Gangl, Selina; Hube , Ma in
A icle — Published Ve sion
F om homemake s o b eadwinne s? How manda o y
kinde ga en a ec s ma e nal labou ma ke ou comes
Jou nal o Popula ion Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Gangl, Selina; Hube , Ma in (2025) : F om homemake s o b eadwinne s?
How manda o y kinde ga en a ec s ma e nal labou ma ke ou comes, Jou nal o Popula ion
Economics, ISSN 1432-1475, Sp inge , Be lin, Heidelbe g, Vol. 38, Iss. 2,
h ps://doi.o g/10.1007/s00148-025-01099-y
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323188
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 p://c ea i ecommons.o g/licenses/by/4.0/
Jou nal o Popula ion Economics (2025) 38:44
h ps://doi.o g/10.1007/s00148-025-01099-y
ORIGINAL PAPER
F om homemake s o b eadwinne s? How manda o y
kinde ga en a ec s ma e nal labou ma ke ou comes
Selina Gangl1·Ma in Hube 2
Recei ed: 18 Oc obe 2023 / Accep ed: 26 Ma ch 2025 / Published online: 16 Ap il 2025
© The Au ho (s) 2025
Abs ac
The majo i y o Swiss child en a end manda o y and cos - ee kinde ga en a age
ou . We examine he e ec o his policy on ma e nal labou ma ke ou comes. Using
adminis a i e da a om Swi ze land, we exploi he bi hday cu -o o kinde ga en
en y in he same o in he ollowing yea and apply a non-pa ame ic eg ession
discon inui y design (RDD). We ind ha manda o y kinde ga en has a s a is ically
signi ican posi i e e ec on he labou ma ke a achmen o p e iously non-employed
mo he s, inc easing hei employmen p obabili y by 4 pe cen age poin s. In con as ,
he e a e no signi ican e ec s on o he g oups o in he o al sample o mo he s.
Keywo ds Manda o y kinde ga en ·Ma e nal employmen ·Reg ession
discon inui y design
JEL Classi ica ion H40 ·J13 ·J18 ·J21 ·J22
1 In oduc ion
The labou o ce a achmen o mo he s has been ecei ing much a en ion in scien i ic
and public discussions, gi en he la ge employmen d ops obse ed in many coun ies
a e gi ing bi h (see, o ins ance, Kle en e al. (2019)). In OECD coun ies, he
a e age ma e nal employmen a e was 71% in 2019, wi h Iceland, Slo enia, and
Responsible edi o : Kompal Sinha
BSelina Gangl
[email p o ec ed]
Ma in Hube
ma in.hube @uni .ch
1Depa men o Economics, F ied ich-Alexande -Uni e si ä E langen-Nü nbe g, Lange Gasse
20, 90403 Nü nbe g, Ge many
2Depa men o Economics, Uni e si y o F ibou g, Bd. de Pé olles 90, 1700 F ibou g, Swi ze land
123
44 Page 2 o 23 S. Gangl, M. Hube
Sweden a he op wi h a es exceeding 85%.1In Swi ze land, which is he coun y
conside edin hiss udy, mo he s’ employmen a edeclinesa e childbi h(He mann
and Mu ie 2016), bu i emains high a almos 78%, on a e age.2Al hough his igu e
is highe han he OECD a e age, mos o i is due o pa - ime wo k (63%), a he
han ull- ime wo k (15%).3
A policy ool o encou age mo he s’ labou ma ke ou comes is he p o ision o
childca e. Fo example, K ap e al. (2020) show ha he oppo uni y o place he
i s child in childca e inc eases he ma e nal labou o ce pa icipa ion in he can on
Be n by 5.1 pe cen age poin s in he long e m. In addi ion o he ex ensi e ma gin,
he bi h o a child also a ec s he in ensi e ma gin o labou supply: mo he s a e
mo e likely han women wi hou child en o wo k pa - ime (He mann and Mu ie
2016). The p o ision o childca e in luences also mo he s’ wo king hou s posi i ely:
An expansion o childca e acili ies aises ma e nal pa - ime hou s (Ra azzini 2018),
while a e -school ca e inc eases mo he s’ ull- ime employmen (Fel e e al. 2016).
Childca e a ailabili y may also inc ease ma e nal ea nings, bu no he ea nings o
he o al household (K ap e al. 2020). An unanswe ed ques ion in his con ex is
how mo he s’ labou ma ke esponses eac when childca e is manda o y. While
kinde ga en o ou -yea -olds has always been cos - ee, i has ecen ly become
manda o y in mos Swiss can ons.
This pape examines he causal e ec o manda o y kinde ga en o ou -yea -old
child en on ma e nal labou beha iou and ea nings in Swi ze land. The aim o he
policy, wi h ega d o mo he s, is o p omo e he compa ibili y o amily and wo k.
We exploi his no el se ing and use he bi hday cu -o o iden i y he causal e ec
o he manda o y kinde ga en.
The pape uses a la ge adminis a i e da ase abou he popula ion o mo he s wi h
ou -yea -old child en li ing in Swi ze land. We link he Popula ion and Households
S a is ics (STATPOP) wi h da a abou he Old-Age and Su i o s’ Insu ance (OASI).
Mo eo e , we ha e collec ed da a abou he implemen a ion yea o he policy, he
bi hday cu -o da es pe yea , and he equi emen s ega ding how many yea s o
kinde ga en mus be o e ed. These da a s em om he can onal depa men s o edu-
ca ion, he Swiss Con e ence o Can onal Depa men s o Educa ion (EDK), and he
can onal laws. The S a e Sec e a ia o Economic A ai s (SECO) p o ides in o ma-
ion abou he can onal unemploymen a e on hei homepage. Again we link his
can onal in o ma ion o he adminis a i e da a. We es ic he sample o mo he s
whose younges child is ou -yea -old and who a e li ing in a can on wi h manda o y
kinde ga en.4Me hodologically, we exploi he andom assignmen a he bi hday
cu -o o en e ing he kinde ga en in he same yea (= ea men ) e sus he ollowing
yea (= con ol) and apply a Reg ession Discon inui y Design (RDD). Since pa en s
may pos pone he kinde ga en en y o hei child by one yea and we canno obse e
he ac ual kinde ga en en y, we a he assess he in en ion- o- ea (ITT) e ec han
he a e age ea men e ec (ATE) a he h eshold.
1Repo ed by OECD. h ps://www.oecd.o g/els/ amily/da abase.h m, e ie ed 2021-08-24.
2Repo ed by OECD. h ps://www.oecd.o g/els/ amily/da abase.h m, e ie ed 2021-08-24.
3Repo ed by OECD. h ps://www.oecd.o g/els/ amily/da abase.h m, e ie ed 2021-08-24.
4See Sec .4Da a o de ails.
123
F om homemake s o b eadwinne s? How manda o y… Page 3 o 23 44
We ind ha manda o y kinde ga en boos s he labou ma ke ou comes o
p e iously non-employed mo he s. In his subg oup, he p obabili y o ma e nal
employmen inc eases by 4 pe cen age poin s on a e age ac oss ou ou come pe i-
ods om 2010 o 2017 (s a is ically signi ican a he 5% le el). Likewise, he annual
income om dependen employmen inc eases by 1030 CHF on a e age (s a is ically
signi ican a he 10% le el) and he p obabili y o ea ning he income om dependen
employmen inc eases by 4 pe cen age poin s (s a is ically signi ican a he 5% le el).
Ou esul s sugges ha manda o y kinde ga en has no signi ican e ec on he labou
ma ke ou comes in he o al sample o mo he s. We a gue ha he high pa icipa ion
a e o mo he s in he labou o ce, he hal -day ca e in manda o y kinde ga en, and
he suppo o g andpa en s o childca e a e likely easons o his inding.
Ou pape i s in o he li e a u e on school policies and i s impac on he amily
(see, e.g. Lande sø e al. (2020)). In pa icula , ou pape belongs o a g oup o s udies
ocusing on ma e nal labou ma ke ou comes. Fo example, Gelbach (2002) in es i-
ga es he e ec o school en olmen and Ba ua (2014) analyses he a ia ion in school
en y laws on ma e nal labou supply in he U.S. Finse aas e al. (2017) examine a
dec ease in he school s a ing age om se en o six yea s on ma e nal labou supply
in No way. The common esul o hese s udies is ha he school en y o he child
aises mo he s’ labou supply. In line wi h ha , Zhu and B adbu y (2015) ind ha he
bi h o a child sho ly a e he bi hday cu -o leads o a nega i e impac on mo he s’
labou supply.
Whe eas he a o emen ioned s udies ocus on child en who a e s a ing school a
age six o se en, ou se ing is unique because all ou -yea -olds a e equi ed o en e
school, o in ou case, manda o y kinde ga en. The e o e, his pape con ibu es o
he exis ing li e a u e by examining he impe us o ma e nal ein eg a ion in o he
wo k o ce wo yea s ea lie due o manda o y kinde ga en a endance o hei child.
Thisispa icula ly ele an because indus ialised coun ies a e cu en ly acing labou
sho ages ha a e expec ed o inc ease due o an ageing popula ion and migh be
mi iga ed by wo king mo he s (Eu opean Commission 2023). Mo eo e , an ea lie
ein eg a ion in he labou o ce migh dec ease he isk o po e y in old age ha
mo he s ace due o in e up ed wo king li es (Uni ed Na ions 2022).
This pape ela es also o he amily policy li e a u e and i s impac on ma e -
nal labou ma ke beha iou (see, e.g. Kle en e al. (2020)). Childca e, speci ically,
may acili a e mo he s’ pa icipa ion in he wo k o ce. Hence, one s and o li e a u e
examines whe he he in oduc ion o subsidised childca e like in Aus ia (Kle en e al.
2020), Canada (Le eb e and Me igan 2008), No way (Ha doy and Schøne, 2015),
Sweden (Lundin e al. 2008), and he Ne he lands (Be endo e al. 2015) impac s
ma e nal employmen . Fu he esea ch analyses he expansion o childca e places as
done in Spain (Nollenbe ge and Rod íguez-Planas 2015), I aly (Ca a and Rizzica
2018), Ge many (Geye e al. 2015), and No way (Kunze and Liu 2019; Eckho
And esen and Ha nes 2019), o he p o ision o a legal en i lemen o childca e, as in
Ge many (Baue nschus e and Schlo e 2015). E alua ions o hese policies sugges
a signi ican ly posi i e e ec on mo he s’ labou o ce pa icipa ion, bu only unde
he condi ion o a low labou o ce pa icipa ion a e and a small ex en o childca e
p io o he e o m. Howe e , he a ailabili y o (subsidised) childca e can lead o a
c owding ou o p i a e childca e, which may o se any inc ease in o e all childca e
123
44 Page 4 o 23 S. Gangl, M. Hube
p o ision and, as a esul , ail o imp o e ma e nal labou ma ke ou comes (Ha nes
and Mogs ad 2011). Fo his eason, Goux and Mau in (2010) ind only an e ec o
childca e a ailabili y on ma e nal labou supply o he subg oup o single mo he s in
F ance.
While his li e a u e analyses he impac o childca e on he labou ma ke beha iou
o a sel -selec ed g oup o mo he s who choose o en ol hei child en in childca e,
i does no encompass he e ec on all mo he s. Ou pape add esses his inqui y by
examining whe he manda o y kinde ga en incen i izes all mo he s o pa icipa e in
he labou ma ke and he esul ing impac on hei wages. This in o ma ion seems o
be c ucial o policy make s seeking o p omo e ma e nal labou o ce pa icipa ion.
The indings o ou s udy align wi h hose o Ha nes and Mogs ad (2011), Goux
and Mau in (2010), and Lundin e al. (2008), indica ing ha dly no impac o childca e
on ma e nal labou ma ke ou comes. In compa ison, Be linski and Galiani (2007)
and Le eb e and Me igan (2008) epo signi ican posi i e e ec s. These di e -
gen conclusions may pa ly s em om a ia ions in he socio-economic se ings o
he s udied coun ies. Fo ins ance, a high a e o ma e nal employmen , such as ha
obse ed in Sweden, could elucida e he nea -negligible e ec o childca e policies
Lundin e al. 2008. Simila ly, in Swi ze land, he ocus o ou s udy, whe e ma e nal
employmen a es ho e a ound 80%, he p esence o manda o y kinde ga en shows
only disce nible e ec s on labou ma ke ou comes o p e iously non-employed mo h-
e s. Ano he ea u e o he Swiss se ing is ha wo king pa en s mainly use he suppo
o g andpa en s o childca e (Jacobs Founda ion 2018). The insigni ican impac o
manda o ykinde ga en onlabou ma ke esponsein he o alsampleo mo he scould
po en ially be a ibu ed o a c owding ou e ec o in o mal childca e, as sugges ed
by Ha nes and Mogs ad (2011). Di e en policy designs may also explain he mixed
esul s in he s udies. Fo example, Le eb e and Me igan (2008) analyse he e ec
o ull- ime childca e and ind signi ican and la ge e ec s. The manda o y kinde -
ga en is only hal -day ca e in mos Swiss can ons, so mo he s a e o en aced wi h
he challenge o balancing hei wo k a ound he limi ed kinde ga en hou s o inding
ano he (cos ly) ca e op ion. This ci cums ance could be a possible explana ion o
he ac ha manda o y kinde ga en does no ha e a a - eaching impac on labou
ma ke ou comes in he o al sample o mo he s.
The emainde o his pape is o ganised as ollows: In he nex sec ion, we p o ide
in o ma ion abou he p eschool policy in Swi ze land. The ea e , we p esen he
da a sou ce, desc ip i e s a is ics, and de ine he sub-sample. Then, we discuss he
empi ical s a egy. In he ollowing sec ion, we epo he esul s and he obus ness
checks. Finally, we d aw a conclusion om he empi ical analysis.
2 Ins i u ional backg ound
Thissec ionp o idesano e iewo hemanda o yand cos - eekinde ga en o ou -
yea -old child en in Swi ze land. Educa ion policy in Swi ze land is decen alised,
wi h each o he 26 can ons ha ing au onomy o e he quali y and s uc u e o educa-
ion. This decen alisa ion leads o a ia ions in aspec s such as he school cu iculum,
he age o en y in o kinde ga en, and he numbe o school yea s (SWI 2006). Hence,
123
F om homemake s o b eadwinne s? How manda o y… Page 5 o 23 44
each can on decides abou he implemen a ion o he manda o y kinde ga en. Table
A.1 in he Online Appendix shows ha Basel-S ad was he i s can on o in oduce
manda o y kinde ga en o ou -yea -olds in 2005. In 2006, a na ional e e endum
abou an educa ional e o m ook place, which was accep ed by 86% o he o e s.
One yea la e , an in e -can onal “Ha moS” conco da was es ablished, 15 can ons
en e ed in o he conco da , while se en can ons ejec ed i and ou can ons a e s ill
indecisi e (as o 2019). The main goal o he conco da is o ha monise he manda o y
school educa ion among he can ons. This ag eemen also includes an age dec ease
o manda o y school a endance. All child en u ning ou be o e he i s o Augus
mus en e kinde ga en (Schweize ische Kon e enz de kan onalen E ziehungsdi ek-
o en 2007). The e o e, o he can ons ollowed Basel-S ad and in oduced manda o y
kinde ga ens. Bu also in his case, nei he he implemen a ion yea no he shi o
he bi hday cu -o da e is egula ed on a cen al le el, see Table A.1 in he Online
Appendix o an o e iew.
In 2017, almos wo- hi ds o he can ons had en o ced manda o y kinde ga en
a endance o ou -yea -olds.5In hese can ons, child en who u n ou be o e o on
he cu o da e mus en e kinde ga en in he same yea , while ou -yea -olds bo n
a e he cu o da e do no en e kinde ga en un il he ollowing yea . Table A.1 in
he Online Appendix shows ha hese bi hday cu -o da es a y om yea o yea
and can on o can on and se le on 31.7. The eason o his shi is he join decision
by he can ons o de e mine a uni o m cu -o da e. Mos can ons ha e decided o
g adually shi he cu -o da e so ha he ush o kinde ga ens emains oughly he
same du ing he ansi ion pe iod and does no inc ease pa icula ly sha ply in one
yea and o e load he in as uc u e.
The wo old aim o he manda o y kinde ga en is o p omo e he de elopmen o
child en, bu also o encou age he compa ibili y o amily and wo k (Schweize ische
Kon e enz de kan onalen E ziehungsdi ek o en 2014). In his pape , we ocus on
he la e and examine whe he he obliga ion o a end kinde ga en a he age o ou
a ec s mo he s’ wo k beha iou . Hence, he leng h o he manda o y kinde ga en may
be c ucial o he ma e nal wo k decision. The kinde ga en hou s di e om can on
o can on: A he lowe bound is he can on Valais, whe e ou -yea -old child en a e
supposed o a end 12 lessons pe week,6each lesson is 45 minu es, i.e. he child en
spend in sum 9hou s pe week in he kinde ga en. A he uppe bound o kinde ga en
hou s is he can on Ticino, whe e child en a end he kinde ga en on ou days om
he mo ning ill he a e noon and an addi ional hal -day. The manda o y kinde ga en
is pa o ea ly childhood educa ion, which akes place p io o p ima y educa ion
(OECD 2015) and co esponds o le el 0 in he In e na ional S anda d Classi ica ion
o Educa ion (ISCED) (Eu ydice 2016/17). In he Ge man- and I alian-speaking pa s
o Swi ze land, kinde ga en is a he a sepa a e ins i u ion, whe eas in he F ench-
speaking pa , kinde ga en is ep esen ed as wo addi ional classes in p ima y school
(Schweize ische Kon e enz de kan onalen E ziehungsdi ek o en 2016).
5Aa gau, Basel-Land, Basel-S ad , Be n, F ibou g, Gene a, Gla us, Ju a, Neuchâ el, S Gallen,
Scha hausen, Solo hu n, Thu gau, Ticino, Vaud, Valais, and Zu ich.
6EDK Kan onsum age 2016/17.
123
44 Page 6 o 23 S. Gangl, M. Hube
Red-shi ing beha iou o he pa en s, i.e. de e ing a ou -yea -old child om
kinde ga en, is possible, bu is associa ed wi h hu dles ha a y om can on o can on.
In some can ons, an applica ion by he pa en s is su icien , in o he can ons a medical
ce i ica e is equi ed which p o es ha he admission o a child o he kinde ga en
would no be easonable e en wi h o he educa ional assis ance.7
3Hypo heses
As he li e a u e e iew e eals, childca e e o ms as well as a lowe school s a ing
age a ec mo he s’ wo k beha iou . In he o me case, a subsidy ei he educes
he inancial bu den o p o ides mo e childca e places. In he la e case, he e ec
comes om he obliga ion o a end school a a younge age. In con as , we e alua e
he impac o he manda o y kinde ga en on mo he s’ wo k beha iou . The e o e,
he analysis combines ea u es om bo h li e a u e s ands: The kinde ga en is an
obliga ion and is ee o cha ge like in he school s a ing age li e a u e. On he o he
side, we e alua e a policy ha a ec s child en in he childca e age. This chap e
p esen s he hypo hesis and he possible channels.
We e alua e he e ec o he manda o y kinde ga en on ma e nal labou ma ke
beha iou wi hin a speci ic age window: We compa e mo he s wi h sligh ly olde
child en who we e bo n p io o a speci ic bi hday cu -o and a e supposed o en e
kinde ga en in he same yea (= ea ed) wi h mo he s whose child en a e sligh ly
younge , bo n a e a speci ic bi hday cu -o and who a e supposed o en e kinde -
ga en in he ollowing yea (= con ol).8Mo he s wi h a sligh ly younge child may
spend hei ime by aking ca e o hei child o mus o ganise a (cos ly) childca e
o alloca e hei ime o wo k. Con a ily, mo he s, whose child is supposed o en e
kinde ga en, can spend hei ime on o he ac i i ies, o example, labou ma ke pa -
icipa ion. Consequen ly, a mo he who would ake ca e o he child in he absence o
he e o m can ea ange he ime: Following he model o he alloca ion o ime Becke
(1965), he mo he can spend mo e ime in ma ke wo k and/o leisu e ac i i ies. Swa
e al. (2019) ind ha mo he s inc ease hei labou supply a e he school en olmen
o hei younges child, gi en ha hey ake ca e o hei child en p io o he school
s a . The e o e, we expec in ou se ing ha only one subg oup o mo he s inc eases
hei labou supply: Mo he s whose child en a e supposed o en e kinde ga en, bu
who would ake ca e o hem i no policy is implemen ed. All in all, we hypo hesise a
posi i e, bu small e ec because he e ec is ze o o mo he s who would send hei
child o kinde ga en anyway.
4 Da a
Ou e alua ion s a egy is a non-pa ame ic eg ession discon inui y design (RDD) o
examine he e ec o he e o m a he bi hday cu -o o en e ing he kinde ga en
7h ps://www.swissmom.ch/de/kind/kinde ga en-und-schule/mi -4-jah en-schon-in-den-kinde ga en-
19063.
8One excep ion is he can on o Ticino ha o e s kinde ga en o h ee-yea -old child en.
123
F om homemake s o b eadwinne s? How manda o y… Page 7 o 23 44
in he same e sus in he ollowing yea . I elies on a la ge adminis a i e da ase
including he uni e se o mo he s wi h ou -yea -old child en li ing in Swi ze land,
he so-called Popula ion and Households S a is ics (STATPOP). We link he la e
wi h da a coming om he old-age and su i o s’ insu ance (OASI) in Swi ze land
by a unique iden i ie . STATPOP p o ides in o ma ion on he s ock and s uc u e o
he Swiss esiden popula ion om 2010 o 2018. This enables us o examine he
popula ion o mo he s whose younges child is ou -yea -old and li ing in a can on
wi h manda o y kinde ga en. Fu he mo e, he da a con ain pe sonal cha ac e is ics
om he mo he , hei child en, and he a he o he ou -yea -old child. Two a iables
a e cen al o ou eg ession discon inui y app oach (RDD): Fi s ly, he can on o
esidence o de e mine whe he he ou -yea -old li es in a can on wi h manda o y
kinde ga en in he cu en yea . Secondly, he exac da e o he bi h o he ou -yea -
old de e mines whe he a child was bo n p io (= ea ed) o a e (= con ol) a speci ic
bi hday cu -o da e. Addi ionally, o he pe sonal cha ac e is ics such as he ma i al
s a us, coun y o bi h, esiden pe mi o he mo he , a he , and he child en a e also
con ained in he da ase .
The OASI da a con ain in o ma ion abou he income, unemploymen bene i s, dis-
abili y indemni ies o e e y insu ed pe son in Swi ze land om 2010 o 2017. Based
on hese da a, we calcula ed he “Income om wo k” as he o al income minus ans-
e paymen s ( o ins ance, unemploymen insu ance and wel a e bene i s) and minus
o he indemni ies (like he compensa ion o mili a y se ice) o he ollowing sub-
g oups: (i) employees whose employe esides in Swi ze land, (ii) olun a y insu ed
employees whose employe is no liable o con ibu ions in Swi ze land, and (iii) sel -
employed pe sons.9Subsequen ly, we gene a ed he a iable “In labou o ce”, which
is 1 i a pe son ea ns income om wo k o i she ecei es unemploymen bene i s and 0
o he wise. We calcula ed he dummy “Ou o labou o ce” as 1 minus he alue o “In
labou o ce”. Fu he mo e, we cons uc ed he dummy a iable “Employed”, which
u ns 1 i he pe son ea ns an income om wo k and pa icipa es in he labou ma ke
and is 0 o he wise. Vice e sa, a pe son is indica ed as “Unemployed” i she pa ici-
pa es in he labou o ce bu is no employed. The a iable “Income om dependen
employmen ” gi es he sum o he income o mo he s wo king in dependen employ-
men , whe eas he a iable “Income om dependen employmen (bina y)” indica es
whe he a mo he wo ks in dependen employmen . In he las s ep, we ma ched his
da abase wi h can onal in o ma ion abou he unemploymen a es, he yea ly bi hday
cu -o da e o he manda o y kinde ga en, and an indica o o whe he a can on joined
he “Ha moS” conco da . The “Ha moS” conco da is an associa ion o some can ons
o ha monise manda o y educa ion like he uni o m school en y age.10
The da abase con ains a o al o 687,749 mo he s wi h a leas one ou -yea -old
child be ween 2010 and 2018. Since he las pe iod in he OASI da a is 2017, we
d op he STATPOP obse a ions om 2018 such ha 606,434 obse a ions emain.
Ano he da a es ic ion is based on he empi ical li e a u e, because childca e seems o
no a ec he mo he s’ labou ma ke ou comes when he e is a younge sibling o he
ou -yea -old child in he household (see, o example, Be linski and Galiani (2007),
9excluding ag icul u e.
10 Mo e in o ma ion abou he ins i u ional backg ound can be ound in Online Appendix B.1.
123
44 Page 8 o 23 S. Gangl, M. Hube
Fig. 1 Timeline o measu ed a iables No e: The imeline shows he ime poin s a which he a iables
we e measu ed ela i e o he ea men
Fi zpa ick (2012), and Gelbach (2002)). We o e come his conce n by es ic ing
he subsample o mo he s whose younges child is ou yea s old, which dec eases
he sample size o 360,234 mo he s. Th ee can ons wi h manda o y kinde ga en did
no de e mine a bi hday cu -o da e, hence we exclude obse a ions om Aa gau,
F ibou g p io o 2013, and all he municipali ies o he can on Be n excep he ed-
e al ci y o Be n, a e ha 289,728 obse a ions a e in he sample. Fu he mo e, we
d op obse a ions wi h nega i e income en ies11 in he OASI da a and keep 289,631
obse a ions. We exclude obse a ions, which li e in can ons wi hou a manda o y
kinde ga en o ou yea s old. Since wo sou ces12 epo ed wo consecu i e days
as bi hday cu -o da es in Basel-S ad , we ook he la e cu -o da e and excluded
obse a ions on he p e ious day om ou analysis.13 Finally, 201,993 obse a ions
emain in ou e alua ion da ase .
Figu e1p o idesa imeline o he measu emen o he key a iables in ou analysis,
wi h deno ing a speci ic yea . The unning a iable will be hence o h deno ed by D
and depic s whe he he child u ns ou p io (D= 1) o a e he cu -o da e (D=
0), measu ed in he baseline pe iod ( = 0). The ou come pe iods s a in he yea he
child u ns ou ( = 0) and end i e yea s la e ( = 5). Pe sonal cha ac e is ics (e.g.
numbe o child en), a e measu ed one yea be o e he child u ns ou ( =−1).
We check he balance o labou ma ke ele an co a ia es as sugges ed in Lee
(2008), by unning se e al RDD es ima ions whe e each co a ia e se es as an ou -
come. Table 1 epo s he sample mean, he balance checks, and he ela i e numbe o
missings o each a iable a he h eshold. Mo he s’, a he s’, and can onal co a ia es
a e gene ally well balanced, which poin s o a quasi- andom assignmen o manda-
o y kinde ga en a he h eshold. We also see om he able ha mo he s close o
he h eshold ha e on a e age 1.87 child en, a e p edomina ely bo n in Swi ze land,
and li e in a ela ionship. Fu he mo e, 73% o hese mo he s a e employed, ea n
on a e age 32,760 Swiss ancs (CHF) pe yea , which mainly s em om depen-
den employmen . Fa he s labou ma ke cha ac e is ics show wi h 90% a e y high
employmen a e,anannualmeang oss income o 94,342CHF,and86%o his income
comes om dependen employmen . The can onal cha ac e is ics e eal a a he low
mean unemploymen a e (3.6%) and a high p obabili y o pa icipa ing in he Ha moS
conco da (94%).
Table 2 epo s he balance checks o can onal dummies a he h eshold. Due
o he di e en bi hday cu -o da es in he Ge man- and he F ench-speaking pa
o he can on Valais, we analyse Uppe Valais and Cen al/Lowe Valais sepa a ely.
11 o example due o adjus men en ies.
12 see “Regie ungs a sbeschluss be e end S ich ag ü die Einschulung ü die Schuljah e 2011/12 bis
2015/16 (§56 Abs. 1 Schulgese z)” and “EDK/IDES-Kan onsum age” o he yea s 2011 o 2016.
13 o un a Donu -RDD o his can on.
123
F om homemake s o b eadwinne s? How manda o y… Page 15 o 23 44
ca e o hei child en (Jacobs Founda ion 2018). The manda o y kinde ga en may
ha e educed he necessi y o in o mal childca e op ions, bu his did no a ec he
o e all impac on mo he s’ labou supply. The e ec o manda o y kinde ga en migh
be also diminished by he p esence o o mal childca e se ices. Howe e , dayca e in
Swi ze land is known o be expensi e compa ed o o he coun ies (S e n and Fel e
2015), and he e is a subs an ial demand su plus o childca e places (Bundesam ü
Sozial e siche ungen 2006). These ac o s con ibu e o a lowe dayca e a endance
a e, wi h only 36% o child en unde he age o ou cu en ly a ending dayca e
(Jacobs Founda ion 2018). Finally, he labou o ce supply o mo he s in Swi ze land
is al eady high, which migh also explain he lack o a sensible e ec o manda o y
kinde ga en on ma e nal labou ma ke pa icipa ion in he o al sample o mo he s.
This inding is in line wi h o he s udies ha ha e ound ei he no e ec o only a mod-
es e ec o childca e on ma e nal employmen in he con ex o high ma e nal labo
o ce pa icipa ion (see, e.g. Lundin e al. (2008) and Ha nes and Mogs ad (2011)).
An impo an ac o o conside when examining he lack o impac on ma e nal
income is he limi ed lexibili y p o ided by hal -day manda o y kinde ga en. Wi h
he majo i y o can ons o e ing hal -day ca e, mo he s a e o en aced wi h he chal-
lenge o balancing hei wo k schedules a ound he limi ed hou s o kinde ga en.
This cons ain can signi ican ly hinde hei abili y o pu sue ull- ime employmen
oppo uni ies. Fu he mo e, he eliance on al e na i e childca e a angemen s o he
es o he day, such as p i a e dayca e, can come a an addi ional cos o mo he s.
These expenses, combined wi h he logis ical challenges o coo dina ing mul iple ca e
p o ide s, can c ea e inancial and logis ical bu dens ha u he discou age mo he s
om pu suing ull- ime employmen . The e o e, he lack o access o ull-day ca e
ollowing manda o y hal -day kinde ga en no only es ic s he hou s a ailable o
ma e nal employmen bu also in oduces inancial and logis ical ba ie s ha impede
he po en ial inc ease in ma e nal income.
We check he obus ness o ou ITT e ec s by es ima ing he RDD app oach wi hou
co a ia es (Table A.2 in he Online Appendix) and ind a he simila poin es ima es,
ye he p ecision dec eases. Ano he check consis s o a ying he op imal bandwid h,
we mul iplied his by 1.5 and 2/3 (Tables A.3 and A.4 in he Online Appendix). Ou
indings indica e ha he e ec sizes emain ela i ely s able, wi h he excep ion o
he e ec s on “Annual income om dependen employmen ”, as shown in Tables A.3
and A.4 in he Online Appendix, and “Annual wo k income” as p esen ed in Table
A.4 in he same Appendix.
6.2 E ec s o e yea s
In he nex s ep, we analyse he e ec s in sepa a e and consecu i e ou come pe iods,
ela i e o he yea he child u ns ou . Figu e3shows he esul s o he di e en
ou comes, s a ing in pe iod 0 (i.e. he yea he child u ns ou ) and ending in pe iod
5. The do s depic he ITT a he cu -o and he bands ep esen he poin -wise 95%
con idence in e als based on obus s anda d e o s. We do no ind any s a is ically
signi ican e ec o he ou comes in compa ison o he pooled es ima es. This migh
be caused by lowe powe due o a dec ease in he numbe o obse a ions. Mo eo e ,
we ind uns able poin es ima es, bo h he sign and he e ec size change o e pe iods.
123
44 Page 16 o 23 S. Gangl, M. Hube
Fig. 3 RDD: E ec s o e yea s wi h co a ia es No e: This igu e shows he e ec o manda o y kinde ga en
on each ou come a iable sepa a ely and o e ime. Do s ep esen ITTs a he h eshold, bands co espond
o 95% con idence in e als. The ollowing con ol a iables a e included: Numbe o child en, bo n in
Swi ze land, esiden pe mi , age, ela ionship s a us, labou ma ke cha ac e is ics o he mo he and he
a he o he ou -yea -old child, can onal cha ac e is ics, and dummies. Da a s ems om STATPOP (2010–
2017) and OASI (2010–2017), calcula ions a e done by he au ho s. Income is de la ed, he base yea is
2011. The o icial cu ency in Swi ze land is he Swiss F anc (CHF), which had an a e age exchange a e
o 1.04 USD/CHF in he las decade
The plo s e eal u he ha he e ec s a e, especially in he la e pe iods, imp ecisely
es ima ed because o he dec ease in he numbe o obse a ions. Figu e A.7 in he
Online Appendix depic s he plo s wi hou co a ia es and shows simila esul s in
e ms o signi icance, poin es ima es, and p ecision.
6.3 He e ogeneous e ec s
F om a policy pe spec i e, i is in e es ing o know whe he he e ec o manda o y
kinde ga en di e s o speci ic subg oups. Non-employed mo he s may eac di e -
en ly o he kinde ga en policy han employed mo he s. The e o e, we spli he sample
acco ding o he ma e nal labou o ce s a us one yea be o e he younges child u ns
ou .
123
F om homemake s o b eadwinne s? How manda o y… Page 17 o 23 44
Fi s , we s a wi h he p esen a ion o he es s o he RDD assump ions. Tables A.5
and A.6 in he Online Appendix show, simila as o he en i e sample in Table 1, ha
he labou ma ke ele an co a ia es a e balanced o bo h sub-samples. Analogous o
he balance checks o can onal dummies o he en i e sample in Table 2, we conclude
om Tables A.7 and A.8 ha in mos cases, ou es s do no ejec he null hypo hesis
ha he can onal dummies a e balanced a ound he bi hday cu -o . In con as o
he esul o he en i e sample in Fig.2, he esul s o he F andsen’s es in Figs. A.9
and A.10 in he Online Appendix sugges s e idence o manipula ion a he cu -o
i we assume ha he dis ibu ion a ound he cu -o mus be p ecisely linea (k=0),
bu does no poin o manipula ion i we allow o non-linea i ies (k= 0.1 and k=
0.2). We ollow he a gumen a ion o Oos e beek e al. (2021), which s a es ha he
RDD becomes in alida ed by manipula ion only i he e exis sys ema ic dispa i ies
be ween mo he s on ei he side o he bi hday cu -o . Howe e , Tables A.5 and A.6
in he Online Appendix e eal no disce nible associa ion be ween being bo n sligh ly
a e e sus be o e he bi hday cu -o and he p ede e mined co a ia es.
Second, we epo he es ima ion esul s. Table 5 ep esen s he esul s o employed
mo he s, whe eas Table 6 epo s he e ec s o non-employed mo he s. The esul s
sugges ha he e ec is d i en by non-employed mo he s who s a wo king when
hei younges , one-yea -old child is equi ed o a end kinde ga en. Th ee e ec s in
Table 6 emain s a is ically signi ican a he 10% le el a e applying he Benjamini-
Hochbe g p ocedu e.17 To u he alida e he obus ness o ou indings, we conduc
an RDD wi h a ake cu -o se 30 days ea lie han he ac ual h eshold, and he esul s
in Table A.9 in he Online Appendix indica e ha he e ec s a e insigni ican .
Addi ionally, we employ a Donu -RDD app oach in Table A.10 in he Online
Appendix, which excludes obse a ions wi h unning a iables close o he cu -o
( alues be ween −2 and 1). The esul s ob ained unde his design con i m ha all
es ima ed e ec s main ain he same di ec ion as in he main analysis. Howe e , he
s a is ical signi icance o he e ec s is gene ally educed. Speci ically, only he e ec s
on employmen , ou o he labou o ce, and income-dependen employmen exhibi
p- alues close o 10%, while he emaining e ec s show p- alues ha a e conside ably
beyond any con en ional le els o s a is ical signi icance.
Thi d, we p o ide numbe s on he e ec o he kinde ga en policy on educing
he labou sho age. As shown in Table 6, he e ec o manda o y kinde ga en on
employmen is an inc ease o 4 pe cen age poin s o he subg oup o p e iously non-
employed mo he s. I we ex apola e his local e ec o all 179,895 non-employed
mo he s wi h ou -yea -olds in ou sample, manda o y kinde ga en esul s in 7196
addi ional employed mo he s. When his numbe is compa ed o he o al numbe o
employees in Swi ze land in he i s qua e o 201018 i co esponds o an inc ease in
agg ega e employmen o 0.002 pe cen age poin s. Compa ing hese 7196 addi ional
employed mo he s o he job acancies in Swi ze land in he i s qua e o 201019 we
17 We inspec also he he e ogeneous e ec s o olde e sus younge mo he s as well as di o ced mo he s
and do no ind s a is ically signi ican e ec s.
18 4,530,917; Fede al S a is ical O ice: h ps://www.b s.admin.ch/b s/en/home/s a is ics/indus y-
se ices/businesses-employmen /jobs-s a is ics/jobs.asse de ail.32008419.h ml,
19 51,162; Fede al S a is ical O ice: h ps://www.b s.admin.ch/b s/en/home/s a is ics/indus y-se ices/
businesses-employmen /jobs-s a is ics/ acancies.asse de ail.32008426.h ml,
123
44 Page 18 o 23 S. Gangl, M. Hube
Table 5 RDD: he e ogeneous e ec s: employed mo he s
Sample mean Coe icien S anda d e o p- alue Bandwid h Obse a ions
wi hin bandwid h
Ou o labou o ce (bina y) 0.06 0.00 0.01 0.58 67.03 183,521
Employed (bina y) 0.93 −0.00 0.01 0.72 67.66 183,521
Unemployed (bina y) 0.01 −0.00 0.00 0.54 51.19 138,858
Annual wo k income (in CHF) 46,460.42 −531.87 698.65 0.45 72.22 196,930
Income dependen employmen (bina y) 0.90 0.00 0.01 0.96 61.60 167,217
Annual income om dependen employmen (in CHF) 44,347.23 −606.55 650.57 0.35 62.65 170,025
No e: This able epo s he local linea es ima es o Eq.5.2 only o p e iously employed mo he s. Bandwid h shows he MSE-op imal bandwid h chosen by (Calonico e al.
2021). The ollowing con ol a iables a e included: Numbe o child en, bo n in Swi ze land, esiden pe mi , age, ela ionship s a us, labou ma ke cha ac e is ics o he
mo he and he a he o he 4-yea -old child, can onal cha ac e is ics, and dummies. The s anda d e o s a e clus e ed a he indi idual le el. Da a s ems om STATPOP
(2010–2017) and OASI (2010–2017), calcula ions a e done by he au ho s. Income is de la ed, he base yea is 2011. The o icial cu ency in Swi ze land is he Swiss F anc
(CHF), which had an a e age exchange a e o 1.04 USD/CHF in he las decade. Numbe o obse a ions in o al: 484,153
123
F om homemake s o b eadwinne s? How manda o y… Page 19 o 23 44
Table 6 RDD: he e ogeneous e ec s: non-employed mo he s
Sample mean Coe icien S anda d e o p- alue Bandwid h Obse a ions
wi hin bandwid h
Ou o labou o ce (bina y) 0.70 −0.04 0.02 0.05 59.46 59,463
Employed (bina y) 0.30 0.04 0.02 0.04 58.26 58,459
Unemployed (bina y) 0.01 −0.00 0.00 0.39 59.39 59,463
Annual wo k income (in CHF) 5582.01 936.74 652.42 0.15 55.40 55,452
Income dependen employmen (bina y) 0.28 0.04 0.02 0.03 57.77 57,634
Annual income om dependen employmen (in CHF) 4789.60 1030.41 596.34 0.08 54.96 54,394
No e: This able epo s he local linea es ima es o Eq.5.2 only o p e iously non-employed mo he s. Bandwid h shows he MSE-op imal bandwid h chosen by (Calonico
e al. 2021). The ollowing con ol a iables a e included: Numbe o child en, bo n in Swi ze land, esiden pe mi , age, ela ionship s a us, labou ma ke cha ac e is ics o
he mo he and he a he o he 4-yea -old child, can onal cha ac e is ics, and dummies. The s anda d e o s a e clus e ed a he indi idual le el. Da a s ems om STATPOP
(2010–2017) and OASI (2010–2017), calcula ions a e done by he au ho s. Income is de la ed, he base yea is 2011. The o icial cu ency in Swi ze land is he Swiss F anc
(CHF), which had an a e age exchange a e o 1.04 USD/CHF in he las decade. Numbe o obse a ions in o al: 179,895
123
44 Page 20 o 23 S. Gangl, M. Hube
seea educ ion in acan posi ions o 0.14 pe cen age poin s. We assume ha he e ec s
can be ex apola ed. The e o e, i should be no ed ha his is only an app oxima ion
subjec o e o s. I ’s also wo h no ing ha he adminis a i e da ase lacks in o ma ion
on wo king hou s o whe he mo he s wo k ull- o pa - ime, making i impossible o
calcula e he employmen inc ease in ull- ime equi alen s.
Addi ionally, we exploi he spa ial a ia ion and he s agge ed implemen a ion o
he manda o y kinde ga en in a Di e ence-in-Di e ences App oach. The desc ip ion
o he da a, me hod, and esul s a e p esen ed in he Online Appendix. We ind ha
labou ma ke pa icipa ion, employmen , and he pa - ime a e o mo he s inc ease
because o manda o y kinde ga en. Howe e , he p ecision o hese esul s is low due
o a small sample size.
7 Conclusion
This pape analyses he impac o manda o y and cos - ee kinde ga en o ou -yea -
old child en on ma e nal labou ma ke beha iou using Swiss adminis a i e da a.
Ou iden i ica ion s a egy exploi s bi hday cu -o da es as andom assignmen o he
manda o y kinde ga en and compa es labou ma ke esponses o mo he s whose
child en en e kinde ga en in he cu en yea e sus he ollowing yea . We apply a
non-pa ame ic eg ession discon inui y design (RDD) and ind s a is ically signi ican
e ec s on he employmen and ea nings o p e iously non-employed mo he s, bu no
o he o al sample o mo he s.
Ou indings con ibu e o he li e a u e on childca e and school en y by analysing
he manda o y and ee o cha ge kinde ga en. Cu en li e a u e sugges s ha
a ending childca e o school a ec s ma e nal labou o ce pa icipa ion posi i ely
in coun ies wi h low ma e nal labou o ce pa icipa ion and low childca e a en-
dance a e (see, e.g. Nollenbe ge and Rod íguez-Planas (2015) o Baue nschus e
and Schlo e (2015)). In Swi ze land, ma e nal labou o ce pa icipa ion exceeds he
OECD a e age. Ou s udy’s esul ha manda o y kinde ga en only sensibly a ec s
labou ma ke decisions o p e iously non-employed mo he s, bu no o o he g oups,
is in line wi h no o only modes e ec s o childca e on ma e nal employmen ound
in coun ies wi h a high ma e nal labou o ce pa icipa ion (see, e.g. Lundin e al.
(2008) and Ha nes and Mogs ad (2011)).
Fu he mo e, wo king pa en s mainly use he suppo o g andpa en s o child-
ca e in Swi ze land (Jacobs Founda ion 2018). The insigni ican impac o manda o y
kinde ga en on labou ma ke esponse in he o al sample o mo he s migh also be
explained by a c owding ou e ec o in o mal childca e (see, e.g. Ha nes and Mogs ad
(2011)). Since manda o y kinde ga en is hal -day ca e in mos can ons, mo he s a e
o en aced wi h he challenge o balancing hei wo k a ound he limi ed kinde ga en
hou s o inding ano he (cos ly) ca e op ion. This ci cums ance migh po en ially
accoun o he absence o a widesp ead impac o manda o y kinde ga en on he
ea nings in he o al sample o mo he s. Fo u u e esea ch, i would be in e es ing
123
F om homemake s o b eadwinne s? How manda o y… Page 21 o 23 44
o examine whe he a manda o y ull-day kinde ga en impac s ma e nal labou o ce
pa icipa ion and ea nings.
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi.
o g/10.1007/s00148-025-01099-y.
Acknowledgemen s We hank he edi o , Kompal Sinha, and h ee anonymous e e ees o help ul com-
men s and sugges ions. The s udy has been ealised using he da a om he Fede al S a is ical O ice (FSO)
and he Cen al Compensa ion O ice (CCO). We a e g a e ul o he da a p o ision as well as he linking
and anonymizing o he da a conduc ed by he FOS. Fu he mo e, we hank Ma yna I e s, Lau a Ra azzini,
Regina T. Riphahn, Jü g Schwe i, and And eas S einmay o hei help ul sugges ions on a p io e sion o
his pape . We ha e bene i ed om commen s by semina pa icipan s in F ibou g, he Ski & Labou Sem-
ina in Engelbe g-Ti lis, he 9 h i o D esden Wo kshop on Labou Economics and Social Policy, he 10 h
In e na ional Con e ence o Panel Da a Use s in Swi ze land in Lausanne, he Swiss Socie y o Economics
and S a is ics Annual Mee ing in Gene a, he EEA in Manches e , he 3 d In e na ional BFH Con e ence on
Disc imina ion in he Labo Ma ke in Be n, he Ve ein ü Socialpoli ik - Annual Con e ence 2019, he 1s
and 4 h Wo kshop o he Swiss Ne wo k on Public Economics (SNOPE), and he Lunch- ime Mee ing a he
Uni e si y o Innsb uck 2021, COMPIE 2022, Eu opean Associa ion o Labou Economis s (EALE) 2024,
37 h Annual Con e ence o he Eu opean Socie y o Popula ion Economics (ESPE), Wo kshop on Labou
Economics 2024, T ie , Ge many and he P e- and Pos doc Semina a he F ied ich-Alexande -Uni e si y
E langen-Nü nbe g.
Funding Open Access unding enabled and o ganized by P ojek DEAL.
Da a a ailabili y This pape uses con iden ial da a om he Fede al S a is ical O ice (FSO) and Old-Age
and Su i o ’s Insu ance (OASI) in Swi ze land. The da a can be ob ained by iling a eques di ec ly wi h
he FSO (h ps://www.b s.admin.ch/b s/de/home/diens leis ungen/da en e knuep ungen/ ue -d i e.h ml).
Decla a ions
Con lic o in e es The au ho s decla e no compe ing in e es s.
Open Access This a icleislicensedunde aC ea i eCommonsA ibu ion4.0In e na ionalLicense,which
pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e
app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence,
and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included
in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I
ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed
by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he
copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
Re e ences
Ba ua R (2014) In e empo al subs i u ion in ma e nal labo supply: E idence using s a e school
en ance age laws. Labou Econ 31:129–140. h p://www.sciencedi ec .com/science/a icle/pii/
S0927537114000852,h ps://doi.o g/10.1016/j.labeco.2014.07.002
Baue nschus e S, Schlo e M (2015) Public child ca e and mo he s’ labo supply–e idence om
wo quasi-expe imen s. J Pub Econ 123:1–16. h ps://www.sciencedi ec .com/science/a icle/pii/
S004727271500002X,h ps://doi.o g/10.1016/j.jpubeco.2014.12.013
Becke GS (1965) A heo y o he alloca ion o ime. Econ J 75:493–517. h p://www.js o .o g/s able/
2228949
Be linski S, Galiani S (2007) The e ec o a la ge expansion o p e-p ima y school acili ies on p eschool
a endance and ma e nal employmen . Labou Econ 14:665–680
123
44 Page 22 o 23 S. Gangl, M. Hube
Be endo LJ, Jongen EL, Mulle P (2015) Childca e subsidies and labou supply — e idence om a la ge
Du ch e o m. Labou Econ 36:112–123
Bundesam ü Sozial e siche ungen 2006 Bo scha zum Bundesbeschluss übe Finanzhil en ü ami-
liene gänzende Kinde be euung. h ps://www. edlex.admin.ch/eli/ ga/2006/357/de
Calonico SD, CM, Fa ell MH, Ti iunik R (2021) Robus da a-d i en s a is ical in e ence in eg ession-
discon inui y designs. h ps://www.google.com/u l?sa= & c =j&q=&es c=s&sou ce=web&
cd=&cad= ja&uac =8& ed=2ahUKEwiyw Qiq yAhWG_ sIHRdnA60QFnoECAgQAQ&
u l=h ps%3A%2F%2Fc an. -p ojec .o g%2Fweb%2Fpackages%2F d obus %2F d obus .pd &
usg=AO Vaw1D9R 0BUIsPFdj2akTO a
Ca a F, Rizzica L (2018) Ea ly kinde ga en, ma e nal labo supply and child en’s ou comes: e i-
dence om i aly. J Public Econ 158:79–102. h p://www.sciencedi ec .com/science/a icle/pii/
S0047272717302141,h ps://doi.o g/10.1016/j.jpubeco.2017.12.012
Eckho And esen M, Ha nes T (2019) Child ca e, pa en al labo supply and ax e enue. Labou Econ
61:101762. h ps://www.sciencedi ec .com/science/a icle/pii/S0927537119300880,h ps://doi.o g/
10.1016/j.labeco.2019.101762
Eu opean Commission 2023 Commission epo inds labou and skills sho ages pe sis and
looks a possible ways o ackle hem. h ps://ec.eu opa.eu/social/main.jsp?langId=en&ca Id=89&
u he News=yes&newsId=10619#na I em-3. Accessed on 05 Oc 2023
Eu ydice (2016/17) Compulso y educa ion in Eu ope – 2016/17. Eu ydice ac s and ig-
u es. h ps://publica ions.eu opa.eu/en/publica ion-de ail/-/publica ion/2 15cd79-9a83-11e6-9bca-
01aa75ed71a1/language-en. Accessed on 13 Aug 2019
Fede al S a is ical O ice, 2022. Umzugsquo e de s ändigen wohnbe ölke ung nach kan on. h ps://
www.b s.admin.ch/b s/en/home/s a is ics/cons uc ion-housing/dwellings/home-mo es.h ml,
no e=Accessed on 14 Feb 2024
Fel e C, Lechne M, Thiemann P (2016) A e -school ca e and pa en s’ labo supply. Labou Econ 42:64–75.
h p://www.sciencedi ec .com/science/a icle/pii/S0927537116300616
Finse aas H, Ha doy I, Schøne P (2017) School en olmen and mo he s’ labo supply: e idence om a
eg ession discon inui y app oach. Re Econ Household 15:621–638
Fi zpa ick MD (2012) Re ising ou hinking abou he ela ionship be ween ma e nal labo supply and
p eschool. J Hum Resou c 47:583–612
Founda ion Jacobs (2018) Whi epape zu Ve einba kei on Familie und Be u : Zwischen Wunsch und
Reali ä . Technical Repo , Jacobs Founda ion
F andsen B (2017) Pa y bias in union ep esen a ion elec ions: Tes ing o manipula ion in he eg ession
discon inui y design when he unning a iable is disc e e. Ad Econ 38:281–315. h ps://doi.o g/10.
1108/S0731-905320170000038012
Gelbach JB (2002) Public schooling o young child en and ma e nal labo supply. Am Econ Re 92:307–
322. h p://www.js o .o g/s able/3083335
Gelman A, Imbens G (2019) Why high-o de polynomials should no be used in eg ession discon inui y
designs. J Bus Econ S a 37:447–456. h ps://doi.o g/10.1080/07350015.2017.1366909
Geye J, Haan P, W ohlich K (2015) The e ec s o amily policy on ma e nal labo supply: combin-
ing e idence om a s uc u al model and a quasi-expe imen al app oach. Labou Econ 36:84–
98. h p://www.sciencedi ec .com/science/a icle/pii/S0927537115000755,h ps://doi.o g/10.1016/j.
labeco.2015.07.001
Goux D, Mau in E (2010) Public school a ailabili y o wo-yea olds and mo he s’ labou supply.
Labou Econ 17:951–962. h p://www.sciencedi ec .com/science/a icle/pii/S0927537110000576,
h ps://doi.o g/10.1016/j.labeco.2010.04.012
Hahn J, Todd P, de Klaauw WV (2001) Iden i ica ion and es ima ion o ea men e ec s wi h a eg ession-
discon inui y design. Econome ica 69:201–209. h p://www.js o .o g/s able/2692190
Ha doy I, Schøne P (2015) En icing e en highe emale labo supply: he impac o cheape day ca e. Re
Econ Household 13:815–836
Ha nesT,Mogs ad M(2011) Money o no hing? Uni e salchildca e andma e nal employmen .JPubEcon
95:1455–1465. h ps://www.sciencedi ec .com/science/a icle/pii/S0047272711000880,h ps://doi.
o g/10.1016/j.jpubeco.2011.05.016. special Issue: In e na ional Semina o Public Economics on
No ma i e Tax Theo y
He mann AB, Mu ie T (2016) Schweize ische A bei sk ä ee hebung - Mü e au dem A bei sma k .
Bundesam ü S a is ik, Neuchâ el h ps://www.b s.admin.ch/b s/de/home/s a is iken/ka aloge-
da enbanken/publika ionen.asse de ail.1061095.h ml
123
F om homemake s o b eadwinne s? How manda o y… Page 23 o 23 44
Kle en H, Landais C, Posch J, S einhaue A, Zweimülle J (2019) Child penal ies ac oss coun ies: e idence
and explana ions. AEA Pape s P oc 109:122–26. h ps://www.aeaweb.o g/a icles?id=10.1257/pandp.
20191078,h ps://doi.o g/10.1257/pandp.20191078
Kle en H, Landais C, Posch J, S einhaue A, Zweimülle J (2020) Do amily policies educe gende
inequali y? E idence om 60 yea s o policy expe imen a ion. Wo king Pape 28082. Na ional Bu eau
o Economic Resea ch. h p://www.nbe .o g/pape s/w28082,h ps://doi.o g/10.3386/w28082
K ap M, Ro h A, Slo winski M (2020) The e ec o childca e on pa en al ea nings ajec o ies.
h ps://www.cesi o.o g/en/publika ionen/2020/wo king-pape /e ec -childca e-pa en al-ea nings-
ajec o ies. cESi o Wo king Pape s
Kunze A, Liu X (2019) Uni e sal childca e o he younges and he ma e nal labou supply. Technical
Repo . CESi o Wo king Pape No. 7509
Lande sø RK, Nielsen HS, Simonsen M (2020) E ec s o school s a ing age on he amily. J Hum
Resou c 55:1258–1286. h ps://jh .uwp ess.o g/con en /55/4/1258,h ps://doi.o g/10.3368/jh .55.4.
1117-9174R1
Lee DS, Lemieux T (2010) Reg ession discon inui y designs in economics. J Econ Li 48:281–355. h ps://
www.aeaweb.o g/a icles?id=10.1257/jel.48.2.281,h ps://doi.o g/10.1257/jel.48.2.281
Lee DS (2008) Randomized expe imen s om non- andom selec ion in U.S. house elec ions. J Econome -
ics 142:675–697. h ps://www.sciencedi ec .com/science/a icle/pii/S0304407607001121,h ps://
doi.o g/10.1016/j.jeconom.2007.05.004. he eg ession discon inui y design: Theo y and applica ions
Le eb e P, Me igan P (2008) Child-ca e policy and he labo supply o mo he s wi h young child en: a
na u al expe imen om Canada. J Law Econ 26:519–548
Lundin D, Mö k E, Öcke B (2008) How a can educed childca e p ices push emale labou supply?
Labou Econ 15:647–659
Nollenbe ge N, Rod íguez-Planas N (2015) Full- ime uni e sal childca e in a con ex o
low ma e nal employmen : Quasi-expe imen al e idence om spain. Labou Econ 36:124–
136. h p://www.sciencedi ec .com/science/a icle/pii/S0927537115000238,h ps://doi.o g/10.1016/
j.labeco.2015.02.008
OECD (2015) Isced 2011 le el 0: ea ly childhood educa ion, in isced 2011 ope a ional manual: guidelines
o classi ying na ional educa ion p og ammes and ela ed quali ica ions. h ps://www.oecd-ilib a y.
o g/docse e /9789264228368-4-en.pd ?expi es=1565703850&id=id&accname=ocid56025002&
checksum=FABD3CBC97D44F307409C640E5CCAC30. Accessed on 13 Aug 2020
Oos e beek H, e Meulen S, an de Klaauw B (2021) Long- e m e ec s o school-s a ing-age ules.
Econ Educ Re 84:102144. h ps://www.sciencedi ec .com/science/a icle/pii/S0272775721000637,
h ps://doi.o g/10.1016/j.econedu e .2021.102144
Ra azziniL(2018) Childca e andma e nalpa - ime employmen :a na u al expe imen using swisscan ons.
Swiss J Econ S a 154:15. h ps://doi.o g/10.1186/s41937-017-0003-x
Schweize ische Kon e enz de kan onalen E ziehungsdi ek o en (2007) In e kan onale Ve einba ung übe
dieHa monisie ungde obliga o ischenSchule(Ha moS-Konko da )h ps://edudoc.ch/ eco d/24711/
iles/Ha moS_d.pd , Accessed on 30 Jan 2019
Schweize ische Kon e enz de kan onalen E ziehungsdi ek o en (2014) Obliga o ische schule: Schulein i
und e s e jah e. h ps://edudoc.ch/ eco d/111988/ iles/schulein i _d.pd . Accessed on 30 Jan 2019
Schweize ische Kon e enz de kan onalen E ziehungsdi ek o en (2016) Ku z-in o: Obliga o ische schule:
Schuls u en, zählweise de schuljah e. h ps://www.edudoc.ch/s a ic/web/a bei en/sp ach_un e /
ku zin o_zaehlweise_d.pd . Accessed on 30 Jan 2019
S e n S, Fel e C (2015) K ippenkos en und - inanzie ung im in e na ionalen Ve gleich. Technical Repo ,
Soziale Siche hei
Swa L, an den Be ge W, an de Wiel K (2019) Do pa en s wo k mo e when child en s a school?
E idence om he Ne he lands. Technical Repo , CPB Discussion Pape
SWI (2006) Wuch iges Ja ü die Bildungs e assung. h ps://www.swissin o.ch/ge /wuch iges-ja- ue -
die-bildungs e assung/5206776. Accessed on 01 Dec 2018
Uni ed Na ions (2022) Old-age po e y has a woman’s ace. h ps://www.un.o g/de elopmen /desa/dspd/
2022/11/old-age-po e y/. Accessed on 06 Oc 2023
Zhu A, B adbu y B (2015) Delaying school en y: sho - and longe - e m e ec s on mo he s’ employmen .
Econ Reco d 91:233–246
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.
123
