The Geog aphy o In e gene a ional Mobili y
in Eu ope
Oli ia G ans öm, Pe Engzell
Documen ype
Pos -p in : This manusc ip has passed pee e iew and been accep ed o publica ion by a jou nal.
I may include inal edi s by he au ho (s) bu no edi ing o o ma ing by he publishe .
Funding in o ma ion
This esea ch was unded by he Eu opean Resea ch Council, G an Ag eemen No. 101165962
Ma ke s and Mobili y: How Employe s S uc u e Economic Oppo uni y.
Sugges ed ci a ion
G ans öm, Oli ia and Pe Engzell. (2025). The Geog aphy o In e gene a ional Mobili y in Eu ope.
Eu opean Socie ies, o hcoming.
Da e o eco d
Oc obe 10, 2025.
Te ms o use
This wo k is licensed unde a C ea i e Commons A ibu ion-NonComme cial-NoDe i a i es 4.0
In e na ional license: h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
The Geog aphy o In e gene a ional Mobili y
in Eu ope
Oli ia G ans öm
S ockholm Uni e si y
Pe Engzell
Uni e si y College London
Oc obe 10, 2025
Abs ac
How do oppo uni ies o in e gene a ional mobili y depend on whe e you
li e? We add ess his ques ion using Eu opean Social Su ey da a, s udying
he associa ion be ween pa en s’ and child en’s occupa ion, and how i a ies
by egion o esidence. Absolu e mobili y, la gely d i en by shi s in occu-
pa ional s uc u e, di e s om ela i e mobili y, which e lec s he ex en o
which oppo uni ies a e equal ac oss social o igins. Capi al egions eme ge as
hubs o absolu e, bu no necessa ily ela i e, mobili y. Absolu e mobili y co -
ela es wi h human capi al, labo ma ke , demog aphic, and spa ial ac o s.
In con as , ela i e mobili y is p ima ily shaped by economic dispa i ies be-
ween social classes. G ea e inequali y en ails less mobili y, e en compa ing
di e en places wi hin a coun y.
1 In oduc ion
In e gene a ional mobili y e lec s he ex en o which economic s a us pe sis s
ac oss gene a ions. Pa en s wi h g ea e skills and esou ces a e be e placed o
in es in hei child en, and coun e ac ing his imbalance is a key objec i e o social
policy in libe al wel a e s a es (Co ak, 2013). I is di icul o de e mine he op i-
mal le el o mobili y in socie y: mos people p obably hold ha pa en s should be
allowed some in luence o e hei child en. Ne e heless, compa ing coun ies and
egions can help iden i y places whe e oppo uni ies a e mo e o less equal (B een
and Jonsson, 2005). Beyond i s link o oppo uni y, mobili y may be impo an o
ou comes such as social cohesion, democ a ic ep esen a ion, o he e icien alloca-
ion o alen . Following a adi ion ha iews he social di ision o labo as he
backbone o s a i ica ion, we ocus on occupa ional mobili y as he basis o ou
analysis (Ganzeboom e al., 1991).
P e ious compa a i e wo k in his a ea has wo key limi a ions. Fi s , i has
ocused on coun y-le el di e ences, a ibu ing a ia ion o na ional policies such
as ax egimes and educa ion sys ems. Howe e , ecen esea ch sugges s ha op-
po uni y a ies almos as much wi hin coun ies as be ween hem (Che y e al.,
2014; DiP e e, 2020; S uhle , 2018). Secondly, schola s ha e sough o loca e he
1
de e minan s o mobili y in childhood, neglec ing how he labo ma ke ha peo-
ple ace may s uc u e hei oppo uni ies (Engzell and Mood, 2023; Engzell and
Wilme s, 2025). We ad ance he li e a u e on bo h on s, by s udying how occu-
pa ional mobili y a ies wi hin and be ween Eu opean coun ies depending on he
local labo ma ke in which people eside.
Ea lie esea ch on in e gene a ional mobili y o en measu es con ex ual mech-
anisms du ing an indi idual’s childhood o cap u e childhood exposu es ha a e
a gued o be he oo causes o mobili y. In his s udy, we map a ia ion by place
o adul esidence o be e unde s and how he labo ma ke aced as adul s may
impac on mobili y chances, as di e en places p o ide di e en economic oppo -
uni ies. Speci ically, access o jobs (Ro hs ein, 2019) and consequen ly mig a ion
o places whe e jobs a e a ailable (Buscha e al., 2021; Sp ung-Keyse e al., 2022)
ha e been associa ed wi h be e chances o upwa d mobili y. In Eu ope, di e en
egions ace d ama ically di e en g ow h p o iles, wi h some s uggling o uphold
and imp o e p ospe i y o hei esiden s (Dieme e al., 2022; Mo is and Oesch,
2025; Vigna, 2023). The link be ween social o igin and occupa ional des ina ions
may be s eng hened when labo ma ke p ospec s de e io a e, as young people
become mo e elian on pa en s’ social capi al o ind wo k (Mo is, 2023; Zwysen,
2016).
We d aw on a la ge, ha monized c oss-na ional da ase , pooling he i s en
ounds (2002–2020) o he Eu opean Social Su ey (ESS). Recen wo k on mobili y
dispa i ies wi hin coun ies has mos ly analyzed adminis a i e popula ion da a
(Accia i e al., 2022; B andén, 2019; Ca nei o e al., 2019; Che y e al., 2014;
Chua d and G assi, 2020; E iksen and Munk, 2020; Güell e al., 2018; Heid ich,
2017; Kenedi and Si ugue, 2023). Such wo k can p o ide a de ailed po ai o a
gi en coun y, e ealing how ou comes a y by egion, ci y, neighbo hood, o e en
block. Ye , he ac ha adminis a i e da a a e collec ed o di e en pu poses
han esea ch makes hem less compa able ac oss coun ies. The e has been limi ed
wo k b idging he li e a u e on wi hin-coun y geog aphic dispa i ies wi h he c oss-
coun y esea ch ha came be o e i . Ou use o an es ablished mul i-pu pose su ey
allows us o do jus ha . Mo eo e , we explo e a ange o po en ial de e minan s
o mobili y and make ou esul s a ailable o o he esea che s o build on.
Occupa ions can be measu ed in se e al ways. We ocus on occupa ional ank in
ou main esul s, and ollow up wi h esul s using social class. The ank app oach
assigns a sco e based on he ypical educa ion and income o an occupa ion, and
hen anks i wi hin each coun y’s dis ibu ion. Doing so le s us concei e o he
social s uc u e as a ladde , whe e each pe son has a p opo ion o he popula ion
abo e and below he . We u he dis inguish be ween absolu e upwa d and ela i e
mobili y. Upwa d mobili y e lec s how a a pe son bo n a he bo om can expec
o ise. Rela i e mobili y e e s o he co ela ion be ween pa en and child occupa-
ion. The wo a e dis inc . I a place unde goes occupa ional upg ading ela i e o
he coun y mean, he e will be ne upwa d mobili y. Howe e , i p i ileged child en
bene i as much o mo e om his upg ading, he ela i e p ospec s o child en o
di e en o igin may no imp o e. Since ela i e mobili y close e lec s inequali y
o oppo uni y, i is cen al o ou analysis.
We ind s iking a ia ion wi hin coun ies in bo h upwa d and ela i e mobili y,
2
and o bo h men and women. Abou wo hi ds o he a iance in upwa d ank
mobili y, and hal o he a iance in ela i e ank mobili y, occu s wi hin coun ies
a he han be ween hem. Capi al egions se e as hubs o upwa d mobili y; some-
imes, bu no always, a he expense o ela i e mobili y. In he highly mobile
No dic coun ies, he e appea s o be no such adeo , as capi als p omo e mobili y
o bo h kinds. Ou analysis also con ibu es o he li e a u e by explici ly inco po-
a ing women. While b oad mobili y pa e ns a e simila ac oss gende s, mobili y
le els di e in mos coun ies. No ably, in pa s o sou he n and cen al Eu ope,
women’s occupa ional a ainmen is shaped mo e s ongly by social o igin han ha
o men’s.
Finally, we p o ide an explo a o y analysis o po en ial mobili y co ela es, in-
cluding human capi al, labo ma ke composi ion, demog aphics and o he socio-
spa ial cha ac e is ics. Upwa d mobili y is g ea e in places wi h a high-skilled
wo k o ce, a la ge public sec o , and bu eauc a ic o ganiza ion. I is lowe whe e
unemploymen is high, o a high p opo ion wo ks in he p ima y sec o , manu ac-
u ing, o se ices. Rela i e immobili y is ha de o p edic , bu he e appea s o
be less mobili y whe e occupa ions di e mo e widely in pay. We he eby add o
he g owing li e a u e showing ha in addi ion o childhood mechanisms, he local
labo ma ke aced as adul ma e s o mobili y chances.
2 P e ious li e a u e
2.1 In e gene a ional occupa ional mobili y
Occupa ional mobili y has ypically been he pu iew o sociologis s (B een and Jon-
sson, 2005), while economis s ha e ad anced he s udy o income mobili y (Bjö k-
lund and Jän i, 2009). Howe e , hese bounda ies a e po ous: o o e come mea-
su emen e o , analyses o income mobili y o en impu e income on occupa ions
(Jácome e al., 2025; Kenedi and Si ugue, 2023), he eby e ec i ely es ima ing
occupa ional mobili y. Mo eo e , ecen wo k on income mobili y has a o ed ank-
based measu es (Che y e al., 2014; Engzell and Mood, 2023), which app oach
sociological no ions o ela i e s a us (Bloome e al., 2018; Hou , 2018).
Two adi ions exis in he ield o occupa ional mobili y esea ch. In he i s ,
occupa ions a e scaled in e ms o g ada ional ank, p o iding a me ic o occupa-
ional s a us (Ganzeboom e al., 1992). In he second app oach, occupa ions a e
g ouped in o social classes. These classes a e belie ed o e lec employmen ela-
ions and migh , in addi ion o cu en income le el, p o ide a be e ep esen a ion
o income s abili y, secu i y and esou ces, han g ada ional measu es (Bukodi and
Gold ho pe, 2018; Gold ho pe, 1982). To ensu e ha ou esul s a e obus , we
p o ide analyses speaking o bo h app oaches.
A undamen al dis inc ion is ha be ween absolu e and ela i e mobili y. Ab-
solu e mobili y ela es o he ac ual mo emen s be ween o igin and des ina ion. I
can be measu ed as he di e ence be ween pa en and child occupa ional ank, o
by whe he you social class is di e en om he one you g ew up in. Upwa d mo-
bili y om he bo om is o en o pa icula in e es , and we ollow his ocus in ou
s udy. Absolu e mobili y is o a la ge ex en d i en by a ia ions in he occupa ional
3
s uc u e, o example h ough changes in he labo ma ke ha c ea e new occu-
pa ions (o en highe anked) o eplace declining indus ies (B een, 2004; E ikson
and Gold ho pe, 1992). In a egional con ex , absolu e mobili y can be achie ed by
mo ing be ween places wi h di e en occupa ional dis ibu ions; o example, om
u al o indus ial egions o mo e knowledge-in ensi e u ban cen e s.
Rela i e mobili y (o social luidi y) e e s o mo emen s be ween o igin and
des ina ion ne o s uc u al changes, cap u ing he dependency s uc u e be ween
pa en and child. I can be measu ed using co ela ion coe icien s o con inuous
measu es o odds a ios o ca ego ical measu es. Rela i e mobili y e lec s social
inequali ies in he chances o eaching a gi en des ina ion and is he e o e consid-
e ed an indica o o he ela i e openness o socie ies (B een and Jonsson, 2005).
Whe eas absolu e mobili y accoun s o mos expe ienced mobili y, ela i e mobili y
is o in e es o he ques ion o equali y o oppo uni y. E en when upwa d mobili y
is high, ela i e immobili y may domina e i oppo uni ies a e disp opo iona ely
concen a ed among child en om p i ileged backg ounds. We assess bo h in ou
analysis below.
2.2 Coun y di e ences
Resea ch on in e gene a ional income mobili y inds highe mobili y in Eu ope
han in he US—sou he n Eu ope and he Uni ed Kingdom being he low-mobili y
excep ions—and he No dic coun ies gene ally anked as he mos mobile (Blan-
den, 2013; B a be g e al., 2017). This pa e n is weake o educa ion and occupa-
ion, hough he No dics emain among he mos mobile (B een and Jonsson, 2005;
S ömbe g and Engzell, 2025). These ends a e o en a ibu ed o comp ehensi e
wel a e s a es ha equalize oppo uni ies ea ly in li e (Co ak, 2013; E ikson and
Jonsson, 1996), al hough ecen esea ch also highligh s labo ma ke ins i u ions
(Lande sø and Heckman, 2017; Mogs ad e al., 2025). Analyses o income mobil-
i y end o con i m he in e se ela ionship be ween inequali y and mobili y, bo h
be ween and wi hin coun ies (Ha a i e al., 2021; Na ayan e al., 2018), whe eas
e idence o occupa ional mobili y is mo e mixed (DiP e e, 2020).
Inequali y is commonly measu ed by he Gini coe icien , o occasionally op
income sha es. Howe e , be ween-class inequali y, indica ing he esou ce dis ance
be ween classes, has been a gued o p o ide a be e measu e o inequali y in e-
la ion o occupa ional mobili y (He el and G oh-Sambe g, 2019). Be ween-class
inequali y be e e lec s s uc u al ba ie s o mobili y, as i cap u es he economic
dis ance be ween social g oups, whe eas he Gini coe icien does no dis inguish
whe he inequali y s ems om di e ences wi hin o be ween classes. We ollow his
a gumen in ou analysis below, by measu ing be ween-class inequali y in income.
The cen al mo i a ion behind much occupa ional mobili y esea ch is he in-
dus ializa ion hesis (Lipse and Ze e be g, 1959; T eiman, 1970), which posi s
ha he inc easing complexi y o mode n p oduc ion will place g ea e emphasis
on o mal c eden ials. As educa ion expands and becomes mo e accessible, social
o igin should become less impo an in de e mining indi idual a ainmen . As a
esul , bo h absolu e and ela i e mobili y should inc ease wi h economic de el-
opmen . The indus ializa ion hesis has been ques ioned and, in many s udies,
4
e u ed (e.g., B een, 1997, 2004; Bukodi e al., 2020; E ikson and Gold ho pe, 1992;
Knigge e al., 2014a) al hough some e idence also suppo s i (e.g., Aydemi and
Yazici, 2019; Ganzeboom e al., 1989; He ala, 2023; Knigge e al., 2014b; Yaish
and Ande sen, 2012).
While e idence on he link be ween occupa ional mobili y and mac o-le el eco-
nomic indica o s is mixed, one aspec o he indus ializa ion hesis appea s o hold:
educa ional expansion is a key d i e o bo h absolu e and ela i e mobili y (B een
and Mülle , 2020). Because educa ion media es he ela ionship be ween social o i-
gins and des ina ions, b oade access can help le el he playing ield. No ably, he
link be ween social o igins and des ina ions weakens a highe educa ion le els, sug-
ges ing ha high-skilled labo ma ke s unc ion mo e me i oc a ically (B een and
Jonsson, 2007; Hou , 1988; Ka lson, 2019).
2.3 Regional a ia ion
Recen wo k on subna ional di e ences in mobili y— he “geog aphy o
oppo uni y”—has b oadened he sea ch o explana ions beyond na ional-le el ac-
o s. Mos o his esea ch s udies income mobili y, spu ed by inc eased a ailabili y
o adminis a i e ax da a (Connelly e al., 2016; G usky e al., 2019; Song and Cole-
man, 2020). In luen ial wo k by Che y e al. (2014) shows s iking dispa i ies in
oppo uni y be ween places in he US. These e o s ha e since been ex ended o
Eu opean coun ies, including Denma k (E iksen and Munk, 2020), F ance (Kenedi
and Si ugue, 2023), I aly (Accia i e al., 2022; Güell e al., 2018), he Ne he lands
(Beeke s, 2024), No way (Ca nei o e al., 2019), Sweden (B andén, 2019; Heid ich,
2017), and Swi ze land (Chua d and G assi, 2020).
Wha hen, a e he local ac o s ha p omo e o hinde mobili y? In line wi h
coun y-le el s udies, egional esea ch highligh s educa ion as a key d i e o mobil-
i y (Biasi, 2023; Co ak, 2020; E iksen and Munk, 2020). The a ailabili y and quali y
o schooling a e c ucial in childhood (Che y e al., 2014; S adelmann-S e en, 2012),
bu educa ion also shapes labo ma ke s a bo h indi idual and collec i e le els.
Highe a e age educa ion can c ea e social ex e nali ies, in luencing he demand o
skilled labo and os e ing mo e dynamic economies (Ce meño, 2019; Liu, 2015).
Addi ionally, high-skilled labo ma ke s may p omo e mobili y h ough knowledge
spillo e s and agglome a ion e ec s, whe e wo ke s gain skills h ough p o essional
in e ac ions, and indus y clus e s gene a e b oade ca ee p ospec s.
Labo ma ke s also ma e in hei own igh : local unemploymen a es a e neg-
a i ely associa ed wi h upwa d mobili y (Accia i e al., 2022; E iksen and Munk,
2020; Kenedi and Si ugue, 2023). Economic luc ua ions in indus ies like manu ac-
u ing, mining, and pe oleum a e known o impac mobili y (Be ge and Engzell,
2022; Bü iko e e al., 2025; Deu sche and Mazumde , 2020). Sec o al compo-
si ion likewise p edic s mobili y, wi h hea y indus y and manu ac u ing showing
co ela ions anging om posi i e o nega i e, depending on he g ow h o hose
indus ies (Buscha e al., 2021; Heid ich, 2017). In Sweden and Canada, lowe mo-
bili y is ins ead ound in places wi h a s agnan p ima y indus y (Co ak, 2020;
Heid ich, 2017). His o ical esea ch simila ly highligh s local indus ial change as
d i ing mobili y, wi h he apid ansi ion om a ming o manu ac u ing being a
5
wa e shed (Be ge e al., 2023; Pa e son e al., 2025).
Ano he se o po en ial de e minan s is demog aphic. Social disad an age o en
clus e s in places whe e immig an s li e, ye hei child en’s mobili y o en su passes
ha o he gene al popula ion (Bo gen e al., 2025; Bous an e al., 2025; He mansen,
2016). Regions wi h la ge sha es o immig an s (Co ak, 2020; E iksen and Munk,
2020) o indigenous popula ions (Connolly e al., 2019; Deu sche and Mazumde ,
2020) a e associa ed wi h lowe le els o mobili y, as a e egions wi h a la ge sha e
o adul s wi hou a high school diploma (Co ak, 2020). Mo eo e , he incidence o
single pa en s nega i ely p edic s mobili y in se e al coun ies (Accia i e al., 2022;
Conno e al., 2023; Co ak, 2020; E iksen and Munk, 2020; Pe e son e al., 2023).
High inequali y can limi mobili y by es ic ing access o quali y educa ion, p o-
essional ne wo ks, and well-paying jobs o hose om disad an aged backg ounds
(B andén, 2019; Co ak, 2013). In he US, highe le els o esiden ial seg ega ion,
bo h by income and ace, along wi h g ea e income inequali y, a e associa ed wi h
lowe mobili y le els (Che y e al., 2014). Beyond economic dispa i ies, social in-
s i u ions also play a c ucial ole in shaping mobili y ou comes. Social capi al— he
s eng h o ne wo ks and communi y ies—can acili a e mobili y by p o iding sup-
po , job e e als, and men o ship (Che y e al., 2022; Coleman, 1988). Likewise,
social us and b oade ci ic capi al, including ins i u ional quali y and commu-
ni y engagemen , can os e en i onmen s whe e upwa d mobili y is mo e a ainable
(Accia i e al., 2022; Pe e son e al., 2023).
U baniza ion p esen s a mixed pic u e: while ci ies o en o e g ea e economic
oppo uni ies and access o di e se ne wo ks, hey can also exace ba e seg ega-
ion and inequali y, limi ing mobili y o lowe -income g oups (B een and In, 2023;
Conno and S o pe , 2020; F iedman and Macmillan, 2017). Resea ch om he
US documen s a u al ad an age in upwa d mobili y, whe e young men in pa icu-
la bene i om s onge household and communi y suppo s, while women emain
cons ained by adi ional gende no ms (Conno e al., 2023). Finally, exposu e o
iolen c ime may hampe mobili y by dis up ing educa ion, weakening social co-
hesion, and educing economic in es men , al hough e idence o his link is mo e
consis en in he US han in Eu ope (Accia i e al., 2022; Kenedi and Si ugue, 2023;
Manduca and Sampson, 2019; Sha key and To a s-Espinosa, 2017).
3 Theo e ical expec a ions
In his sec ion, we ou line ou expec a ions ega ding he in luence o a ious mea-
su ed p edic o s on mobili y. The e is a di e ence be ween absolu e upwa d mo-
bili y, measu ing he a e a which child en om poo e amilies ise abo e hei
o igins, and ela i e mobili y, measu ed as ank co ela ions o odds a ios and cap-
u ing he o e all simila i y in s a us be ween pa en s and child en. Gi en ha
mos esea ch on egional mobili y dispa i ies ocuses on upwa d mobili y, we o -
mula e ou hypo heses wi h his in mind while no ing ha hese expec a ions may
o may no ex end o ela i e mobili y.
We expec ha highe le els o human capi al in a egion will co ela e wi h
highe upwa d mobili y. Educa ion and aining expand he kinds o wo k a ail-
6
able, which can help indi iduals imp o e hei social s anding. Fu he mo e, access
o be e educa ion esou ces, as indica ed in ou analysis by he numbe o each-
e s pe capi a, should be bene icial o upwa d mobili y. Con e sely, in places
whe e mos people ha e limi ed schooling, oppo uni ies o his kind a e likely o
be sca ce . These pa e ns may e lec bo h indi idual-le el ad an ages o an edu-
ca ion and b oade local economic condi ions. Gi en he explo a o y na u e o ou
analysis, we do no a emp o disen angle hese explana ions.
We expec ha high unemploymen a es will be nega i ely associa ed wi h
upwa d mobili y. Mo eo e , egions wi h a highe p opo ion o employmen in
low-skill sec o s such as ag icul u e o manual labo (e.g., manu ac u ing) a e likely
o expe ience lowe upwa d mobili y, as hese indus ies end o o e ewe ca-
ee oppo uni ies. Con e sely, egions wi h a highe p opo ion o employmen in
p o essional, echnical, and se ice occupa ions should exhibi highe upwa d mo-
bili y. Fo upwa d mobili y, hese co ela ions a e o some ex en mechanical: mo e
high-s a us jobs simply c ea e mo e oom a he op. I emains an open ques ion
whe he hese pa e ns ex end o ela i e mobili y, which depends on who ac ually
gains access o hose oppo uni ies. We also measu e he con ibu ion o he public
sec o and bu eauc a ic oles o employmen , which we expec o ela e posi i ely o
mobili y. Classically, Webe (1978) a gued ha bu eauc acy could acili a e social
mobili y by p o iding clea ules and p ocedu es o ad ancemen .
The ela ionship be ween immig a ion and mobili y is ambiguous. Regions wi h
la ge immig an popula ions may expe ience lowe mobili y o e all, due o po en-
ial social and economic ma ginaliza ion. This is wha p e ious li e a u e, ocusing
on childhood condi ions, has ended o ind (Co ak, 2020; E iksen and Munk, 2020).
Howe e , we measu e loca ion in adul hood, which complica es his expec a ion.
Immig an s o en seek upwa d mobili y o hei child en and clus e in u ban cen-
e s whe e oppo uni ies a e mo e abundan . The e o e, immig an p esence may
co ela e posi i ely o nega i ely wi h mobili y.
We expec highe le els o single pa en hood and la ge household sizes o co -
ela e wi h lowe upwa d mobili y. Single-pa en households o en ace economic
ha dships ha limi hei child en’s oppo uni ies. Simila ly, la ge amilies a e as-
socia ed wi h less in es men pe child (Gibbs e al., 2016), and ypically also a
socie y whe e women shoulde a la ge pa o he household bu den (Men a and
Lepin eu , 2021). We may expec hese amily ac o s o be especially impo an
o women, al hough esea ch also inds ha young men a e mo e ulne able o
dis up ed amily s uc u es (Conno e al., 2023). Mo eo e , we in es iga e he ex-
en o h ee-gene a ion households and he median age o he popula ion. Th ee-
gene a ion households may os e s onge amily bonds bu could also inc ease
occupa ional simila i y be ween pa en s and child en. Addi ionally, aging egions
may indica e economic s agna ion, u he hinde ing mobili y.
Highe le els o inequali y wi hin a egion, as measu ed by income di e ences
be ween social classes (He el and G oh-Sambe g, 2019), a e expec ed o co ela e
nega i ely wi h mobili y ou comes. G ea e inequali y no only aises he incen-
i es o hose a he op o main ain hei posi ion bu also ein o ces social ba ie s
h ough di e ences in esou ces. In con as , s ong social capi al and high le els
o social us a e expec ed o be posi i ely associa ed wi h upwa d mobili y. So-
7
cial ne wo ks and us ha e he po en ial o dispe se oppo uni ies mo e widely
h ough job e e als, men o ship, and social suppo . Simila ly, we expec a eas
wi h high ci ic engagemen (e.g., poli ical pa icipa ion) o os e mo e inclusi e
en i onmen s, whe e upwa d mobili y is mo e a ainable.
U baniza ion may ha e a mixed e ec on mobili y. While ci ies o en o e
g ea e access o di e se job ma ke s and educa ional oppo uni ies, hey can also
ampli y inequali ies and social seg ega ion, which may limi mobili y o some
g oups. We expec u baniza ion o be associa ed wi h highe upwa d mobili y,
bu no necessa ily wi h g ea e ela i e mobili y, due o he pe sis ence o u ban
po e y and inequali y. Addi ionally, highe le els o iolen c ime a e expec ed o
co espond wi h lowe upwa d mobili y, as c ime can unde mine social cohesion,
weaken us , and discou age in es men in educa ion and local communi ies.
4 Measu ing in e gene a ional mobili y
We measu e in e gene a ional mobili y in e ms o occupa ional ank and social
class. Fo each, we dis inguish be ween absolu e upwa d and ela i e mobili y, and
be ween women and men. To measu e ank mobili y, we i s c ea e pe cen ile sco es
in he na ional dis ibu ion o pa en and child occupa ional s a us, desc ibed in
g ea e de ail below. We hen es ima e he ollowing linea model:
Ri =αj+βjRi −1+εi .(1)
whe e Riis he esponden ’s ank, Ri −1is he pa en al ank, αjand βja e
pa ame e s ha a y a he le el o local a eas and εi is a esidual e o e m. Bo h
αjand βja e o in e es as hey cap u e, espec i ely, upwa d and ela i e mobili y.
In he ollowing, we will call αjupwa d ank mobili y and βj ank immobili y.
Upwa d ank mobili y αjis he expec ed ank o esponden s whose pa en s
a e a he bo om o he dis ibu ion. Highe alues indica e ha child en bo n
in he bo om can expec o a ain a highe ank, and hus a highe alue means
mo e mobili y. Rank immobili y βj, also known as he ank- ank slope, cap u es
he s eng h o he associa ion be ween pa en and child s a us. A slope o 0.20 in-
dica es ha pa en s on a e age pass on 20% o hei (dis)ad an age in occupa ional
ank. The s onge βjis, he lowe he in e gene a ional mobili y, as i means ha
indi iduals can expec o end up on a ung close o whe e hey we e bo n.
Wi hin a coun y as a whole, o in he ull sample, he ela ionship be ween he
in e cep and slope is de e minis ic, since he expec ed alue a he median o pa en
ank mus be app oxima ely 50. This cons ain does no hold a he subna ional
( egional) le el, whe e he e may be meaning ul a ia ion in he dis ibu ion o
na ional occupa ional anks.
To ob ain eliable es ima es a a local le el, we es ima e Equa ion 1 using mul-
ile el, o mixed, models (Snijde s and Boske , 2011). In p inciple, i is possible o
ob ain es ima es o ou pa ame e s by unning sepa a e eg essions o each egion.
The p oblem is ha his will yield ex eme es ima es o some egions due o sam-
pling a ia ion (Heid ich, 2017). The mul ile el solu ion s a s om he assump ion
ha he ela ionship o in e es consis s o a “ ixed” componen , common ac oss all
8
(a) Upwa d ank mobili y, men (b) Upwa d ank mobili y, women
Figu e 2: Geog aphic a ia ion in upwa d ank mobili y ( ank- ank in e cep ).
Highe alues indica e mo e mobili y. Resul s o class mobili y in Appendix Figu e
A8.
p o iding an al e na i e ep esen a ion o he in e gene a ional associa ion.
6.1 The geog aphy o in e gene a ional mobili y
Nex , we u n o egional a ia ion. Because he geog aphical pa e ns a e simila
o occupa ional ank and social class mobili y, we concen a e on ank mobili y,
sepa a ing be ween upwa d mobili y and ela i e immobili y. Simila esul s o
social class a e shown in Figu es A8–A9 o he Appendix.
Figu e 2 plo s he a ia ion in upwa d mobili y o men and women, measu ed
as he in e cep (upwa d ank mobili y). Highe alues indica e mo e upwa d mo-
bili y. Fo men, he highes le els o upwa d ank mobili y a e ound in some o
he No dic coun ies, wi h a es o a ound 46. Fo women, he No dic coun ies
a e he mos upwa dly mobile, ollowed closely by pa s o I eland and some o he
Bal ic coun ies, wi h a es o a ound 43. The lowes a es o men a e ound in
I eland, a ound 28, and o women in sou he n Spain, a ound 20. These pa e ns
ollow known coun y di e ences in mobili y (Bukodi e al., 2020). Ye , a ia ion
is mos ly con inuous, a he han clea ly delinea ed by na ional bo de s.
Upwa d ank mobili y is sys ema ically mo e common in capi al egions, espe-
cially so in he No dic coun ies, whe e Oslo, S ockholm, and Helsinki consis en ly
ank among he en mos mobile egions in Eu ope. Be lin and Pa is also s and
ou om hei su ounding egions, and ank among he 20 mos mobile egions.
This ela ionship is likely explained in pa by mig a ion om p o inces o capi al
egions, some hing we a e unable o es wi h ou da a. In se e al o he coun ies,
he capi al is isibly mo e mobile han he p o inces, bu does no en e among
15
(a) Rela i e ank immobili y, men (b) Rela i e ank immobili y, women
Figu e 3: Geog aphic a ia ion in ela i e ank immobili y ( ank- ank slope).
Highe alues indica e less mobili y. Resul s o class mobili y in Appendix Fig-
u e A9.
op anking egions o e all, by i ue o he gene al le el o mobili y in he coun y
being lowe .
In Figu e 3, we u n o ela i e ank immobili y, measu ed as he slope o he
ank- ank associa ion. In con as o upwa d ank mobili y, highe alues he e
indica e less mobili y. Rank immobili y anges om a low o 0.23 o men in pa s
o he Bal ic s a es and he Ne he lands, wi h simila alues o women in he
Ne he lands and Scandina ia. The highes ank immobili y, a ound 0.45, is ound
in pa s o Poland o men, and o women in Poland, C oa ia, Hunga y, I aly and
sou he n Spain. The ma kedly highe ela i e immobili y among women in much
o sou he n and cen al Eu ope is he mos s iking gende di e ence in ou da a.
How do hese numbe s compa e o social class mobili y? Figu e A8 in he Ap-
pendix plo s he a ia ion in upwa d class mobili y, measu ed as he p opo ion
upwa dly mobile om he wo king class. Highe alues indica e mo e upwa d mo-
bili y. The pa e n e lec s he esul s shown in ank mobili y, wi h highe le els
o upwa d class mobili y in he Ne he lands and he No dic coun ies (a ound 60%
o bo h men and women). Addi ionally, he London egion shows high le els o
upwa d class mobili y o men (59%), bu no as high o women (50%). The lowes
le els o men a e ound in pa s o Bulga ia and Hunga y (12%) and o women
in pa s o Bulga ia and Po ugal (a ound 20%). Simila o ank mobili y, upwa d
class mobili y is mo e common in capi al egions, and especially so in he No dic
capi als.
Figu e A9 in he Appendix plo s class immobili y, measu ed as odds a ios,
whe e highe alues indica e mo e immobili y. The lowes le els o class immobili y
a e ound mainly in he No dic coun ies, as well as Es onia and he Ne he lands,
16
wi h odds a ios a ound 2. The highes le els o immobili y a e ound in Spain,
Hunga y and Poland (odds a io a ound 6–7). The pa e ns o class immobili y a e
simila o hose o ank immobili y, bu wi h he excep ion o No he n Po ugal
being an ou lie in class immobili y o women, wi h odds a ios as high as 8.
In mapping he landscape o ela i e ank and class mobili y, capi als no longe
s and ou as hubs o mobili y. In ac , in se e al examples such as Pa is, Mad id,
and Lisbon, inequali y o oppo uni y is app eciably s onge in he capi al han
in he su ounding egions. He ein lies a po en ial dilemma o policy-make s: how
can ci ies unc ion as engines o oppo uni y wi hou he a endan inequali ies in
who seizes hose oppo uni ies? In e es ingly, his in e se ela ionship is no e iden
among No he n Eu opean capi als, which manage o achie e high upwa d mobili y
wi hou sac i icing on ela i e mobili y.
How does his wi hin-coun y a ia ion compa e o coun y di e ences ha ha e
been he ocus o much p e ious esea ch? Appendix Table A7 p esen s in a-
class co ela ions e lec ing how simila ou comes a e o di e en egions wi hin a
coun y, ha is, he p opo ion o a iance cap u ed by he be ween-coun y le el.
Fo ank mobili y, his igu e a ies om abou a hi d o upwa d ank mobili y
(men 0.29, women 0.38) o close o hal o ela i e ank immobili y (men 0.37,
women 0.54). This indica es ha as much as hal o a ia ion in ela i e mobili y,
and wo hi ds o a ia ion in absolu e mobili y, occu s be ween egions wi hin
coun ies. Class mobili y a ies mo e sys ema ically be ween coun ies, bu e en
he e, be ween a hi d and wo i hs o a ia ion is wi hin coun y.
6.2 Co ela es o in e gene a ional mobili y
Wha explains local a ia ion in in e gene a ional mobili y? To answe his ques-
ion, we es ima e bi a ia e a ea-le el co ela ions be ween each mobili y me ic
and he comp ehensi e se o co a ia es desc ibed abo e. We es ima e wo se s o
models, i s using all geog aphic a ia ion and hen pu ging all be ween-coun y
a ia ion om explana o y and ou come a iables, o cap u e wi hin-coun y co -
ela ions only. He e, oo, we concen a e on ank mobili y, while showing simila
esul s o social class in Figu e A12 o he Appendix. We s ess ha hese asso-
cia ions a e ecological and sugges i e, and we make no a emp a causal o e en
mul i a ia e modeling. Ou aim is o p o ide en a i e e idence o u u e esea ch
o explo e in g ea e dep h.
Figu e 4, op panels, shows co ela es o upwa d ank mobili y ( he ank- ank
in e cep ) o men and women, espec i ely. Upwa d mobili y is signi ican ly co e-
la ed wi h mos o he measu es in ou da a and se e al human capi al indica o s
show a subs an ial posi i e co ela ion wi h upwa d mobili y. Mos o hese do no
diminish in size o signi icance once we make use o he wi hin-coun y a ia ion
only, hus con i ming ou hypo hesis o a link be ween high le els o human capi al
and upwa d mobili y. In places wi h a la ge p opo ion o high-school d opou s,
le els o upwa d mobili y a e lowe .
Labo ma ke composi ion is ano he signi ican p edic o ha , ollowing ou
expec a ions, shows lowe le els o upwa d mobili y in places wi h a la ge sha e o
unemployed, and la ge p ima y and manu ac u ing sec o s. Highe upwa d mobili y
17
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-1 -.5 0 .5 1
Co ela ion To al Wi hin-coun y
(a) Upwa d ank mobili y, men
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-1 -.5 0 .5 1
Co ela ion To al Wi hin-coun y
(b) Upwa d ank mobili y, women
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-.5 -.25 0 .25 .5
Co ela ion To al Wi hin-coun y
(c) Rela i e ank immobili y, men
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-.5 -.25 0 .25 .5
Co ela ion To al Wi hin-coun y
(d) Rela i e ank immobili y, women
Figu e 4: Co ela es o in e gene a ional mobili y. Co ela ion coe icien and 95%
con idence in e al, s anda d e o s clus e ed a he coun y le el. Resul s o class
mobili y in Appendix Figu e A12.
18
is ound in places wi h la ge public sec o s and mo e bu eauc a ic o ganiza ions. Fo
men, he co ela ion wi h public-sec o employmen is only signi ican be ween and
no wi hin coun ies, whe eas o women i is signi ican ly co ela ed bo h be ween
and wi hin coun ies al hough subs an i ely small. The size o he se ice sec o
shows an inconsis en pa e n: posi i ely co ela ed wi h upwa d mobili y o men,
bu only be ween coun ies, whe eas i is nega i ely co ela ed wi hin coun ies o
women.
Tu ning o demog aphic co a ia es, upwa d mobili y is posi i ely co ela ed wi h
immig a ion and nega i ely co ela ed wi h median age; he la e only wi hin coun-
ies, which migh be an indica ion o s agna ing egional de elopmen . In e es -
ingly, he sha e o single mo he s is posi i ely co ela ed wi h women’s upwa d
mobili y bo h wi hin and ac oss coun ies, a pa e n no obse ed o men. One
possible in e p e a ion is ha egions suppo ing single mo he s may also p o ide
mo e inclusi e ins i u ions o se ices ha acili a e women’s mobili y, echoing ea -
lie esul s ha he dominance o wo-pa en households bene i he mobili y o men
o e women (Conno e al., 2023).
Nex , we u n o indica o s o inequali y, u baniza ion, social capi al, and us .
We ha e al eady no ed ha upwa d mobili y is highe in capi al egions, and his is
e iden in a posi i e co ela ion be ween u baniza ion and upwa d mobili y, which
aligns wi h ou expec a ions. Somewha su p isingly, iolen c ime is also posi-
i ely co ela ed wi h upwa d mobili y, mos likely explained by c ime being mo e
p e alen in u ban a eas. Be ween-class income inequali y is un ela ed o upwa d
mobili y o ei he men o women. Las ly, and ollowing ou p edic ions, social cap-
i al, ci ic capi al, and social us a e posi i ely co ela ed wi h upwa d mobili y,
bu mos ly o men and be ween coun ies. We cau ion he eade ha he amoun
o wi hin-coun y a ia ion in hese a iables is low (Appendix Table A7).
In sum, upwa d mobili y is associa ed wi h a ange o human capi al, labo ma -
ke , demog aphic, and socio-spa ial co a ia es. This pic u e changes once we u n
o he ank- ank slope, measu ing ank immobili y o he s eng h o he associa ion
be ween pa en and child occupa ion (Figu e 4, bo om panels). Al hough some o
he abo e co a ia es explain a ia ion be ween coun ies, hese associa ions u n
non-signi ican when es ic ing a ia ion o ha wi hin coun ies. One no able ex-
cep ion, howe e , is be ween-class income inequali y which is nega i ely associa ed
wi h ela i e mobili y wi hin coun ies o bo h men and women. Thus ou esul s
con i m p e ious esea ch showing ha he in e se ela ionship be ween inequali y
and ( ela i e) mobili y holds wi hin and no jus be ween coun ies (Che y e al.,
2014; Güell e al., 2018; Ha a i e al., 2021).
While human capi al a iables show no clea associa ion wi h ela i e immobil-
i y among men, hey e eal a su p ising pa e n o women. As expec ed, ac oss
coun ies, highe educa ional a ainmen is linked o g ea e equali y o oppo u-
ni y. Howe e , wi hin coun ies, egions wi h highe educa ion le els exhibi g ea e
in e gene a ional pe sis ence o women, a pa e n ha challenges he con en ional
iew o educa ion as an equalize (B een and Jonsson, 2007; Hou , 1988). I is
mo e consis en , howe e , wi h obse a ions om gende schola ship sugges ing
ha women’s en y in o adi ionally male-domina ed domains like highe educa-
ion and p o essional occupa ions has been une en, disp opo iona ely bene i ing
19
hose om p i ileged backg ounds (England, 2010). In con as , men’s ela i e mo-
bili y appea s less in luenced by egional educa ional composi ion, sugges ing ha
hei in e gene a ional ep oduc ion in he labo ma ke depends less on educa ion.
6.3 Compa ison wi h p e ious wo k
How do ou esul s compa e wi h p e ious wo k on egional mobili y a ia ion wi hin
coun ies? A ela ed s udy is Be häuse e al. (2021), which examines educa ional
mobili y using he same da ase and egions. They es ima e in e gene a ional isk
a ios o e ia y educa ion, mos compa able o he odds a ios o class immobili y
in ou analysis. Some wi hin-coun y pa e ns align ac oss s udies; o example,
lowe mobili y in wes e n Ge many and no he n Po ugal appea s in bo h educa ion
and occupa ion. In con as , coun ies like F ance and Poland show di e gen o
e en e e sed pa e ns. The e a e also no able c oss-coun y di e ences. B i ain
and Ge many, o ins ance, a e high-mobili y cases in Be häuse e al. (2021),
bu appea mo e middling o low-mobili y in ou esul s. A likely explana ion
is ha e ia y a ainmen cap u es limi ed a ia ion, o e looking di e ences in
ield o s udy, ins i u ional p es ige, o p og am du a ion, all o which ma e o
occupa ional ou comes (Ge be and Cheung, 2008; Reime and Thomsen, 2019).
Se e al s udies ha e also es ima ed egional a ia ion in income mobili y wi hin
one coun y (e.g., Accia i e al., 2022; B andén, 2019; Bü iko e e al., 2025; Güell
e al., 2018; Heid ich, 2017; Kenedi and Si ugue, 2023) o , in some cases, occupa-
ional mobili y (Bell e al., 2023; B een and In, 2024a,b; Buscha e al., 2021). These
s udies ypically examine a ia ion by egion o o igin a he han esidence, and
ypically agg ega e esul s o men and women o epo only on men.
Ou indings o capi al egions ecei e suppo in some, bu no all, ea lie wo k.
P e ious s udies ind ha capi al egions can be si es o high absolu e (Bell e al.,
2023; Kenedi and Si ugue, 2023) and low ela i e mobili y (B andén, 2019; Heid ich,
2017), bu he pa e ns by coun y a e no always he same as ou s. Some o
hese di e ences may be d i en by he dis inc ion be ween occupa ion and income.
Capi al egions end o be economically dynamic, o e ing di e se labo ma ke s
and oppo uni ies ha can help indi iduals mo e beyond hei geog aphic o class
o igins. A he same ime, hey o en ha e a highly s a i ied uppe ie , which may
be be e cap u ed by de ailed income da a han by occupa ional classi ica ions.
Mo e esea ch is needed on unde wha condi ions, and o whom, capi als se e as
engines o mobili y (B een and In, 2023; F iedman and Macmillan, 2017).
In se e al cases, we ind ha ea lie esul s in he li e a u e eplica e mainly wi h
espec o men and upwa d mobili y. This applies o p e iously known no h–sou h
g adien s in England (high mobili y in he sou h, Bell e al., 2023) and I aly (low
mobili y in he sou h, Accia i e al., 2022; Güell e al., 2018), bu also o he lowe
mobili y in eas e n Ge many (Dodin e al., 2024) o highe mobili y in No wegian
oil egions (Bü iko e e al., 2025). In each o hese cases, we ep oduce pa e ns in
upwa d mobili y o men, bu no in ela i e mobili y o o women. Fu u e wo k on
egional mobili y di e ences would he e o e be well-ad ised o pay close a en ion
o ela i e mobili y and gende di e ences. Ou inding o highe ela i e mobili y
in eas Ge many is, howe e , consis en wi h p e ious esea ch using social class
20
o p edic child en’s educa ion o occupa ion (Be häuse , 2019; Pollak and Mülle ,
2002), showing ha occupa ional and income mobili y a e di e en concep s ha
need no ollow he same pa e ns (B een e al., 2016).
7 Discussion and conclusion
In his pape , we ha e examined local a ia ion in in e gene a ional mobili y by
using he Eu opean Social Su ey. We show ha in e gene a ional mobili y a ies
a leas as much wi hin coun ies as be ween hem. Capi al egions a e hubs o
upwa d mobili y, bu no gene ally o ela i e mobili y. While upwa d mobili y
is ela ed o a ange o cha ac e is ics o local labo ma ke s, including indus ial
composi ion and human capi al, ela i e mobili y u ns ou o be mo e idiosync a ic
and ha de o p edic . Howe e , ela i e mobili y is ela ed o class di e ences in
income, such ha high le els o be ween-class segmen a ion make i ha de o climb
he social ladde . Pa e ns also di e no ably o women and men, wi h be ween-
coun y di e ences gene ally being mo e ma ked o women’s oppo uni ies.
Much p e ious esea ch has ocused on he mechanisms o mobili y in childhood.
This pape adds o moun ing e idence showing ha adul esidence and labo ma -
ke condi ions also play a ole. Ou indings, along wi h his wide li e a u e, sugges
impo an implica ions o social policy. To imp o e mobili y, i is essen ial o ocus
no only on ensu ing equal oppo uni ies o child en bu also on s eng hening labo
ma ke s ac oss Eu ope. Howe e , u he esea ch is needed o disen angle he ole
o childhood and adul esidence, especially aking sel -selec ion in o mig a ion in o
accoun . This may also help explain why mobili y ou comes a y mo e wi hin coun-
ies o men, as couples a e mo e likely o eloca e o he man’s ca ee (B andén
and Haand ikman, 2019; Jayachand an e al., 2024), making selec i e mig a ion a
po en ially mo e in luen ial ac o he e.
A he same ime, women’s mobili y shows la ge di e ences be ween coun ies,
pa icula ly in ela i e immobili y. While we canno conclusi ely de e mine he
easons o his, i may be ela ed o di e ences in wo k and gende no ms, as
well as he a ailabili y o ins i u ional suppo o women’s ca ee s. In pa icula ,
ela i e immobili y o women is ex eme in pa s o sou he n and cen al Eu ope.
Wi h mo e adi ional gende oles, women ace s onge social expec a ions o s ay
close o home, ake on ca egi ing esponsibili ies, o p io i ize amily o e ca ee .
In such con ex s, women may also encoun e mo e segmen ed labo ma ke s wi h
limi ed access o jobs, pa icula ly in male-domina ed sphe es. This explains why
women’s a ainmen would be lowe in gene al, bu no he s ong dependence
on occupa ional o igins. Fu he esea ch should aim o be e unde s and hese
gende ed mobili y pa e ns and hei unde lying ac o s.
To unde s and local a ia ion in in e gene a ional mobili y, we explo e how a ea-
le el mobili y me ics co ela e wi h a ange o egional cha ac e is ics. By exam-
ining bo h be ween- and wi hin-coun y a ia ion, we ind ha upwa d mobili y is
linked o indica o s o human capi al, such as low d opou a es and high educa ional
a ainmen . These pa e ns pe sis e en when ocusing on wi hin-coun y a ia ion.
Labo ma ke composi ion also plays a ole: a eas wi h mo e bu eauc a ic oles and
21
( o women) la ge public sec o s end o ha e highe upwa d mobili y, while egions
domina ed by p ima y and manu ac u ing sec o s show lowe mobili y. Addi ionally,
demog aphic ac o s such as younge popula ions and highe le els o immig a ion
a e associa ed wi h mo e upwa d mobili y. We ha e no a emp ed o disen angle
whe he hese associa ions e lec he cha ac e is ics o indi iduals who a e mo e o
less mobile, o he cha ac e is ics o places hemsel es. Fo ins ance, highe edu-
ca ion may bene i indi idual mobili y bu could also in luence local labo ma ke s
h ough inc eased p oduc i i y and demand o skilled labo .
When shi ing ocus o ela i e immobili y, he associa ions weaken wi hin coun-
ies, excep o a clea nega i e ela ionship wi h be ween-class income inequali y.
This suppo s he idea, well documen ed in p io esea ch, ha g ea e inequal-
i y ends o educe mobili y, e en a he egional le el. Compa isons wi h ea lie
s udies show ha some pa e ns eplica e, especially o men and upwa d mobili y,
such as no h–sou h di ides in he UK and I aly o highe mobili y in No way’s
oil- ich a eas. Howe e , pa e ns a e di e en o ela i e mobili y and he mobili y
o women. De ia ions om p e ious s udies may also esul om he a ying ocus
on occupa ion and income, illus a ing he need o s udy hese wo dimensions no
only sepa a ely bu also in combina ion.
Fu u e esea ch should also aim o be e unde s and he in e play be ween
na ional ins i u ions and egional labo ma ke s. While na ional ac o s may no
di ec ly explain egional di e ences in a s a is ical sense, his does no mean ha
na ional policies do no in luence egional ou comes. Jus as decen alized school
policy can lead o la ge a ia ions in educa ional quali y and equali y ac oss egions
(Biasi, 2023; S adelmann-S e en, 2012), na ional indus ial policies o labo ma -
ke in e en ions can c ea e signi ican egional di e ences in oppo uni ies. These
kinds o na ional policies may no lend hemsel es o s a is ical modeling o wi hin-
coun y a ia ion, bu hey can s ill ha e an impac on egional di e ences.
Ou con ibu ion b idges ecen esea ch based on adminis a i e da a ha ex-
plo es subna ional a ia ion in de ail, wi h ea lie wo k compa ing coun ies using
la ge-scale ep esen a i e su ey da a. Adminis a i e da a a e collec ed by na-
ional s a is ical agencies wi h a ying egula ions, p o ocols, co e age, ime spans,
and quali y. As a esul , ecen esea ch on egional a ia ion o en uses “islands
o da a,” which capi alize on he speci ics o a gi en da ase o p o ide a de ailed
iew o one coun y, bu wi hou he compa a i e elemen . Su ey esea ch, on he
o he hand, has un il ecen ly lacked su icien sample sizes o examine de ailed sub-
na ional a ia ion, bu he ma u a ion o long- unning su eys like he Eu opean
Social Su ey has made his inc easingly possible. Ou wo k o e s a empla e o
how esea che s can combine he s eng hs o each app oach. To acili a e u he
wo k, we make ou code and es ima es a ailable o esea che s wishing o explo e
in e gene a ional mobili y and i s co ela es.
Acknowledgemen s Funding was gene ously p o ided by he Eu opean Re-
sea ch Council g an no. 101165962 (MaMo). We hank Bas ian Be häuse , Caspa
Kaise , and Nha An T inh o sha ing R code ha aided ou analysis. Ea lie e -
sions o his wo k we e p esen ed a he 2024 Wo kshop on Spa ial Inequali ies in
Eu ope a Uni e si y College London and he 2024 Eu opean Conso ium o So-
22
ciological Resea ch (ECSR) con e ence in Ba celona. We hank he audiences on
hese occasions, as well as he edi o ial boa d and h ee anonymous e iewe s o
Eu opean Socie ies whose commen s g ea ly imp o ed he manusc ip .
AI use disclosu e AI-based ools we e used o some copyedi ing asks. All
analyses and da a p epa a ion we e pe o med wi hou he use o AI.
Au ho con ibu ions Bo h au ho s con ibu ed equally o his wo k.
Da a a ailabili y Sou ce da a a e a ailable o download om he Eu opean
Social Su ey. The au ho s’ da ase con aining agg ega e egional le el s a is ics,
as well as code o ep oduce all ables and igu es, a e a ailable a h ps://os .
io/24w36/.
Decla a ion o in e es s a emen The au ho s decla e no con lic s o in e es .
E hical app o al This s udy uses seconda y da a om he Eu opean Social
Su ey, a publicly a ailable da ase collec ed unde s ic e hical guidelines and
in o med consen p o ocols es ablished by he ESS. Only anonymized da a we e
accessed, and no pe sonal iden i ie s we e used in any pa o he analysis. No
addi ional e hical app o al was needed.
Supplemen s Supplemen al da a o his a icle can be accessed online.
Re e ences
Accia i, P., Polo, A., and Violan e, G. L. (2022). And ye i mo es: In e gene a ional
mobili y in I aly. Ame ican Economic Jou nal: Applied Economics, 14(3):118–63.
Ahsan, M. N., Em an, M. S., Jiang, H., Mu phy, O., and Shilpi, F. (2022). When
measu es con lic : Towa ds a be e unde s anding o in e gene a ional educa-
ional mobili y. A ailable a SSRN.
Aydemi , A. B. and Yazici, H. (2019). In e gene a ional educa ion mobili y and he
le el o de elopmen . Eu opean Economic Re iew, 116:160–85.
Beeke s, L. (2024). Sou ces o egional a ia ion in in e gene a ional mobili y: E i-
dence om he Ne he lands. Cen ER Discussion Pape 2024-015, Cen ER, Cen e
o Economic Resea ch.
Bell, B., Blundell, J., and Machin, S. (2023). Whe e is he land o hope and glo y?
The geog aphy o in e gene a ional mobili y in England and Wales. Scandina ian
Jou nal o Economics, 125(1):73–106.
Belle , E. (2009). B inging in e gene a ional social mobili y esea ch in o he
wen y- i s cen u y: Why mo he s ma e . Ame ican Sociological Re iew,
74(4):507–528.
23
Be ge , T. and Engzell, P. (2022). Indus ial au oma ion and in e gene a ional
income mobili y in he Uni ed S a es. Social Science Resea ch, 102686.
Be ge , T., Engzell, P., E iksson, B., and Molinde , J. (2023). Social mobili y in
Sweden be o e he wel a e s a e. Jou nal o Economic His o y, 83(2):431–463.
Be häuse , B. A. (2019). The e ec o he pos -socialis ansi ion on inequal-
i y o educa ional oppo uni y: E idence om Ge man uni ica ion. Eu opean
Sociological Re iew, 35(4):461–473.
Be häuse , B. A., Kaise , C., and T inh, N. A. (2021). Regional a ia ion in inequal-
i y o educa ional oppo uni y ac oss Eu ope. Socius, 7(23780231211019890).
Biasi, B. (2023). School inance equaliza ion inc eases in e gene a ional mobili y.
Jou nal o Labo Economics, 41(1):1–38.
Bjö klund, A. and Jän i, M. (2009). In e gene a ional income mobili y and he
ole o amily backg ound. In Sal e da, W., Nolan, B., and Smeeding, T., edi-
o s, Ox o d Handbook o Economic Inequali y, pages 491–521. Ox o d Uni e si y
P ess.
Blanden, J. (2013). C oss-coun y ankings in in e gene a ional mobili y: A com-
pa ison o app oaches om economics and sociology. Jou nal o Economic Su -
eys, 27(1):38–73.
Bloome, D., Dye , S., and Zhou, X. (2018). Educa ional inequali y, educa ional ex-
pansion, and in e gene a ional income pe sis ence in he Uni ed S a es. Ame ican
Sociological Re iew, 83(6):1215–1253.
Bo gen, S. T., Bo gen, N. T., and Zach isson, H. D. (2025). E hnic mino i ies
in disad an aged neighbo hoods: Di e en ial e ec s on educa ional ou comes.
Social Science Resea ch, 127:103152.
Bous an, L., Jensen, M. F., Ab ami zky, R., Jácome, E., Manning, A., Pé ez, S.,
Wa ley, A., Ade mon, A., A ellano-Bo e , J., Åslund, O., e al. (2025). In e gen-
e a ional mobili y o immig an s in 15 des ina ion coun ies. Technical epo ,
Na ional Bu eau o Economic Resea ch.
B andén, M. and Haand ikman, K. (2019). Who mo es o whom? Gende di e -
ences in he dis ance mo ed o a sha ed esidence. Eu opean Jou nal o Popula-
ion, 35:435–458.
B andén, G. (2019). Does inequali y educe mobili y? The G ea Ga sby cu e and
i s mechanisms. Wo king Pape , Ins i u e o E alua ion o Labou Ma ke and
Educa ion Policy (IFAU), Uppsala.
B a be g, E., Da is, J., Mazumde , B., Nybom, M., Schni zlein, D. D., and Vaage,
K. (2017). A compa ison o in e gene a ional mobili y cu es in Ge many, No -
way, Sweden, and he US. Scandina ian Jou nal o Economics, 119(1):72–101.
24
S ömbe g, E. and Engzell, P. (2025). How obus a e coun y ankings in educa-
ional mobili y? Sociological Science, o hcoming.
S uhle , J. (2018). A e iew o in e gene a ional mobili y and i s d i e s. Publica-
ions O ice o he Eu opean Union, Luxembou g.
Thaning, M. and Hälls en, M. (2020). The end o dominance? E alua ing measu es
o socio-economic backg ound in s a i ica ion esea ch. Eu opean Sociological
Re iew, 36(4):533–47.
T eiman, D. J. (1970). Indus ializa ion and social s a i ica ion. Sociological In-
qui y, 40:207–34.
Vigna, N. (2023). Subjec i e social s a us in places ha don’ ma e : Geog aphical
inequali ies in F ance and Ge many. Eu opean Socie ies, 25(5):693–720.
Webe , M. (1978). Economy and Socie y: An Ou line o In e p e i e Sociology.
Uni e si y o Cali o nia P ess, Be keley, CA. O iginally published in Ge man in
1922.
Winship, C. and Radbill, L. (1994). Sampling weigh s and eg ession analysis.
Sociological Me hods & Resea ch, 23(2):230–257.
Yaish, M. and Ande sen, R. (2012). Social mobili y in 20 mode n socie ies: The
ole o economic and poli ical con ex . Social Science Resea ch, 41(3):527–38.
Zwysen, W. (2016). C owding ou o disad an aged young adul s in Ge many:
Backg ound ma e s depending on local labou ma ke . Eu opean Sociological
Re iew, 32(5):662–674.
31
Appendix o :
The Geog aphy o In e gene a ional Mobili y in Eu ope ( o
online publica ion only)
1
Table A1: A ailabili y o ESS ounds by coun y.
Coun y R1 R2 R3 R4 R5 R6 R7 R8 R9 R10
Albania ✓
Aus ia ✓ ✓ ✓ ✓ ✓ ✓
Belgium ✓ ✓ ✓ ✓ ✓ ✓
Bulga ia ✓ ✓ ✓ ✓ ✓ ✓
C oa ia ✓ ✓ ✓ ✓
Cyp us ✓ ✓ ✓ ✓ ✓
Czech Republic ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Denma k ✓ ✓ ✓ ✓ ✓ ✓ ✓
Es onia ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Finland ✓ ✓ ✓ ✓ ✓ ✓ ✓
F ance ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Ge many ✓✓✓✓✓✓✓✓✓
G eece ✓ ✓ ✓ ✓
Hunga y ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Iceland ✓ ✓ ✓ ✓ ✓
I eland ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
I aly ✓ ✓ ✓ ✓
Koso o ✓
La ia ✓ ✓
Li huania ✓ ✓ ✓ ✓ ✓ ✓
Luxembou g ✓ ✓
Mon eneg o ✓ ✓
Ne he lands ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
No way ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Poland ✓✓✓✓✓✓✓✓✓
Po ugal ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Romania ✓
Se bia ✓
Slo ak Republic ✓ ✓ ✓ ✓ ✓ ✓ ✓
Slo enia ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Spain ✓ ✓ ✓ ✓ ✓ ✓ ✓
Sweden ✓✓✓✓✓✓✓✓✓
Swi ze land ✓ ✓ ✓ ✓ ✓ ✓ ✓
Uk aine ✓ ✓ ✓ ✓ ✓
Uni ed Kingdom ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
2
Table A2: Pe cen se ice class, obse a ions and egions pe coun y.
Pe cen se ice
Coun y Pa en s Child Obse a ions Regions
Albania 20 37 585 1
Aus ia 35 47 10310 9
Belgium 35 55 7687 11
Bulga ia 22 34 10622 28
C oa ia 26 40 3970 6
Cyp us 18 43 3491 1
Czech Republic 32 43 13890 14
Denma k 43 52 9311 5
Es onia 35 41 13919 7
Finland 27 48 11699 8
F ance 35 53 12519 19
Ge many 34 54 20051 16
G eece 20 37 5383 13
Hunga y 18 35 12500 9
Iceland 45 57 3442 1
I eland 26 46 12621 8
I aly 23 46 5289 5
Koso o 22 45 420 1
La ia 25 37 2426 6
Li huania 18 37 8605 17
Luxembou g 29 48 1999 1
Mon eneg o 24 42 1046 1
Ne he lands 39 59 14645 12
No way 45 54 13796 10
Poland 22 39 11266 16
Po ugal 17 32 10950 5
Romania 15 36 1124 8
Se bia 25 40 1211 1
Slo ak Republic 25 40 8655 8
Slo enia 29 48 8103 16
Spain 21 38 8427 7
Sweden 42 52 13506 8
Swi ze land 36 60 9133 7
Uk aine 28 40 7695 26
Uni ed Kingdom 39 50 16172 12
To al 31 46 296468 323
3
Table A3: Coding o esponden social class.
ESeC 2002, 2004, 2006, 2008, 2010 ESeC 2012, 2014, 2016, 2018 This s udy
•La ge employe s, highe
manage s / p o essional
•Lowe manage s / p o es-
sional, highe supe iso s
•In e media e occupa ions
•La ge employe s, highe
manage s / p o essional
•Lowe manage s / p o es-
sional, highe supe iso s
•In e media e occupa ions
Se ice class
•Lowe supe iso s and
echnicians
•Lowe sales and se ice
•Lowe echnical
•Rou ine
•Small employe s and
sel -employed (ag icul u e
incl.)
•Lowe supe iso s and
echnicians
•Lowe sales and se ice
•Lowe echnical
•Rou ine
Wo king class
4
Table A4: Coding o pa en social class.
Pa en occupa ion 2002, 2004,
2006
Pa en occupa ion 2008, 2010,
2012, 2014, 2016, 2018
This s udy
•T adi ional p o essional
•Mode n p o essional
•Cle ical and in e media e
•Senio manage o adminis-
a o s
•Middle o junio manage s
•P o essional and echnical
•Highe adminis a o
•Cle ical
Se ice class
•Technical and c a
•Fa me
•Semi- ou ine/manual/ se -
ice
•Rou ine manual and se ice
•Skilled wo ke
•Semi-skilled wo ke
•Unskilled
•Se ice
•Sales
•Fa me
Wo king class
5
Table A5: Coding o a he ’s occupa ional ank.
2002 2004-2006 2008-2018 I eland, 2002 This s udy
Semi-
ou ine/manual/
se ice occupa-
ions
Semi-
ou ine/manual/
se ice occupa-
ions
Fa m wo ke Fa me 1
Rou ine manual
and se ice occu-
pa ions
Rou ine manual
and se ice occu-
pa ions
Unskilled wo ke Rou ine manual
and se ice occu-
pa ions
2
Technical and
c a occupa ions
Technical and
c a occupa ions,
a me
Semi-skilled
wo ke
Semi-
ou ine/manual/
se ice occupa ion
3
Middle o junio
manage s
Middle o junio
manage s
Skilled wo ke Technical and
c a occupa ions
4
Cle ical and in-
e media e occu-
pa ions
Cle ical and in-
e media e occu-
pa ions
Se ice occupa-
ions
T adi ional p o-
essional occupa-
ions
5
Senio manage o
adminis a o s
T adi ional p o-
essional occupa-
ions
Sales occupa ions Cle ical and in-
e media e occu-
pa ions
6
Mode n p o es-
sional occupa ions
Mode n p o es-
sional occupa ions
Cle ical occupa-
ions
Mode n p o es-
sional occupa ions
7
T adi ional p o-
essional occupa-
ions
Senio manage o
adminis a o s
Highe adminis-
a o occupa ions
Middle o junio
manage s
8
— — P o essional and
echnical occupa-
ions
Senio manage o
adminis a o s
9
6
Table A6: Coding o mo he ’s occupa ional ank.
2002 2004-2006 2008-2018 I eland, 2002 This s udy
Semi-
ou ine/manual/
se ice occupa ion
Technical and
c a occupa ions,
a me
Fa m wo ke Fa me 1
Rou ine manual
and se ice occu-
pa ions
Semi-
ou ine/manual/
se ice occupa-
ions
Unskilled wo ke Technical and
c a occupa ions
2
Technical and
c a occupa ions
Rou ine manual
and se ice occu-
pa ions
Semi-skilled
wo ke
Rou ine manual
and se ice occu-
pa ions
3
Middle o junio
manage s
Middle o junio
manage s
Skilled wo ke Cle ical and in-
e media e occu-
pa ions
4
Senio manage o
adminis a o s
Cle ical and in-
e media e occu-
pa ions
Se ice occupa-
ions
Semi-
ou ine/manual/
se ice occupa ion
5
Cle ical and in-
e media e occu-
pa ions
Senio manage o
adminis a o s
Sales occupa ions Senio manage o
adminis a o s
6
Mode n p o es-
sional occupa ions
T adi ional p o-
essional occupa-
ions
Cle ical occupa-
ions
Mode n p o es-
sional occupa ions
7
T adi ional p o-
essional occupa-
ions
Mode n p o es-
sional occupa ions
Highe adminis-
a o occupa ions
T adi ional p o-
essional occupa-
ions
8
— — P o essional and
echnical occupa-
ions
Middle o junio
manage s
9
7
Table A7: In a-class co ela ion coe icien s a he coun y le el o in e gene a-
ional mobili y and co a ia es.
(a) Unweigh ed
Va iable ICC
Rank- ank in e cep , men 0.291
Rank- ank in e cep , women 0.378
Rank- ank slope, men 0.376
Rank- ank slope, women 0.541
Upwa d class mobili y, men 0.665
Upwa d class mobili y, women 0.538
Class immobili y (odds), men 0.350
Class immobili y (odds), women 0.541
A e age schooling 0.774
Pe cen college 0.691
P o essionals 0.493
STEM wo ke s 0.542
Teache s pe capi a 0.580
Pe cen less han HS 0.913
Unemployed 0.785
P ima y sec o 0.550
Manu ac u ing 0.538
Se ices 0.602
Public sec o 0.657
Bu eauc acy 0.702
Immig a ion 0.549
Single mo he s 0.424
3-gen households 0.816
Household size 0.913
Median age 0.480
Inequali y 0.505
Social capi al 0.882
Social us 0.894
Ci ic capi al 0.775
U baniza ion 0.000
Violen c ime 0.601
(b) Weigh ed
Va iable ICC
Rank- ank in e cep , men 0.288
Rank- ank in e cep , women 0.304
Rank- ank slope, men 0.264
Rank- ank slope, women 0.522
Upwa d class mobili y, men 0.624
Upwa d class mobili y, women 0.534
Class immobili y (odds), men 0.298
Class immobili y (odds), women 0.488
A e age schooling 0.809
Pe cen college 0.735
P o essionals 0.523
STEM wo ke s 0.516
Teache s pe capi a 0.519
Pe cen less han HS 0.906
Unemployed 0.504
P ima y sec o 0.562
Manu ac u ing 0.433
Se ices 0.579
Public sec o 0.565
Bu eauc acy 0.657
Immig a ion 0.387
Single mo he s 0.333
3-gen households 0.709
Household size 0.577
Median age 0.421
Inequali y 0.518
Social capi al 0.894
Social us 0.889
Ci ic capi al 0.733
U baniza ion 0.012
Violen c ime 0.560
8
Table A8: Co ela ion ma ix be ween in e gene a ional mobili y and co a ia es (unweigh ed).
Va iables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31)
Rank- ank in e cep , men 1.00
Rank- ank in e cep , women 0.23 1.00
Rank- ank slope, men -0.03 -0.18 1.00
Rank- ank slope, women -0.11 -0.38 0.33 1.00
Upwa d class mobili y, men 0.66 0.24 0.19 -0.32 1.00
Upwa d class mobili y, women 0.38 0.68 0.04 -0.31 0.68 1.00
Class immobili y (odds), men -0.15 -0.30 0.61 0.46 -0.25 -0.27 1.00
Class immobili y (odds), women -0.09 -0.38 0.37 0.68 -0.18 -0.33 0.55 1.00
A e age schooling 0.38 0.56 0.08 -0.26 0.49 0.63 -0.24 -0.24 1.00
Pe cen college 0.33 0.47 -0.05 -0.13 0.30 0.46 -0.22 -0.25 0.48 1.00
P o essionals 0.58 0.54 0.15 -0.22 0.71 0.78 -0.13 -0.20 0.69 0.56 1.00
STEM wo ke s 0.43 0.25 0.15 -0.13 0.52 0.50 -0.07 -0.08 0.43 0.26 0.57 1.00
Teache s pe capi a 0.32 0.28 0.03 -0.29 0.50 0.51 -0.18 -0.25 0.44 0.37 0.69 0.28 1.00
Pe cen less han HS -0.04 -0.25 -0.09 -0.09 0.02 -0.22 0.03 0.03 -0.45 -0.14 -0.19 -0.29 -0.02 1.00
Unemployed -0.25 -0.23 -0.14 0.21 -0.45 -0.43 0.04 0.09 -0.39 -0.16 -0.41 -0.41 -0.37 0.11 1.00
P ima y sec o -0.41 -0.33 -0.10 0.06 -0.37 -0.40 0.01 -0.02 -0.49 -0.19 -0.48 -0.46 -0.28 0.36 0.36 1.00
Manu ac u ing -0.34 -0.19 -0.09 0.20 -0.56 -0.44 0.11 0.13 -0.16 -0.29 -0.41 0.03 -0.32 -0.31 0.06 -0.06 1.00
Se ices 0.19 -0.05 -0.07 -0.41 0.17 -0.07 -0.14 -0.26 0.13 0.04 0.15 0.02 0.31 0.30 -0.13 -0.10 -0.27 1.00
Public sec o 0.36 0.27 0.02 -0.39 0.58 0.56 -0.25 -0.29 0.53 0.37 0.73 0.35 0.84 -0.13 -0.28 -0.34 -0.31 0.35 1.00
Bu eauc acy 0.46 0.23 0.24 -0.24 0.74 0.57 -0.08 -0.06 0.51 0.11 0.57 0.48 0.41 -0.04 -0.48 -0.47 -0.36 0.16 0.44 1.00
Immig a ion 0.26 0.27 0.09 -0.23 0.46 0.41 -0.10 -0.13 0.26 0.35 0.46 0.30 0.27 0.07 -0.21 -0.32 -0.36 0.12 0.26 0.46 1.00
Single mo he s -0.12 0.14 -0.13 0.12 -0.34 -0.16 -0.04 0.03 0.09 0.30 -0.10 -0.05 -0.10 -0.17 0.11 -0.06 0.27 -0.09 -0.13 -0.22 0.09 1.00
3-gen households -0.30 -0.18 -0.13 0.33 -0.55 -0.48 0.10 0.11 -0.48 -0.02 -0.50 -0.39 -0.46 -0.12 0.59 0.38 0.26 -0.36 -0.47 -0.60 -0.33 0.19 1.00
Household size -0.17 -0.11 0.03 0.25 -0.21 -0.24 0.08 0.15 -0.33 -0.07 -0.35 -0.32 -0.33 -0.01 0.47 0.21 0.02 -0.35 -0.37 -0.37 -0.20 -0.04 0.69 1.00
Median age -0.11 -0.17 -0.24 -0.08 -0.28 -0.32 -0.13 -0.09 -0.26 -0.06 -0.21 -0.17 -0.08 0.15 0.28 0.24 0.27 0.11 0.08 -0.27 -0.18 0.20 0.16 -0.25 1.00
Inequali y 0.13 -0.05 0.20 0.09 0.22 0.10 0.16 0.18 0.02 -0.15 0.13 0.10 -0.01 0.34 0.04 -0.10 -0.22 0.10 0.00 0.27 0.09 -0.15 -0.25 -0.07 -0.16 1.00
Social capi al 0.36 -0.02 0.02 -0.34 0.46 0.17 -0.12 -0.14 0.09 0.07 0.34 0.29 0.34 0.18 -0.02 -0.27 -0.30 0.49 0.48 0.41 0.25 -0.24 -0.27 -0.10 0.05 0.19 1.00
Social us 0.36 0.29 0.11 -0.46 0.60 0.57 -0.21 -0.30 0.57 0.22 0.64 0.45 0.59 -0.08 -0.54 -0.34 -0.26 0.42 0.65 0.53 0.34 -0.21 -0.73 -0.49 -0.18 0.10 0.41 1.00
Ci ic capi al 0.26 0.20 0.03 -0.42 0.56 0.43 -0.21 -0.26 0.34 0.10 0.44 0.25 0.48 0.10 -0.38 -0.17 -0.35 0.33 0.44 0.48 0.29 -0.33 -0.49 -0.21 -0.29 0.11 0.38 0.65 1.00
U baniza ion 0.37 0.28 0.06 0.08 0.13 0.20 0.09 0.03 0.27 0.28 0.33 0.20 -0.02 -0.07 -0.03 -0.25 -0.28 0.05 -0.03 0.18 0.24 0.15 -0.05 -0.12 -0.22 0.08 0.01 -0.03 -0.17 1.00
Violen c ime 0.36 0.21 -0.09 -0.27 0.38 0.33 -0.17 -0.22 0.31 0.28 0.43 0.35 0.32 0.25 -0.17 -0.16 -0.28 0.35 0.32 0.34 0.36 -0.01 -0.41 -0.29 -0.16 0.23 0.37 0.34 0.31 0.29 1.00
9
(a) Upwa d ank mobili y, a he -son (b) Upwa d ank mobili y, mo he -daugh e
Figu e A10: Geog aphic a ia ion in gende -speci ic upwa d ank mobili y ( ank-
ank in e cep ). Highe alues indica e mo e mobili y.
(a) Rela i e ank immobili y, a he -son (b) Rela i e ank immobili y, mo he -daugh e
Figu e A11: Geog aphic a ia ion in gende -speci ic ela i e ank immobili y ( ank-
ank slope). Highe alues indica e less mobili y.
16
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-1 -.5 0 .5 1
Co ela ion To al Wi hin-coun y
(a) Upwa d class mobili y, men
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-1 -.5 0 .5 1
Co ela ion To al Wi hin-coun y
(b) Upwa d class mobili y, women
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-.5 -.25 0 .25 .5
Co ela ion To al Wi hin-coun y
(c) Rela i e class immobili y, men
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-.5 -.25 0 .25 .5
Co ela ion To al Wi hin-coun y
(d) Rela i e class immobili y, women
Figu e A12: Co ela es o in e gene a ional mobili y, social class. Co ela ion coe -
icien and 95% con idence in e al, s anda d e o s clus e ed a he coun y le el.
17
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-1 -.5 0 .5 1
Co ela ion To al Wi hin-coun y
(a) Upwa d ank mobili y, men
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-1 -.5 0 .5 1
Co ela ion To al Wi hin-coun y
(b) Upwa d ank mobili y, women
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-.5 -.25 0 .25 .5
Co ela ion To al Wi hin-coun y
(c) Rank immobili y, men
A. Human capi al
A e age schooling
Pe cen college
P o essionals
STEM wo ke s
Teache s pe capi a
Pe cen less han HS
B. Labo ma ke
Unemployed
P ima y sec o
Manu ac u ing
Se ices
Public sec o
Bu eauc acy
C. Demog aphics
Immig a ion
Single mo he s
3-gen households
Household size
Median age
D. Socio-spa ial
Inequali y
Social capi al
Social us
Ci ic capi al
U baniza ion
Violen c ime
-.5 -.25 0 .25 .5
Co ela ion To al Wi hin-coun y
(d) Rank immobili y, women
Figu e A13: Co ela es o in e gene a ional mobili y (unweigh ed). Co ela ion
coe icien and 95% con idence in e al, s anda d e o s clus e ed a he coun y
le el.
18
