Wiedne , Jonas
A icle — Accep ed Manusc ip (Pos p in )
Unde - o o e expansion o educa ion? T ends in
quali ica ion misma ch in he Uni ed Kingdom and
Ge many, 1984-2017
Social Science Resea ch
P o ided in Coope a ion wi h:
WZB Be lin Social Science Cen e
Sugges ed Ci a ion: Wiedne , Jonas (2024) : Unde - o o e expansion o educa ion? T ends in
quali ica ion misma ch in he Uni ed Kingdom and Ge many, 1984-2017, Social Science Resea ch,
ISSN 1096-0317, Else ie , Ams e dam, Vol. 117, pp. 1-15,
h ps://doi.o g/10.1016/j.ss esea ch.2023.102948
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/313713
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
1
Unde - o o e expansion o educa ion?
T ends in quali ica ion misma ch in he Uni ed Kingdom
and Ge many, 1984-2017
Jonas Wiedne
WZB Be lin Social Science Resea ch Cen e
Abs ac
P ominen heo e ical posi ions in sociology and labo economics disag ee whe he educa-
ional expansion has ou s ipped he demand o quali ied labo (o e expansion), o whe he
economies ace a skill sho age despi e inc eases in educa ion (unde expansion). Focusing
on he Uni ed Kingdom and Wes Ge many, wo coun ies wi h dissimila skill o ma ion ins i-
u ions, pa e ns o expansion, and labo ma ke s, his pape asks o wha deg ee expansion
o educa ion has been abso bed. I poin ou sho comings o wage-cen e ed analyses and
de elop an app oach ha ocuses on ends in sel -assessed o e - and unde quali ica ion.
Using epea ed su eys among wo ke s and o icial labo ma ke s a is ics, I es ima e eg es-
sion models ha link he coho -le el expansion o educa ion o he coho -le el p e alence o
misma ch. Resul s sugges o e expansion in he Uni ed Kingdom, wi h o e quali ica ion in-
c easing and unde quali ica ion dec easing o e his o ical imes and coho s. Wes Ge many,
on he o he hand, shows signs o unde expansion. While dominan heo e ical accoun s ocus
on he unde -/o e expansion o e ia y educa ion, my esul s show ha misma ch-dynamics
in bo h con ex s a e s onges o wo ke s wi hou uni e si y deg ees.
Keywo ds: educa ional expansion, misma ch, occupa ional change, compa a i e esea ch
This a icle was published by Else ie in
Social Science Resea ch, Vol. 117 (2024), A . 102948:
h ps://doi.o g/10.1016/j.ss esea ch.2023.102948
© 2024 This manusc ip e sion is made a ailable
unde he CC-BY-NC-ND 4.0 license
h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
2
1. In oduc ion
O e he pas cen u y, comple ion o educa ional p og ams a all le els has inc eased d ama -
ically. In 1953, jus 19% o 17-yea -old B i ons a ended ull ime schooling o aining, bu in
he 2010s close o 90% did (House o Commons 2012). By 2018, hal o 25- o 34-yea -olds
had unde gone e ia y p og ams (OECD 2019). In Wes Ge many, 79% o 7 h g ade s a -
ended Volksschulen in 1952 and we e hus bound o g adua e by he age o 15. As a esul ,
jus 8% o a coho en olled in e ia y educa ion in 1960. In 2010, by con as , almos 40% did
(Geißle 2014, p. 335 .). This d ama ic expansion o educa ion has been ecognized by soci-
ologis s as one o he majo o ces shaping social change (Collins 1979; Bake 2014).
Whe he hese g owing numbe s o highe -educa ed wo ke s ha e been able o ind adequa e
wo k is cen al o unde s anding s a i ica ion sys ems in mode n socie ies. Ye , social science
schola ship o e s wo con adic o y answe s o his ques ion, each backed up by empi ical
li e a u e. The i s , mainly associa ed wi h sociological heo ies o social closu e and c eden ial
in la ion, a gues ha he e a e now oo many highe quali ied wo ke s o labo ma ke s o
abso b in i ing wo k (F eeman 1976; Collins 1979; Wol 2002). The second, oo ed in a ian s
o skill-biased echnological change (SBTC) a gumen s in economics, a gues ha , gi en
changes in skill demand, he e a e s ill oo li le (Goldin and Ka z 2010; Acemoglu and Au o
2011; Ca ne ale and Rose 2014). I e e o hese con as ing assessmen s as he diagnosis
o o e - o unde expansion o educa ion, espec i ely.
An impo an implica ion o ei he o e - o unde expansion is ha hey inc ease he numbe s
o e ical educa ion misma ches: in he case o unde expansion, wo ke s whose educa ion
and aining all sho o wha is usual o nominally equi ed in hei occupa ion (unde quali i-
ca ion), o , in he case o o e expansion, whose educa ion and aining a e in excess o wha
is usual o equi ed (o e quali ica ion) (Capsada-Munsech 2019). T ends in quali ica ion mis-
ma ch a e he e o e indica i e o he deg ee o labo ma ke abso p ion o educa ional expan-
sion.
1
1
I use he e m “quali ica ion“ o e e o ce i ica es o g adua ion om educa ion o aining p og ams
and o skill-ce i ica es mo e b oadly. I p e e he e m “quali ica ion” o e “educa ion” because i includes
3
Agains his backg ound and ocusing on he Uni ed Kingdom and Ge many, his a icle se s
ou o answe he ollowing wo ques ions: Fi s , o wha deg ee has expansion o educa ion in
hese wo socie ies been abso bed by labo ma ke s? Is he e e idence o o e expansion, wi h
highe le els o o e quali ica ion, o do we see unde expansion, wi h inc easing unde quali i-
ca ion? Second, o which ex en does his di e be ween he wo coun ies? A ocus on Ge -
many and he Uni ed Kingdom, as, wi hin he Eu opean con ex , e y dissimila cases, allows
p obing he ex en o a ia ion in di e en coun ies’ expe ience. The Uni ed Kingdom and Ge -
many ep esen di e en a ie ies o capi alism wi h as ly di e en educa ion sys ems (choice-
d i en p og ession in he Uni ed Kingdom, igidly acked in Ge many), indus y s uc u es,
labo ma ke ules and pa e ns o expansion (s ong expansion ac oss all le els in he Uni ed
Kingdom; a mode a e expansion o e ia y educa ion in Ge many) (Mülle and Gangl 2003;
Hall and Soskice 2001).
My analysis adds o an eme ging li e a u e ha uses al e na i e measu es o assess he labo
ma ke abso p ion o educa ional expansion di ec ly. E idence o unde expansion o educa ion
has adi ionally been a byp oduc o s udies o inc easing wage inequali y, and he e o e o-
cused on wage-di e ences be ween educa ion g oups (Goldin and Ka z 2010). He e, I de elop
a new app oach which ocuses on pe iod and coho ends in sel - epo ed e ical mis-
ma ches be ween indi iduals’ educa ion and hei jobs, ha is, on o e - and unde quali ica ion.
I use ime-se ies o misma ch-incidence based on epea ed su eys (Uni ed Kingdom: Skills
and Employmen Su ey), o a panel s udy (Ge many: Socioeconomic Panel S udy), espec-
i ely, as well as con ex ual in o ma ion om o icial labo o ce su eys. My o mal models use
di e ences in educa ion be ween coho s in he same egion and du ing he same ime o
es ima e he ela ionship be ween a coho s’ incidence o misma ch and i s quali ica ion s uc-
u e.
ce i ied skills ha ha e no been acqui ed in a o mal educa ional se ing, which ha e been cen al o
aining-policy deba es pa icula ly in he Uni ed Kingdom (Wol 2002). My usage o he e m “quali ica-
ion” does, howe e , no include non-ce i ied skills, abili ies, and expe ience, such as gene al cogni i e
abili y, physical s eng h, o wo k expe ience, which a e some imes subsumed unde he e m.
4
As an empi ical con ibu ion, I p o ide es ima es o e ical misma ch- ends o he pe iod
1986-2017 and o coho s bo n be ween he 1920s and 1980s o he Uni ed Kingdom and
Ge many, and o e a no el, coho -based pe spec i e on he ela ionship be ween inc easing
educa ional le els and quali ica ion misma ch (bu c . Ho owi z 2018; and Pollmann-Schul and
Maye 2004). In con as o mos p e ious wo k (c. ., Shin and Bills 2021), my analyses include
unde quali ica ion and sys ema ically conside expansion a di e en quali ica ion le els. I
show ha he ela ionship be ween educa ional expansion and quali ica ion misma ch is highly
con ingen on ins i u ional egimes. Educa ional expansion wen hand in hand wi h inc easing
o e - and alling unde quali ica ion in he Uni ed Kingdom, bu no in Ge many. This inding
suppo s ecen a gumen s ha global o ces o echnological and social change ha e e y
di e en e ec s in di e en ins i u ional con ex s (Oesch and Rod iguez Menes 2011; Fe nán-
dez-Macías and Hu ley 2017). Finally, my esul s documen ha changes in misma ch- a es
wi h educa ional expansion a e mos p onounced o wo ke s wi h quali ica ions below he uni-
e si y-le el.
In he nex sec ion, I e iew he wo mos impo an heo ies o he changing labo ma ke
balance o educa ion and demand o i , skill-biased echnological change and c eden ial in la-
ion, and examine he ole o ins i u ional con ex . I hen in oduce he da a and he analy ical
app oach I use. In sec ion 4, I p esen desc ip i e and mul i a ia e esul s be o e I summa ize
my indings and d aw conclusions.
2. Theo e ical backg ound and p e ious esea ch
Pe spec i e 1: echnological change esul s in an unde expansion o
educa ion
The dominan iew on he changing balance be ween educa ional expansion and demand o
i in economics is ha o an inc easing skill sho age (a leas a s able p ices, i.e. wages).
While SBTC heo y’s main conce n is wi h inc easing wage-inequali y, he mechanism i as-
sumes is ha o an unde expansion o educa ion. SBTC heo y he e o e has implica ions o
5
misma ch- ends. SBTC claims a e oo ed in he idea ha echnological inno a ions comple-
men he p oduc i i y o wo ke s wi h highe le els o educa ion, bu o en subs i u e o less
educa ed wo ke s (Tinbe gen 1956; Goldin and Ka z 2010; Acemoglu and Au o 2012). Tech-
nological p og ess hus inc eases he ela i e demand o skilled s. unskilled wo ke s. Edu-
ca ion's alue on he labo ma ke he e o e depends on a “ ace be ween educa ion and ech-
nology”: I he a e o echnological inno a ion ou paces he expansion o educa ion, he econ-
omy’s skill demand will ou s ip supply, inc easing highly educa ed ela i e o less educa ed
wages (Goldin and Ka z 2010). F om an SBTC pe spec i e, inc easing wage inequali y hus
e lec s a sho age o skilled labo , ha is unde expansion o educa ion (Ca ne ale and Rose
2014).
A newe e inemen o SBTC a gues ha he kinds o asks ha a e being eplaced by au o-
ma ion a e no low-skill pe se, bu ou ine in na u e. These, i s p oponen s a gue, p edomina e
no in low, bu in middle-skill occupa ions (Au o e al. 2003; Au o and Do n 2013). The con-
sequence o such ou ine-biased echnological change (RBTC) would hus no be an upg ad-
ing, bu a pola iza ion o skill-demand. While SBTC’s and RBTC’s implica ions o educa ional
expansion abso p ion coincide in he uppe hal o he educa ion dis ibu ion, RBTC would ex-
pec no an unde - bu an o e expansion o middling ce i ica es — assuming ha educa ional
upg ading is linea .
Wage-based s udies o unde expansion
Wo k by economis s on he Uni ed S a es documen s d ama ically inc easing wage inequali y
be ween, bu also wi hin quali ica ion g oups, which is linked o he s agna ion o college com-
ple ion a es since he 1970s (Ka z and Mu phy 1992; Goldin and Ka z 2010). Va ian s o he
skill sho age hesis o SBTC a e hence widely accep ed among US economis s (bu c . Ca d
and DiNa do 2002; Acemoglu and Au o 2011).
2
In Ge many, inc eases in wage inequali y
2
I should be no ed, howe e , ha dissiden labo ma ke analys s ha e challenged he uncomplica ed
applicabili y o he pola iza ion- hesis o he US labo ma ke (Holze and Le man 2007, 2009; Holze
2015). D awing on disagg ega ed occupa ion-le el wage and employmen ends, hey show ha he
supposed wide decline o middle-skill wo k is all bu uni o mly e iden . Ra he han expec ing a secula
decline o middle-skill wo k, hey a gue ha ins i u ional sho comings in wo ke aining a e in pa e-
sponsible o obse ed declines in some sub-segmen s o he skill-middle. As a consequence, and in
6
ha e been mo e mode a e, which migh e lec he subs an ial expansion o he educa ion sys-
em since he 1950s. Ne e heless, s udies conclude ha SBTC, and hence a ela i e unde -
expansion o educa ion, had a pa in inc easing wage inequali y in Ge many (Dus mann e al.
2009). In he Uni ed Kingdom, educa ional expansion was e en mo e p onounced. A he same
ime, howe e , inequali y be ween educa ion g oups – and wage inequali y o e all – ha e
isen d ama ically, sugges ing ha ins i u ional changes may ha e played a dominan ole he e
(Gosling and Lemieux 2004).
S udies in e ing changes in he a io o demand o highe educa ed wo ke s o i s supply om
wage da a ace c i icism, howe e (Kallebe g 2011; Cappelli 2015). In o de o p o ide e i-
dence in a o o SBTC, hey assume ha obse able wage changes a e due o shi ing de-
mand o supply, and no o o he ac o s. Bu much esea ch demons a es ha ac o s such
as changing minimum wages, de-unioniza ion and o he o ms o en -des uc ion and -c ea ion
ha e con ibu ed signi ican ly o inc eased wage inequali y ( o examples, see K is al and Co-
hen 2017; Weeden and G usky 2014; Fi zenbe ge e al. 2013; Gosling and Lemieux 2004).
A leas in simple wage analyses, hese o ces ac as omi ed a iables and bias es ima es o
excess demand o educa ion. The SBTC conclusion ha wage inequali y ose because o a
sho age o highly educa ed labo may he e o e be p ema u e. Vice e sa, his also pu s im-
po an e idence in a o o he unde expansion hesis in o ques ion.
Pe spec i e 2: C eden ial in la ion, social closu e and ela i e educa-
ion esul in an o e expansion o educa ion
An impo an assump ion o bo h a ian s o he echnological change a gumen , SBTC and
RBTC, is ha skill and educa ion a e o a la ge ex en synonymous. SBTC and RBTC a e
based on he human capi al heo y o educa ion, which holds ha educa ion and quali ica ions
a e alued by employe s and s uden s because hey c ea e and ce i y ele an skills, which
con as o pola iza ion heo is s, hey ecommend policies o encou ages skill- o ma ion and be e
ma ching among middle-skill wo ke s (Ca ne ale and Des oche s 2002).
7
hen ansla e in o high p oduc i i y and ea nings. By con as , many sociological app oaches
highligh ha educa ion’s ole o s uden s and socie y goes beyond echnical know-how: o -
mal educa ion and p oduc i e skills a e no he same hing. While hese heo ies di e in he
causal mechanisms hey ocus on, hey ag ee ha an inc easing alue a ached o educa ion
o non- echnical easons has esul ed in i s expansion beyond labo ma ke s’ abso p ion ca-
paci y. P ominen heo ies a ound he no ion o c eden ial in la ion, o ins ance, a gue ha
schooling is impo an o policy make s, pa en s, s uden s and employe s because i ins ils
obedien wo k a i udes (Bowles and Gin is 1977), legi imizes social s a i ica ion (Bou dieu
and Passe on 1970), equips s uden s wi h cul u al cu ency ha gi es access o closed occu-
pa ional posi ions (Collins 1979), o is a means o sel - ealiza ion and a symbol o amily p es-
ige (Bake 2014). Such ideological, poli ical, o cul u al easons o expansion bea li le con-
nec ion o echnological equi emen s. Expansion o hese non- echnical easons is he e o e
likely o exceed employe s’ demand o mo e highly quali ied wo ke s and esul in o e expan-
sion.
O he social scien is s poin o mic o-le el p ocesses which lead a ional ac o s o acqui e ed-
uca ion beyond labo ma ke demand and esul in o e expansion. The cen al a gumen in
such analyses is ha educa ion has a ela i e a he han an absolu e alue. Acco ding o he
labo queue model, educa ion is a signal o u u e p oduc i i y o ainabili y and employe s
ank applican s acco ding o hei le el o quali ica ion (Thu ow 1975). The alue o a gi en
pe son’s quali ica ion he e o e depends on he le els o educa ion o o he s in he queue. This
mechanism is desc ibed as he ela i e o posi ional educa ion hypo hesis (Ho owi z 2018). I
educa ion is posi ional, educa ional expansion ansla es in o a heigh ened indi idual incen i e
o ob ain ye mo e educa ion in o de o p ese e one’s ela i e posi ion (Bol 2015). Since
educa ion essen ially becomes a ace o he highes deg ee, o e quali ica ion, ela i e o jobs’
ac ual skill equi emen , will be e y p e alen as companies’ hi ing s anda ds escala e.
To sum up: I o ces o he han echnology-induced demand a e behind inc eased quali ica ion
up ake, as a gued by many sociological app oaches, he e will be o e expansion. In his
8
si ua ion, holde s o highe deg ees will ind i inc easingly ha d o ind i ing wo k. Expansion
will lead o highe a es o o e quali ica ion.
Misma ch and he abso p ion o educa ional expansion
As e iewed abo e, he e idence in a o o he unde expansion hesis has adi ionally been
a byp oduc o s udies o inc easing wage inequali y, and aces me hodological conce ns. In
line wi h his c i icism, s udies using o he app oaches o e whelmingly ind suppo o he
o e expansion hesis. The p incipal al e na i e s a egy ocuses on misma ches be ween wo k-
e s’ quali ica ions and he equi emen s o hei jobs. Ra es o ( e ical) quali ica ion misma ch
can se e as an immedia e measu e o he deg ee o abso p ion o educa ional expansion by
labo ma ke s.
3
In con as o in e ence based on ela i e wage- ends, misma ches o e a
di ec way o gauge he abso p ion o someone’s educa ion on he labo ma ke . The main
ad an age o misma ch- end-based assessmen s o o e - and unde expansion is ha unob-
se ed p ocesses (like changing minimum wages, de-unioniza ion, o inc easingly oligopsonic
labo ma ke s) do no di ec ly a ec he measu emen o a quali ica ion imbalance on he labo
ma ke in he way hey do in ela i e-wage-based s a egies. E en hough misma ch-measu e-
men may in oduce ambigui ies o i s own (see Appendix B in he Online Supplemen ), sub-
s an i e conclusions in a deba e as cen al o policy and social science as he unde -/o e ex-
pansion ques ion should no es on a single me hodological app oach.
F om a misma ch-pe spec i e, a es o o e quali ica ion should dec ease unde SBTC, as em-
ploye s seek o exploi he capaci ies o he ela i ely sca ce highly quali ied wo ke s. A he
same ime, a es o unde quali ica ion should inc ease, as i ms hi e wo ke s o posi ions
abo e hei nominal le el o educa ion o a oid in e up ions o hei business. C eden ial in la-
ion, social closu e and ela i e educa ion app oaches would expec he in e se pa e n: ising
o e quali ica ion and sinking unde quali ica ion o e his o ical ime, and in coho s subjec o
3
Acco ding o s anda d unde s andings in he li e a u e, misma ch can be ho izon al (be ween ield o
s udy/ aining and job) o e ical (be ween skill le els o aining and skill le el o employmen ) (Di S asio
2017). While educa ional expansion has implica ions o he ho izon al ma ching be ween wo ke s and
jobs (Elbe s e al. 2021), unde - and o e expansion a e mos clea ly exp essed by a es o e ical
misma ch ha is by o e - and unde quali ica ion.
15
hea o he o e expansion deba e. In line wi h my heo e ical easoning, I in e p e a posi i e
ela ionship as e idence o o e expansion.
Independen a iables
My main independen a iables a e indica o s o he con ex ual sha e o people o a gi en
quali ica ion le el a a gi en poin in ime, in a gi en egion, in a gi en ex ended coho (de ined
as people bo n wi hin +- 3 yea s om he coho base yea ). I measu e hem using o icial labo
ma ke su eys. A his ex ended-coho - egion-pe iod-le el, I dis inguish six ca ego ies (below
seconda y educa ion, any non-minimal seconda y educa ion quali ica ion, lowe oca ional
quali ica ions, oca ional quali ica ions including ade app en iceships, lowe e ia y quali ica-
ions, and e ia y quali ica ions) in he Uni ed Kingdom, and in Ge many h ee ca ego ies (sec-
onda y o lowe , oca ional quali ica ions, and e ia y quali ica ions). This analy ical choice
e lec s he consis en ly dis inguishable le els o educa ion in he di e en ounds o he labo
o ce su eys I use.
Con ol a iables
I adjus my es ima es o indi iduals’ own highes quali ica ion, which ules ou ha composi ion
e ec s in luence esul s. Based on he a ailable indi idual-le el su ey da a, I dis inguish six
(Uni ed Kingdom) and i e (Ge many) g oups, acco ding o a esponden ’s highes quali ica-
ion:
• no quali ica ions,
• (non-minimal) seconda y quali ica ions ( hese wo a e combined in Ge many),
• wo ca ego ies o oca ional quali ica ions
o in he Uni ed Kingdom:
▪ lowe oca ional quali ica ions om sho p og ams, i.e. le el 1
o 2 in he NVQ classi ica ion, and
▪ mo e ad anced oca ional quali ica ions (NVQ le el 3), e.g. ap-
p en iceships o SCOTEC/SCOTBEC quali ica ions;
o in Ge many:
▪ wo ke s wi h oca ional aining, and
▪ wo ke s wi h highe oca ional aining, i.e. Meis e and Techni-
ke ,
• lowe e ia y ce i ica es
16
o Uni ed Kingdom: NVQ le el 4, e.g. uni e si y ce i ica es o nu sing de-
g ees;
o Ge many: uni e si ies o applied sciences (Fachhochschule), and
• uni e si y g adua es.
I limi he compa ison o coho a ia ion by including su ey yea - egion ixed-e ec s in o my
model. Such ixed e ec s id my es ima es o any pe iod and egion a ia ion. This s a egy
he e o e holds changes on he demand side cons an (i hey do no di e ac oss coho s) and
allows me o ocus in e p e a ion on supply-side changes, i.e., he ole o educa ional expan-
sion. I also con ols o supply-side con ounde s such as immig a ion. I u he discuss he
assump ions o his app oach in he Online Appendix C.
In addi ion o hese ixed-e ec s, I adjus o a ange o o he a iables ha migh con ound
es ima es o he e ec s o educa ional expansion. A he indi idual le el, I con ol o espond-
en s’ mig a ion backg ound (Ge many), o hei e hnic g oup (Uni ed Kingdom), espec i ely,
and an in e ac ion o gende wi h hei pa ne ship s a us. Gende -speci ic esul s can be ound
in Sec ion D o he Online Supplemen . A he pe iod- egion-coho le el, I adjus o he size
o wo ke s’ ex ended coho (as abo e de ined as people bo n wi hin +- 3 yea s om he base
yea ), ela i e o he en i e wo king age popula ion 30 o 65 in ha yea in ha egion. This is
o accoun o he possibili y ha membe s o ela i ely la ge coho s migh ace inc eased
compe i ion in accessing ma ching jobs. In Sec ion F o he Online Supplemen , I epo spec-
i ica ion cu es o he main esul s, which show he ull ange o es ima es ob ained o all
plausible speci ica ions.
Analy ical s a egy
In o de o es ima e he ela ionship be ween educa ional expansion, measu ed as he ex-
ended coho -speci ic sha e o quali ica ion 𝑄𝑙, and he p obabili y o be misma ched, I use he
ollowing andom-e ec s linea p obabili y model:
𝑝(𝑀𝑀𝑖,𝑟,𝑐,𝑡 =1)=∑𝛼𝑙𝑄𝑟,𝑐,𝑡,𝑙
𝑙=𝐿
𝑙=1 + ∑(𝛽𝑘𝑋𝑘,𝑖)
𝑘=𝐾
𝑘=1 +δ𝑟×𝑡 + 𝑢𝑐.
(Eq. 1)
17
This model es ima es he p obabili y ha an indi idual 𝑖, membe o coho 𝑐, su eyed in e-
gion 𝑟 a his o ical ime 𝑡 will be misma ched. In his model, δ𝑟×𝑡 ep esen s he pe iod- egion
ixed-e ec s, 𝑢𝑐 is a coho -le el andom e ec o accoun o he clus e ing o obse a ions
and measu es wi hin coho s, and he e m ∑(𝛽𝑘𝑋𝑘,𝑖)
𝑘=𝐾
𝑘=1 ep esen s 𝐾 con ol a iables includ-
ing indi idual educa ion. I es ima e Eq. 1 sepa a ely o unde - and o e quali ica ion as he
misma ch ou come. When es ima ing he p obabili y o o e quali ica ion, unde quali ied wo k-
e s o m pa o he e e ence ca ego y o non-o e quali ied wo ke s, and ice e sa. Wo ke s
who canno indica e a misma ch in he su ey ques ionnai e (uni e si y g adua es, o ins ance,
canno be o mally unde quali ied and people wi hou quali ica ions canno be o e quali ied)
a e excluded om he espec i e models. The quan i y o in e es in bo h kinds o models a e
he 𝛼𝑙, he pa ial ela ionship be ween he coho -speci ic sha e o quali ica ion 𝑄𝑙 in 𝑟, a 𝑡 (𝑙
indexes he quali ica ion le els) and he co esponding unde - o o e quali ica ion p obabili y.
The Online Supplemen C discusses he p incipal condi ions unde which a causal in e p e a-
ion can be assigned o his coe icien . The e a e wo mo e speci ic h ea s o iden i ying he
causal e ec s o educa ional expansion in my amewo k: endogenous mig a ion in esponse
o labo ma ke condi ions and sample selec ion bias a ising om selec i e unemploymen . I
desc ibe my s a egy o ackle hese issues, and he esul ing es ima es in Appendix G o he
Online Supplemen .
4. Resul s
Educa ional expansion, occupa ional upg ading, and misma ch
ends
Be o e I u n o he ques ion o how a es o e ical misma ch ha e e ol ed, Figu e 1 gi es a
desc ip i e imp ession o he supply o quali ica ions, he change o he occupa ional s uc u e,
and he esul ing demand o quali ica ions on B i ish and Ge man labo ma ke s. Panel A o
Figu e 1 akes a bi h-coho pe spec i e and shows he highes quali ica ion epo ed by 30-
yea -olds in la ge-scale labo o ce su eys. I documen s signi ican educa ional expansion in
18
he Uni ed Kingdom, and, o a lesse deg ee, in Ge many o coho s bo n be ween he mid-
1950s and he mid-1980s. While oughly a i h o a coho bo n in he 1950s has unde gone
e ia y educa ion in bo h socie ies, he igu e o Ge many s ood a 25% in he 1983 coho
and a mo e han 35% in he same bi h-coho in he Uni ed Kingdom. A second di e ence
be ween he wo coun ies is ha Ge man expansion was limi ed o he op, i.e., o e ia y
educa ion, wi h he sha e o hose wi hou any oca ional quali ica ions no showing any clea
end. B i ish educa ion, on he o he hand, expanded a all le els, he eby d ama ically de-
c easing he sha e o hose wi hou o mal oca ional educa ion om o e 60% o a coho bo n
in he 1950s o a ound 30% o coho s bo n in he 1980s. This igu e is, howe e , s ill mo e
han wice as high as he sha e o 30-yea -old Ge mans wi hou oca ional quali ica ions.
Panel B o Figu e 1 changes pe spec i es and shows he e alua ion o he occupa ional s uc-
u e o he Ge man and he B i ish Labo ma ke s o e his o ical ime. I uses he i s digi o
he ISCO88 classi ica ion o de ine occupa ional g oups. In bo h coun ies, highe skilled occu-
pa ions (manage s, p o essionals, echnicians) expanded, while lowe -skilled ones (c a wo k-
e s, machine ope a o s, elemen a y occupa ions) con ac ed be ween he 1980s and he
2010s. Howe e , he e a e di e ences, when i comes o he o e all equency o occupa ions:
by he end o he 2010s, o ins ance, I es ima e he sha e o wo ke s in he manage ca ego y
o be app oxima ely 3.3 imes highe in he Uni ed Kingdom han in Ge many, (20% s. 6%),
while he opposi e is ue o echnicians (15% s 25%). The e is hus clea e idence o occu-
pa ional upg ading and no e idence o pola iza ion—a leas a he ela i ely coa se le el o
occupa ional g oups—in bo h coun ies, bu he desc ip i e analysis also documen s pe sis en
di e ences in he composi ion o he wo labo ma ke s.
Panel C mo es o he co e o my analysis and shows he quali ica ions wo ke s epo ed we e
needed o hei cu en job in a gi en su ey-yea . I p esen s e idence ha he occupa ional
upg ading appa en in Panel B ansla ed in o a signi ican ly inc easing demand o (highe )
quali ica ions on bo h B i ish and Ge man labo ma ke s. This is ue e en hough my subjec-
i e measu e o misma ch likely unde es ima es he ex end o ac ual demand shi s, as quali i-
ca ion equi emen s o un illed acancies a e no coun ed.
19
In Ge many, he sha e o jobs equi ing no o mal quali ica ions a all has hal ed be ween 1984
and 2018, om mo e han 44% o 22%, while e ia y-educa ion- equi emen s expanded om
11% o jobs o 26%. This shows ha , agains RBTC-a gumen s, demand upg ading was linea :
no only did demand o highly quali ied wo ke s g ow, bu he ela i e demand o no o mally
quali ied wo ke s sh ank. In e es ingly, his is di e en om he o m o educa ional expansion.
Panel A documen s ha i was limi ed o e ia y deg ees. This hin s a a i s sou ce o unde -
expansion in he Ge man case: he collapsing demand o un ained wo ke s.
In he Uni ed Kingdom, no oca ional educa ion beyond seconda y school diplomas was e-
qui ed o ge in o hal o jobs in 1986, in 2017, his was only ue o a hi d. A e ia y-educa ion
equi emen ose om a qua e o jobs in he 1980s o jus unde hal in 2017. In con as o
Ge many, hese wo ends imply ha he demand o middling pos -seconda y, non- e ia y
oca ional quali ica ions declined ma kedly, om 37% o 18% in 2017. In line wi h Oesch and
Pici o (2019), and in con as o he ends a he le el o coa se occupa ional g oups in Panel
B, he Uni ed Kingdom he e o e shows some signs o pola ized upg ading.
20
Figu e 1: Educa ional expansion, occupa ional change, and upg ading o jobs’ quali ica ion
equi emen s in he Uni ed Kingdom and in Ge many
Sou ce: Own calcula ion based on GMZ 1976-2013, UKLFS 1979-2017, GSOEP 1984-2017,
and UKSESS 1986-2017. The i s ound o he UKSESS was no ca ied ou in all egions.
1986 igu es a e hence no s ic ly compa able o la e pe iods.
21
Combining demand- and supply-side pe spec i e, I now mo e o an examina ion o misma ch
a es. I i s examine wha desc ip i e ends in misma ch p e alence can ell us abou he
alidi y o he wo uni e salis heo e ical pe spec i es, be o e I u n o a o mal assessmen o
he ela ionship be ween expansion and misma ch a he coho le el.
Figu e 2 shows he o e all and coho -speci ic end in sel - epo ed misma ch a es o e su -
ey-yea s in he wo coun ies and he eby o e s a i s desc ip i e assessmen , limi ing he
analysis o wo ke s be ween he ages o 30 and 60. I documen s ha misma ch ends in he
Uni ed Kingdom and in Ge many ha e been s ikingly di e en . While he e a e signs o o e -
expansion in he Uni ed Kingdom, Ge many’s labo ma ke seems o be mo ing in o a mild
quali ica ion sho age. O e all, o e quali ica ion a es ose in he Uni ed Kingdom bu declined
in Ge many. The opposi e is ue o unde quali ica ion. A second no ewo hy obse a ion is
ha he e a e only small di e ences be ween coho s (lines in g ay) in Ge many, whe eas
B i ish coho s ace as ly di e en si ua ions, e en a he same poin in ime. Respec i e
younge coho s aced he highes isk o be o e quali ied, and he lowes isk o be unde qual-
i ied du ing almos all su ey yea s. The ising a e o o e quali ica ion and he declining a e
o unde quali ica ion hus appea o be mainly d i en by coho eplacemen . T ends in Ge -
many, on he o he hand, appea o ake place mainly a he pe iod le el, wi h misma ch a es
o di e en coho s being almos indis inguishable. I any hing, younge coho s appea o show
lowe sel -assessed o e quali ica ion a es om he la e 2000s onwa d. No e ha coun y di -
e ences, hus, e ol e in opposing di ec ions o any possible bias in oduced by he sligh ly
di e ing i em wo ding in he Uni ed Kingdom and in Ge many.
22
Figu e 2 sugges s ha o e - and unde expansion a e highly con ex -dependen phenomena.
Howe e , i does no accoun o he ac ha wi h educa ional expansion, a highe sha e o
wo ke s becomes eligible o o e quali ica ion in he i s place, and likewise is no longe a isk
o unde quali ica ion. Mo eo e , a his poin , i is unclea whe he misma ch ends a e p e-
dominan ly d i en by he demand o he supply side. I hus now u n o models desc ibed in
Eq. 1 ha link o e - and unde quali ica ion incidence o coho -le el quali ica ion sha es — he
main indica o o educa ional expansion.
Figu e 2 T ends in misma ch a es in he Uni ed Kingdom and in Ge many
23
E idence om eg ession models
Table 1 p o ides a global assessmen o he link be ween educa ional expansion o e bi h-
coho s and coho le el-o e -/unde quali ica ion incidence. I answe s he co e ques ion a
s ake in he o e expansion deba e whe he we obse e mo e o less o e -/unde quali ica ion
in coho s whe e highe le els o educa ion a e mo e widesp ead. Using a concep ually di e -
en app oach, i con i ms he signi ican di e ence in he ela ionship be ween expansion and
misma ch in he wo coun ies sugges ed by Figu e 1: o e expansion in he Uni ed Kingdom,
bu an o e all balance in Ge many. Table 1 sugges s ha B i ish expansion a bo h he lowe
and he uppe end o he quali ica ion hie a chy was only pa ially abso bed by labo ma ke s.
E en ne o own quali ica ions, and ne o pe iod- egion di e ences in demand, o e quali ica-
ion is highe in egion-coho s, whe e he sha e o uppe e ia y and seconda y educa ed
wo ke s is highe . Expansion also se e ely limi ed unde quali ica ion: his ype o misma ch is
d ama ically lowe in egion-coho s whe e uppe e ia y and seconda y educa ion is mo e
p e alen . Expansion o middling quali ica ion le els, howe e , is no associa ed wi h changed
misma ch a es.
I do no ind any e idence o a compa able o e expansion in he Ge man da a. On he con a y,
column 3 shows ha in Ge many o e quali ica ion dec eased in coho egions, whe e e ia y
educa ion inc eased. The e ec sizes I ind a e la ge: I es ima e ha a 10-pe cen age poin
inc ease in he numbe o g adua es has his o ically been associa ed wi h a 2.6-pe cen age
poin inc ease in o e quali ied and a 5-pe cen age poin educ ion in unde quali ied wo ke s in
he Uni ed Kingdom. In Ge many, o e quali ica ion educed a mo e han hal he a e a which
e ia y educa ion expanded.
24
Uni ed Kingdom
Ge many
O e quali ica ion
Unde quali ica ion
O e quali ica ion
Unde quali ica ion
(1)
(2)
(3)
(4)
Expansion o …
…seconda y
0.192*
-0.195*
(1.98)
(-2.29)
…lowe oc.
-0.264
0.151
(-0.96)
(0.61)
… oca ional
0.0293
0.0953
0.173
0.0636
(0.14)
(0.54)
(1.14)
(0.61)
…lowe e .
-0.175
0.382
(-0.58)
(1.40)
… e ia y
0.259*
-0.495***
-0.535***
0.0879
educa ion
(2.41)
(-4.89)
(-3.55)
(0.85)
Indi idual con ols
Yes
Yes
Yes
Yes
Coho con ols
Yes
Yes
Yes
Yes
N
17878
16560
21489
17971
NYea s
7
7
29
29
NRegions
12
12
10
10
NCoho s
61
61
56
56
Va iance componen s
Va In e cep
0.0000668
2.22e-20
1.51e-15
3.18e-17
Va Residual
0.227
0.170
0.153
0.0610
s a is ics in pa en heses. Coho andom e ec s. Indi idual con ols: In e ac ion gende and pa ne ship s a us, and
e hnici y. Coho con ols: ela i e size o coho . Resul s con olled o pe iod- egion ixed-e ec s.
* p < 0.05, ** p < 0.01, *** p < 0.001
Table 1 Rela ionship be ween educa ional expansion and quali ica ion misma ch, RE-LPM
esul s
Table 1 epo s esul s, when all quali ica ion sha es a e en e ed simul aneously, omi ing he
“no quali ica ion” ca ego y. Resul s o using o he e e ences ca ego ies can be ound in Sec-
ion E o he Online Supplemen . These addi ional es ima es unde line he obus ness o my
esul s. As I discuss in mo e de ail in he Online Supplemen C, no causal in e p e a ion can
be applied o he Ge man esul s. A alid, bu mo e cau ious in e p e a ion o columns 3 and 4
in Table 1 is ha e ia y expansion in Ge many was no enough o s abilize a es o o e qual-
i ica ion, which we e alling o o he easons. E en wi hou a causal in e p e a ion, his pa e n
ules ou he o e expansion hesis o he Ge man case.
How did he appa en o e expansion o B i ish seconda y and e ia y educa ion play ou
ac oss di e en le els o educa ion? And wha kinds o wo ke s ound i easie o mo e ou o
o e quali ica ion in Ge many? Figu es 3 and 4 b eak down he associa ion be ween expansion
31
A hi d con ibu ion o his s udy is o ake a holis ic app oach by s udying misma ch and ex-
pansion ac oss le els o educa ion. This pe spec i e a o ds new insigh s. I ind ha e ia y
expansion in he Uni ed Kingdom was no associa ed wi h o e quali ica ion o uni e si y g ad-
ua es, bu o g adua es o lowe e ia y ins i u ions and oca ional p og ams – sugges ing a
labo queue model in which expansion a he op uels o e quali ica ion among wo ke s wi h
middling quali ica ions, as “ hei ” jobs a e inc easingly illed wi h g adua es. My analysis o he
Ge man da a sugges s ha any possible unde expansion is no o be ound among he mos
highly quali ied wo ke s, uni e si y g adua es. Supplemen a y analyses e en yielded some
e idence ha g adua es’ access o execu i e posi ions has become somewha ha de wi h
e ia y expansion. Ins ead, unde expansion seems o be loca ed among wo ke s wi h (ad-
anced) oca ional quali ica ions. I his in e p e a ion is co ec , educa ional policy in Ge many
should ocus on sho ing up he oca ional sec o , a he han expanding uni e si y access u -
he .
Re e ences
Acemoglu, Da on, and Da id Au o . 2011. “Skills, Tasks and Technologies: Implica ions o
Employmen and Ea nings.” Pp. 1043–1171 in Handbook o Labo Economics, edi ed
by O. Ashen el e , R. Laya d, and D. Ca d. Ams e dam; New Yo k: Else ie .
———. 2012. “Wha Does Human Capi al Do? A Re iew o Goldin and Ka z’s The Race be-
ween Educa ion and Technology.” Jou nal o Economic Li e a u e 50(2): 426–63.
Allen, Jim, and Rol an de Velden. 2001. “Educa ional Misma ches e sus Skill Misma ches:
E ec s on Wages, Job Sa is ac ion, and on- he-Job Sea ch.” Ox o d Economic Pape s
53(3): 434–52.
Au o , Da id H, and Da id Do n. 2013. “The G ow h o Low-Skill Se ice Jobs and he Pola i-
za ion o he US Labo Ma ke .” Ame ican Economic Re iew 103(5): 1553–97.
Au o , Da id H., F ank Le y, and Richa d J. Mu nane. 2003. “The Skill Con en o Recen
Technological Change: An Empi ical Explo a ion.” The Qua e ly Jou nal o Economics
118(4): 1279–1333.
Bake , Da id. 2014. The Schooled Socie y: The Educa ional T ans o ma ion o Global Cul u e.
S an o d Uni e si y P ess.
Bol, T. 2015. “Has Educa ion Become Mo e Posi ional? Educa ional Expansion and Labou
Ma ke Ou comes, 1985-2007.” Ac a Sociologica 58(2): 105–20.
Bol, Thijs, and Kim A. Weeden. 2015. “Occupa ional Closu e and Wage Inequali y in Ge many
and he Uni ed Kingdom.” Eu opean Sociological Re iew 31(3): 354–69.
Bou dieu, Pie e, and Jean-Claude Passe on. 1970. La Rep oduc ion: Élémen s pou une
héo ie du sys ème d’enseignemen . Minui .
Bowles, Samuel, and He be Gin is. 1977. Schooling In Capi alis Ame ica: Educa ional Re-
o m And The Con adic ions O Economic Li e. New edi ion. New edi ion. New Yo k:
Basic Books.
Bu is, Val. 1983. “The Social and Poli ical Consequences o O e educa ion.” Ame ican Soci-
ological Re iew 48(4): 454.
32
Busemeye , Ma ius R. 2014. “The Poli ics o Educa ion and T aining Re o m.” Pp. 58–122 in
Skills and Inequali y: Pa isan Poli ics and he Poli ical Economy o Educa ion Re o ms
in Wes e n Wel a e S a es. Camb idge: Camb idge Uni e si y P ess.
Cappelli, Pe e H. 2015. “Skill Gaps, Skill Sho ages, and Skill Misma ches E idence and A -
gumen s o he Uni ed S a es.” ILR Re iew 68(2): 251–90.
Capsada-Munsech, Que al . 2015. “The Role o Social O igin and Field o S udy on G adua es’
O e educa ion: The Case o I aly.” Highe Educa ion 69(5): 779–807.
———. 2019. “Measu ing O e educa ion: Incidence, Co ela ion and O e laps Ac oss Indica-
o s and Coun ies.” Social Indica o s Resea ch 145(1): 279–301.
Ca d, Da id, and John E. DiNa do. 2002. “Skill-Biased Technological Change and Rising
Wage Inequali y: Some P oblems and Puzzles.” Jou nal o Labo Economics 20(4):
733–83.
Ca ne ale, An hony P, and Donna M Des oche s. 2002. “The Missing Middle:Aligning Educa-
ion and he Knowledge Economy.” 25(1).
Ca ne ale, An hony P, and S ephen J Rose. 2014. “The Unde educa ed Ame ican.”
Geo ge own Uni e si y Cen e on Educa ion and he Wo k o ce.
Collins, Randall. 1979. The C eden ial Socie y : An His o ical Sociology o Educa ion and S a -
i ica ion. New Yo k : Academic P ess.
C oce, Giuseppe, and Emanuela Ghignoni. 2012. “Demand and Supply o Skilled Labou and
O e educa ion in Eu ope: A Coun y-Le el Analysis.” Compa a i e Economic S udies
54(2): 413–39.
Da ia, Ma ia A., Seamus McGuinness, and Philip J. O’Connell. 2017. “De e minan s o Re-
gional Di e ences in Ra es o O e educa ion in Eu ope.” Social Science Resea ch 63:
67–80.
Delaney, Judi h, Seamus McGuinness, Kons an inos Pouliakas, and Paul Redmond. 2020.
“Educa ional Expansion and O e educa ion o Young G adua es: A Compa a i e Anal-
ysis o 30 Eu opean Coun ies.” Ox o d Re iew o Educa ion 46(1): 10–29.
Di Pie o, Gio gio. 2002. “Technological Change, Labo Ma ke s, and ‘Low-Skill, Low-Technol-
ogy T aps.’” Technological Fo ecas ing and Social Change 69(9): 885–95.
Di S asio, Valen ina. 2017. “Who Is Ahead in he Labo Queue? Ins i u ions’ and Employe s’
Pe spec i e on O e educa ion, Unde educa ion, and Ho izon al Misma ches.” Sociol-
ogy o Educa ion 90(2): 109–26.
Di S asio, Valen ina, Thijs Bol, and He man G. Van de We ho s . 2016. “Wha Makes Educa-
ion Posi ional? Ins i u ions, O e educa ion and he Compe i ion o Jobs.” Resea ch in
Social S a i ica ion and Mobili y 43: 53–63.
Dus mann, Ch is ian, Johannes Luds eck, and U a Schönbe g. 2009. “Re isi ing he Ge man
Wage S uc u e.” Qua e ly Jou nal o Economics 124(2): 843–81.
Elbe s, Benjamin, Thijs Bol, and Thomas A DiP e e. 2021. “T aining Regimes and Skill Fo -
ma ion in F ance and Ge many An Analysis o Change Be ween 1970 and 2010.” So-
cial Fo ces 99(3): 1113–45.
Esping-Ande sen, Gos a. 1999. Social Founda ions o Pos indus ial Economies. Ox o d Uni-
e si y P ess.
Fels ead, Gallie, G een, and Inanc. 2014. “Skills and Employmen Su eys Se ies Da ase ,
1986, 1992, 1997, 2001, 2006 and 2012.” Colches e , Essex: UK Da a A chi e.
Fe nández-Macías, En ique, and John Hu ley. 2017. “Rou ine-Biased Technical Change and
Job Pola iza ion in Eu ope.” Socio-Economic Re iew 15(3): 563–85.
Fi zenbe ge , Be nd, Ka s en Kohn, and Alexande C. Lembcke. 2013. “Union Densi y and
Va ie ies o Co e age: The Ana omy o Union Wage E ec s in Ge many.” ILR Re iew
66(1): 169–97.
F eeman, Richa d. 1976. The O e educa ed Ame ican. Academic P ess.
Fu ey, Jane. 2021. “Rela i e o Whom? Commen on ‘Rela i e Educa ion and he Ad an age
o a College Deg ee.’” Ame ican Sociological Re iew 86(5): 1000–1010.
Geißle , Raine . 2014. Die Sozials uk u Deu schlands. Wiesbaden: Sp inge Fachmedien
Wiesbaden.
Glendinning, Richa d, Vi Young, Alexand a Bogdan, and G K UK Social Resea ch. 2018.
Skills and Employmen Su ey 2017. Technical Repo .
33
Goebel, Jan, Ma kus M. G abka, S e an Liebig, Ma in K oh, Da id Rich e , Ca s en Sch öde ,
and Jü gen Schupp. 2019. “The Ge man Socio-Economic Panel (SOEP).” Jah büche
ü Na ionalökonomie und S a is ik 239(2): 345–60.
Goldin, Claudia, and Law ence F. Ka z. 2010. The Race be ween Educa ion and Technology.
2/28/10 edi ion. 2/28/10 edi ion. Camb idge, Mass.: Belknap P ess.
Gosling, Amanda, and Thomas Lemieux. 2004. “Labo Ma ke Re o ms and Changes in Wage
Inequali y in he Uni ed Kingdom and he Uni ed S a es.” in Seeking a P emie Econ-
omy: The Economic E ec s o B i ish Economic Re o ms, 1980-2000, edi ed by D.
Ca d, R. Blundell, and R. B. F eeman. Uni e si y o Chicago P ess.
Hall, Pe e A., and Da id Soskice. 2001. “In oduc ion o Va ie ies o Capi alism.” Pp. 1–68 in
Va ie ies o Capi alism. The Ins i u ional Founda ions o Compa a i e Ad an age. Ox-
o d, New Yo k: Ox o d Uni e si y P ess.
Heisig, Jan Paul. 2018. “Measu ing he Signaling Value o Educa ional Deg ees: Seconda y
Educa ion Sys ems and he In e nal Homogenei y o Educa ional G oups.” La ge-scale
Assessmen s in Educa ion 6(1): 9.
Holze , Ha y. 2015. “Job Ma ke Pola iza ion and U.S. Wo ke Skills: A Tale o Two Middles.”
Economic S udies a B ookings.
Holze , Ha y J, and Robe I Le man. 2007. “Ame ica’s Fo go en Middle-Skill Jobs: Educa ion
and T aining Requi emen s in he Nex Decade and Beyond.” Wo k o ce Alliance.
———. 2009. “The Fu u e o Middle-Skill Jobs.”
Ho owi z, Jona han. 2018. “Rela i e Educa ion and he Ad an age o a College Deg ee.” Ame -
ican Sociological Re iew 83(4): 771–801.
House o Commons. 2012. Educa ion: His o ical S a is ics. London.
Kallebe g, A ne L. 2011. Good Jobs, Bad Jobs : The Rise o Pola ized and P eca ious Em-
ploymen Sys ems in he Uni ed S a es, 1970s-2000s. New Yo k: Russell Sage Foun-
da ion.
Ka z, L. F., and K. M. Mu phy. 1992. “Changes in Rela i e Wages, 1963-1987: Supply and
Demand Fac o s.” The Qua e ly Jou nal o Economics 107(1): 35–78.
Ko pi, Tomas, and Michael Tåhlin. 2009. “Educa ional Misma ch, Wages, and Wage G ow h:
O e educa ion in Sweden, 1974–2000.” Labou Economics 16(2): 183–93.
K is al, Tali, and Yinon Cohen. 2017. “The Causes o Rising Wage Inequali y: The Race be-
ween Ins i u ions and Technology.” Socio-Economic Re iew 15(1): 187–212.
Maye , Ka l Ul ich, Wal e Mülle , and Reinha d Pollak. 2007. “Ge many: Ins i u ional Change
and Inequali ies o Access in Highe Educa ion.” in S a i ica ion in Highe Educa ion:
A Compa a i e S udy, edi ed by Y. Sha i , R. A um, and A. Gamo an. S an o d, Cali-
o nia: S an o d Uni e si y P ess.
McGuinness, Seamus, Adele Be gin, and Adele Whelan. 2018. “O e educa ion in Eu ope:
T ends, Con e gence, and D i e s.” Ox o d Economic Pape s 70(4): 994–1015.
Mülle , Wal e , and Ma kus Gangl. 2003. T ansi ions om Educa ion o Wo k in Eu ope: The
In eg a ion o You h In o Eu Labou Ma ke s. Ox o d ; New Yo k: Ox o d Uni P .
Nolan, B ian, Wieme Sal e da, Daniele Checchi, I e Ma x, Abigail McKnigh , Is án Gyö gy
Tó h, and He man G. an de We ho s (eds.). 2014. Changing Inequali ies and Socie-
al Impac s in Rich Coun ies: Thi y Coun ies’ Expe iences. Ox o d, New Yo k: Ox o d
Uni e si y P ess.
OECD. 2019. Educa ion a a Glance 2019: OECD Indica o s. OECD.
Oesch, D., and J. Rod iguez Menes. 2011. “Upg ading o Pola iza ion? Occupa ional Change
in B i ain, Ge many, Spain and Swi ze land, 1990-2008.” Socio-Economic Re iew 9(3):
503–31.
Oesch, Daniel, and Gio gio Picci o. 2019. “The Pola iza ion My h: Occupa ional Upg ading in
Ge many, Spain, Sweden, and he UK, 1992–2015.” Wo k and Occupa ions 46(4):
441–69.
Pollmann-Schul , Ma hias, and Ka l Ul ich Maye . 2004. “Re u ns o Skills: Voca ional T aining
in Ge many 1935-2000.” Yale Jou nal o Sociology 4: 24.
Powell, Jus in J. W., and Heike Solga. 2011. “Why A e Highe Educa ion Pa icipa ion Ra es
in Ge many so Low? Ins i u ional Ba ie s o Highe Educa ion Expansion.” Jou nal o
Educa ion and Wo k 24(1–2): 49–68.
34
Rossen, Anja, Ch is ina Boll, and And é Wol . 2019. “Pa e ns o O e educa ion in Eu ope:
The Role o Field o S udy.” IZA Jou nal o Labo Policy 9(1).
Shin, Dong-Hoon, and Da id Bills. 2021. “T ends in Educa ional and Skill Misma ch in he
Uni ed S a es.” Social Sciences 10(10): 395.
Thu ow, Les e C. 1975. Gene a ing Inequali y: Mechanisms o Dis ibu ion in he U.S. Econ-
omy. Basic Books.
Tinbe gen, Jan. 1956. “On he Theo y o Income Dis ibu ion.” Wel wi scha liches A chi 77:
22.
Vaisey, S ephen. 2006. “Educa ion and I s Discon en s: O e quali ica ion in Ame ica, 1972–
2002.” Social Fo ces 85(2): 835–64.
Ve haes , Die e , and Rol Van de Velden. 2013. “C oss-Coun y Di e ences in G adua e
O e educa ion.” Eu opean Sociological Re iew 29(3): 642–53.
Wagne , Ge G., Joachim R. F ick, and Jü gen Schupp. 2007. “The Ge man Socio-Economic
Panel S udy (SOEP)-E olu ion, Scope and Enhancemen s.”
Weeden, Kim A., and Da id B. G usky. 2014. “Inequali y and Ma ke Failu e.” Ame ican Be-
ha io al Scien is 58(3): 473–91.
Wiedne , Jonas. 2021. “Poli ical and Social Consequences o Quali ica ion Misma ches: A
Bounding App oach o S a us Inconsis ency.” Social Fo ces (soab120).
Wol , Alison. 2002. Does Educa ion Ma e ? My hs Abou Educa ion and Economic G ow h.
Penguin Books Limi ed.
W en, Anne (ed.). 2013. The Poli ical Economy o he Se ice T ansi ion. Ox o d, UK: Ox o d
Uni e si y P ess.
