Inequali ies in ea ly
childhood and
amilies
Jani E ola
Uni e si y o Tu ku
Lau a Heiskala
Uni e si y o Tu ku
Esa Ka onen
Uni e si y o Tu ku
Elina Kilpi-Jakonen
Uni e si y o Tu ku
No embe 2025
Mapineq deli e able D2.4
Repo : Inequali ies in ea ly childhood and amilies
2
Mapineq – Mapping inequali ies h ough he li e cou se– is a h ee-yea p ojec (2022-
2025) ha s udies he ends and d i e s o in e gene a ional, educa ional, labou ma ke ,
and heal h inequali ies o e he li e cou se du ing he las decades. The esea ch is un by
a conso ium o eigh pa ne s: Uni e si y o Tu ku, Uni e si y o G oningen, Na ional
Dis ance Educa ion Uni e si y, WZB Be lin Social Science Cen e , S ockholm Uni e si y,
Tallinn Uni e si y, Popula ion Eu ope, and Uni e si y o Ox o d
Websi e: www.mapineq.eu
The Mapineq p ojec has ecei ed unding om he Eu opean Union’s Ho izon Eu ope
esea ch and inno a ion p og amme unde he g an ag eemen No. 101061645.
Views and opinions exp essed a e howe e hose o he au ho (s) only and do no
necessa ily e lec hose o he Eu opean Union, he Eu opean Resea ch Execu i e Agency,
o hei a ilia ed ins i u ions. Nei he he Eu opean Union no he g an ing au ho i y can
be held esponsible o hem.
Publica ion in o ma ion:
This wo k is licensed unde he C ea i e Commons A ibu ion-NonComme cial-Sha eAlike
4.0 In e na ional (CC BY-NC-SA 4.0) license. You a e ee o sha e and adap he ma e ial
i you include p ope a ibu ion (see sugges ed ci a ion), indica e i changes we e made,
and do no use o adap he ma e ial in any way ha sugges s he licenso endo ses you
o you use. You may no use he ma e ial o comme cial pu poses.
Summa y his o y
Ve sion
Da e
Commen s
1.0
24.11.2025
Manusc ip o e iew
1.1.
27.11.2025
Manusc ip e iewed o submission
Sugges ed ci a ion:
E ola, J., Heiskala, L., Ka onen, E., & Kilpi-Jakonen, E. (2025). Inequali ies in ea ly
childhood and amilies. Mapineq deli e ables. Tu ku: INVEST Resea ch Flagship Cen e /
Uni e si y o Tu ku. DOI: 10.5281/zenodo.17732967
Repo : Inequali ies in ea ly childhood and amilies
3
Execu i e summa y
This wo k package examined how inequali ies in ea ly childhood and amily en i onmen s
shape child en’s li e chances ac oss Eu ope, wi h a pa icula ocus on educa ion as a
cen al li e-cou se ou come. The aim was o unde s and how amily s uc u e, pa en al
esou ces, ea ly ins i u ional se ings, and gene ic and social ac o s in e ac o p oduce o
mi iga e educa ional and socioeconomic inequali ies om ea ly childhood onwa d.
Resea ch combined c oss-na ional su ey da a wi h high-quali y No dic popula ion egis e
da a and gene ically in o med designs, enabling bo h b oad compa isons and causal
in e ence.
Ac oss all coun ies s udied, pa en al educa ion eme ged as he mos powe ul and
consis en p edic o o child en’s economic and educa ional ou comes. Child en g owing
up in highly educa ed amilies expe ience sys ema ically be e economic condi ions han
hose in lowe -educa ed amilies. Al hough household s uc u e di e s by pa en al
educa ion — wi h mo e s able wo-adul households and ewe siblings among highly
educa ed amilies — hese s uc u al di e ences explain only a ela i ely small sha e o
inequali ies. Mos dispa i ies a e d i en by di e ences in pa en al ea nings and labou
ma ke e u ns o educa ion, a he han by amily composi ion i sel .
Analyses o pa en al income in Sweden and Finland sugges ha childhood economic
esou ces ha e a causal impac on child en’s chances o comple ing highe educa ion.
Resea ch designs using ins umen al a iables and sibling ixed e ec s demons a e ha
s anda d co ela ional models unde es ima e hese e ec s. S able pa en al employmen ,
especially a achmen o he labou ma ke , is a leas as impo an as income i sel ,
indica ing ha wo k- ela ed s abili y p o ides bo h ma e ial and psychosocial ad an ages.
Economic shocks a he egional le el, such as ecessions, can c ea e long-las ing
inequali ies when hey dis up pa en al employmen du ing child en’s o ma i e yea s.
The wo k also ea ed amily human capi al as dynamic a he han ixed. Family
dis up ions — no ably pa en al sepa a ion o dea h — a e common and a e s ongly
associa ed wi h lowe educa ional a ainmen . Con a y o p e ious indings, he e was no
e idence o e-pa ne ing compensa ing o hese losses, and i may, in some con ex s,
deepen disad an age. Pa en al educa ional upg ading du ing a child’s upb inging shows
only weak and non-causal associa ions wi h child en’s own educa ional ou comes once
amily-le el selec ion is p ope ly con olled. This indica es ha while pa en al educa ion
ma e s g ea ly, la e imp o emen s in pa en s’ educa ion a ely ansla e di ec ly in o
imp o ed ou comes o child en.
Beyond he amily, he p ojec examined he ole o geog aphical mobili y and local
oppo uni y s uc u es. Li ing close o highe educa ion ins i u ions subs an ially inc eases
he likelihood o comple ing e ia y educa ion. Howe e , esiden ial mobili y i sel is
dis up i e: child en who mo e du ing childhood pe o m wo se educa ionally han hose
who emain in s able en i onmen s, e en when mo es imp o e ins i u ional access.
Impo an ly, place-based expansion o oppo uni y s uc u es — such as he Finnish
poly echnic expansion — bene i s child en who g ow up locally, educing he need o
Repo : Inequali ies in ea ly childhood and amilies
4
dis up i e mobili y. Family ins abili y, pa icula ly pa en al unemploymen and sepa a ion,
ampli ies he nega i e e ec s o mo ing.
A inal con ibu ion o his wo k package conce ns ea ly childhood educa ion and ca e
(ECEC/dayca e) and gene ic–social in e plays. The indings show ha g ea e dayca e
a ailabili y is associa ed wi h a s onge co ela ion be ween gene ic p edisposi ions and
obse ed educa ional ou comes, ega dless o social backg ound. In o he wo ds, dayca e
does no educe o widen social gaps di ec ly. Ra he , i appea s o unc ion as an
ins i u ional con ex in which indi idual gene ic po en ial is mo e ully ealized. Dayca e
hus shapes he deg ee o gene–en i onmen co ela ion, inc easing he p edic i e powe
o gene ic di e ences o la e ou comes wi hou al e ing a e age achie emen le els o
selec i ely bene i ing pa icula social s a a.
Taken oge he , he e idence leads o a clea conclusion: inequali ies o igina e ea ly and
a e p ima ily s uc u ed by pa en al educa ion, economic s abili y, amily con inui y,
ins i u ional quali y, and hei in e ac ion wi h gene ic p edisposi ions. Policies aimed a
educing inequali y should he e o e p io i ise ea ly-li e in e en ions: s eng hening
educa ional and labou ma ke oppo uni ies o pa en s, educing nega i e consequences
o amily ins abili y, in es ing in high-quali y uni e sal dayca e, and imp o ing local
oppo uni y s uc u es. In e ening ea ly is a mo e e ec i e han a emp ing o
compensa e o disad an age la e in he li e cou se, as inequali ies es ablished in
childhood end o pe sis in o adul hood.
Repo : Inequali ies in ea ly childhood and amilies
5
Abb e ia ions
SES Socioeconomic s a us
Repo : Inequali ies in ea ly childhood and amilies
6
Con en
EXECUTIVE SUMMARY 3
ABBREVIATIONS 5
1. INTRODUCTION 7
2. ECONOMIC CIRCUMSTANCES OF CHILDREN LIVING IN HIGHER AND LOWER-EDUCATED
FAMILIES AND THE CONTRIBUTION OF HOUSEHOLD STRUCTURE 8
3. PARENTAL INCOME AND THE ACHIEVEMENT OF HIGHER EDUCATION IN SWEDEN AND
FINLAND 10
4. CHANGE IN HUMAN CAPITAL OF THE IMMEDIATE FAMILY 11
5. CHILDHOOD GEOGRAPHICAL MOBILITY, LOCAL OPPORTUNITY STRUCTURES, AND THE
PURSUIT OF HIGHER EDUCATION 14
6. CHANGES IN WIDER SOCIAL ENVIRONMENTS: DAYCARE 17
7. CONCLUDING REMARKS 18
REFERENCES 20
Tables
TABLE 1. LINEAR REGRESSION ANALYSIS OF YEARS OF EDUCATION (N=2,008) ............................................................... 18
Figu es
FIGURE 1. MEAN YEARS OF EDUCATION BY PARENTAL EDUCATION QUARTILES AND FAMILY (IN)STABILITY (N=702 038).
ADAPTED FROM HEISKALA ET AL. 2024. ........................................................................................................... 13
FIGURE 2. MEAN YEARS OF EDUCATION BY PARENTAL EDUCATION QUARTILES AND PARENTAL EDUCATIONAL UPGRADE (N=439
906). ADAPTED FROM HEISKALA ET AL. 2024. ................................................................................................. 14
FIGURE 3. CHANGE IN THE DISTRIBUTION OF FINNISH HIGHER EDUCATION INSTITUTIONS FROM 1982–2017. CIRCLES ARE
UNIVERSITY INSTITUTIONS, AND BLACK DOTS REPRESENT POLYTECHNICS. ............................................................... 15
FIGURE 4. PREDICTED PROBABILITIES OF HIGHER EDUCATION ATTAINMENT AT AGE 30 BY MOVING TYPE DURING CHILDHOOD
(AGES 5–15). ............................................................................................................................................... 16
Repo : Inequali ies in ea ly childhood and amilies
7
Inequali ies in ea ly childhood
and amilies
The ea ly childhood amily con ex s a e conside ed o ma i e o adul educa ional and
socioeconomic ou comes. We s udy he in e gene a ional impac s o pa en al educa ion
and economic esou ces, as well as amily s uc u e, and conside he modi ying e ec s o
local educa ional oppo uni y s uc u es on social and gene ic in luences on educa ional
and socioeconomic a ainmen .
Childhood amily s uc u e
has a modes impac on
adul a ainmen compa ed
o pa en s' educa ional and
socioeconomic esou ces,
independen ly o he
coun y con ex .
While pa en al educa ion
emains he mos
consequen ial ac o o
child en’s a ainmen ,
pa en al economic
esou ces emain c ucial
and a e o en empi ically
unde es ima ed.
Uni e sally a ailable local
educa ional oppo uni y
s uc u es inc ease
educa ional a ainmen
and can e en boos he
chances o mee ing gene ic
po en ial o e e yone.
________
________
________
1. In oduc ion
The WP2 in he MAPINEQ p ojec ocused on ea ly childhood and amily p ocesses ela ed
o educa ional and socioeconomic a ainmen in adul hood, especially di e ences and
changes in amilies and b oade social en i onmen s. Rega ding ea ly childhood, we e e
o he ini ial yea s o he li e cou se, om bi h o he s a o compulso y educa ion. In he
li e a u e on social inequali ies, his p e-ma ke pe iod is iewed as he mos in luen ial o
la e achie emen : i is when he ounda ions o cogni i e and non-cogni i e skill
de elopmen a e es ablished (e.g., Duncan & B ooks-Gunn, 2000; Heckman, 2006).
We di e en ia e he e ec s o amily backg ound om hose o he non- amily
en i onmen . The e ec s o amily backg ound a e ansmi ed om pa en s o child en
h ough social o gene ic in luences. Social in luences a e usually associa ed wi h pa en s’
socioeconomic and o he social g oup- ela ed cha ac e is ics, such as hei educa ional
a ainmen , occupa ional s a us and income (E ola & Kilpi-Jakonen, 2017). Pa en al
esou ces ela ed o hese cha ac e is ics end o be c ucial o he kind o immedia e
amily en i onmen in which child en a e b ough up. Fo ins ance, pa en al income la gely
de e mines he amily's economic ci cums ances, including i s ma e ial weal h and e en
he neighbou hood o he amily home. Howe e , di e en ypes o amily esou ces end
o co ela e. As educa ion is he mos impo an con ibu o o occupa ional a ainmen ,
Repo : Inequali ies in ea ly childhood and amilies
8
and jobs end o be he main sou ces o income o mos amilies, he e ec s o pa en al
educa ion, occupa ion s a us and income on child en’s u u e educa ional and
socioeconomic a ainmen end o o e lap conside ably (E ola e al., 2016).
While he genes child en acqui e om hei pa en s a e ixed, hei e ec s on la e
a ainmen (and subsequen ends in social inequali ies) a e no , bu also a y acco ding
o bo h amily- and non- amily- ela ed social in luences (Mills & T op , 2020). Mos amily
backg ound e ec s obse ed in social scien i ic s udies a e in ac a mix o bo h, as well
as he non- amily en i onmen : o ins ance, educa ional a ainmen ypically depends on
pa en al in luences on he educa ional choices o child en, child en’s own abili ies linked
wi h hei gene ically inhe i ed cogni i e skills, and he ins i u ional en i onmen p o iding
he con ex o educa ional oppo uni ies (e.g., access o academic ack educa ion).
The s uc u e o he immedia e amily en i onmen is a c ucial aspec o many
de elopmen s du ing ea ly childhood. I has been a gued ha he di usion o non-s anda d
amily o ms among low-educa ed and low SES pa en s is s eng hening in e gene a ional
inequali ies in Wes e n socie ies (Che lin, 2014; Esping-Ande sen, 2007; McLanahan &
Pe cheski, 2008). The bene i s o pa en s’ economic esou ces and he ime alloca ed o
child ea ing a e likely o depend on he age o he child and he sibship composi ion,
whe eas he ole models o he alues pa en s p o ide a e no expec ed o su e as much
om dilu ion.
Bo h ypes o amily in luences a e suscep ible o a ia ion in he non- amily en i onmen .
Fo he la e a ia ion, we ocus on local-le el di e ences in educa ional oppo uni y
s uc u es, namely he a ailabili y o ea ly childhood educa ion and highe educa ion
ins i u ions.
2. Economic ci cums ances o child en li ing in
highe and lowe -educa ed amilies and he
con ibu ion o household s uc u e
Lau a Heiskala, Minna Tuominen, Jani E ola, and Elina Kilpi-Jakonen
In his deli e able (D2.1), la e sligh ly e ised and published as a pee - e iewed a icle
(Heiskala e al., 2025), we assessed he ex en o which household s uc u e explains
di e ences in child en’s economic ci cums ances ac oss highe - and lowe -educa ed
households. We also s udied whe he he impo ance o household s uc u e o his
income gap a ies be ween Eu opean coun ies and egions. While p e ious esea ch has
ypically used adul s o households as he uni o analysis, his s udy adop s a child-cen ed
pe spec i e, ocusing on income dispa i ies be ween child en li ing in highe - and lowe -
educa ed amilies. The analysis d aws on c oss-na ionally compa able da a om he
Gene a ions and Gende Su ey, co e ing mo e han 60,000 child en in 11 Eu opean
coun ies, and employs bo h desc ip i e me hods and Blinde –Oaxaca decomposi ion
echniques. In ou s udy, highe -educa ed households a e de ined as hose in which a
leas one adul has comple ed a e ia y-le el deg ee.
Repo : Inequali ies in ea ly childhood and amilies
9
Ac oss all coun ies s udied, child en in highly educa ed households enjoy be e economic
ci cums ances han hose in lowe -educa ed households. These di e ences a e
subs an ial: in some coun ies, he gap in coun y-speci ic income pe cen ile based on
equi alence-scaled household income eaches mo e han 30 pe cen ile poin s. The size
o he educa ional g oup income gap a ies conside ably be ween coun ies, being
smalles (a ound 15 pe cen age poin s) in he Ne he lands, Poland, and No way and
la ges in Romania. Despi e hese a ia ions, he o e all pa e n is clea : pa en al
educa ion is s ongly associa ed wi h child en’s economic ci cums ances. Fu he mo e, we
assessed whe he di e ences in household s uc u e explain he gap in child en’s
economic ci cums ances be ween highe - and lowe -educa ed amilies, o whe he i
mainly e lec s e u ns o educa ion.
Ou desc ip i e esul s show ha child en o highly educa ed pa en s a e mo e likely o li e
in wo-adul households, ha e ewe siblings, and ha e pa en s who en e ed pa en hood
a olde ages han child en in lowe -educa ed households. Bu do hese di e ences in
household s uc u e explain hei be e economic ci cums ances, and i so, o wha
ex en ? Ou decomposi ion analyses e eal ha hese household cha ac e is ics accoun
o only a modes sha e o he income gap: ypically a ound 5 pe cen age poin s in absolu e
e ms, and up o 20 pe cen in ela i e e ms in some coun ies. Somewha unexpec edly,
he di e ences ac oss coun ies we e a he small, pa icula ly ega ding he con ibu ion
o household s uc u e o he income gap be ween child en om lowe - and highe -
educa ed households in absolu e e ms. In o he wo ds, while amily composi ion ma e s,
mos o he ad an age enjoyed by child en in highly educa ed households s ems om
di e ences in pa en al ea nings and e u ns o educa ion, a he han om di e ging
des inies due o socioeconomic di e ences in household s uc u es, and his pa e n
appea s o be consis en ac oss all Eu opean coun ies s udied.
The MapIneq 2.1. deli e able also explo ed egional a ia ion wi hin coun ies, inding ha
di e ences in he ole o household s uc u e o he educa ional g oup income gap a e
e en smalle a he egional le el han ac oss coun ies. Thus, he pa e ns obse ed
ac oss egions la gely mi o hose obse ed be ween coun ies. This sugges s ha
na ional con ex s, such as wel a e egimes and social policies, play a mo e impo an ole
han local en i onmen s in shaping hese pa e ns.
Taken oge he , hese indings unde sco e he impo ance o educa ional inequali ies in
shaping child en’s li ing condi ions. While di e ences in household s uc u e con ibu e o
dispa i ies, hey do so only ma ginally. The economic ci cums ances o child en in Eu ope
a e s ongly ied o pa en al educa ion, and his associa ion pe sis s ac oss di e se
demog aphic and policy con ex s. Household s uc u e plays a seconda y ole, explaining
only a ac ion o he obse ed dispa i ies. The policy implica ions a e clea : in e en ions
aimed a educing child po e y o inequali y should ocus p ima ily on imp o ing
educa ional oppo uni ies a he han on in luencing amily composi ion. These esul s
unde line he well-es ablished inding ha educa ion is a key de e minan o
in e gene a ional inequali y.
Repo : Inequali ies in ea ly childhood and amilies
16
mo e s wi h access', 'non-mo e s gaining access', 'mo e s o imp o ed oppo uni y', o
'mo e s o wo se access'. Ma ginal s uc u al models (MSMs) wi h in e se-p obabili y
weigh ing adjus o ac o s such as pa en al sepa a ion and unemploymen , which bo h
in luence and a e in luenced by mo ing decisions.
Figu e 4 summa ises he main indings. Child en who con inuously li ed in a eas wi h
nea by highe educa ion ins i u ions had he highes a ainmen a es (≈42%). Those who
emained in a eas ha la e gained access, h ough he opening o a new poly echnic o
uni e si y, also pe o med s ongly. In con as , child en who mo ed, e en in o a eas wi h
be e access, had no ably lowe p obabili ies o comple ing a deg ee (≈27%). The ac o
mo ing i sel appea s mo e de imen al han he bene i o he imp o ed oppo uni y
gained.
Figu e 4. P edic ed p obabili ies o highe educa ion a ainmen a age 30 by mo ing ype
du ing childhood (ages 5–15).
In addi ion, amily ci cums ances s ongly shape hese pa e ns: pa en al sepa a ion and
unemploymen in ensi y he nega i e impac o mobili y, while s able amily condi ions can
mi iga e i . Child en who mo e amid amily dis up ion a e pa icula ly poo ly. Thus, place
and amily s abili y join ly unde pin educa ional success. The Finnish expansion o
poly echnics demons a es ha upg ading local oppo uni y s uc u es bene i s hose who
Repo : Inequali ies in ea ly childhood and amilies
17
g ew up nea by, including hose who would ha e o he wise had o mo e o access
poly echnics. Mo eo e , he indings highligh ha social policies suppo ing amily
s abili y, such as measu es o cushion he impac s o unemploymen o sepa a ion, could
indi ec ly enhance educa ional a ainmen by educing he impac o mobili y-d i en
dis up ion.
6. Changes in wide social en i onmen s: Dayca e
Elina Kilpi-Jakonen, Minna Tuominen, and Jani E ola
In his ask, we ocused on how he local ins i u ional con ex , in his case, he a ailabili y
o ea ly childhood educa ion, is connec ed o he in e play be ween social and gene ic
in luences in educa ional a ainmen . The exis ing empi ical e idence on he bene i s o
ea ly childca e on la e a ainmen has been mixed and a ies be ween coun y con ex s.
O en, s udies ha e ound i o be bene icial o child en om disad an aged amily
backg ounds (Ruhm & Wald ogel, 2012; Kulic e al., 2019), while o he s ha e ound
indica ions o he mul iplica i e bene i s o he high-SES amily child en in some coun y
con ex s (D ange & Telle, 2020, also o No way Laaninen e al., 2024), and o he s ha e
ound ba ely e ec s a all (Ka hula e al., 2017; G ube e al., 2023).
In his s udy, we conside ed whe he he mixed indings may esul om he dayca e
a ailabili y mode a ing he gene-en i onmen in e play. Beha iou al gene ics s udies
sugges ha amily backg ound e ec s in educa ion a e pa ly shaped by passi e gene-
en i onmen co ela ions (Plomin e al., 1977). Pa en s pass on genes and choose he
ea ing en i onmen , bu genes also ma e o ea ing en i onmen choices. Fo ins ance,
highe cogni i e skills ha a e highly he i able a e likely con ibu ing o he
in e gene a ional ansmission o highe educa ion, bu also imp o e pa en al ea nings
ha allow hem o make mo e eely choices ega ding he amily home. Because o his,
many e ec s assumed o esul om he ea ing en i onmen may in ac be due o
unobse ed gene ic e ec s. The g ea e a ailabili y o dayca e can be expec ed o
homogenise ea ing en i onmen s and mode a e passi e gene-en i onmen in e ac ions.
A he same ime, his li e a u e assumes ha educing social ba ie s should inc ease
child en’s oppo uni ies o ealize hei gene ic po en ial (Sca -Salapa ek, 1971; Guo &
S ea ns, 2002; Nielsen, 2006). When his is he case, uni e sally a ailable dayca e should
s eng hen he impo ance o child en’s own genes in educa ional a ainmen .
We used Finnish egis e -linked da a on bi h coho s 1973–78, wi h gene ic in o ma ion
(polygenic sco es o educa ion) a ailable om su eys, and u he ma ched wi h
municipali y-le el in o ma ion on dayca e a ailabili y in 1976–81 (as digi ized by Silliman,
2023). Dayca e a ailabili y was measu ed a age 3, pa en al educa ion a ound age 10–
15, and child en’s educa ion a age 40. Full in o ma ion on all o hese (including polygenic
sco es) was a ailable o 2,008 indi iduals. The pe iod was c ucial o he expansion o
he dayca e in Finland. The dayca e law was in oduced in 1973, and while he se ice was
al eady widely a ailable in he la ges ci ies, i was only spa sely a ailable in u al
municipali ies and expanded only a e ha .
Repo : Inequali ies in ea ly childhood and amilies
18
The indings sugges ha he inc easing a ailabili y o dayca e was no associa ed wi h
imp o ed educa ional a ainmen , eplica ing Silliman's (2023) p e ious Finnish indings.
Howe e , dayca e a ailabili y appea ed o boos he associa ion be ween genes and
educa ional a ainmen , as indica ed by a posi i e in e ac ion be ween dayca e a ailabili y
and he polygenic sco e (Table 1). The e we e no subs an ial di e ences in his in e ac ion
be ween he child en o highe -educa ed pa en s and hose o o he pa en s, sugges ing
ha he bene i was la gely uni e sal.
Table 1. Linea eg ession analysis o yea s o educa ion (N=2,008)
M1
M2
M3
Dayca e a ailabili y (in owns)
1.262
1.278
1.028
(1.669)
(1.668)
(1.641)
Ru al a ea * dayca e a ailabili y
1.627
1.455
0.534
(3.127)
(3.131)
(3.110)
Polygenic sco e (PS)
0.653***
0.662***
0.564***
(0.064)
(0.063)
(0.092)
PS * dayca e a ailabili y
1.052***
0.670*
(0.355)
(0.363)
Pa en al educa ion
X
All models include municipali y ixed e ec s and con ol o sex and he p incipal componen s o
gene ic a ia ion. M3, which adds pa en al educa ion, also adds he in e ac ion be ween pa en al
educa ion and he polygenic sco e. Dayca e a ailabili y heo e ically a ies be ween 0–1 and has
been cen ed in he analyses a 0.3. S a is ical signi icance: *p<0.05, **p<0.01, ***p<0.001.
The e idence sugges s ha dayca e a ailabili y did no make much o a di e ence o he
passi e gene-en i onmen in e ac ions. Howe e , hey do sugges ha dayca e may ha e
a boos ing e ec : he mo e ex ensi e he a ailabili y o dayca e, he s onge he
associa ion be ween polygenic sco e and educa ional a ainmen . Thus, he dayca e
expansion may indeed ha e inc eased he likelihood ha child en om di e en
backg ounds would mee hei gene ic po en ial. Howe e , he ac ha his boos ing e ec
appea ed o be simila ac oss he social s a a also sugges s ha i did no esul om he
dayca e sys em demolishing he ba ie s o educa ional achie emen o speci ic
sociodemog aphic g oups, bu a he ha he bene i s o he se ice made uni e sally
a ailable we e also mos ly uni e sal.
7. Concluding ema ks
The indings o Wo k Package 2 ha e been ex ensi e, bo h con i ming p e ious
unde s andings, cla i ying o he s, and e u ing o he s. All in all, he indings sugges ha
while amily s uc u e is consequen ial o child en’s adul educa ional and socioeconomic
ou comes, i is much less impo an han ac o s ela ed o pa en s’ educa ion,
occupa ional s a us, and income. This appea s o be ue ac oss mul iple coun y con ex s.
This inding u he sugges s ha any policy e o s o imp o e child en's adul ou comes
by in luencing amily a angemen s (e.g., di o ce laws) a e likely o be much less
consequen ial han policies ha in luence educa ional and labou -ma ke oppo uni ies.
Repo : Inequali ies in ea ly childhood and amilies
19
Fu he mo e, while he esul s also unde line he impo ance o pa en al educa ion o
child en's u u e socioeconomic a ainmen , hey also unde sco e he limi s o he bene i s
o imp o emen s in amily human capi al. Nei he a pa en acqui ing a highe educa ional
deg ee du ing childhood no a new, be e -educa ed s ep-pa en en e ing he amily
imp o ed he child en's educa ional a ainmen . E en local educa ional oppo uni y
s uc u es appea ed o ha e only a limi ed impac on ou comes. Mo ing o a new place
du ing childhood ended o ha e de imen al e ec s on u u e educa ion, e en i he new
place o e ed be e highe educa ion oppo uni ies han he a eas he amilies we e
mo ing away om. O e all, e en hough p e ious s udies sugges ha pa en al income is
less consequen ial o adul ou comes han pa en al occupa ional s a us and educa ion,
he indings indica e ha he ypically applied empi ical models unde es ima e he
impo ance o amily income, e en in No dic egali a ian wel a e-s a e con ex s wi h
ex ensi e income edis ibu ion.
Howe e , he uni e sally a ailable educa ional sys ems can s ill ha e bene i s, e en i he
e ec s may be mo e limi ed han is o en conside ed. Those g owing up in a eas wi h be e
access o highe educa ion ended o go u he han o he s. Finally, inc easing access o
ea ly childhood educa ion appea s o imp o e child en's chances o mee ing hei gene ic
po en ial, independen o hei backg ound.
Repo : Inequali ies in ea ly childhood and amilies
20
Re e ences
Che lin, A. J. (2014). Labo ’s Lo e Los : The Rise and Fall o he Wo king-Class Family in
Ame ica. Russell Sage Founda ion.
D ange, N., & Telle, K. (2020). Seg ega ion in a Uni e sal Child Ca e Sys em:
Desc ip i e Findings om No way. Eu opean Sociological Re iew, 36(6), 886–
901. h ps://doi.o g/10.1093/es /jcaa026
Duncan, G. J., & B ooks-Gunn, J. (2000). Family Po e y, Wel a e Re o m, and Child
De elopmen . Child De elopmen , 71(1), 188–196.
E ola, J., Jalonen, S., & Leh i, H. (2016). Pa en al Educa ion, Class and Income o e
Ea ly Li e Cou se and Child en’s Achie emen . Resea ch in Social S a i ica ion
and Mobili y, 44, 33–43.
E ola, J., & Kilpi-Jakonen, E. (2017). Chap e 1: Compensa ion and o he o ms o
accumula ion in in e gene a ional social inequali y. In J. E ola & E. Kilpi-Jakonen
(Eds.), Social Inequali y Ac oss he Gene a ions. The Role o Compensa ion and
Mul iplica ion in Resou ce Accumula ion (pp. 3–25). Edwa d Elga .
h ps://www.elga online.com/edcollchap/edcoll/9781786432551/978178643
2551.00006.xml
Esping-Ande sen, G. (2007). Equal Oppo uni ies and he Wel a e S a e. Con ex s, 6(3),
23–27. h ps://doi.o g/10.1525/c x.2007.6.3.23
G ube , J., Kosonen, T., & Hu unen, K. (2023). Paying Moms o S ay Home: Sho and
Long Run E ec s on Pa en s and Child en (Wo king Pape No. 30931). Na ional
Bu eau o Economic Resea ch. h ps://doi.o g/10.3386/w30931
Guo, G., & S ea ns, E. (2002). The Social In luences on he Realiza ion o Gene ic
Po en ial o In ellec ual De elopmen . Social Fo ces, 80(3), 881–910.
h ps://doi.o g/10.1353/so .2002.0007
Heckman, J. J. (2006). Skill Fo ma ion and he Economics o In es ing in
Disad an aged Child en. Science, 312(5782), 1900–1902.
h ps://doi.o g/10.1126/science.1128898
Heiskala, L., Kailaheimo-Lönnq is , S., Ka onen, E., & E ola, J. (2024). Change in
human capi al o he immedia e amily. Mapineq deli e ables. Tu ku: INVEST
Resea ch Flagship Cen e / Uni e si y o Tu ku. DOI:
10.5281/zenodo.14355337
Heiskala, L., Tuominen, M., E ola, J., & Kilpi-Jakonen, E. (2025). Economic
Ci cums ances o Child en Li ing in Highe and Lowe -Educa ed Families and he
Con ibu ion o Household S uc u e: A C oss-Coun y Compa ison wi h a Child's
Pe spec i e. S udies o T ansi ion S a es and Socie ies, 17, 4–23.
Jän i, M., & Ka onen, E. (2025). Pa en al income and he achie emen o he highe
educa ion in Sweden and Finland. Mapineq deli e ables. Tu ku: INVEST
Resea ch Flagship Cen e / Uni e si y o Tu ku.
h ps://doi.o g/10.5281/zenodo.15267325
Ka hula, A., E ola, J., & Kilpi-Jakonen, E. (2017). Home swee home? Long- e m
educa ional ou comes o childca e a angemen s in Finland. In H.-P. Bloss eld,
N. Kulic, J. Skopek, & M. T i en i (Eds.), Childca e, Ea ly Educa ion and Social
Inequali y – A C oss-na ional Pe spec i e (pp. 268–285). Edwa d Elga .
Kulic, N., Skopek, J., T i en i, M., & Bloss eld, H. P. (2019). Social backg ound and
child en's cogni i e skills: The ole o ea ly childhood educa ion and ca e in a
Repo : Inequali ies in ea ly childhood and amilies
21
c oss-na ional pe spec i e. Annual Re iew o Sociology, 45(1), 557–579.
h ps://doi.o g/10.1146/annu e -soc-073018-022401
Laaninen, M., Kulic, N., & E ola, J. (2024). Age o en y in o ea ly childhood educa ion
and ca e, li e acy and educ ion o educa ional inequali y in No dic coun ies.
Eu opean Socie ies, 26(5), 1333–1362.
h ps://doi.o g/10.1080/14616696.2024.2310694
McLanahan, S., & Pe cheski, C. (2008). Family S uc u e and he Rep oduc ion o
Inequali ies. Annual Re iew o Sociology, 34(1), 257–276.
Mills, M. C., & T op , F. C. (2020). Sociology, Gene ics, and he Coming o Age o
Sociogenomics. Annual Re iew o Sociology, 46(1), 553–581.
h ps://doi.o g/10.1146/annu e -soc-121919-054756
Nielsen, F. (2006). Achie emen and Asc ip ion in Educa ional A ainmen : Gene ic and
En i onmen al In luences on Adolescen Schooling. Social Fo ces, 85(1), 193–
216. h ps://doi.o g/10.1353/so .2006.0135
Plomin, R., DeF ies, J. C., & Loehlin, J. C. (1977). Geno ype-en i onmen in e ac ion and
co ela ion in he analysis o human beha io . Psychological Bulle in, 84(2),
309–322. h ps://doi.o g/10.1037/0033-2909.84.2.309
Ruhm, C., & Wald ogel, J. (2012). Long- e m e ec s o ea ly childhood ca e and
educa ion. No dic Economic Policy Re iew, 1(1), 23–51.
Sca -Salapa ek, S. (1971). Race, Social Class, and IQ. Science, 174(4016), 1285–
1295.
Silliman, M. I. (2023). Essays in educa ion policy and labo economics. Doc o al
disse a ion, Ha a d Uni e si y G adua e School o A s and Sciences.
