and Men in
Ge many, KZ SS Kölne Zei sch i ü Soziologie und Sozialpsychologie, ISSN 1861-891X, Sp inge
Fachmedien Wiesbaden GmbH, Wiesbaden, Vol. 77, Iss. 3, pp. 321-354,
h ps://doi.o g/10.1007/s11577-025-01017-w
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/330228
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/deed.de
ABHANDLUNGEN
h ps://doi.o g/10.1007/s11577-025-01017-w
Köln Z Soziol (2025) 77:321–354
Does a Ci il Se ice Job Ma e ? The In luence o Ci il
Se ice Employmen on he T ansi ion o he Fi s and
he Second Child o Women and Men in Ge many
Paul Se e in Löwe
Recei ed: 6 Ap il 2025 / Accep ed: 18 Augus 2025 / Published online: 6 Oc obe 2025
© The Au ho (s) 2025
Abs ac This s udy compa es he in luence o ci il se ice and p i a e sec o
employmen on he ansi ion a es o he i s and he second child. The ci il se ice
is o en associa ed wi h a amily- iendly wo k en i onmen , o e ing he possibili y
o e e se he p oblem o low bi h a e. An in luence has been demons a ed in
some Eu opean coun ies, bu Ge many lacks an indi idual-le el analysis. Using
da a om he Ge man Family Panel (Pai am), we an disc e e e en his o y models
in a piecewise-cons an speci ica ion o calcula e he ansi ions o he i s and he
second child condi ional on ci il se ice o p i a e sec o employmen . Ou indings
indica e a subs an ial and s a is ically signi ican posi i e in luence o ci il se ice
employmen on he ansi ion o he second child o women. Fo men, we obse e
a subs an ial nega i e bu s a is ically insigni ican in luence on he ansi ion o he
second child. No signi ican o subs an ial in luence was ound o ei he women o
men ega ding he ansi ion o he i s child. The esul s sugges ha , especially
o he si ua ion a e he bi h o a child, employmen in he ci il se ice can be an
impo an s a ing poin o in luencing e ili y.
Keywo ds Fe ili y · Public sec o · Wo king condi ions · Family iendliness ·
Pai am
Online Appendix: h ps://kz ss.uni-koeln.de/si es/kz ss/pd /Loewe.pd
P. S. Löwe
S a e Ins i u e o Family Resea ch, Uni e si y o Bambe g
Hein ichsdamm 4, 96047 Bambe g, Ge many
E-Mail: paul.loewe@i b.uni-bambe g.de
K
322 P. S. Löwe
Mach ein Job im ö en lichen Diens einen Un e schied? De Ein luss
de Beschä igung im ö en lichen Diens au den Übe gang zum e s en
und zwei en Kind ü F auen und Männe in Deu schland
Zusammen assung In diese S udie wi d de Ein luss eine Beschä igung im ö -
en lichen Diens im Ve gleich zu P i a wi scha au die Übe gangs a e zum e s en
und zwei en Kind un e such . De ö en liche Diens wi d häu ig mi einem amili-
en eundlichen A bei sum eld in Ve bindung geb ach , das die Möglichkei bie e ,
einen Bei ag zu Lösung des gesellscha lichen P oblems de nied igen Gebu en-
a e zu leis en. In ande en eu opäischen Lände n wu de ein solche Ein luss nach-
gewiesen, abe in Deu schland ehl e bishe eine Analyse au indi iduelle Ebene.
Un e Ve wendung on Da en aus dem Deu schen Familienpanel (Pai am) nu zen
wi disk e e E eignis e lau smodelle in eine Piecewise-cons an -Spezi ika ion, um
den Übe gang zum e s en Kind in Abhängigkei on de Beschä igung im ö en li-
chen Diens ode in de P i a wi scha zu be echnen. Wi inden einen inhal lich
bedeu enden und s a is isch signi ikan en posi i en Ein luss de Beschä igung im
ö en lichen Diens au den Übe gang zum zwei en Kind ü F auen. Fü Männe
wi d ein inhal lich bedeu ende , abe s a is isch nich signi ikan e nega i e Ein luss
au den Übe gang zum zwei en Kind es ges ell . Wede ü F auen noch ü Män-
ne läss sich ein inhal lich bedeu ende ode s a is isch signi ikan e Ein luss au
den Übe gang zum e s en Kind es s ellen. Die E gebnisse deu en da au hin, dass
die Beschä igung im ö en lichen Diens einen Ansa zpunk zu Beein lussung de
Fe ili ä , insbesonde e nach de Gebu , bie e .
Schlüsselwö e Fe ili ä · Ö en liche Sek o · A bei sbedingungen ·
Familien eundlichkei · Pai am
1 In oduc ion
The objec i e o his pape is o ex end he s a e o esea ch on amily o ma ion
and expansion by in es iga ing whe he he e is a posi i e in luence o ci il se ice1
e sus p i a e sec o employmen on he ansi ion a e o he bi hs o i s and
second child en in Ge many. Ou analysis o he ci il se ice goes beyond p e ious
wo k by making ci il se ice employmen he ocus o he examina ion.
In Ge many, as in many o he Eu opean coun ies, low e ili y a es a e pe sis-
en , and he di e ence be ween he desi ed and he ac ual numbe o child en is
high, wi h ewe bi hs han desi ed (Sobo ka and Beaujouan 2014;Tes a2012).
The employmen si ua ion has been ecognized as an inc easingly signi ican ac o
in p omo ing econcilia ion and ensu ing a su icien and secu e income o aising
a amily, as he majo i y o bo h women and men engage in paid wo k (F ode mann
e al. 2013; G eulich e al. 2017; K eyen eld 2010; Lau ijssen and Glo ieux 2013;
1We use he e m “ci il se ice” ins ead o “public sec o ” because i is mo e p ecise. Ci il se ice in-
cludes only people who wo k di ec ly o he s a e unde special condi ions, no employees who wo k o
a p i a e company ha is (pa ly) owned by he s a e.
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 323
Schulze 2009; Seyda 2003; S a is ik de Bundesagen u ü A bei 2021). S udies
ha e shown ha coun ies wi h highe e ili y a es a e ypically cha ac e ized by
ewe di icul ies in econciling wo k and amily li e, as well as by a posi i e co e-
la ion be ween income le els and e ili y a es (Doepke e al. 2023; Fox e al. 2019;
G eulich e al. 2017; Luci-G eulich and Thé enon 2013,2014).
In dis inc ion o he p i a e sec o , he ci il se ice o e s speci ic condi ions,
as i is less subjec o p o i o ien a ion, compe i i e pe o mance, and exploi a ion
(Becke 1993). This low ma ke o ien a ion a o s he eme gence o ins i u ional
a angemen s ha decouple he ci il se ice mo e om sho - e m economic cos
p essu es and compe i ion (Go schall e al. 2015; Saue e al. 2022) and acili a e
he eme gence o ea u es such as he p o ision and p omo ion o amily- iendly
policies and he o e o secu e jobs wi h su icien income (Go schall e al. 2015;
Löwe 2022; Saue e al. 2019). These ac o s ha e been shown o in luence e ili y
posi i ely (G eulich e al. 2017; Köppen 2006; K eyen eld e al. 2023; K eyen eld
and Zabel 2005; Ma ysiak and S einme z 2008;OECD2007). Wi h o e i e mil-
lion employees in he ci il se ice (S a is isches Bundesam 2023), he s a e has
a powe ul ool o p omo e e ili y by p o iding suppo i e wo king condi ions.
So a , only a ew s udies ha e looked a he case o Ge many. Ci il se ice
employmen is only conside ed as an in e ac ion o measu ed indi ec ly, and no
in luence (Gebel and Giesecke 2009) o a nega i e in luence (Oppe mann 2014)on
he ansi ion o he i s child has been ound. Howe e , indings on he impac o
ci il se ice employmen on e ili y in o he Eu opean coun ies ha e been incon-
clusi e. In a s udy wi h a Eu opean ocus, Adse a (2011) iden i ied an associa ion
be ween a la ge ci il sec o and as e ansi ions o i s , second, and hi d bi hs.
A he indi idual le el, posi i e e ec s o ci il se ice employmen on he ansi ion
o he i s child ha e been demons a ed in I aly (Con i and Se e 2013)andSpain
(Ma in-Ga cia and Cas o-Ma in 2013) and on he ansi ion o he second child in
he Uni ed Kingdom (K eyen eld and Zabel 2005). No e ec was ound o he i s
child in he Ne he lands (Begall and Mills 2013). Gi en he indica ion o coun y-
speci ic in luences in p e ious indings, Ge many p esen s an in e es ing case due
o i s highly egula ed and coo dina ed economy in bo h he ci il se ice and p i-
a e sec o s, as well as i s ela i ely a o able economic condi ions o young and
e ile indi iduals compa ed o o he Eu opean coun ies. Agains his backg ound,
i emains unclea whe he he di e ences be ween jobs in he p i a e sec o and he
ci il se ice a e dis inc enough o in luence e ili y.
We shed ligh on his unsol ed puzzle by concen a ing on he unc ion ha
gain ul employmen can ha e in suppo ing e ili y. D awing on Becke ’s (1981)
amily economic heo y, we examine whe he ci il se ice employmen is mo e
likely han p i a e sec o employmen o p o ide a secu e and su icien income
while acili a ing he econcilia ion o amily li e and gain ul employmen .
Ou analysis ocuses on he ansi ions o he i s and he second child, as hese
accoun o o e 80% o bi hs in Ge many (S a is isches Bundesam 2024). On one
hand, demog aphic pa e ns a e ma ked by a delayed ansi ion o he i s child, ac-
companied by ising childlessness (K eyen eld and Konie zka 2008b). On he o he
hand, e ili y di e ences ac oss Eu ope a e la gely d i en by a decline in second
bi hs (B e on and P ioux 2009; F ejka and Sa don 2007; F ejka and Sobok a 2008;
K
324 P. S. Löwe
G eulich e al. 2017). Schola s a gue ha he employmen si ua ion is ele an o
bo h ansi ions. Howe e , con lic s be ween paid employmen and amily esponsi-
bili ies, as well as he need o secu e a s able and su icien income, become mo e
p onounced o he second child, and enabling ac o s, such as wo king condi ions,
migh become mo e in luen ial (Adse a 2004;AhnandMi a2002;B ews e and
Rind uss 2000; Vlasblom and Schippe s 2006).
We conduc ed all analyses sepa a ely by gende , as Becke ’s (1981) amily eco-
nomic heo y sugges s ha di e en job cha ac e is ics in luence he e ili y deci-
sions o women and men and ha empi ical di e ences in he labo o ce beha io o
women and men a e s ill p esen (Cooke 2004; Lewin-Eps ein e al. 2006;Sh e le
e al. 2010).
We used he longi udinal da a o he Panel Analysis o In ima e Rela ionships
and Family Dynamics (Pai am) 2008–2020, wa es 1–12. Pai am is a unique da a
sou ce, co e ing in de ail he opics o amily planning, household con ex , and
p econdi ions o ha ing child en. We es ima ed s epwise mul i a ia e disc e e e en
his o y da a models wi h a piecewise cons an speci ica ion o calcula e he ansi ion
a e o he i s and he second child condi ional on a job in he ci il se ice.
Ou indings indica e ha ci il se ice employmen signi ican ly in luences he
ansi ion o he second child, whe eas no subs an ial in luence is obse ed o he
ansi ion o he i s child. Rega ding he i s child, we iden i y a sligh ly posi i e bu
s a is ically insigni ican in luence o men and no in luence o women. In con as ,
o he ansi ion o he second child, he in luence o men is subs an ially nega i e
bu s a is ically insigni ican , while o women, ci il se ice employmen —compa ed
o p i a e sec o employmen —has a subs an ial and s a is ically signi ican posi i e
in luence. These esul s sugges ha in Ge many, ci il se ice employmen may
se e as a policy le e o in luence e ili y, pa icula ly in he phase ollowing
he bi h o he i s child, when he need o balance paid employmen and amily
esponsibili ies in ensi ies.
2 Theo y and S a e o Resea ch
In his sec ion, we examine how speci ic wo king condi ions in he ci il se ice
a e expec ed o in luence he ansi ion o he i s and he second child and de i e
co esponding hypo heses. D awing on amily economic heo y (Becke 1981)and
empi ical esea ch, we explo e he employmen condi ions and job cha ac e is ics
ha inc ease he likelihood o ha ing child en (Sec . 2.1). We hen ou line he speci ic
cha ac e is ics o ci il se ice employmen and discuss hei po en ial impac on
e ili y (Sec . 2.2). Fu he mo e, we conside di e ences be ween he i s and he
second child as well as be ween women and men (Sec . 2.3) be o e o mula ing
hypo heses (Sec . 2.4).
2.1 Theo e ical Explana ions and Empi ical Resul s
The amily economic model is one o he mos in luen ial app oaches o explain he
in e play be ween job si ua ion and e ili y beha io . Becke (1981) p ima ily a -
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 325
ibu es he economic in luence on e ili y o a cos –bene i analysis in which pa en s
make a ional decisions ega ding whe he o ha e child en and how many. Child en
a e seen as a inancial and ime-in ensi e in es men . The cos s o ha ing child en
include bo h di ec expenses, such as hose o ood and clo hing, and oppo uni y
cos s incu ed by pa en s. Oppo uni y cos s a ise om o egoing al e na i e uses
o ime and esou ces, such as ca ee ad ancemen . Fac o s ha inc ease income
(posi i e income e ec ) o educe oppo uni y cos s (by mi iga ing nega i e subs i-
u ion e ec s) a e gene ally conduc i e o e ili y. To maximize household income
and minimize oppo uni y cos s, he di ision o labo is conside ed economically
ad an ageous (Becke 1985). Acco dingly, he indi idual wi h he lowes ma ke
wage is mo e likely o assume pa en ing esponsibili ies as a a ional economic
choice. Al hough income e ec s and oppo uni y cos s apply o bo h women and
men, women ha e adi ionally aken on he p ima y ca egi ing ole, as hey end o
ea n lowe incomes han hei male pa ne s, while men ha e ocused p ima ily on
paid employmen (Köppen 2006).
Howe e , in he amewo k o amily economic heo y, o he ac o s o a ange-
men s ha aise income and educe oppo uni y cos s a e possible o include. We
i s concen a e on oppo uni y cos s. Wi h he inc easing labo ma ke pa icipa-
ion o women in ecen decades, emale employmen oppo uni ies and wages ha e
isen, leading o highe oppo uni y cos s associa ed wi h ha ing child en (Köppen
2006). As a esul , he adi ional gende -speci ic di ision o household and ca egi -
ing esponsibili ies has become mo e cos ly and less p e alen . A he same ime,
legal egula ions and s a e p o isions ha e imp o ed he econcilia ion o amily and
paid wo k, he eby educing oppo uni y cos s. S udies ha e shown ha app oaches
such as a highe le el o child ca e p o ision (G eulich e al. 2017;K ap 2013)o
he ex ension o pa en al bene i s and lea e egula ions (Thomas e al. 2022)ha e
allowed o a be e econcilia ion and had a posi i e in luence on e ili y (Luci-
G eulich and Thé enon 2013,2014).
In addi ion o hese b oade app oaches linked o legal egula ions, his a icle
ocuses on job- ela ed ac o s and how speci ic job cha ac e is ics can educe op-
po uni y cos s. Key aspec s o wo k– amily econcilia ion include he abili y o
emain in he wo k o ce a e s a ing a amily and ensu ing ha pa en hood does
no hinde ca ee p og ession. Wo k–li e compa ibili y is expec ed o be highe when
job cha ac e is ics minimize he oppo uni y cos s o aising child en by sa egua d-
ing human capi al in es men s and acili a ing he alignmen o paid employmen
wi h amily esponsibili ies (Begall and G unow 2015; Voydano 2005). While he
o e all posi i e impac o imp o ed wo k– amily econcilia ion has been well doc-
umen ed (Doepke e al. 2023; G eulich e al. 2017; Luci-G eulich and Thé enon
2013,2014), s udies speci ically examining single job- ela ed ac o s ha educe
oppo uni y cos s emain su p isingly sca ce. An excep ion a e s udies add essing
he posi i e in luence o pa - ime wo k (A iza e al. 2005) and a mo e amily-
iendly business cul u e caused by a gende ed s uc u e o he speci ic labo ma ke
and social in e ac ions in he wo kplace (Cook and Minno e 2008).
The second cen al in luence in he amily economic heo y is income (Becke
1981). The in es men in child en equi es he p o ision o a secu e and su icien
income, and a highe le el o income has been shown o in luence e ili y posi-
K
326 P. S. Löwe
i ely (Doepke e al. 2023; Fox e al. 2019). Besides he le el o income, he le el o
employmen secu i y is c ucial. The secu i y o ha ing a long- e m wo king a ange-
men ha includes income secu i y, anspa en and s able ca ee de elopmen , and
plannabili y as well as p edic abili y is seen as impo an , as long- e m binding com-
mi men s such as pa en hood equi e a secu e economic basis (Oppenheime 1988).
Se e al s udies examining he impac o uns able and insecu e employmen , pa icu-
la ly ixed- e m con ac s, epo nega i e e ec s on e ili y (Aue and Danze 2014;
B ose 2008; Kind and Kleib ink 2013), whe eas o he s ind no signi ican e ec s
(Ba on and Rapp 2019).
Al hough income e ec s and oppo uni y cos s a ec bo h women and men, and
he inc easing labo ma ke pa icipa ion o women has al e ed income and oppo u-
ni y s uc u es (Köppen 2006), he di ision o amily and economic esponsibili ies
emains gende ed (Cooke 2004; Lewin-Eps ein e al. 2006;Sh e le e al.2010).
Mos women as well as men pa icipa e in paid employmen and he labo ma ke ,
bu women wo k ewe hou s (S a is ik de Bundesagen u ü A bei 2021). Men
p o ide he majo i y o he income (Dieckho e al. 2020), and women do he majo -
i y o he amily wo k (BMFSFJ 2019). In his con ex , wo k– amily econcilia ion
measu es a e pa icula ly bene icial o women, while he p o ision o a high and
secu e income is ad an ageous o bo h sexes, hough especially o men.
Agains he discussed heo e ical backg ound, we explain he speci ic condi ions
o employmen in he ci il se ice (Sec . 2.2) and discuss how hey a e likely o
in luence e ili y p ocesses.
2.2 The Ge man Ci il Se ice and Fe ili y
How does ci il se ice employmen impac income le els and oppo uni y cos s by
acili a ing he econcilia ion o wo k and amily li e?
The ci il se ice p o ides jobs ha , unlike he p i a e sec o , a e ela i ely
shielded om he ope a ion o ma ke o ces (Esping-Ande sen 1993). The si u-
a ion in he ci il se ice is no cha ac e ized by a capi alis logic o pe o mance
and exploi a ion bu ollows a bu eauc a ic logic (Becke 1993). Unlike in he p i-
a e sec o , he p ima y objec i e o he ci il se ice is no p o i gene a ion bu
he p o ision o public se ices (Cze wick 2007). Many public se ices, in p inci-
ple, canno be p o ided (s a e-de ining a ea, e.g., adminis a i e sys em, police) o
should no be p o ided in a comple ely ma ke able way (wel a e s a e, e.g., educa-
ion, social a ai s; s a e-economic a ea, e.g., public anspo , was e managemen ;
Becke 1993).
The low ma ke o ien a ion acili a es he de elopmen o s uc u es ha insu-
la e he ci il se ice om sho - e m economic p essu es and compe i ion, he eby
p omo ing job and income s abili y (Adse a 2004;Löwe2022; Mills and Bloss eld
2005). I p omo es he exis ence o s anda dized, secu e, and long- e m employmen
ela ionships wi h anspa en ca ee pa hs and opens up oom o mo e employee-
o ien ed a angemen s o secu e coope a ion (Go schall e al. 2015; Saue e al.
2022). As a esul , ca ee pa hs a e clea and highly dependen on o mal quali i-
ca ions and senio i y (Becke 1993; an de We ho s 2011). Al hough he Ge man
economy is highly egula ed and coo dina ed, hese cha ac e is ics a e e en mo e
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 327
p onounced in he ci il se ice. Wi h en ance in o he ci il se ice, he pa h o
u he ca ee de elopmen ega ding posi ions and income is known o mos o
he employees. Collec i e ag eemen s o legally binding egula ions o malize hese
s anda dized ca ee pa hs in he ci il se ice. In con as o he p i a e sec o , col-
lec i e egula ions a e alid o he whole ci il se ice s a (Becke 1993; Ellgu h
and Kohau 2011;S uck2006).
The lowe ma ke o ien a ion also has consequences o wage dis ibu ion in he
ci il se ice. The dependence on economic down u ns is educed, bu he possibili y
o pa icipa e in economic up u ns is hampe ed. This suppo s wage s abili y bu
nega i ely a ec s high income po en ials, e.g., lack o pa icipa ion in p o i s (Becke
1993; Löwe and Vale 2023). The la ened wage dis ibu ion in he ci il se ice
is ein o ced by he equalizing e ec o collec i e wage ag eemen s (Hay e and
Weinbe g 2011) and he s a e’s ole as an employe commi ed o social balance
(Tepe and K oos 2010). S udies highligh he impac o hese s uc u al di e ences,
showing sligh ly highe incomes o he s eadily declining g oup o low- and middle-
skilled employees, while highly quali ied employees ea n signi ican ly less in he
ci il se ice compa ed o he p i a e sec o (Co neo 2014; Tepe and K oos 2010;
Wal e 2007).
O e all, he in luence o ci il se ice employmen on income is ambiguous. On
one hand, a ca ee in he ci il se ice is cha ac e ized by a high le el o s abili y
and plannabili y, ac o s ha make he long- e m in es men o becoming pa en s
o ex ending a amily mo e likely because he esou ce income is pe manen ly
a ailable and insecu i ies a e educed. On he o he hand, especially o high-income
employees who can be expec ed o p o ide a high sha e o income o he household,
incomes a e lowe compa ed o he p i a e sec o , which implies a nega i e income
e ec on e ili y. Wi h he compa a i ely good and s able economic si ua ion o
young people in Ge many (Die ich 2012), we expec he income le el e ec o
domina e, which implies a nega i e in luence o ci il se ice employmen on e ili y.
In addi ion o ensu ing a secu e and su icien income, a second key ac o in lu-
encing e ili y is a job’s abili y o educe he oppo uni y cos s o ha ing a amily.
Reconcilia ion ac o s o he job include bo h he abili y o s ay in he wo k o ce
a e s a ing a amily and he assu ance ha pa en hood will no comp omise ca ee
de elopmen .
The abili y o emain in he wo k o ce is acili a ed by mo e gene ous een y
igh s in he ci il se ice compa ed o he p i a e sec o . Fo each child, pe sonnel
in he ci il se ice ha e he (minimum) op ion o aking i e yea s o pa en al lea e,
wi h a legal en i lemen o he old job posi ion. Nega i e consequences o ca ee
de elopmen a e educed by he ac ha he e u n igh s also include g ouping in
e ms o sala y, which ex ends he pe iod in which een y in o he job a e pa en al
lea e is possible wi hou losing he senio i y wage en i lemen s acqui ed be o e
pa en al lea e (§ 28; § 17 pa a. 3 sen ence 2 TVöD). In con as , legal en i lemen s
in he p i a e sec o a e limi ed o h ee yea s (§ 15 pa a. 2 BEEG; § 5 no. 1 RL
2010/18/EU).
A e he e u n o paid employmen , wo k– amily econcilia ion is u he a-
cili a ed by a highe a ailabili y o pa - ime posi ions. The ci il se ice posi ions
i sel as a amily- iendly employe and ac i ely p omo es pa - ime wo k (Al maie
K
328 P. S. Löwe
2019). Unlike p i a e sec o wo ke s, ci il se ice s a ha e long had mo e gene ous
legal igh s o wo k pa - ime and o a ange hei wo king hou s mo e lexibly i
hey ha e a child (Bundesminis e ium des Inne en 2014). This is e lec ed in a highe
sha e o pa - ime employmen in he ci il se ice (32%) han in he economy as
a whole (S a is isches Bundesam 2017a, b).
Nega i e ca ee consequences a e also educed by he ull co e age o collec-
i e ba gaining ag eemen s and legally binding egula ions in he ci il se ice. The
po en ial o disc imina ion agains pa en s is limi ed, as supe io s ha e es ic ed
disc e ion o e p omo ions and sala y inc eases, wi h o mal quali ica ions and se-
nio i y se ing as key c i e ia (Ellgu h and Kohau 2011; Go schall e al. 2015).
Ano he ac o is a mo e amily- iendly wo k en i onmen . The highe pe cen -
ages o emales in a business sec o occupa ion ha e been shown o esul in a mo e
amily- iendly cul u e (Cook and Minno e 2008). In he ci il se ice, women domi-
na e in indi idual sec o s, such as heal h (62%) and educa ion (64%), and he o e all
ci il se ice also has a much highe p opo ion o women (56%) han he economy
as a whole (46%; S a is isches Bundesam 2017a, b).
O e all, employmen in he ci il se ice is expec ed o educe bo h he oppo uni y
cos s associa ed wi h ca ee in e up ions and he cos s incu ed upon e u ning o
wo k. Mo e gene ous een y igh s mi iga e he inancial and p o essional impac
o ca ee b eaks, while he g ea e a ailabili y o pa - ime posi ions, he highe
p opo ion o women in he wo k o ce, and s onge p o ec ions agains pa en al
disc imina ion h ough collec i e ag eemen s u he suppo wo k– amily balance
a e eemploymen . Consequen ly, hese ac o s a e likely o lowe oppo uni y cos s
and ha e a posi i e e ec on e ili y.
2.3 The Numbe o Child en and Gende Di e ences
The impo ance o oppo uni y cos s, income le el, and income secu i y a ies de-
pending on whe he he ocus is on he ansi ion o he i s o he second child.
A close examina ion o he unde lying p ocesses in luencing each ansi ion is
he e o e essen ial.
When income secu i y and wo k– amily econcilia ion a e no ensu ed, he an-
si ion o pa en hood may be pos poned, leading o an inc ease in he age a i s bi h
(Begall and Mills 2011; Be nha d 1993). We p e iously discussed how ci il se ice
employmen should a ec hese ac o s. While he in luence o ci il se ice employ-
men on employmen secu i y, income secu i y, and he econcilia ion o amily and
wo k can be expec ed o be posi i e, he le el o income should ha e a nega i e
impac .
Fo he second child he same ac o s a e ele an , bu p ocesses ha we e al-
eady challenging wi h one child in ensi y (Vlasblom and Schippe s 2006). The
ime in es men equi ed o ca egi ing inc eases wi h he a i al o a second child,
along wi h ising inancial expendi u es. S a egies ha may ha e been iable wi h
one child, such as educing leisu e o couple ime, become mo e cons ained wi h
he a i al o a second child (B ews e and Rind uss 2000; G eulich e al. 2017;
Vlasblom and Schippe s 2006). On one hand, he g ea e ime in es men limi s he
possibili y o simply inc easing wo king hou s o co e highe cos s, he eby plac-
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 335
We chose a me hod ha can add ess he issue such ha he p ocess ime ( he
ime a isk o expe ience a i s o second bi h; o a desc ip ion o p ocess ime,
see Sec . 3.1) and he obse ed p ocess ime h ough he panel su ey Pai am a e
no iden ical. Fi s and second bi hs aking place be o e he obse a ion pe iod o
Pai am and ime- a ying co a ia es a e no a ailable e ospec i ely. Al hough we
know he s a o he p ocess ime, some esponden s en e ed he su ey a olde
ages and could no be obse ed in he i s yea s o po en ial pa en hood. In his
se ing, a me hod ha is based on a “condi ional likelihood,” such as disc e e his o y
models, is ecommended (A ánz Becke and Lois 2015; Jenkins 1995).
Fo he es ima ion o he e en his o y models, we p epa ed he da a in long o ma
as p oposed by Jenkins (1995) on a yea ly base, and each esponden con ibu ed
as many obse a ions o his eo ganized da a se as he e we e yea s be ween he
in e iew in wa e one and he bi h o a i s /second child o being igh -censo ed.
Because c ucial in o ma ion on employmen s a us is a ailable only on a yea ly
basis, he e en his o y models a e speci ied wi h a piecewise cons an speci ica ion.
A piecewise cons an unc ion allows he es ima ion o indi idual haza d a es o
di e en ime in e als. This e lec s he di e en magni udes o ime-dependen
e ec s and p e en s o e es ima ion o unde es ima ion o e ec s. Fo he ansi ion
o he i s child, we c ea ed i e ime in e als based on age (14–20, 21–25, 26–30,
31–35, 36–49). Fo he ansi ion o he second child, we c ea ed se en ime in e als
based on he yea s a e he bi h o he i s child (1, 2, 3, 4, 5, 6–10, 11–32).
We epo a e age ma ginal e ec s a he han log-odds o odds a ios because
hey allow a mo e in ui i e in e p e a ion o p obabili ies (Bes and Wol 2010).
They show he a e age e ec on p obabili y in pe cen age poin s. In ou se ing,
his ep esen s he p obabili y (di e ence) o he ansi ion o he i s child being
condi ional on employmen in he ci il se ice o he p i a e sec o .
3.5 Analy ical S a egy
We s a wi h a desc ip i e o e iew o he a iables used in he mul i a ia e e en
his o y models o highligh po en ial di e ences be ween he ci il se ice and he
p i a e sec o ha a e accoun ed o in ou analyses. Table 1 ollows he s uc u e
o he e en his o y models and dis inguishes be ween ci il se ice and he p i a e
sec o , as well as be ween women and men and he i s and second bi hs.
Nex , we c ea ed mul i a ia e e en his o y models o he ansi ion o he i s
(Table 2) and o he second child (Table 3) o women and men. We illus a e he
esul s wi h ma gin-plo s showing he le el and he e ec (including signi icance)
o ci il se ice s. p i a e sec o employmen o he ime in e als o he piecewise
cons an speci ica ion (Figs. 1,2,3,4,5,6,7,and8).
In he Online Appendix we p o ide se e al obus ness checks o he e ogenei ies
based on ixed- e m and pe manen con ac s, he wo king s a us o he pa ne and
eas e n and wes e n Ge many (Sec . 2 in Online Appendix, Tables 2–13).
K
336 P. S. Löwe
4Resul s
4.1 Desc ip ion o he Sample
The desc ip i e esul s o Table 1show ha employees in he ci il se ice and
he p i a e sec o di e ed subs an ially ega ding he a iables conside ed in he
e en his o y models. The p esen ed pe cen ages indica e ha mos obse ed spells
and ansi ions occu ed in he p i a e sec o , e lec ing he la ge p opo ion o
employees wo king in he p i a e sec o compa ed o he ci il se ice.
The educa ion le el in he ci il se ice was much highe o esponden s and
hei pa ne s. Mo e employees had Ge man ci izenship. The pe cen age o pa ne s
who we e in paid employmen was highe in he ci il se ice (excep o women’s
pa ne s o he second child). The sha e o women in he ci il sec o was highe han
in he p i a e sec o . The age o he ansi ions o he i s and o he second child
was sligh ly highe in he ci il se ice. The pa ne ship s a us di e ed ega ding
gende and be ween he samples o he i s and he second child. While in bo h
samples bo h sexes mo e o en had a pa ne , women in he ci il se ice we e less
o en ma ied in he i s -child sample and mo e o en ma ied in he second-child
sample. In con as , men in he ci il se ice exhibi ed pa ne ship pa e ns mo e
simila o hose o men in he p i a e sec o . These desc ip i e indings highligh
subs an ial di e ences be ween ci il se ice and p i a e sec o employmen , which
a e accoun ed o in he mul i a ia e e en his o y models in he ollowing sec ion.
4.2 Resul s o Disc e e E en His o y Analyses
The mul i a ia e disc e e e en his o y analyses in es iga ed sec o al di e ences in
he ansi ions o he i s and o he second child o bo h women and men. We
p esen sepa a e ables o he ansi ion o he i s child and he ansi ion o he
second child. To enhance he in e p e a ion o ou indings, we supplemen he epo
wi h g aphical illus a ions ha p o ide addi ional insigh s in o he esul s.
4.2.1 T ansi ion o he Fi s Child
Table 2p esen s he baseline model and he model wi h con ol a iables o he e en
his o y analyses o he ansi ion o he i s child sepa a ely o women (models 1
and 2) and men (models 3 and 4).
Women. The bi a ia e esul s (Table 2, model 1) o women show a null inding.
The esul s indica e nei he a subs an i ely ele an e ec (0.00 pe cen age poin s)
no a s a is ically signi ican e ec (p<0.995), sugges ing no associa ion be ween
employmen in he ci il se ice and ansi ion o he i s child. In he second model
o Table 1, which accoun s o socios uc u al cha ac e is ics discussed in Sec . 3.3,
he e ec o ci il se ice employmen on e ili y beha io emains la gely un-
changed (–0.2 pe cen age poin s) and is s a is ically insigni ican (p< 0.702).
Figu es 1and 2illus a e he di e ences and le els in he p obabili ies o an-
si ioning o he i s child, no only in e ms o he o e all e ec bu also ac oss
each ime in e al wi hin he piecewise-cons an speci ica ion. These igu es allow
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 337
Table 1 Desc ip ion o he sample o he e en his o y models
Fi s child Second child
Women Men Women Men
Ci il
se ice
P i a e
sec o
Ci il
se ice
P i a e
sec o
Ci il
se ice
P i a e
sec o
Ci il
se ice
P i a e
sec o
Va iables o he mul i a ia e e en his o y models
Sec o (ci il se ice and p i a e sec o ) (%) 27.30 72.70 13.28 86.72 22.80 77.20 15.45 84.55
Sha e o ansi ions ( i s /second bi hs) (%) 26.77 73.23 15.46 84.54 31.45 68.52 15.93 84.01
Gende (sha e o women/men in he sec o ) (%) 60.96 40.61 39.04 59.39 69.09 55.23 30.91 44.77
Educa ion le el
In educa ion – – – – – – – –
Low (%) 7.23 11.51 11.35 14.64 8.78 14.06 6.54 12.49
Middle (%) 55.99 62.85 51.28 66.41 62.58 67.69 53.22 68.26
High (%) 36.78 25.63 37.37 18.95 28.63 18.25 40.24 19.26
Ci izenship (Ge man compa ed o non-Ge man) (%) 95.26 93.02 96.66 93.85 93.33 89.77 94.37 90.37
Pa ne ship s a us
Ma ied and cohabi a ing (%) 18.64 21.48 16.47 16.22 68.01 60.21 66.87 67.73
Cohabi a ion wi hou ma iage (%) 31.70 28.46 25.56 22.01 16.80 17.59 24.42 20.19
Li ing apa oge he (%) 18.35 16.99 16.76 16.99 5.49 7.48 2.52 4.34
No pa ne (%) 31.31 33.07 41.20 44.77 9.70 14.72 6.18 7.73
Educa ion le el o pa ne
In educa ion (%) 3.03 2.76 5.52 5.52 0.59 0.57 1.93 1.19
Low (%) 5.83 6.37 4.40 6.45 8.88* 10.81 6.28% 13.41
Middle (%) 39.22 39.65 30.57 29.61 53.69 56.86 53.52 59.59
High (%) 20.62 18.16 18.31 13.65 27.19 17.53 32.08 19.39
No pa ne 31.31 33.07 41.20 44.77 9.70 14.72 6.18 7.73
Wo king s a us o pa ne
No paid employmen (%) 7.80 8.04 10.76 13.45 6.60 4.49 26.53 32.37
K
338 P. S. Löwe
Table 1 (Con inued)
Fi s child Second child
Full- ime paid employmen (%) 56.69 56.30 39.37 35.39 78.97 77.87 27.17 27.85
Pa - ime paid employmen (%) 4.21 2.59 8.67 6.39 4.73 2.54 40.12 32.04
No pa ne 31.31 33.07 41.20 44.77 9.70 14.72 6.18 7.73
Bi h coho
1991–1993 (%) 21.64 26.50 18.44 22.92 3.32 5.82 5.14 6.46
1981–1983 (%) 49.44 48.35 45.58 47.91 40.82 41.18 37.61 41.46
1971–1973 (%) 28.92 25.15 35.99 29.17 55.85 53.00 57.26 52.08
2001–2003 (%) 00000000
Piecewise cons an ime in e als
Fi s child: age g oups
a isk i s child
Second child: yea s a e he
bi h o he i s child
14–20 (%) 1(%) 2.39 4.03 2.80 3.67 9.44 8.03 10.52 13.22
21–25 (%) 2(%) 14.28 17.20 11.08 14.67 10.39 9.31 12.65 14.50
26–30 (%) 3(%) 38.71 38.00 32.96 33.88 9.53 7.89 13.61 11.01
31–35 (%) 4(%) 13.91 13.26 15.07 14.38 8.44 7.13 7.69 8.39
36–49 (%) 5(%) 30.73 27.51 38.09 33.40 6.34 6.40 7.03 07.14
–6–10 (%) – – – – 26.72 27.28 22.95 22.61
–11–32 (%) – – – – 29.14 33.96 25.55 23.14
Age a he ansi ion o i s /second child; mean alues
in yea s
30.15 29.52 31.53 31.51 31.87 31.10 33.86 33.67
N(numbe o spells) 2027 5591 1298 8178 1207 3941 540 3195
The numbe s e e o spells in which employees we e wo king in he ci il se ice o he p i a e sec o . The able includes only cases in which he esponden was in he labo
o ce s a us o “employed dependan .” As a consequence, he labo o ce s a us is no shown, and o he a iable educa ion le el, no esponden could be “in educa ion.”
Sou ce: Pai am 2008–2020; au ho ’s calcula ions, wi h weigh s
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 339
Table 2 T ansi ion o i s child o women and men: e en his o y models, a e age ma ginal e ec s
Women Men
Basic model
(1)
Wi h con ols
(2)
Basic model
(3)
Wi h con ols
(4)
Sec o ( e e ence: p i a e sec o )
Ci il se ice 0.000
(0.005)
–0.002
(0.005)
0.008
(0.005)
0.005
(0.005)
Piecewise-cons an ime in e als ( e e ence: 14–20)
21–25 0.020***
(0.006)
0.009
(0.024)
0.012**
(0.004)
0.014
(0.022)
26–30 0.036***
(0.008)
–0.015
(0.024)
0.028***
(0.004)
0.005
(0.021)
31–35 0.062***
(0.008)
–0.004
(0.024)
0.060***
(0.007)
0.019
(0.022)
36–49 0.006
(0.005)
–0.036
(0.024)
0.016***
(0.004)
–0.013
(0.021)
Labo o ce s a us ( e e ence: inac i e/unemployed)
In educa ion –0.022**
(0.007)
–0.224***
(0.006)
–0.000
(0.003)
0.004
(0.007)
Sel -employed 0.029
(0.025)
0.025
(0.021)
0.040**
(0.012)
0.027**
(0.010)
Employed 0.019
(0.007)
0.018**
(0.006)
0.029***
(0.003)
0.024***
(0.004)
Con ols
Educa ion le el ( e e ence: middle)
In educa ion – 0.731***
(0.003)
– –0.020**
(0.007)
Low – –0.007
(0.008)
– 0.000
(0.006)
High – 0.004
(0.005)
– 0.005
(0.004)
Ci izenship ( e e ence: non-Ge man)
Ge man – –0.002
(0.009)
– 0.004
(0.006)
Pa ne ship s a us ( e e ence: no pa ne )
Ma ied and cohabi a ing – 0.049***
(0.012)
– 0.031**
(0.009)
Cohabi a ing wi hou
ma iage
– 0.031***
(0.009)
– 0.018**
(0.007)
Li ing apa oge he – 0.010
(0.007)
– 0.006
(0.007)
Educa ion le el o pa ne ( e e ence: in educa ion)
Low – 0.019
(0.013)
– 0.014*
(0.008)
Middle – 0.010
(0.010)
– 0.016**
(0.008)
High – 0.027
(0.012)
– 0.024**
(0.008)
K
340 P. S. Löwe
Table 2 (Con inued)
Women Men
Basic model
(1)
Wi h con ols
(2)
Basic model
(3)
Wi h con ols
(4)
Wo king s a us o pa ne ( e e ence: no paid employmen )
Full- ime paid employ-
men
– –0.002
(0.007)
– 0.014**
(0.005)
Pa - ime paid employ-
men
– 0.002
(0.013)
– 0.005
(0.005)
Bi h coho ( e e ence: 1991–1993)
1981–1983 – 0.036***
(0.004)
– 0.018***
(0.003)
1971–1973 – 0.017**
(0.006)
– 0.017**
(0.006)
N(spells) 12241 12241 14484 14484
McFadden adjus ed R20.090 0.169 0.079 0.171
AIC 2612.868 2387.483 2934.855 2641.173
*p< 0.05, **p< 0.01, ***p< 0.001
Wi h weigh s; coe icien s as a e age ma ginal e ec s; s anda d e o s in pa en heses. Sou ce: Pai am
2008–2020; au ho ’s calcula ions
o he iden i ica ion o po en ial di e ences wi hin speci ic ime in e als, e en i
he o e all e ec emains null.
Figu e 1 ocuses on he di e ences be ween ci il se ice and p i a e sec o em-
ploymen , p o iding in o ma ion on whe he hese di e ences a e s a is ically sig-
Fig. 1 Women: di e ences in p obabili y o ansi ion o i s child be ween ci il se ice s. p i a e
sec o by ime, wi h con ols; 95% con idence in e als; wi h weigh s. Sou ce: Pai am 2008–2020; own
calcula ions
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 341
Fig. 2 Women: p obabili y o ansi ion o i s child by ime and sec o , wi h con ols; 95% con idence
in e als; wi h weigh s. Sou ce: Pai am 2008–2020; own calcula ions
ni ican . The igu e con i ms ha he null indings emain obus . Despi e some
luc ua ions, he e ec sizes emain close o ze o ( anging om –0.09 o 0.29 pe -
cen age poin s). The 95% con idence in e als include ze o o all ime pe iods,
indica ing ha he e ec s a e s a is ically insigni ican .
Figu e 2 ocuses on he p obabili y le els o each sec o , ci il se ice and p i a e,
illus a ing he likelihood o ansi ioning o pa en hood wi hin speci ic ime in e -
als. Unlike Fig. 1, which solely examines he exis ence o sec o al e ec s, Fig. 2
p o ides insigh s in o he absolu e p obabili y le els. This allows o an assessmen
o whe he obse ed e ec sizes a e subs an ial in ela ion o he o e all p obabili y
o ansi ioning in bo h sec o s. The p obabili y o ansi ioning o he i s child is
highes o indi iduals aged 21–25 (app oxima ely 6%) and lowes o hose aged
36–49 (app oxima ely 1.5%). In ela ion o he p obabili ies o making he ansi ion
o he i s child in he wo sec o s, he di e ences be ween ci il se ice and p i a e
sec o a e nei he subs an ial no s a is ically signi ican .
The expec ed posi i e in luence o employmen in he ci il se ice on he an-
si ion o he i s child among women canno be con i med. Ou indings e eal
nei he subs an ial no s a is ically signi ican coe icien sizes. The analyses sugges
ha employmen in he ci il se ice, compa ed o he p i a e sec o , does no ha e
a disce nible impac on he likelihood o ansi ioning o he i s child o women.
The e o e, H1, whe eby women employed in he ci il se ice a e expec ed o ha e
a highe ansi ion a e o he i s child han women employed in he p i a e sec o ,
is ejec ed.
Men. Fo men, model 3 and 4 o Table 2as well as Figs. 3and 4illus a e he
esul s o he e en his o y analyses. The bi a ia e esul s (Table 2, model 3) o men
K
342 P. S. Löwe
Fig. 3 Men: Di e ences in p obabili y o ansi ion o i s child be ween ci il se ice s. p i a e sec o by
ime, wi h con ols; 95% con idence in e als; wi h weigh s. Sou ce: Pai am 2008–2020; own calcula ions
Fig. 4 Men: p obabili y o ansi ion o i s child by ime and sec o , wi h con ols; 95% con idence
in e als; wi h weigh s. Sou ce: Pai am 2008–2020; own calcula ions
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 343
show a null inding. The esul s show a sligh ly nega i e (–0.8 pe cen age poin ) bu
s a is ically insigni ican (p< 0.175) e ec , indica ing no ela ionship be ween he
ansi ion o he i s child and a wo king posi ion in he ci il se ice. In model 4,
which accoun s o socios uc u al cha ac e is ics, he e ec o ci il se ice employ-
men on e ili y beha io shi s o a small posi i e e ec (0.5 pe cen age poin ) bu
emains s a is ically insigni ican (p< 0.353).
When examining he ime in e als sepa a ely, we obse e a sligh ly posi i e e ec
size ( anging om 0.27 o 0.70 pe cen age poin s, as shown in Fig. 3), sugges ing ha
he ansi ion a e o he i s child is highe o men employed in he ci il se ice.
Figu e 4indica es ha he p obabili y o ansi ioning o he i s child is highes
o he age g oup 31–35 (app oxima ely 3%) and lowes o he age g oup 36–49
(app oxima ely 0.5%). In ela ion o he o e all p obabili ies o making he ansi ion
in bo h sec o s, he di e ences be ween ci il se ice and p i a e sec o employmen
appea no ewo hy. Howe e , he 95% con idence in e als in Fig. 3all in e sec
wi h ze o, indica ing ha , ac oss all ime pe iods, he obse ed di e ences emain
s a is ically insigni ican .
The expec ed nega i e in luence o employmen in he ci il se ice on he an-
si ion o he i s child o men canno be con i med. Ra he , con a y o ou ex-
pec a ions, he coe icien sizes indica e a sligh ly posi i e in luence o ci il se ice
employmen . Howe e , i is impo an o emphasize ha hese di e ences a e no
s a is ically signi ican a any poin in ime. The e o e, H2, whe eby men employed
in he ci il se ice a e expec ed o ha e a lowe ansi ion a e o he i s child han
men employed in he p i a e sec o , is ejec ed.
4.2.2 T ansi ion o he Second Child
Fo he ansi ion o he second child, he esul s imply a highe ele ance o he
employmen si ua ion in he ci il se ice and he p i a e sec o .
Women. Fo he bi a ia e case o women (Table 3, model 1), we ind a subs an ial
posi i e (2.27 pe cen age poin s) and s a is ically signi ican (p< 0.008) ela ionship
be ween he ansi ion o he second child and a wo king posi ion in he ci il se ice.
The e ec emains obus a e adding con ols (Table 3; model 2). The ansi ion
a e o he second child is 2.0 pe cen age poin s highe in he ci il se ice han in
he p i a e sec o . The e ec is signi ican on an alpha le el o 5% (p< 0.040).
The analysis o he e ec ac oss ime in e als wi hin he piecewise cons an
speci ica ion (Fig. 5) e eals consis en ly posi i e e ec s, hough e ec sizes a y
conside ably. The di e ences be ween he ci il se ice and he p i a e sec o a e
smalles in he yea ollowing he bi h o he i s child (app oxima ely 1 pe cen age
poin ) and in he 11–32 yea s a e childbi h (0.9 pe cen age poin ). In con as ,
du ing he hi d and ou h yea s a e he bi h o he i s child, he p obabili y
o ansi ioning o a second child is mo e han 3 pe cen age poin s highe in he
ci il se ice. The di e ences be ween he ci il se ice and he p i a e sec o a e
s a is ically signi ican a he 5% alpha le el o ime in e als 3, 4, 5, and 6–10 and
a he 10% alpha le el o ime in e als 1, 2, and 11–32. Figu e 6 u he illus a es
ha an e ec size o 3 pe cen age poin s is subs an ial, gi en ha he p obabili y
K
344 P. S. Löwe
Table 3 T ansi ion o second child o women and men: e en his o y models, a e age ma ginal e ec s
Women Men
Basic model
(1)
Wi h con ols
(2)
Basic model
(3)
Wi h con ols
(4)
Sec o ( e e ence: p i a e sec o )
Ci il se ice 0.027**
(0.011)
0.020*
(0.010)
–0.012
(0.012)
–0.018
(0.011)
Piecewise-cons an ime in e als ( e e ence: 1)
2 0.085**
(0.025)
0.050**
(0.015)
0.095***
(0.020)
0.068***
(0.015)
3 0.122***
(0.026)
0.083***
(0.016)
0.145***
(0.023)
0.121***
(0.019)
4 0.115***
(0.027)
0.097***
(0.019)
0.148***
(0.026)
0.139***
(0.023)
5 0.063
(0.024)
0.066***
(0.018)
0.073**
(0.022)
0.078***
(0.022)
6–10 –0.038
(0.016)
0.029*
(0.012)
0.027*
(0.012)
0.039**
(0.013)
11–32 –0.038
(0.015)
–0.003
(0.011)
–0.013
(0.012)
–0.005
(0.011)
Labo o ce s a us ( e e ence: inac i e/unemployed)
In educa ion 0.056*
(0.025)
0.073**
(0.028)
0.032
(0.026)
0.049
(0.034)
Sel -employed 0.002
(0.019)
–0.001
(0.021)
0.005
(0.023)
0.002
(0.028)
Employed 0.048***
(0.009)
0.040***
(0.010)
0.065***
(0.012)
0.055**
(0.017)
Con ols
Educa ion le el ( e e ence: middle)
In educa ion – –0.035
(0.020)
– –0.050*
(0.026)
Low – –0.015
(0.011)
– –0.011
(0.016)
High – 0.031**
(0.010)
– 0.013
(0.013)
Ci izenship ( e e ence: non-Ge man)
Ge man – 0.013
(0.012)
– 0.023
(0.015)
Pa ne ship s a us ( e e ence: no pa ne )
Ma ied and cohabi a ing – 0.070***
(0.010)
– 0.072***
(0.012)
Cohabi a ing wi hou
ma iage
– 0.036***
(0.010)
– 0.039**
(0.013)
Li ing apa oge he – 0.027*
(0.014)
– 0.015
(0.021)
Educa ion le el o pa ne ( e e ence: in educa ion+low)
Middle – 0.002
(0.011)
– 0.017
(0.013)
High – 0.039**
(0.014)
– 0.032*
(0.016)
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 351
Doepke, Ma hias, Anne Hannusch, Fabian Kinde mann and Michèle Te il . 2023. The Economics o
Fe ili y: a New E a. In Handbooks in economics, Handbook o he economics o he amily, eds.
Shelly Lundbe g and Alessand a Voena, 151–254. Ams e dam, Ox o d: No h-Holland an imp in o
Else ie .
Ellgu h, Pe e , and Susanne Kohau . 2011. De S aa Als A bei gebe : Wie un e scheiden sich die A bei s-
bedingungen zwischen ö en lichem Sek o und de P i a wi scha ? Indus ielle Beziehungen 18(1-
2):11–38.
Esping-Ande sen, Gøs a. 1993. Pos -Indus ial Class S uc u es: An Analy ical F amewo k. In Changing
Classes. S a i ica ion and Mobili y in Pos -Indus ial Socie ies, ed. Gøs a Esping-Ande sen, 1–31.
London: Sage.
Esse , Ha mu . 2006. Mig a ion, language and in eg a ion. AKI Resea ch Re iew No. 4 (WZB) (4):1–117.
h ps://www.wzb.eu/www2000/al /aki/ iles/aki_ esea ch_ e iew_4.pd (Accessed: 3 Ma ch 2017).
Fox, Jona han, Sebas ian Klüsene and Mikko My skylä. 2019. Is a Posi i e Rela ionship Be ween Fe ili y
and Economic De elopmen Eme ging a he Sub-Na ional Regional Le el? Theo e ical Conside a-
ions and E idence om Eu ope. Eu opean Jou nal o Popula ion 35(3):487–518.
F ejka, Tomas, and Jean-Paul Sa don. 2007. Coho Bi h O de , Pa i y P og ession Ra io and Pa i y Dis-
ibu ion T ends in De eloped Coun ies. Demog aphic Resea ch 16:315–374.
F ejka, Tomas, and Tomáš Sobok a. 2008. Fe ili y in Eu ope: Di e se, delayed and below eplacemen . In
Childbea ing ends and policies in Eu ope. Demog aphic Resea ch, Special Collec ion, eds. Tomas
F ejka, Tomáš Sobo ka, Jan M. Hoem and Lau en Toulemon,15–46. Ros ock: Max Planck Ins i u e
o Demog aphic Resea ch.
F ode mann, Co inna, Dana Mülle and Ma in Ab aham. 2013. De e minan en des Wiede eins iegs on
Mü e n in den A bei sma k in Vollzei ode Teilzei . Kölne Zei sch i ü Soziologie und Sozialpsy-
chologie 65(4):645–668.
Gebel, Michael, and Johannes Giesecke. 2009. Ökonomische Unsiche hei und Fe ili ä . Die Wi kung on
Beschä igungsunsiche hei und A bei slosigkei au die Familieng ündung in Os - und Wes deu sch-
land. Zei sch i ü Soziologie 38(5):399–417.
Giesecke, Johannes. 2006. A bei sma k lexibilisie ung und Soziale Ungleichhei : Sozio-ökonomische Kon-
sequenzen be is e e Beschä igungs e häl nisse in Deu schland und G oßb i annien. Wiesbaden:
VS Ve lag ü Sozialwissenscha en.
Golds ein, Joshua R., and Michaela K eyen eld. 2011. Has Eas Ge many O e aken Wes Ge many? Re-
cen T ends in O de -Speci ic Fe ili y. Popula ion and De elopmen Re iew 37(3):453–742.
Go schall, Ka in, Be nha d Ki el, Kend a B iken, Jan-Ocko Heue , Syl ia Hills, Sebas ian S eb and
Ma kus Tepe. 2015. Public Sec o Employmen Regimes. T ans o ma ion o he S a e as an Employe .
London: Palg a e Macmillan.
G eulich, Angela, Ma hilde Gue goa -La i iè e and Oli ie Thé enon. 2017. Employmen and second
childbi hs in Eu ope. Popula ion 72(4):625–648
Hay e , Susan, and B adley Weinbe g. 2011. Mind he Gap: Collec i e Ba gaining and Wage Inequali y.
In The ole o collec i e ba gaining in he global economy: Nego ia ing o Social Jus ice,ed.Susan
Hay e ,136-186. Chel enham e c.: Edwa d Elga .
Hohendanne , Ch is ian, and Hans-Die e Ge ne . 2010. Die Übe nahme be is e Beschä ig e im Kon ex
be iebliche Pe sonalpoli ik. Soziale Wel 61(1):27–50.
Hohendanne , Ch is ian, Es he Os meie and Philipp Ramos-Loba o. 2015. Be is e e Beschä igung
im ö en lichen Diens : En wicklung, Mo i e und ech liche Umse zung. IAB-Fo schungsbe ich
(12):1–178.
Huinink, Johannes, Jose B üde l, Be nha d Nauck, Sabine Walpe , Lau a Cas iglioni and Michael Feld-
haus. 2011. Panel analysis o in ima e ela ionships and amily dynamics (pai am): Concep ual
amewo k and design. Zei sch i ü Familien o schung/Jou nal o Family Resea ch 23(1):77–101.
Jenkins, S ephen P. 1995. Easy Es ima ion Me hods o Disc e e-Time Du a ion Models. Ox o d Bulle in
o Economics and S a is ics 57(1):129–136.
Kelle , Be nd , and Ha mu Sei e . 2014. A ypische Beschä igungs e häl nisse im ö en lichen Diens .
WSI-Mi eilungen 67(8):628–638.
Kind, Michael, and Jan Kleib ink. 2013. Soone o La e : Economic Insecu i y and he Timing o Fi s
Bi h. Ruh Economic Pape s (422):1–23.
Kohle , Hans-Pe e , F ancesco C. Billa i and José A. O ega. 2006. Low Fe ili y in Eu ope: Causes, Im-
plica ions and Policy Op ions. In The Baby Bus : Who will do he Wo k? Who Will Pay he Taxes? Ed.
F ed R. Ha is, 48–109. Lanham, MD: Rowman and Li le ield.
Köppen, Ka ja. 2004. The compa ibili y be ween wo k and amily li e—an empi ical s udy o second bi h
isks in Wes Ge many and F ance. Ros ock: Max Planck Ins i u e o Demog aphic Resea ch.
K
352 P. S. Löwe
Köppen, Ka ja. 2006. Second Bi hs in Wes e n Ge many and F ance. Demog aphic Resea ch 14:295–330.
K ap , Sand a. 2013. Public Childca e P o ision and Fe ili y Beha io : A Compa ison o Sweden and
Ge many. 1s ed. Opladen: Ba ba a Bud ich.
K eyen eld, Michaela. 2003. C isis o Adap a ion Reconside ed: A Compa ison o Eas and Wes Ge -
man Fe ili y Pa e ns in he Fi s Six Yea s a e he ‘Wende’. Eu opean Jou nal o Popula-
ion 19:303–329.
K eyen eld, Michaela. 2010. Unce ain ies in emale employmen ca ee s and he pos ponemen o pa en -
hood in Ge many. Eu opean Sociological Re iew 26(3):351–366.
K eyen eld, Michaela, and Di k Konie zka. 2008a. Bleib alles ande s. Gebu en- und Familienen wicklung
in Os - und Wes deu schland. In Demog a ische Wandel, ed. Nikkolaus We z, 50–70. Baden-Baden:
Nomos.
K eyen eld, Michaela, and Di k Konie zka. 2008b. Educa ion and Fe ili y in Ge many. In Demog aphic
Change in Ge many, eds. Ing id Hamm, Helmu Sei z and Ma in We ding, 165–187. Be lin:
Sp inge .
K eyen eld, Michaela, and Co dula Zabel. 2005. Female Educa ion and he Second Child: G ea B i ain
and Wes e n Ge many Compa ed. Schmolle s Jah buch – Zei sch i ü Wi scha s- und Sozialwis-
senscha en. Jou nal o Applied Social Science S udies 125(1):145–156.
K eyen eld, Michaela, Di k Konie zka, Philippe Lambe and Vincen J. Ramos. 2023. Second Bi h Fe -
ili y in Ge many: Social Class, Gende , and he Role o Economic Unce ain y. Eu opean Jou nal o
Popula ion 39(1):5.
Ku z, Ka in, Nikolei S einhage and Ka in Golsch. 2005. Case s udy Ge many: Global compe i ion, unce -
ain y and he ansi ion o adul hood. In Globaliza ion, Unce aini y and You h in Socie y, eds. Hans-
Pe e Bloss eld, E ik Klijzing, Melinda Mills and Ka in Ku z, 51–82. New Yo k: Rou ledge.
Lau ijssen, Ilse, and Ignace Glo ieux. 2013. Ca ee T ajec o ies o Women A e Childbi h: Job Quali y
and Wo k-Family Balance. Eu opean Sociological Re iew 29(3):426–436.
Lechne , Michael. 2001. The Empi ical Analysis o Eas Ge man Fe ili y a e Uni ica ion: An Upda e.
Eu opean Jou nal o Popula ion 17:61–74.
Lewin-Eps ein, Noah, Haya S ie and Michael B aun. 2006. The Di ision o Household Labo in Ge many
and Is ael. Jou nal o Ma iage and Family 68(5):1147–1164.
Löwe, Paul S. 2022. A e i s jobs in he Ge man public sec o mo e s able? An examina ion unde consid-
e a ion o he ins i u ional s uc u e. Soziale Wel 73(2):379–417.
Löwe, Paul S., and Pe e Vale . 2022. Re o men im ö en lichen Diens – Auswi kungen au Beam *innen
und A bei nehme *innen on 1993 bis 2018. WSI-Mi eilungen 75(2):119–128.
Löwe, Paul S., and Pe e Vale . 2023. Is de ö en liche Diens noch ein a ak i e A bei gebe ? Ve än-
de e Wah nehmungen zu Jobsiche hei , lang is ige Beschä igung und Ka ie epe spek i en on
Beschä ig en im ö en lichen Diens und de P i a wi scha on 1985 bis 2018. Zei sch i ü
Sozial e o m 69(1):1–30.
Luci-G eulich, Angela, and Oli ie Thé enon. 2013. The Impac o Family Policies on Fe ili y T ends in
De eloped Coun ies. Eu opean Jou nal o Popula ion 29(4):387–416.
Luci-G eulich, Angela, and Oli ie Thé enon. 2014. Does Economic Ad ancemen ‘Cause’ a Re-inc ease
in Fe ili y? An Empi ical Analysis o OECD Coun ies (1960–2007). Eu opean Jou nal o Popula-
ion 30(2):187–221.
Ma in-Ga cia, Te esa, and Te esa Cas o-Ma in. 2013. Do Women Wo king in he Public Sec o Ha e i
Easie o Become Mo he s in Spain? Popula ion Re iew 52(1):149–171.
Ma ysiak, Anna, and S ephanie S einme z. 2008. Finding Thei Way?: Female Employmen Pa e ns in
Wes Ge many, Eas Ge many, and Poland. Eu opean Sociological Re iew 24(3):331–345.
Mills, Melinda, and Hans-Pe e Bloss eld. 2005. Globaliza ion, Unce ain y and Changes in Ea ly Li e
Cou ses. Zei sch i ü E ziehungswissenscha (6):188–218.
Mülle , Be ina, and Lau a Cas iglioni. 2020. Do Tempo a y D opou s Imp o e he Composi ion o Panel
Da a? An Analysis o Gap In e iews in he Ge man Family Panel Pai am. Sociological Me hods &
Resea ch 49(1):193–215.
OECD. 2007. Babies and Bosses—Reconciling Wo k and Family Li e. Pa is: OECD Publishing h ps://
www.oecd.o g/con en /dam/oecd/en/publica ions/ epo s/2007/11/babies-and-bosses- econciling-
wo k-and- amily-li e_g1gh7 03/9789264032477-en.pd .
Oppenheime , Vale ie K. 1988. A Theo y o Ma iage Timing. Ame ican Jou nal o Sociology 94(3):563–
591.
Oppe mann, Anja. 2014. Explo ing he Rela ionship be ween Educa ional Field and T ansi ion o Pa -
en hood—An Analysis o Women and Men in Wes e n Ge many. Eu opean Sociological Re iew
30(6):728–749.
K
Does a Ci il Se ice Job Ma e ? The In luence o Ci il Se ice Employmen on he T ansi ion... 353
Saue , Ca s en, Pe e Vale and Vincen J. Roscigno. 2019. Neue Ungleichhei en im ö en lichen Sek o .
In Komplexe Dynamiken globale und lokale En wicklungen. Ve handlungen des 39. Kong esses de
Deu schen Gesellscha ü Soziologie in Gö ingen 2018, ed. Nicole Bu zan, 1–11.
Saue , Ca s en, Pe e Vale , Vincen J. Roscigno and Geo ge Wilson. 2022. Neolibe alism’s Impac on
Public-Sec o Job Quali y: The US and Ge many Compa ed. In The Ox o d Handbook o Job Quali y,
eds. Ch is Wa hu s , Ch is Ma hieu and Rachel E. Dwye , 469–484. Ox o d: Ox o d Uni e si y P ess.
Schmid, Susanne, and Ma in Kohls. 2009. Rep oduc i e beha iou o mig an women in Ge many: Da a,
pa e ns and de e minan s. Vienna Yea book o Popula ion Resea ch 7:39–61.
Schulze, Alexande . 2009. Sozioökonomische Konsequenzen de Fe ili ä : Folgen de Gebu on Kinde n
ü den Wohls and on Paa haushal en. 1s . ed. Wiesbaden: VS Ve lag ü Sozialwissenscha en.
Seyda, Susanne. 2003. F auene we bs ä igkei und Gebu en e hal en. IW-T ends – Vie eljah essch i zu
empi ischen Wi scha s o schung 30(2):26–36.
Sh e le , Ka ina M., Amy E. Pi e i and Robe D ago. 2010. Wo k–Family Con lic and Fe ili y In en-
ions: Does Gende Ma e ? Jou nal o Family and Economic Issues 31(2):228–240.
Sobo ka, Tomáš, and É a Beaujouan. 2014. Two Is bes ? The pe sis ence o a wo-child amily ideal in
Eu ope. Popula ion and De elopmen Re iew 40(3):391–419.
S a is isches Bundesam . 2016. Finanzen und S eue n – Pe sonal des ö en lichen Diens es: Fachse ie 14
Reihe 6. Wiesbaden: S a is isches Bundesam .
S a is isches Bundesam . 2017a. Be ölke ung und E we bs ä igkei E we bsbe eiligung de Be ölke ung
E gebnisse des Mik ozensus zum A bei sma k : Fachse ie 1 Reihe 4.1. Wiesbaden: S a is isches Bun-
desam .
S a is isches Bundesam . 2017b. Finanzen und S eue n Pe sonal des ö en lichen Diens es: Fachse ie 14
Reihe 6. Wiesbaden: S a is isches Bundesam .
S a is isches Bundesam . 2022. Quali ä de A bei – Be is e Beschä ig e. h ps://www.des a is.de/DE/
Themen/A bei /A bei sma k /Quali ae -A bei /Dimension-4/be is e -beschae ig e.h ml.
S a is isches Bundesam . 2023. Finanzen und S eue n. Pe sonal des ö en liches Diens es 2021: Fachse ie
14 Reihe 6. Wiesbaden: S a is isches Bundesam .
S a is isches Bundesam . 2024. Zahl de Gebu en im Jah 2023 au nied igs em S and sei 2013.h ps://
www.demog a ie-po al.de/DE/Poli ik/Meldungen/2024/240502-bund-zahl-de -gebu en-im-jah -
2023-au -nied igs em-s and-sei -2013.h ml (Accessed: 28 Ma ch 2025).
S ein, Pe a, Sebas ian Willen and Monika Pa e ic. 2014. Couples’ e ili y decision-making. Demog aphic
Resea ch 30(63):1697–1732.
S olzenbe g, Ross M., and Linda J. Wai e. 1977. Age and he ela ionships be ween young women’s plans
o childbea ing and employmen . Ame ican Sociological Re iew 42:769–783.
S uck, Ola . 2006. Flexibili ä und Siche hei : Empi ische Be unde, heo e ische Konzep e und ins i u-
ionelle Ges al ung on Beschä igungss abili ä . Wiesbaden: VS.
Tepe, Ma kus, and Daniela K oos. 2010. Luk a i e S aa sdiens ? Lohndi e enzen zwischen ö en lichem
Diens und P i a wi scha . WSI-Mi eilungen (1):3–10.
Tes a, Ma ia R. 2012. Family sizes in Eu ope: e idence om he 2011 Eu oba ome e su ey. Vienna:
Vienna Ins . o Demog aphy.
Thomas, Jac, F ancisco Rowe, Paul Williamson and E ic S. Lin. 2022. The E ec o Lea e Policies on In-
c easing Fe ili y: A Sys ema ic Re iew. Humani ies and Social Sciences Communica ions 9(1):1–16.
Vlasblom, Jan D., and Joop Schippe s. 2006. Changing Dynamics in Female Employmen A ound Child-
bi h. Wo k, Employmen & Socie y 20(2):329–347.
Voydano , Pa icia. 2005. Towa d a Concep ualiza ion o Pe cei ed Wo k-Family Fi and Balance: A De-
mands and Resou ces App oach. Jou nal o Ma iage and Family 67:822–836.
Wal e , Ul ike. 2007. Löhne und Gehäl e in Deu schland: ih e En wicklung in Wi scha und S aa on
1960–2000. Wiesbaden: Deu sche Uni e si ä s-Ve lag.
an de We ho s , He man G. 2011. Skills, posi ional good o social closu e? The ole o educa ion ac oss
s uc u al–ins i u ional labou ma ke se ings. Jou nal o Educa ion and Wo k 24(5):521–548.
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps
and ins i u ional a ilia ions.
K
354 P. S. Löwe
Paul Se e in Löwe 1988, D . e . pol, esea ch associa e and p ojec manage a he S a e Ins i u e o
Family Resea ch a he Uni e si y o Bambe g (i b). Fields o esea ch: Sociology o amilies, sociol-
ogy o labou ma ke s, wel a e s a es. Publica ions: De ö en liche Diens im Wandel: Konsequenzen ü
Beschä igungsbedingungen und Beschä ig e in Deu schland. Bambe g 2024; Is de ö en liche Diens
noch ein a ak i e A bei gebe ? Ve ände e Wah nehmungen zu Jobsiche hei , lang is ige Beschä i-
gung und Ka ie epe spek i en on Beschä ig en im ö en lichen Diens und de P i a wi scha on
1985 bis 2018. Zei sch i ü Sozial e o m 69(1), 2023 (mi P. Vale ).
K
