Wilke, Ch is ina Beni a (Ed.); FOM Hochschule ü Oekonomie & Managemen ,
Kompe enzCen um ü angewand e Volkswi scha sleh e (KCV) (Ed.)
P oceedings
Pension Re o ms in he No dic Coun ies and Ge many: Con e ence P oceedings o he 2024 Annual
Mee ing o he Ge man Socie y o Demog aphy (DGD) by he Wo king G oup on Demog aphic and
Social De elopmen s
KCV Sch i en eihe de FOM, No. 8
P o ided in Coope a ion wi h:
KCV Kompe enzCen um ü angewand e Volkswi scha sleh e, FOM Hochschule ü Oekonomie
& Managemen
Sugges ed Ci a ion: Wilke, Ch is ina Beni a (Ed.); FOM Hochschule ü Oekonomie & Managemen ,
Kompe enzCen um ü angewand e Volkswi scha sleh e (KCV) (Ed.) (2024) : Pension Re o ms
in he No dic Coun ies and Ge many: Con e ence P oceedings o he 2024 Annual Mee ing o
he Ge man Socie y o Demog aphy (DGD) by he Wo king G oup on Demog aphic and Social
De elopmen s, KCV Sch i en eihe de FOM, No. 8, ISBN 978-3-89275-387-2, MA Akademie Ve lags-
und D uck-Gesellscha mbH, Essen
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/307058
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
KCV Kompe enzCen um
ü angewand e Volkswi scha sleh e
de FOM Hochschule ü Oekonomie & Managemen
KCV Sch i en eihe
Vol.
8
Pension Re o ms in he No dic Coun ies and
Ge many
Con e ence P oceedings o he 2024 Annual
Mee ing o he Ge man Socie y o Demo-
g aphy (DGD) by he Wo king G oup on
Demog aphic and Social De elopmen s
Ch is ina Beni a Wilke (Ed.)
Ma hias Die meie / Ma in D ees / Ila i Ilmakunnas / Ja i Kannis o / Sa u Ni alainen /
Ru h Ma ia Schüle / Mika Vidlund
Pension Re o ms in he No dic Coun ies and Ge many
Con e ence P oceedings o he 2024 Annual Mee ing o he Ge man Socie y o Demog aphy (DGD)
by he Wo king G oup on Demog aphic and Social De elopmen s
KCV Sch i en eihe de FOM, Vol. 8
Essen 2024
ISBN (P in ) 978-3-89275-386-5 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-387-2 ISSN (eBook) 2701-9411
Dieses We k wi d he ausgegeben om KCV Kompe enzCen um ü angewand e Volkswi scha sleh e
de FOM Hochschule ü Oekonomie & Managemen gGmbH
Ve lag:
MA Akademie Ve lags- und D uck-Gesellscha mbH, Leimkugels aße 6, 45141 Essen
[email p o ec ed]
Die Deu sche Na ionalbiblio hek e zeichne diese Publika ion in de Deu schen Na ionalbibliog aphie;
de aillie e bibliog aphische Da en sind im In e ne übe h p://dnb.d-nb.de ab u ba .
Dieses We k is lizenzie un e CC BY 4.0:
C ea i e Commons Namensnennung 4.0 In e na ional.
Diese Lizenz e laub un e den Vo ausse zungen de Lizenzbedingungen, u. A. de Namens-
nennung de U hebe in ode des U hebe s, de Angabe de CC-Lizenz (inkl. Link) und de gg .
o genommenen Ände ungen die Bea bei ung, Ve iel äl igung und Ve b ei ung des Ma e ials
in jedem Fo ma ode Medium ü beliebige Zwecke. Die Rech e und P lich en in Zusam-
menhang mi de Lizenz e geben sich ausschließlich aus dem Lizenzinhal : CC BY 4.0 Deed |
Namensnennung 4.0 In e na ional | C ea i e Commons | h ps://c ea i ecommons.o g/licenses/
by/4.0/legalcode.de.
Die Bedingungen de C ea i e-Commons-Lizenz gel en nu ü O iginalma e ial. Die Wiede -
e wendung on Ma e ial aus ande en Quellen (gekennzeichne mi Quellenangabe) wie z. B.
on Schaubilde n, Abbildungen, Fo os und Tex auszügen e o de gg . wei e e Nu zungsge-
nehmigungen du ch den jeweiligen Rech einhabe .
Ch is ina Beni a Wilke (Ed.)
Pension Re o ms in he No dic Coun ies
and Ge many
Con e ence P oceedings o he 2024 Annual Mee ing o
he Ge man Socie y o Demog aphy (DGD) by he
Wo king G oup on Demog aphic and
Social De elopmen s
Ma hias Die meie / Ma in D ees / Ila i Ilmakunnas / Ja i Kan-
nis o / Sa u Ni alainen / Ru h Ma ia Schüle / Mika Vidlund
Con ac In o ma ion
P o . D . Ch is ina Beni a Wilke
FOM Uni e si y o Applied Sciences
P o esso o Economics
Scien i ic Di ec o o he Compe ence Cen e o Applied Economics
(KCV Kompe enzCen um ü angewand e Volkswi scha sleh e)
Head o Academic S udies, FOM Uni e si y Cen e B emen
ch is ina.wilke@ om.de
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
III
Fo ewo d by he Chai s o he Wo king G oup on Demog aphic and Social
De elopmen s o he Ge man Socie y o Demog aphy (DGD)
Dea eade s,
This olume documen s exce p s om ou session on pension e o ms in he No -
dic coun ies and Ge many a he Annual Mee ing o he DGD a he Uni e si y o
Hambu g in Ma ch 2024.
Pension sys ems in Ge many as well as he No dic coun ies ha e been subjec
o subs an ial pension e o ms in he pas decades, o emos in o de o espond
o he demog aphic challenges hese na ions ace. Bo h egions a e expe iencing
aging popula ions, which pu s p essu e on public pension sys ems and equi es
inno a i e e o m s a egies o ensu e sus ainabili y. Al hough Ge many and he
No dic coun ies sha e a his o y o s ong wel a e s a es, hei app oaches o pen-
sions a y – o e ing aluable lessons in policy design, om he No dic coun ies’
a he lexible pension ages o Ge many’s e ol ing mul ipilla sys em. Looking
o wa d, he e is a need o u he e o ms ha adap o e ol ing demog aphic
ends, ensu e in e gene a ional ai ness, and main ain iscal sus ainabili y, while
also add essing new challenges such as digi aliza ion and changing wo k pa -
e ns.
Empi ical esea ch has ecei ed many new impulses om deba es on demo-
g aphic challenges like hose on pension sys ems. Ou Wo king G oup on Demo-
g aphic and Social De elopmen s o he Ge man Socie y o Demog aphy (DGD)
is aimed a expe ienced scien is s, doc o al s uden s, and p ac i ione s om ad-
minis a ion, poli ics, and business. I you a e in e es ed in joining us, please ha e
a look a ou websi e: h ps://dgd-online.de/a bei sk eise/demog a ische-und-ge-
sellscha liche-en wicklungen/. We a e always happy o welcome new membe s.
We would like o hank he FOM Uni e si y as well as he Compe ence Cen e o
Applied Economics (KCV Kompe enzCen um ü angewand e Volks-
wi scha sleh e) o he oppo uni y o publish he con e ence con ibu ions o ou
session in his an hology. A special hank you goes o he FOM Publica ion Team
and in speci ic o Ms. Sa ah Be ndsen o he ho ough o mal e ision o all pa-
pe s.
We wish you a s imula ing and exci ing ead!
B emen and Cologne, Oc obe 2024
P o . D . Ch is ina Beni a Wilke (Edi o ) D . Philipp Desche meie
DGD Wo king G oup Chai DGD Wo king G oup Chai
FOM Uni e si y o Applied Sciences, Ge man Economic Ins i u e,
Essen and B emen Cologne
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
IV
Table o Con en s
Fo ewo d by he chai s o he wo king g oup on demog aphic and social
de elopmen s o he Ge man Socie y o Demog aphy (DGD): .................... III
Abou he Edi o ................................................................................................ V
Abou he Au ho s ............................................................................................ VI
1 Expec ed E ec i e Re i emen Age and Exi age in he No dic Coun ies
Ja i Kannis o / Mika Vidlund ........................................................................... 1
2 Labo Ma ke E ec s o Inc easing S a u o y Re i emen Age: The 2017
Pension Re o m in Finland
Sa u Ni alainen / Ila i Ilmakunnas ............................................................... 25
3 Re o ming he Pension Sys em in Ge many – An Empi ical S udy o Re o m
A e sions
Ma hias Die meie / Ru h Ma ia Schüle .................................................... 61
4 Non inancial De ined Re u n (NDR) Pension F amewo k and a New
Pe spec i e on Pension Sys em Sus ainabili y
Ma in D ees ................................................................................................ 79
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
V
Abou he Edi o
P o . D . Ch is ina Beni a Wilke has been P o esso o Economics a he FOM
Uni e si y o Applied Sciences since 2016. She is also Head o Academic S udies
a he FOM Uni e si y Cen e B emen and Scien i ic Di ec o o he KCV Compe-
ence Cen e o Applied Economics. P e iously, she wo ked as a senio e-
sea che and managing di ec o a he Mannheim Resea ch Ins i u e o he Eco-
nomics o Ageing (MEA) as well as senio expe and managing di ec o o he
B emen b anch o he Hambu g Ins i u e o In e na ional Economics (HWWI). He
esea ch ocuses on demog aphic change, social and labou ma ke policy and
heal h economics.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
VI
Abou he Au ho s
D . e . pol. Ma hias Die meie is head o he esea ch uni Democ acy, Soci-
e y, Ma ke Economy since 2022, and since 2024 managing di ec o o IW Ge-
sellscha s o schung gGmbH. He ob ained his Bachelo ’s deg ee in economics
a he Uni e si y o S . Gallen and he Uni e sidad de San And és, A gen ina. He
comple ed his Mas e ’s in Economics a he Uni e si y o Zu ich and he Sciences
Po Pa is. F om 2015 o 2022 he was pe sonal assis an o he di ec o a he
Ge man Economic Ins i u e. He comple ed his Doc o a e a he Ins i u e o Poli i-
cal Science o he Uni e si y o Duisbu g-Essen (NRW School o Go e nance) in
2021.
Ma in D ees s a ed his PhD in Ma hema ics a he Uni e si y o Bonn in he
a ea o Chip Design in 2022. He has s udied ma hema ics a he Uni e si y o
Bonn. As a p i a e p ojec , he is wo king on pension sys em design wi h a heo-
e ic ocus.
D Ila i Ilmakunnas is cu en ly wo king as a senio esea che a he Finnish
Cen e o Pensions. His cu en esea ch ocuses on he e alua ion o pension
e o ms, pa ial e i emen , he adequacy o pensions, and he measu emen o
po e y. He ob ained a PhD om he Uni e si y o Tu ku in 2019.
Ja i Kannis o (Lic. Phil.) wo ks wi h s a is ics a he Planning Depa men o he
Finnish Cen e o Pensions (ETK). He s a ed in ETK in 1986. Fo he las 25
yea s he has wo ked as a de elopmen manage , and be o e ha as a s a is ician.
The ocus has always been on examining pensione s, e i emen , and wo king
ca ee s.
D Sa u Ni alainen is cu en ly wo king as an economis a he Finnish Cen e
o Pensions. She has a long and many-sided expe ience in pension- ela ed e-
sea ch. He a eas o expe ise include e i emen on an old-age pension, con in-
ued wo king, wo king ca ee s, pa ial old-age pension, and beha io al e ec s o
pension e o ms. She ob ained a PhD om he Uni e si y o Jy äskylä in 2011.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
VII
D . e . pol. Ru h Ma ia Schüle is economis o social secu i y sys ems and
income and weal h dis ibu ion wi h a esea ch ocus on pension secu i y a he
Ge man Economic Ins i u e since 2021. She holds a Mas e ’s deg ee in Econom-
ics om he Uni e si y o Müns e and a Mas e ’s deg ee in Economic His o y
om he Uni e si y o Lund, Sweden. She ob ained he Bachelo ’s deg ee in Go -
e nance and Public Policy a he Uni e si y o Passau. F om 2012 o 2018 she
was esea che a he i o Ins i u e Munich Cen e o Educa ional Economics. She
ob ained he Doc o a e in Economics a he Ludwig Maximilian Uni e si y o Mu-
nich in 2016. F om 2018 o 2021 she wo ked as an economis o educa ion and
poli ical pa icipa ion a he Ge man Economic Ins i u e.
Mika Vidlund (Lic.Soc.Sc.) wo ks as eam leade o he In e na ional Analysis
Team wi hin he Planning Depa men a he Finnish Cen e o Pensions (ETK),
which p oduces se ices needed in he implemen a ion and de elopmen o ea n-
ings- ela ed pensions. He has wo ked a he ETK since 1999, i s as a Re-
sea che and Special Ad ise ocusing on compa a i e pension esea ch. He has
also wo ked as an expe o Finland suppo ing he wo k o he Eu opean Com-
mission DG Employmen , Social A ai s, and Inclusion in 2010-2014. He has
g adua ed (M.Soc.Sc) om he Uni e si y o Tu ku in 2000 and has a Licen ia e
deg ee on Social Sciences om 2006.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
7
1.2 F ames o e i emen in he No dic coun ies
Re i emen age is he single mos impo an ac o in e i emen . Changes in pen-
sion ac s ha e, among o he hings, aised he e i emen age and closed ea ly
exi ou es in an a emp o pos pone e i emen , e en hough e o ms may ha e
long ansi ion pe iods.
The s a u o y ea ning- ela ed old-age pension sys ems in Finland, No way, and
Sweden a e lexible in e ms o he e i emen age, and he sys em gi es indi id-
uals mo e choice in hei e i emen decisions. Indi iduals ha e a igh o choose
when hey will e i e on an old-age pension, a e hey ha e eached he lowes
possible e i emen age (Figu e 1).
Possibili y o e i e lexibly on he ea nings- ela ed pension was es ablished in
Sweden in he 1999-2002 s uc u al pension e o m and in Finland in he 2005
pension e o m (see e.g., Ba , 2013a; 2013b). In Sweden he lowes e i emen
age was hen se a 61 and in Finland a 63 yea s o age. The p ocess o inc eas-
ing he e i emen ages wi hin he Finnish pension sys em was ini ia ed in 2017
and in he Swedish pension sys em in 2020. Bo h coun ies ha e linked e i e-
men age o li e-expec ancy (e.g., OECD, 2023; Ande sen, 2021b).
In No way he 2011 pension e o m in oduced a lexible wi hd awal age be ween
ages 62 and 75. In he old sys em he wi hd awal age was ixed a 67 yea s.
Re i emen a he age o 62 is an op ion o hose wi h pension en i lemen s a
leas as high as he minimum pension. Be o e he e o m only ce ain g oups had
access o ea ly e i emen om age 62, and one had o wi hd aw comple ely om
he labou ma ke o ecei e an ea ly e i emen pension (e.g., Ande sen 2021a;
Eu opean Commission, 2024). A e 2010 one could con inue wo king ull- ime
and s ill wi hd aw a pension a age 62. This has u ned ou o be a popula op ion.
One goal o he e o m was o p o ide good wo k incen i es o pensione s. Fo h-
coming e o m in 2026 will link he lowes e i emen age o he de elopmen o
li e-expec ancy (Minis y o Labou and Social Inclusion, 2023a).
In Denma k he mos impo an e o m has been o link he e i emen age o 67
o li e expec ancy in 2011. Ea ly e i emen is p o ided by a speci ic olun a y
ea ly e i emen scheme (i.e., VERP, e e løn). The e i emen age o he VERP
was aised om 2013 and onwa ds, and he leng h o ime ha i was possible o
ecei e VERP be o e he e i emen age o old-age pension was educed om
i e o h ee yea s. Ano he impo an e o m was he e o m o he disabili y pen-
sion scheme in 2012 (Kannis o & Vidlund, 2022).
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
8
In Iceland gene al e i emen age has been ixed o decades o 67 wi h a possi-
bili y o e i e ea ly a he age o 65. Howe e , unde he ac - egula ed manda o y
occupa ional pension scheme ea ly e i emen ules may a y be ween unds.
The no mal e i emen age is 67 bu he supplemen a y pension sa ings can be
claimed ea lies om he age o 60 (see e.g. IPFA 2024; Danielsson e al. 2023).
Re i emen age is 60 yea s also o seamen i hey ha e been wo king in his
occupa ion o a leas 25 yea s.
In all he No dic coun ies e i emen age o minimum pension se s a kind o no m
o e i emen . Many social insu ance bene i s a e also paid up o his age. Fo
example, in Sweden he e has been a s ong endency o claim pension a age
65 ega dless o he lexibili y in he e o med ea nings- ela ed pension sys em.
In Finland and Sweden e i emen age o minimum pension was 65 whe eas in
Iceland, No way and Denma k i was 67 in 2022. These ages a e linked o he
de elopmen o li e-expec ancy in Sweden, Finland and Denma k.
Figu e 1: Re i emen ages and li e-expec ancy in 2022
Sou ce: Finnish Cen e o Pensions; Li eindenma k; No wegian Labou and Wel a e Ad-
minis a ion; Social Insu ance Adminis a ion Iceland; Swedish Pensions Agency.
60
62 62 64 64
80
68
75
69 67
50
55
60
65
70
75
80
85
90
Iceland Sweden No way Finland Denma k
Lowe limi Highe limi Li e-expec ancy a age 50
67 No m
65 No m 67 No m 65 No m
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
9
1.3 Da a and me hods
1.3.1 De ini ion o employmen a es
The employmen a e is he pe cen age sha e o employed pe sons in he popu-
la ion o he same age. The e iew is based on he annual a e age alues o he
labou o ce su ey by Eu os a co e ing yea s 2013-2022. A pe son is consid-
e ed employed, when, du ing he su ey week, hey ecei ed a mone a y sala y
o inge bene i s o p o i o a leas an hou o gain ul employmen , o someone
who has been empo a ily o wo k. Mo e de ailed de ini ions a e a ailable om
he wo k o ce esea ch o Eu os a (2024).
1.3.2 De ini ion and calcula ion o he labou ma ke exi age
The labou ma ke exi age measu es he a e age age o exi om he labou
ma ke . The calcula ion is based on indi iduals who a e in he labou o ce a he
age o 50. The calcula ion is made by compa ing he labou o ce pa icipa ion o
e e y ollowing age g oup a e he age o 50 wi h he labou o ce pa icipa ion
a 50 yea s o age. By doing so, an expec ed exi age om he labou o ce is
ob ained.
50-yea -old indi iduals a e used as a baseline o he calcula ions because o he
assump ion ha he labou o ce pa icipa ion is a i s peak a he age o 50.
The calcula ion o he exi age is pe o med by using a s a ic me hod ha includes
da a o only one yea a he han compa ing wo yea s. The basis o he calcula-
ion is he changes in labou o ce pa icipa ion be ween successi e ages in he
gi en yea .
A s a ic me hod says no hing abou he a e age age when a ce ain coho wi h-
d aws om wo k; i shows a snapsho gi en he cu en cha ac e is ics o he
labou ma ke . The esul om he s a ic me hod shows he age a which people
who a e 50 yea s old would exi he wo k o ce i he labou o ce pa e ns e-
mained he same as in he gi en yea .
The da a is based on Eu os a employmen s a is ics (Labou Fo ce Su ey). A
minimum o one hou o wo k pe week is equi ed o be included in he labou
o ce. The labou o ce also consis s o pe sons who a e unemployed bu a aila-
ble o employmen . In hese s a is ics pensione s a e also employed i hey do a
li le wo k. Exi age measu es he endpoin o wo k. The e o e, he e i emen
age in calcula ion o exi age usually ises highe han e ec i e e i emen age,
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
10
because i also includes he wo king ime a e i emen . I also makes compa i-
sons be ween coun ies mo e di icul , as he impo ance o wo k in he li elihoods
o pensione s a ies om coun y o coun y.
Resul s a e ob ained by in e pola ing labou o ce pa icipa ion a es in one-yea
age g oups om i e-yea age g oups da a. I we used one-yea age g oups in-
s ead o in e pola ion, he andom a iabili y in he esul s could be g ea e .
The ma hema ical desc ip ion o he calcula ion o exi age can be ound om
he Appendix 1.1.
1.3.3 De ini ion and calcula ion o expec ed e ec i e e i emen age
The expec ed e ec i e e i emen age is calcula ed in a way ha is simila o he
me hod used in calcula ing a e age emaining li e expec ancy. The indica o de-
sc ibes he e ec i e e i emen age, assuming ha he e i emen isk and mo -
ali y o each age g oup emains a he le el o he yea o obse a ion. A key
poin is ha he indica o is no a ec ed by he age s uc u e o he popula ion.
The expec ed e ec i e e i emen age can be calcula ed o pe sons a any age.
This a icle p esen s he es ima ed e ec i e e i emen age in each coun y o
50-yea -olds based on cohe en da a. The da a o he expec ed e ec i e e i e-
men age has mainly been ob ained om he employmen da a o each coun y’s
s a is ics o ice, and hey include he s a u o y (o public) pensions o he pe sons
esiden in he coun y. Fo No way, he da a has been ob ained om he No we-
gian Labou and Wel a e Adminis a ion (NAV). Mode o calcula ion is p esen ed
mo e de ailed in he Appendix 1.2.
1.4 Resul s
1.4.1 Employmen a e o 55-64-yea -olds has p og essed a ou ably
A common a ge in pension policy o a leas 20 yea s o mo e has been o in-
c ease employmen a es o hose aged 55 o 64 yea s. Employmen a e o he
younge pa o his age g oup, meaning 55 o 59-yea -olds, has al eady suc-
cess ully inc eased o he same le el as gene al employmen a e. Employmen
has also imp o ed signi ican ly among he 60 o 64-yea -olds.
Iceland has he highes employmen a e in he No dics (Figu e 2). Denma k
has clea ly app oached he le el o he leading coun ies bu no as apidly in
2022 as be o e. A he same ime, Finland has na owed he gap o Denma k.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
11
The employmen a es in hese wo coun ies a e s ill clea ly below Sweden and
Iceland. In gene al, he le el in he No dic coun ies is highe han he EU-a e -
age.
Figu e 2: 55-64-yea -olds employmen a e in he No dic coun ies and
EU
Sou ce: Eu os a , Employmen , Labou Fo ce Su eys
A p esen , he ocus should be inc easingly on he o e -60-yea -olds. The o e -
60s a e e en mo e in e es ing because hey a e o en he a ge o cu en pen-
sion policy measu es and he e is s ill mo e o be achie ed. The employmen
a e among he 60 o 64-yea -olds is lowe han o younge age g oups bu a
posi i e end can be obse ed he e as well (Figu e 3). Pension e o ms ha e
been one o he key d i e s con ibu ing o he inc eased labou o ce pa icipa-
ion o ageing people. O cou se, i is impo an o ealize ha he e o ms o
pension policy a e only one enabling ac o o encou age longe wo k ca ee s.
Much depends on he economic g ow h bu also on he e o ms in wo king li e
and quali y o wo k li e.
60
61
62
63
64
65
66
67
68
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
Age
Iceland
No way
Sweden
Denma k
Finland
EU27
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
12
Figu e 3: 60-64-yea -olds employmen a e in he No dic coun ies and EU
Sou ce: Eu os a , Employmen , Labou Fo ce Su eys
Employmen a e o men is gene ally highe han o women. Women can be
seen o be a he le el a which men we e a decade ago (Figu e 4). In gene al,
men ha e a i e o en pe cen age poin s highe employmen a e han women.
This di e ence has sligh ly dec eased in las en yea s. Howe e , he e is one
exemp ion in compa ison. In Finland olde women ha e highe employmen a e
han men which is also globally qui e excep ional. The employmen a e o olde
men in Finland is ha dly a he le el o EU-a e age and much lowe han in he
o he No dic coun ies.
Figu e 4: 55-64-yea -olds’ employmen a es by gende in he No dic coun-
ies and EU
Sou ce: Eu os a , Employmen , Labou Fo ce Su eys
30
40
50
60
70
80
90
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
%
Iceland
No way
Sweden
Denma k
Finland
EU27
30
40
50
60
70
80
90
2013 2016 2019 2022
%
Iceland
Sweden
No way
Denma k
Finland
EU27
2013 2016 2019 2022
Females Males
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
13
1.4.2 Labou ma ke exi age is highes in Iceland
The exi age has isen in he No dic coun ies and h oughou he EU (Figu e 5).
Only in Iceland he exi age has no inc eased in he 2010s, bu he e he exi age
has been clea ly highe han in he o he coun ies al eady a decade ago, and s ill
is. Al hough he di e ences be ween coun ies a e s ill clea , hey ha e na owed.
Denma k and Finland belong o a g oup wi h a lowe exi age han he o he No -
dic coun ies. Bu hey a e s ill mo e han one yea highe han he EU-a e age.
Be ween 2011 and 20222 Finland has expe ienced he la ges inc ease in he
exi age o mo e han wo yea s, ollowed by Denma k wi h an inc ease o almos
wo yea s. In No way he ise has been 1.2 yea s and in Sweden less han a yea .
The a e age inc ease in he exi age o he whole o he EU has been wo yea s.
Figu e 5: Labou ma ke exi age o 50-yea -olds
Sou ce: Own compu a ion
The exi age has inc eased du ing he pe iod unde e iew o bo h men and
women. Exi age is clea ly highe o men han o women. In Iceland exi age is
e y high, especially o men – i is e en highe han gene al e i emen age 67.
In 2022, he a e age di e ence in a ou o men was mo e han one yea . In
Denma k, i was almos wo yea s and in Iceland and No way a yea and a hal .
In Sweden he di e ence was one yea and in Finland less han a yea .
60
61
62
63
64
65
66
67
68
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
Age
Iceland
No way
Sweden
Denma k
Finland
EU27
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
14
Figu e 6: Di e ence o he exi age be ween males and emales
Sou ce: Own compu a ion
1.4.3 Expec ed e ec i e e i emen age – Sweden a he op
Th ee main poin s can be obse ed when ocusing on he expec ed e ec i e e-
i emen age o 50-yea -olds. Fi s , Sweden has emained on a high le el wi h
mino changes and has he highes e ec i e e i emen age. Secondly, Den-
ma k and Finland ha e a signi ican ise in his indica o in 15 yea s. Thi dly, in
No way and Iceland he expec ancy has dec eased; in No way mainly because
o he 2011 pension e o m which dec eased he e i emen age. Howe e , a e
ha i has inc eased.
In Iceland, he e is a high deg ee o luc ua ion, which is pa ly explained by a
compa a i ely small size o popula ion and he yea ly s a is ics on mo ali y and
numbe o pensione s may he e o e unde go subs an ial andom a ia ion. How-
e e , he s a e o he economy is s ongly e lec ed in he e ec i e e i emen age.
In addi ion o he 2008 inancial c isis, Iceland su e ed om an in e nal bank c isis
in he ea ly 2010s, which is seen in he swaying igu es depic ing he coun y’s
e ec i e e i emen age. The consequences o he all o he banks we e mo e o
less immedia e in la e 2008, and people could ei he e i e immedia ely o s ay on
unemploymen bene i s o up o wo and hal yea s (see also Danielsson e al.,
2023).
60
61
62
63
64
65
66
67
68
69
70
Age
Iceland
No way
Sweden
Denma k
Finland
EU27
Females Males
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
15
Recen decline a e COVID-19 is ela ed o he ac ha senio ci izens ha e
s a ed aking paymen s ea lie om pension unds be o e hey s a ecei ing a
pension om social secu i y sys em. Figu es om S a is ics Iceland (2024) show
an inc ease in he numbe o old-age pensione s who claim ea ly pension ins ead
o no mal e i emen age 67. Fo example, abou a qua e o people aged 65-66
had s a ed d awing a pension in 2007, bu his p opo ion had isen o 43% in
2022. The igu es also show ha he majo i y o his g oup only d aws pensions
om pension unds, and he e a e indica ions ha many who do so a e s ill in he
labou ma ke (see also Danielsson e al., 2023).
Figu e 7: Expec ed e ec i e e i emen age o 50-yea -old*
* Finland: p o isional da a o 2021-2023.
Sou ce: Own compu a ion
1.4.4 Di e ence be ween he e ec i e e i emen age and he labou ma ke
exi age
When we compa e he labou ma ke exi age o he e ec i e e i emen age, we
no ice ha he di e ence be ween hese wo indica o s a ies g ea ly be ween
he compa ed coun ies. In his espec , he No dic coun ies a e di ided in o wo
g oups.
In he i s g oup, Finland, Sweden, and Denma k, he di e ence is small: only
a ound hal a yea . In he second g oup, No way and Iceland, he e is a clea gap
be ween indica o s, abou wo yea s o e en mo e. This means ha in he i s
60
61
62
63
64
65
66
Age
Sweden
Denma k
Finland
No way
Iceland
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
16
g oup wo king in e i emen is no as common as in he second g oup. In Finland,
Sweden and Denma k wo k ends usually when you e i e, bu in No way and
Iceland i is usual o con inue wo king a e e i emen .
Figu e 8: Labou ma ke exi age and expec ed e ec i e e i emen age o
50-yea -olds in 2011-2022
Sou ce: Own compu a ion
In gene al, he exi age is highe han he expec ancy, hough in Denma k he
si ua ion has been somewha di e en . The e a e p obably se e al unde lying
easons o he di e ence be ween he exi age and he expec ancy, and he phe-
nomenon can be app oached om di e en di ec ions. Di e en easons unde ly
he a ious beha iou in he coun ies o compa ison, including lexible labou
ma ke s, pension le els, and o he sa e y ne s, as well as employmen oppo u-
ni ies o e i ees. As a ule, pension in he No dic coun ies secu e a basic in-
come o e i ees, meaning people a e no o ced o wo k in e i emen .
T adi ionally, pa - ime wo king has been ex ensi e in Sweden. This has helped
people con inue wo king un il hei e i emen age. Pension legisla ion also plays
a cen al ole. The pension e o ms in Finland and Denma k ha e educed he
oppo uni ies o ea ly e i emen and aised he e i emen age.
In No way he new, lexible e i emen age allows o e i emen a an ea lie age
han be o e. Many ha e aken his oppo uni y o e i e a an ea lie age al hough
hey ha e con inued wo king. I is common, especially among men, o wi hd aw
he manda o y pension as soon as possible (a age 62) and s ill con inue wo king.
Acco ding o s a is ics om NAV (2024) o e 60 pe cen o new old age pension-
Sweden Denma k Finland
60
61
62
63
64
65
66
67
68
Iceland No way
Dashed line = exi age Solid line = effec i e e i emen age
Sweden, Denma k, Finland Age Iceland, No way
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
23
IPFA, Icelandic Pension Funds Associa ion (2024): Old-age pension.
h ps://www.li ey ismal.is/en/qa/old-age-pension
Li eindenma k (2024): The o icial guide o li e in Denma k. h ps://li einden-
ma k.bo ge .dk/
Minis y o Labou and Social Inclusion (2023a): E bed e pensjonssys em med
en s y ke sosial p o il. Meld. S . 6 (2023-2024). h ps://www. egje -
ingen.no/no/ak uel /e -bed e-pensjonssys em-med-en-s y ke -sosial-p o-
il/id3018783/
Minis y o Labou and Social Inclusion (2023b): A beid, pensjon og inn ek i den
eld e be olkningen. Rappo a Pensjonspoli isk a beidsg uppe 30. mai
2023.h ps://www. egje ingen.no/con en asse s/ 1094d6dda2 44 ab02 24d8
3c34e454/ppag-endelig- appo -2023.pd
No wegian Labou and Wel a e Adminis a ion (NAV) (2024): Alde spensjon –
S a is ikkno a e . h ps://www.na .no/no/na -og-sam unn/s a is ikk/pensjon-
s a is ikk/ ela e -in o masjon/alde spensjon-s a is ikkno a e
OECD (2023): Pensions a a Glance: 2023. OECD and G20 Indica o s. OECD
Publishing, Pa is. h ps://doi.o g/10.1787/678055dd-en
Social Insu ance Adminis a ion Iceland (2024): h ps://island.is/s/ yggingas -
o nun
S a is ics Iceland (2024): Pe cen o old age pension bene icia ies by ype o pay-
men , inance s, esidence, gende and age in Decembe 2007-2022.
h ps://px.hags o a.is/pxen/pxweb/en/Sam elag/Sam elag__ elags-
mal__li ey is hega /FEL02000.px/
Swedish Pensions Agency (2024): h ps://www.pensionsmyndighe en.se/
Vidlund, M. (2017): Flexible e i emen – a model o he u u e? Lessons om
Sweden, No way and Finland. Finnish Cen e o Pensions. Helsinki.
h ps://u n. i/URN:ISBN:978-951-691-276-2
Vidlund, M. (2023): Pa ial old-age pension gaining g ound in Finland – mo e
ea ly old-age pension ecipien s in Finland han in No way o Sweden.
h ps://www.e k. i/en/blogs/pa ial-old-age-pension-gaining-g ound-in- in-
land-mo e-ea ly-old-age-pension- ecipien s-in- inland- han-in-no way-o -
sweden/
Labo Ma ke E ec s o Inc easing S a u o y Re i e-
men Age: The 2017 Pension Re o m in Finland
Sa u Ni alainen / Ila i Ilmakunnas
Con ac In o ma ion
D Sa u Ni alainen
Economis a he Finnish Cen e o Pensions, Helsinki, Finland
E-Mail: sa u.ni alainen@e k. i
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
26
Abs ac
We s udy he labo ma ke e ec s o he 2017 pension e o m in Finland. The
s a u o y e i emen age o he s udied coho s inc eased om 63 yea s o he
1954 coho o 63 yea s and six mon hs o he 1956 coho . Using o al egis e
da a eco ded a a mon hly le el and a di e ences-in-di e ences app oach, we
es ima e he e ec o his e o m on e i emen , employmen , unemploymen , dis-
abili y, sickness, and inac i i y. We ind an inc ease in he employmen a e o 19
pe cen age poin s be ween he old and he new e i emen age, bu also a no able
inc ease in unemploymen , inac i i y, and disabili y. Fo he mos pa – bu no
en i ely – his is explained by he pe sis ence o he p e ious labo ma ke s a e.
The e a e no la ge gende di e ences, bu he e ec s a y conside ably by edu-
ca ion and employmen sec o .
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
27
Table o Con en s
Lis o Figu es .................................................................................................... 28
Lis o Tables ..................................................................................................... 28
2.1 In oduc ion ................................................................................................ 29
2.2 Ins i u ional se up and he 2017 pension e o m ........................................ 32
2.3 Da a and empi ical me hodology ............................................................... 37
2.3.1 Da a ................................................................................................. 37
2.3.2 Empi ical me hodology .................................................................... 39
2.4 Desc ip i e e idence .................................................................................. 41
2.5 Reg ession esul s ..................................................................................... 44
2.5.1 Main esul s ..................................................................................... 44
2.5.2 Passi e o ac i e subs i u ion? ........................................................ 47
2.5.3 He e ogenei y o e ec s by gende and socioeconomic g oup ....... 50
2.6 Conclusion ................................................................................................. 55
Re e ences ........................................................................................................ 58
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
28
Lis o Figu es
Figu e 1: Sha e o e i ed indi iduals in di e en coho s by age, %.
Ve ical lines indica e he SRA o di e en coho s. ........................ 41
Figu e 2: Sha e o employed indi iduals in di e en coho s by age, %.
Ve ical lines indica e he SRA o di e en coho s. ........................ 42
Figu e 3: Sha e o unemployed, disabled, sick and o he wise inac i e
indi iduals in di e en coho s by age, %. Ve ical lines
indica e he SRA o di e en coho s ............................................... 44
Lis o Tables
Table 1: Changes in pension ules in 2017 pension e o m ........................ 34
Table 2: Numbe o indi iduals in each coho and o al
numbe o obse a ions in da a ..................................................... 38
Table 3: Desc ip i e s a is ics by coho ...................................................... 39
Table 4: A e age e ec o inc ease in SRA on di e en labo
ma ke s a es, pe cen age poin s .................................................. 45
Table 5: Sensi i i y es s o he employmen e ec o he inc ease in
SRA ............................................................................................... 47
Table 6: A e age e ec o inc ease in SRA on labo ma ke ansi ions
om employmen , unemploymen , disabili y and sickness,
pe cen age poin s ......................................................................... 48
Table 7: Employmen a e jus be o e he SRA, bunching a he SRA
and he haza d a e by subg oup, coho 1954 ............................. 50
Table 8: A e age e ec o inc ease in SRA on di e en labo ma ke
s a es by gende , pe cen age poin s ............................................. 51
Table 9: A e age e ec o inc ease in SRA on di e en labo ma ke
s a es by educa ion, pe cen age poin s ........................................ 52
Table 10: A e age e ec o inc ease in SRA on di e en labo
ma ke s a es by sec o , pe cen age poin s .................................. 54
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
29
2.1 In oduc ion
In esponse o apidly aging popula ions, many coun ies ha e e o med hei
pension sys ems in ecen decades. In mos cases, he e o ms ha e included
ising pension eligibili y ages. The aim has been o secu e he inancial sus aina-
bili y o pension sys ems by ex ending wo king li es and inc easing he employ-
men a e o olde pe sons (Hin ichs, 2021).
Finland is no excep ion: he Finnish pension sys em unde wen a e o m in 2017.
The e a e no ea ly e i emen ages (ERA) in he Finnish pension sys em, and
be o e he e o m he s a u o y e i emen age (SRA; o en e e ed o as he No -
mal Re i emen Age, NRA), he minimum age a which ull pension bene i s could
be claimed, was 63 yea s. I was belie ed ha inancial incen i es would encou -
age people o con inue a wo k a e eaching hei SRA. Howe e , he inancial
incen i es did no ha e he desi ed e ec (e.g., G ube e al., 2022). The e o e,
in he 2017 e o m i was ag eed ha he SRA will inc ease by h ee mon hs o
each bi h coho un il i is a 65 yea s. The i s coho a ec ed was ha bo n in
1955. Fo hose bo n in 1962-1964, he SRA is 65 yea s, a e which i will be
linked o changes in li e expec ancy. Only i he e o m succeeds in pos poning
e i emen and ex ending wo king li es can he sus ainabili y o he pension sys-
em be secu ed.
Unde s anding indi iduals’ beha io al esponse o pension e o ms aising he
eligibili y age o he old-age pension is c ucial o de e mining hei success. In
his s udy, we in es iga e he e ec s o he ising SRA on employmen and o he
labo ma ke ou comes o olde indi iduals in Finland. We u ilize he coho -based
changes in e i emen age esul ing om he 2017 pension e o m and s udy he
coho s bo n in 1954–1956. Fo hose bo n in 1954, he SRA was 63 yea s (con-
ol g oup), and o hose bo n in 1955 and 1956, i was 63 yea s h ee mon hs
and 63 yea s six mon hs, espec i ely ( ea men g oups). We exploi he exoge-
nous a ia ion ha he e o m has induced in he e i emen age and iden i y he
impac o he ising e i emen age on labo ma ke ou comes by compa ing co-
ho s who ace di e en e i emen ages. We use a di e ences-in-di e ences de-
sign, which allows us o isola e causal e ec s ha mani es be ween he old and
new s a u o y e i emen age.
Ou analysis ela es o he li e a u e examining beha io al e ec s o changes in
he e i emen age on labo ma ke ou comes. Due o nume ous pension e o ms
in Eu ope and o he coun ies in ecen decades, he esea ch in e es owa ds
he e ec s o pension e o ms has inc eased. Fi s s udies in es iga ed he e ec s
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
30
o ising NRA in he US and mainly ocused on he e i emen decision (Mas o-
buoni, 2009; Behagel & Blau, 2012), bu nowadays a g owing li e a u e especially
in Eu ope has e alua ed he e ec s o he ising e i emen age on employmen
and o he labo ma ke ou comes (e.g., S aubli & Zweimülle , 2013; A alay & Ba -
e , 2015; Manoli & Webe , 2016; C ibb e al., 2016; Raba é & Rochu , 2020; A a
e al., 2019; 2021; Geye e al., 2020; Geye & Wel eke, 2021; Soosaa e al.,
2021; Mo is, 2021; 2022). The o e all conclusion is ha he ising e i emen age
pos pones e i emen and posi i ely a ec s he employmen a e, bu he size o
he e ec a ies g ea ly be ween s udies. This is due o, o example, di e ences
in age limi s o he old-age pension, inancial incen i es o he pension sys em,
p e- e o m employmen a es and bunching a he ERA/SRA.
We build on p e ious s udies in se e al ways. Many o hem ha e ocused on
changes in he ERA, and due o he na u e o he e o ms, he e ec s ha e been
es ima ed among women (A alay & Ba e , 2015; C ibb e al., 2016; Geye e al.,
2020; Geye & Wel eke, 2021; Mo is, 2021; 2022). The e o e, examining he im-
pac s sepa a ely o men and women has no been possible. We p o ide an e al-
ua ion o he labo ma ke impac s o he ising SRA in a No dic wel a e s a e –
Finland – whe e men and women ha e he same e i emen age. Finland makes
an in e es ing case since, unlike in mos Eu opean coun ies and he US, he
employmen a e o 55-64-yea -old Finnish women is as high as ha o men (66
pe cen s 67 pe cen ), and olde women a e signi ican ly mo e o en highly
educa ed han men (50 pe cen s 36 pe cen ) (Eu os a , 2024). While many
exis ing s udies ha e elied on samples o su eys, we use indi idual-le el o al
popula ion egis e da a including all Finns bo n in 1954–1956. In o ma ion on
e i emen and se e al o he labo ma ke ou comes is a ailable a a mon hly
le el, and we can p ecisely ollow indi iduals’ labo ma ke ajec o ies om he
mon h hey u n 60 un il he mon h hey u n 64. Hence, we can isola e he e ec
o inc easing SRA on di e en labo ma ke ou comes mo e comp ehensi ely and
wi h g ea e accu acy han mos p e ious s udies. We a e also able o in es iga e
he e ec o inc easing SRA on he use o al e na i e pa hways o e i emen . In
addi ion o gende di e ences, ou ich da a allows us o s udy socioeconomic
di e ences in he e ec s o inc easing SRA, an issue dis ega ded in mos s udies
analyzing he e ec s o pension e o ms.
We es ima e he e ec o he e o m on mul iple labo ma ke ou comes: e i e-
men , employmen , unemploymen , disabili y, sickness, and inac i i y. By ocus-
ing on se e al labo ma ke ou comes, we can gain insigh in o hese e ec s in
mo e de ail han mos p e ious s udies. E en hough he aim o he e o m is o
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
31
ex end wo king li es and inc ease employmen , i is also likely o impac he use
o al e na i e pa hways o e i emen , as all indi iduals a e no able o willing o
wo k un il he highe e i emen age. In ha sense he e o m has some po en ial
ad e se e ec s, which a e called subs i u ion e ec s (e.g., Raba é & Rochu ,
2020; Geye & Wel eke, 2021). Impo an ly, subs i u ion e ec s may s em om
passi e o ac i e subs i u ion. Passi e subs i u ion e ec s occu , o example, i
indi iduals who a e al eady ou o employmen s ay longe in hei cu en labo
ma ke s a e due o he inc eased e i emen age. On he o he hand, as he e-
i emen age inc eases, he ime window du ing which ansi ions be ween di e -
en labo ma ke s a es a e possible ex ends. Fo example, someone who is em-
ployed may become unemployed be o e eaching hei new e i emen age.
These ansi ions a e called ac i e subs i u ion e ec s. We inspec he e o m-
induced pe sis ence o he p e ious labo ma ke s a e as well as ansi ions be-
ween labo ma ke s a es. Ou s udy hence also ela es o he s and o li e a u e
which in es iga es he subs i u ion be ween social insu ance p og ams (e.g.,
Duggan e al., 2007; Ka ls öm e al., 2008; S aubli, 2011; Inde bi zin e al., 2016).
A gene al inding in hese s udies is ha igh ening he c i e ia o one pa hway
leads o subs i u ion e ec s in o al e na i e ea ly exi pa hways such as unem-
ploymen , disabili y, o sickness bene i s. In he con ex o ising e i emen age,
he use o al e na i e ou es o e i emen has a ely been ho oughly examined
(see howe e Raba é & Rochu , 2020).
Ou s udy also con ibu es o he li e a u e conce ning socioeconomic inequali ies
in he labo ma ke . We in es iga e he e ec he e ogenei y no only acco ding o
gende , bu also by le el o educa ion. When designing and e o ming he pension
sys em i is impo an o unde s and how di e en g oups espond o hese e-
o ms. In case o ising e i emen age, he possibili ies o wo k un il old-age pen-
sion depend on he employmen oppo uni ies and indi idual abili ies o con inue
o wo k longe (Ebbinghaus & Ho äcke , 2013; Oude Hengel e al., 2021). As
men ioned ea lie , Finland makes an in e es ing case o s udying gende di e -
ences. Mo eo e , i is known ha he e is a social g adien in he labo ma ke
exi : low-educa ed wo ke s lea e he labo ma ke ea lie han high-educa ed
wo ke s, and o en his is in olun a y (Mäcken e al., 2022). This is no only due
o mo e demanding wo king condi ions and weake heal h which inc ease he isk
o disabili y, bu also o unemploymen (Pol inen e al., 2013; Rob oek e al.,
2015; OECD, 2015). As a esul , low-educa ed wo ke s a e likely o ha e mo e
di icul ies in eaching hei new e i emen age, which may lead o inc easing
social and economic inequali ies be ween di e en socioeconomic g oups. Ea lie
e idence on he e ec s acco ding o gende o educa ion is limi ed. A a e al.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
32
(2019) ound a la ge employmen e ec o men, and Geye e al. (2020) ound
ha o women wi h low o medium educa ion, he employmen e ec o he ERA
inc ease was accompanied by equally la ge subs i u ion e ec s o o he labo
ma ke s a es.
Ou esul s show ha he SRA inc ease esul ed in a la ge employmen e ec o
19 pe cen age poin s be ween he old and he new e i emen age. Employmen
esponse is qui e subs an ial, as he employmen e ec ep esen s 60 pe cen o
he dec ease in he e i emen a e induced by he e o m. We also ind a la ge
inc ease in unemploymen , inac i i y and disabili y, and a smalle inc ease in he
use o sickness bene i s. Fo he mos pa , his is due o passi e subs i u ion, bu
we also e idence ac i e subs i u ion om employmen o o he labo ma ke
s a es. E en hough ansi ions om unemploymen and sickness o employmen
also occu , he ansi ions ou o employmen clea ly ou weigh he ansi ions o-
wa ds employmen . The e a e p ac ically no di e ences in he e ec s acco ding
o gende . Fo he low-educa ed indi iduals, he subs i u ion e ec s o labo ma -
ke s a es ou side employmen a e la ge han o he high-educa ed indi iduals,
which means ha he inc ease in he SRA is less e icien in e ms o employmen
o his mo e ulne able g oup in he labo ma ke .
The emainde o his pape p oceeds as ollows. Sec ion 2 desc ibes he ins i u-
ional se ing and he e o m o 2017. The da a and he empi ical me hodology
a e p esen ed in sec ion 3. Sec ion 4 p o ides desc ip i e e idence on how em-
ploymen and o he labo ma ke s a es de eloped a he ime o he e o m. The
eg ession esul s a e p esen ed in sec ion 5, and sec ion 6 concludes.
2.2 Ins i u ional se up and he 2017 pension e o m
The Finnish pension sys em consis s o h ee pilla s. The s a u o y pension sys-
em ( i s pilla ) includes he employmen -based ea nings- ela ed pension and
esidence-based na ional and gua an ee pensions. The na ional and gua an ee
pensions aim a ensu ing a basic income secu i y. The ea nings- ela ed pension
sys em is a de ined bene i sys em whe e he pension le el is de e mined by he
leng h o wo k his o y and he amoun o pas ea nings. The ea nings- ela ed sys-
em is manda o y and co e s all wo ke s and i ually all ea nings. Pension ac-
c ues h oughou wo king li e wi h con ibu ions paid by employees and employ-
e s. S a u o y pensions a e b oad in scope, and he e is no ceiling o he pension-
able ea nings o he pension amoun . Con a y o wha is he case in many o he
coun ies, he ole o second-pilla o hi d-pilla supplemen a y pensions is mino
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
33
in Finland. In he p i a e sec o , ea nings- ela ed pensions a e mainly p o ided
by pension insu ance companies. The public sec o has i s own pension p o ide .
The Finnish ea nings- ela ed pension sys em was e o med in 2017. Be o e he
e o m, he s a u o y e i emen age (SRA) was 63 yea s o all, independen o
coho . Like he US and a couple o o he coun ies, Finland has a lexible e i e-
men age, which means ha an indi idual can ake ou hei pension wi hin a
ce ain age ange. The SRA is he minimum age a which he ull pension bene i s
can be claimed (o en e e ed as he NRA). The e a e no ERAs in he Finnish
pension sys em.1 In he 2017 e o m, i was ag eed ha he SRA will be inc eased
by h ee mon hs o each coho . The i s coho a ec ed by he e o m was ha
bo n in 1955. I s SRA inc eased o 63 yea s and h ee mon hs. Those bo n in
1956 had an SRA o 63 yea s and six mon hs, and so on. The e i emen age
inc eases by h ee mon hs un il i eaches 65 o he coho bo n in 1962. Fo
hose bo n in 1963–1964, he h eshold is also 65 yea s. As o hose bo n in 1965,
he old-age e i emen age will be linked o li e expec ancy. A e ha , he ise in
he e i emen age can be wo mon hs pe coho a mos . The key aspec s o he
Finnish pension sys em be o e and a e he 2017 e o m a e gi en in Table 1.
1 Pa ial old-age pension allows o ake ou 25 pe cen o 50 pe cen o acc ued ea n-
ings- ela ed pension a age 61 (wi h an ac ua ial adjus men ).
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
40
due o he ising SRA, one coho has al eady eached he SRA a a ce ain age
while he o he coho has no . We es ima e he ollowing di e ences-in-di e -
ences model:
𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖=𝛽𝛽0+𝜆𝜆𝑖𝑖+𝜃𝜃𝑖𝑖+𝛾𝛾𝑖𝑖+𝛽𝛽1𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖+𝑋𝑋i′𝛽𝛽2+𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 (1)
whe e 𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 is a labo ma ke s a e dummy equal o 1 i he indi idual i, a age a,
om coho c, a ime is in he ele an s a e. The labo ma ke s a es unde
inspec ion a e “ e i ed”, “employed”, “unemployed”, “on disabili y pension”, “on
sickness bene i ”, and “o he wise inac i e”.
In his speci ica ion, 𝜆𝜆𝑖𝑖 a e coho dummies, 𝜃𝜃𝑖𝑖 a e age dummies (age in mon hs)
and 𝛾𝛾𝑖𝑖 a e ime e ec s. 𝑋𝑋𝑖𝑖 includes a se o indi idual con ols (gende , educa-
ion, sec o , leng h o ca ee and ype o municipali y).
The key explana o y a iable is a dummy, 𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖, which is equals o one
i he indi idual is below he SRA applicable wi hin hei coho , and ze o o he -
wise. Because he inc ease in he SRA a ec s indi iduals a di e en ages de-
pending on hei coho , he alue o his a iable changes wi h age, coho , and
ime. Fo hose bo n in 1954, his a iable is equal o one un il age 63, and ze o
a e ha . Fo hose bo n in 1955, i is equal o one un il age 63 yea s h ee
mon hs, and ze o a e ha . Fo hose bo n in 1956, i is equal o one un il age 63
yea s six mon hs, and ze o a e ha .
The main pa ame e o In e es is 𝛽𝛽1 which cap u es he SRA inc ease-induced
di e ence in he p obabili y o being in a ce ain s a e be ween di e en coho s.
In p ac ice we compa e he p obabili y o a ce ain s a e a he same age o pe -
sons belonging o di e en coho s wi h a a ying SRA. The cen al iden i ying
assump ion is ha , wi hou he inc ease in he SRA, he p obabili y o be in a
ce ain s a e a he same age would be simila ac oss coho s, con olling o he
coho - and age- ixed e ec s, indi idual-le el a iables, and ime e ec s (so-
called common ends assump ion). I his assump ion holds, 𝛽𝛽1 can be in e -
p e ed as he causal e ec o he SRA inc ease. We es ima e his pa ame e wi h
a linea p obabili y model and, as a esul , 𝛽𝛽1 can be in e p e ed as a pe cen age
poin di e ence in he p obabili y ha an indi idual is in a ce ain s a e in he co-
ho ha has no ye eached i s SRA ( ea men g oup) compa ed o he coho
ha has al eady eached i s SRA (con ol g oup).
The e o m o 2017 has a ec ed di e en coho s a di e en ages, and di e en
coho s each a ce ain age a di e en poin s in ime. The e o e, we also wan o
con ol o he ime e ec s (o he business cycle) o minimize hei impac on
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
41
labo ma ke s a es. To ci cum en he iden i ica ion issue o pe ec linea i y be-
ween age, pe iod, and coho , we use di e en ime s eps in each. We measu e
coho a a yea ly le el and age a a mon hly le el, and we con ol o he business
cycle a a qua e -yea le el. This means ha we assume ha he business cycle
e ec s a e he same o indi iduals obse ed in he same qua e .
2.4 Desc ip i e e idence
Figu es 1 and 2 p esen he sha e o e i ed and employed indi iduals om he
mon h he indi iduals u ned 60 o he mon h hey u ned 64. The SRA o he
con ol coho 1954 was 63, and he SRAs o he ea men coho s 1955 and
1956 we e 63 yea s and h ee mon hs and 63 yea s and six mon hs, espec i ely.
This means ha indi iduals bo n in 1954 can claim hei pension in he mon h
a e hey u n 63 and hose bo n in 1955 and 1956 in he mon h a e hey u n
63 yea s and h ee mon hs and 63 yea s and six mon hs, espec i ely. The e i-
cal lines in he Figu es ma k he SRAs o he di e en coho s.
Figu e 1: Sha e o e i ed indi iduals in di e en coho s by age, %. Ve ical
lines indica e he SRA o di e en coho s.
Sou ce: calcula ions based on egis e da a o he Finnish Cen e o Pensions.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
42
Figu e 2: Sha e o employed indi iduals in di e en coho s by age, %. Ve -
ical lines indica e he SRA o di e en coho s.
Sou ce: calcula ions based on egis e da a o he Finnish Cen e o Pensions.
We obse e a clea pa allel end in bo h e i emen and employmen be o e age
63. Due o occupa ional e i emen ages in he public sec o , some indi iduals
ha e e i ed be o e age 62. The e is also a small jump in he e i emen a e a
age 62, pa ly due o he occupa ional e i emen ages in he public sec o , and
pa ly due o he en i lemen o he long- e m unemployed o a ull old-age pension
a age 62. A e age 63 we see a conside able inc ease in e i emen and de-
c ease in employmen among indi iduals bo n in 1954. Fo indi iduals bo n in
1954, bunching a he SRA is a ound 22 pe cen age poin s (haza d a e, he d op
in he sha e o employed pe sons a he SRA o e he sha e o employed pe sons
jus be o e he SRA, is 0.43). I should be no ed ha he sha e o e i ed indi id-
uals inc eases igh a e age 63 also among hose bo n in 1955–1956. This is o
he mos pa due o disabili y pensions whe e he disabili y has s a ed be o e
2017, as hese pensions au oma ically ansi ion in o an old-age pension a age
63 (see Figu e 3). Fu he mo e, he e is a small d op in employmen a age 63
also among hose bo n in 1955–1956. This is mainly due o wo ac o s: some
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
43
public sec o employees ha e a pe sonal e i emen age o 63 yea s, and a small
numbe o p i a e sec o employees ha e an SRA o 63 yea s due o collec i e
labo ag eemen s. A e age 63 he de elopmen in coho s acing di e en e i e-
men ages di e ges. The inc ease in e i emen and he d op in employmen shi
o he igh co esponding wi h he inc ease in he SRA, ha is, by h ee mon hs
o he coho o 1955 and again by h ee mon hs o he coho o 1956.
In Figu e 3, we also obse e a a he pa allel end in he p opo ion o indi iduals
in unemploymen , disabili y, sickness, and inac i i y be o e age 63, and a clea
d op a e his age o hose bo n in 1954. The e a e small di e ences in he sha e
o unemployed a ound age 62 and he ea e especially be ween hose bo n in
1956 and o he coho s, bu hese di e ences a e ai ly small, a mos 2 pe cen -
age poin s. In case o disabili y, due o p e- e o m disabili y pensions, he e is a
conside able d op a e age 63 also o hose bo n in 1955 and 1956. I should be
no ed ha a e age 63, he sha e o indi iduals on disabili y pension e lec s
hose disabili y pensions whe e he disabili y has s a ed in 2017 o la e . The
pe iod om 2017 o he new SRA is sho e o hose bo n in 1955, and he sha e
on disabili y pension among hose bo n 1955 has only inc eased modes ly. The
sha e among hose bo n in 1956 has inc eased mo e. O he wise o hose bo n
in 1955, he la ges d op in hese s a es occu s a e age 63 yea s and h ee
mon hs, and o hose bo n in 1956, a e age 63 yea s and six mon hs.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
44
Figu e 3: Sha e o unemployed, disabled, sick and o he wise inac i e indi-
iduals in di e en coho s by age, %. Ve ical lines indica e he
SRA o di e en coho s
Sou ce: calcula ions based on egis e da a o he Finnish Cen e o Pensions.
The Figu es gi e an ini ial idea o he e ec s we aim o analyze. They also suppo
he assump ion o common ends o di e en coho s be o e he inc ease in e-
i emen age, which is a he cen e o ou iden i ica ion s a egy.
2.5 Reg ession esul s
2.5.1 Main esul s
Table 4 p esen s linea eg ession es ima es o he impac o he inc ease in SRA
on e i emen , employmen , unemploymen , disabili y, sickness, and inac i i y us-
ing equa ion (1). We only p esen he di e ences-in-di e ences coe icien s. All
es ima es a e signi ican a he 0.1 pe cen le el. We obse e ha he SRA in-
c ease has dec eased e i emen by 32 pe cen age poin s on a e age o he
mon hs be ween he old and new SRA. Be o e he e o m 64 pe cen we e e-
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
45
i ed. The SRA inc ease is accompanied by a ise in employmen o 19 pe cen -
age poin s. Compa ed o he p e- e o m baseline (29 pe cen ), he ela i e in-
c ease in employmen a e amoun s o nea ly 70 pe cen .
A he same ime, he SRA e o m has inc eased unemploymen by 6 pe cen age
poin s on a e age. This is a la ge e ec gi en ha he p opo ion o unemployed
indi iduals be o e he inc ease in he SRA was unde 5 pe cen . Likewise, he e
has been an inc ease in disabili y pension ecipiency o 2 pe cen age poin s. The
ac ion o o he wise inac i e pe sons has inc eased by o e 4 pe cen age poin s,
which means ha his sha e has almos ipled because o he e o m. Sickness
plays a mino ole as i has inc eased by only 0.5 pe cen age poin s (bu again,
compa ed o he p e- e o m baseline, he inc ease is qui e la ge).
Table 4: A e age e ec o inc ease in SRA on di e en labo ma ke
s a es, pe cen age poin s
Re i emen Employ-
men
Unemploy-
men
Disabili y Sickness Inac i i y
Unde
SRA
-32.13*** 19.37*** 6.01*** 2.06*** 0.48*** 4.20***
(0.205) (0.149) (0.069) (0.075) (0.029) (0.096)
P e- e-
o m,
%
63.66 29.08 4.80 0 0.30 2.16
Obse -
a-
ions
10,465,567
10,465,56
7
10,465,567
10,465,56
7
10,465,567
10,465,56
7
P e- e o m: hose bo n in 1954 a age 63 yea s and one mon h. Linea eg ession including
coho dummies, age dummies, qua e dummies and sociodemog aphic and egional
con ols (gende , educa ion, sec o , ca ee leng hs, egion), s anda d e o s clus e ed by
mon h o bi h. Robus s anda d e o s in pa en heses. Coe icien s and s anda d e o s
a e mul iplied by 100 and should be in e p e ed as pe cen age poin s. ***p<0.001
Sou ce: Own compu a ion
Employmen ep esen s 60 pe cen o he dec ease in he e i emen a e
(19.4/32.1), which means ha his sha e o hose indi iduals who delay claiming
pension bene i s as a esponse o he SRA inc ease a e wo king. A he same
ime, 19 pe cen o hose who would ha e al eady e i ed in he absence o he
SRA inc ease a e unemployed and 7 pe cen a e on a disabili y pension. In o al,
15 pe cen a e on sickness bene i s o o he wise inac i e due o he e o m.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
46
The SRA e o m has hence inc eased employmen conside ably bu , a he same
ime, i has gene a ed la ge subs i u ion e ec s om e i emen o o he labo
ma ke s a es. We can expec ha pa o hese e ec s is due o passi e subs i-
u ion: when he SRA inc eases, indi iduals ha would ha e o he wise e i ed,
s ay longe a hei espec i e s a es, o example unemployed. Ano he pa o
he e ec s may de i e om ac i e subs i u ion, which means ha , o example,
some employed indi iduals may end up as unemployed be o e eaching he new
e i emen age. We will in es iga e subs i u ion e ec s in mo e de ail la e in Sec-
ion 5.2.
Nex , we will p esen obus ness es s o ou esul s. We i s es he sensi i i y
o he employmen e ec s o al e na i e speci ica ions. The esul s a e p esen ed
in Table 5. In he i s speci ica ion he e a e no con ols and no clus e ing o
s anda d e o s. In he second speci ica ion, he s anda d e o s a e clus e ed by
mon h o bi h. The hi d speci ica ion includes qua e dummies o con ol o he
business cycle. In he ou h speci ica ion, sociodemog aphic and egional con ol
a e added o he model. This is he p e e ed speci ica ion ha was used in Table
5. The i h speci ica ion p esen s he employmen e ec only o he i s inc ease
in he SRA, ha is, when only hose bo n in 1954 and 1955 a e included. The
es ima ed e ec s o he change o he employmen a e emain e y simila inde-
penden o he speci ica ion used. A u he es o he alidi y o ou model is o
conduc a placebo es – ha is, o es whe he he e is an e ec when we would
no expec o see one. The e o e, we es ima e he same speci ica ion as in col-
umn 4 using hose bo n in 1953 and 1954 and assume (inco ec ly) ha he in-
c ease in he SRA s a ed om he coho o 1954 (speci ica ion six). The es ima e
o his pseudo e o m is nega i e and e y small in magni ude.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
47
Table 5: Sensi i i y es s o he employmen e ec o he inc ease in SRA
No con ols,
no clus e -
ing
(1)
No con ols,
clus e ing
by mon h o
bi h (2)
2+ Qua e
dummies
(3)
3+ Socio-
demo-
g aphic and
egional
con ols (4)
4+ Only
hose bo n
in 1954
and 1955
(5)
4+
Pseudo
e o m
(6)
Unde
SRA
20.18*** 20.18*** 19.39*** 19.37*** 18.58*** -0.89***
(0.094) (0.203) (0.148) (0.149) (0.234) (0.162)
R2 0.077 0.077 0.077 0.222 0.224 0.219
Obse -
a ions
10,465,567 10,465,567 10,465,567 10,465,567 6,976,130
5,462,939
Coe icien s and s anda d e o s a e mul iplied by 100 and should be in e p e ed as pe -
cen age poin s. Pseudo e o m includes hose bo n in 1953 and 1954 and assumes ha
he SRA s a ed o inc ease om he coho o 1954. ***p<0.001
Sou ce: Own compu a ion
2.5.2 Passi e o ac i e subs i u ion?
As men ioned ea lie , he employmen and subs i u ion e ec s shown in Table 4
can esul ei he om he pe sis ence o a ce ain labo ma ke s a e (passi e
subs i u ion) o om a ansi ion be ween s a es (ac i e subs i u ion) due o an
inc ease in he SRA. To in es iga e his issue in mo e de ail, we es ima e equa-
ion (1) condi ioning by he ini ial labo ma ke s a e a age 62 yea s and six
mon hs. A his age, 54 pe cen o he s udy popula ion was employed, 11 pe
cen unemployed, 17 pe cen on a disabili y pension and 1 pe cen on sickness
bene i . Linea eg ession esul s a e p esen ed in Table 6. To inspec he sensi-
i i y o he choice ega ding he ini ial s a e, we es ima ed he models measu ing
he ini ial s a e also a age 62 yea s and nine mon hs. The esul s a e e y simila
o hose epo ed he e.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
48
Table 6: A e age e ec o inc ease in SRA on labo ma ke ansi ions
om employmen , unemploymen , disabili y and sickness, pe -
cen age poin s
Re i e-
men
Employ-
men
Unemploy-
men
Disabili y Sickness Inac i i y
Employmen
Unde SRA -39.42*** 34.53*** 1.58*** 0.46*** 0.70*** 2.14***
(0.253) (0.188) (0.074) (0.040) (0.026) (0.076)
P e- e o m,
%
44.54 53.90 0.68 0 0.43 0.45
Obse a ions 2,210,859 2,210,859 2,210,859 2,210,859 2,210,859 2,210,859
Unemploy-
men
Unde SRA -44.82*** 1.97*** 40.17*** 0.22*** 0.39*** 2.07***
(0.413) (0.089) (0.388) (0.039) (0.043) (0.265)
P e- e o m, %
60.76 2.34 35.81 0 0.16 0.92
Obse a ions 462,498 462,498 462,498 462,498 462,498 462,498
Disabili y
Unde SRA -5.95*** 0.05 0.01 5.90*** 0.00 -0.01
(0.449) (0.034) (0.011) (0.454) (0.004) (0.006)
P e- e o m, %
99.98 0 0.01 0 0 0.02
Obse a ions 691,923 691,923 691,923 691,923 691,923 691,923
Sickness
Unde SRA -67.72*** 20.02*** 4.54*** 35.76*** 5.37*** 2.04***
(1.662) (0.940) (0.283) (1.240) (0.978) (0.413)
P e- e o m, %
81.45 12.29 1.81 0 3.98 0.48
Obse a ions 50,160 50,160 50,160 50,160 50,160 50,160
P e- e o m: hose bo n in 1954 a age 63 yea s and one mon h. O iginal s a us measu ed
a age 62 yea s and six mon hs. Linea eg ession including coho dummies, age dum-
mies, qua e dummies and sociodemog aphic and egional con ols (gende , educa ion,
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
49
sec o , ca ee leng hs, egion), s anda d e o s clus e ed by mon h o bi h. Robus s and-
a d e o s in pa en heses. Coe icien s and s anda d e o s a e mul iplied by 100 and
should be in e p e ed as pe cen age poin s. ***p<0.001
Sou ce: Own compu a ion
We no ice ha he ising SRA has inc eased employmen pe sis ence on a e age
by 35 pe cen age poin s. This means ha abou 88 pe cen (34.5/39.4) o hose
employed indi iduals who would ha e e i ed gi en he op ion con inue o wo k
due o he e o m. A he same ime, ac i e subs i u ion has also aken place, as
we see an inc ease in ansi ions om employmen o inac i i y (2.1 pp) and un-
employmen (1.6 pp) and, o a smalle deg ee, o sickness (0.7 pp) and disabili y
(0.5 pp). The sha e o o iginally employed indi iduals who a e ou side wo k has
inc eased in o al by 5 pe cen age poin s be ween he old and new SRA. This
means ha , due o he e o m, in o al a ound 12 pe cen (4.9/39.4) o employed
indi iduals who would ha e al eady e i ed we e i no o he SRA inc ease a e
inac i e, unemployed, on sickness bene i , o on a disabili y pension.
In he case o unemploymen , he dec ease in he e i emen a e induced by he
e o m has mos ly ansla ed in o con inua ion o unemploymen . As a esul o
he SRA inc ease, unemploymen pe sis ence has g own by 40 pe cen age
poin s. This indica es ha 90 pe cen (40.2/44.8) o hose unemployed who delay
claiming pension bene i s due o he SRA inc ease s ay unemployed longe be-
cause o he e o m. A small po ion o unemployed ha e e u ned o employmen
o become inac i e. The inc ease in he SRA has also impac ed disabili y pension
o sickness en olmen among unemployed indi iduals o some deg ee.
Disabili y is a e y abso bing s a e, as he e a e p ac ically no ansi ions om a
disabili y pension o any o he labo ma ke s a es ollowing he e o m. Recei ing
sickness bene i s, on he o he hand, seems o be a less abso bing s a e; he
inc ease in pe sis ence in his s a e due o he ising SRA is only less han 6
pe cen age poin s, which accoun s o 8 pe cen (5.4/67.7) o he dec ease in he
e i emen a e. Then again, a conside able pa (53 pe cen ) o hose on a sick-
ness bene i who would ha e e i ed wi hou he e o m end up on a disabili y
pension. Due o he SRA inc ease, 30 pe cen o indi iduals on sickness bene i
e u n o employmen ins ead o e i ing.
Since he numbe o employed indi iduals is signi ican ly la ge han he numbe
o indi iduals ha a e unemployed o on sickness bene i , ansi ions om em-
ploymen o o he labo ma ke s a es clea ly ou weigh he ansi ions om unem-
ploymen o sickness o employmen . This means ha he SRA inc ease esul s
in mo e nega i e (ou o wo k) han posi i e ( owa ds wo k) ac i e subs i u ion.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
56
Fu he analyses showed ha he obse ed subs i u ion e ec s a e mos ly
caused by passi e a he han ac i e subs i u ion. Tha is, hose acing a highe
SRA s ay longe in hei espec i e labo ma ke s a es a he han e i e. Sick-
ness o ms an excep ion: majo i y o hose on sickness bene i s ansi ion o o he
labo ma ke s a es, mos equen ly o disabili y pension bu also o employmen .
As a whole, he esul s indica e ha o a conside able pa o he popula ion, he
inc ease in SRA means ha unemploymen o disabili y pe sis s. Inc eased em-
ploymen hus comes wi h he social cos o hose in a weake posi ion emaining
in ha posi ion o an ex ended pe iod.
Howe e , we also ound inc eased en y o inac i i y, unemploymen , sickness
bene i s and disabili y pension among o iginally employed indi iduals. In o he
wo ds, he e has been some ac i e p og am subs i u ion om employmen o
o he labo ma ke s a es due o he inc ease in he SRA. In o al, 12 pe cen o
employed indi iduals who would ha e e i ed we e i no o he e o m exi ed
wo k be o e eaching hei new e i emen age. The esul indica es ha in he
case o ising SRA some employed indi iduals end up using al e na i e pa hways
o e i emen . This can be olun a y o in olun a y. E en hough ansi ions om
unemploymen and sickness o employmen also occu ed, he ansi ions ou o
employmen clea ly ou weigh he ansi ions owa ds employmen .
The abili y o wo k longe may a y acco ding o, o example, socioeconomic
s a us. To e eal po en ial he e ogeneous beha io al e ec s, we examined he
e ec o he SRA inc ease sepa a ely by gende , educa ion, and sec o . We did
no ind la ge di e ences be ween men and women. This is a new inding in he
li e a u e and mos likely e lec s gende equali y in employmen a es in Finland.
In addi ion, we ound ha , o he low-educa ed indi iduals, he subs i u ion e -
ec s o labo ma ke s a es ou side wo k we e la ge han o he high-educa ed.
The same applies o he p i a e s he public sec o . The esul s indica e ha ,
among he low-educa ed, he SRA inc ease is less e icien in e ms o employ-
men in ha employmen ep esen s less o he dec ease in he e i emen a e
induced by he e o m. The low-educa ed indi iduals use unemploymen and dis-
abili y as al e na i e pa hways o e i emen mo e o en. This may lead o ising
social and economic inequali ies be ween he low- and high-educa ed indi iduals
ia nega i e e ec s on hei inancial posi ion, hei pension acc ual and, conse-
quen ly, hei old-age secu i y, no o men ion po en ial ad e se e ec on hei
men al wellbeing (e.g., Ba schke e al., 2022). I is also wo h poin ing ou ha
when he SRA inc eases, indi iduals who die ea lie lose a highe p opo ion o
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
57
hei li e ime pensions. This e ec is pa icula ly p onounced among indi iduals
wi h a lowe socioeconomic s a us, as hei li e-expec ancy is sho e .
Ou esul s e eal impo an in o ma ion on he e ec s o he ising SRA and com-
plemen he exis ing li e a u e on his subjec . I is wo h poin ing ou , howe e ,
ha a limi a ion is ha in his s udy we in es iga ed he sho - e m e ec s o he
SRA inc ease, ha is, a si ua ion in which he SRA has no ye inc eased much.
Then again, he la ges e ec o he SRA inc ease seems o mani es i sel exac ly
a he poin when he SRA inc eases so, in his sense, e en a sho - e m inspec-
ion p o ides ele an insigh s in o he e ec s o he 2017 pension e o m. Keep-
ing ha in mind, i seems ha in Finland, aising he SRA has gene a ed an em-
ploymen e ec ha is among he la ges ound in he li e a u e, and has also so
a been a mo e e icien way o imp o e employmen han in many o he coun-
ies. Howe e , i should be no ed ha he e o m-induced dec ease in he e i e-
men a e in Finland has been un il now a enua ed in he coho s unde inspec-
ion because disabili y pensions whe e he disabili y had s a ed be o e 2017 au-
oma ically changed in o an old-age pension a age 63, he p e- e o m SRA. This
means ha he p opo ion o employmen o e he e o m-induced dec ease in
e i emen will mos likely dec ease in he u u e. This e ec will be p onounced
in he g oups whe e he incidence o disabili y is high: among indi iduals in lowe
socioeconomic posi ion.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
58
Re e ences
A alay, K. & Ba e , G.F. (2015): The impac o age pension eligibili y age on
e i emen and p og am dependence: e idence om an Aus alian expe i-
men . The Re iew o Economics and S a is ics 97(1), 71–87.
A a , T., Jongen, E. & Raba é, S. (2019): The E ec s o he Inc ease in he Re-
i emen Age in he Ne he lands. CPB discussion pape .
A a , T., Jongen, E. & Raba é, S. (2021): Inc easing he E ec i e Re i emen
Age: Key Fac o s and In e ac ion E ec s. IZA DP No. 14150.
Ba schke , M., Geye , J., Haan, P. & Hamme schmid, A. (2022): The e ec s o
an inc ease in he e i emen age on heal h – E idence om adminis a i e
da a. Jou nal o he Economics o Ageing, 23, 100403.
h ps://doi.o g/10.1016/j.jeoa.2022.100403
Behaghel, L., & Blau., D.M. (2012): F aming social secu i y e o m: Beha io al
esponses o changes in he ull e i emen age. Ame ican Economic Jou -
nal: Economic Policy 4(4), 41-67.
C ibb, J., Emme son, C. & Te low, G. (2016): Signals ma e ? La ge e i emen
esponses o limi ed inancial incen i es. Labou Economics 42, 203–2012.
h ps://doi.o g/10.1016/j.labeco.2016.09.005
Duggan, M., Single on, P. & Song, J. (2007): Aching o e i e? The ise in he ull
e i emen age and i s impac on he social secu i y disabili y olls. Jou nal o
Public Economics 91(7), 1327–1350.
Ebbinghaus, B. & Ho äcke , D. (2013): Re e sing Ea ly Re i emen in Ad anced
Wel a e Economies. A Pa adigm Shi o O e come Push and Pull Fac o s.
Compa a i e Popula ion S udies 38(4), 807–840.
Eu os a (2024): Popula ion by educa ional a ainmen le el, sex and age (%).
h ps://ec.eu opa.eu/eu os a /da ab owse / iew/EDAT_LFS_9903__cus-
om_9813243/de aul / able?lang=en&page= ime:2022 [accessed Feb ua y
7 h, 2024].
Finnish Cen e o Pensions s a is ical Da abase (2024): Numbe o new e i ees
on ea nings- ela ed pension. h ps:// ilas-
o .e k. i/pxweb/en/ETK/ETK__120 yoelakkeensaaja __40 yoelakkeelle_si-
i yneiden_lkm/elsi_ 07_ k_laji.px/ able/ ableViewLayou 1/ [accessed Feb-
ua y 19 h ,2024].
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
59
Geye , J., Haan, P., Hamme schmid, A. & Pe e s, M. (2020): Labo Ma ke and
Dis ibu ional E ec s o an Inc ease in he Re i emen Age. Labou Econom-
ics 65. h ps://doi.o g/10.1016/j.labeco.2020.101817
Geye , J. & Wel eke, C. (2021): Closing Rou es o Re i emen o Women How
Do They Respond? Jou nal o Human Resou ces 56(1), 311–341.
muse.jhu.edu/a icle/780193
G ube , J., Kanninen, O. & Ra aska, T. (2022): Relabeling, e i emen and eg e .
Jou nal o Public Economics 211(July): 104677.
h ps://doi.o g/10.1016/j.jpubeco.2022.104677
Hin ichs, K. (2021): Recen pension e o ms in Eu ope: Mo e challenges, new
di ec ions. An o e iew. Social Policy & Adminis a ion 55(3), 409–422.
h ps://doi.o g/10.1111/spol.12712
Inde bi zin, L., S aubli, S. & Zweimülle , J. (2016): Ex ended unemploymen ben-
e i s and ea ly e i emen : P og am complemen a i y and p og am subs i u-
ion. Ame ican Economic Jou nal: Economic Policy 8(1), 253-288.
Ka ls öm, A., Palme, M. & S ensson, I. (2008): The employmen e ec o s ic e
ules o eligibili y o DI: E idence om a na u al expe imen in Swe-
den. Jou nal o Public Economics 92(10-11), 2071-2082.
Manoli, D.S. & Webe , A. (2016): The E ec s o he Ea ly Re i emen Age on
Re i emen Decisions. Na ional Bu eau o Economic Resea ch Wo king Pa-
pe 22561.
Mas obuoni, G. (2009): Labo supply e ec s o he ecen social secu i y bene i
cu s: Empi ical es ima es using coho discon inui ies. Jou nal o Public Eco-
nomics 93(11–12), 1224–1233.
Mo is, T. (2021): Re‐examining emale labo supply esponses o he 1994 Aus-
alian pension e o m. Re iew o Economics o he Household.
h ps://doi.o g/10.1007/s11150-021-09561-y
Mo is, T. (2022): The unequal bu den o e i emen e o m: E idence om Aus-
alia. Economic Inqui y 60, 592-619.
Mäcken, J., P äg, P., Hess, M. & Ellwa d , L. (2022): Educa ional inequali ies in
labo ma ke exi o olde wo ke s in 15 Eu opean coun ies. Jou nal o Social
Policy 51(2), 435–459. h ps://doi.o g/10.1017/S0047279421000258
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
60
OECD (2015). Educa ion a a Glance In e im Repo : Upda e o Employmen and
Educa ional A ainmen Indica o s. OECD, Pa is. www.oecd.o g/edu/EAG-
In e im- epo .pd
Oude Hengel, K.M., Riumallo-He l, C., Sch am, J.L., Nieboe , D., an de Beek,
A.J. & Bu do , A. (2021): E ec s o changes in ea ly e i emen policies on
labo o ce pa icipa ion: he di e en ial e ec s o ulne able g oups. Scan-
dina ian Jou nal o Wo k, En i onmen & Heal h 47(3):224-32.
h ps://doi.o g/10.5271/sjweh.3946
Pol inen, A., Gould, R., Lahelma, E. & Ma ikainen, P. (2013): Socioeconomic
di e ences in disabili y e i emen in Finland: he con ibu ion o ill-heal h,
heal h beha iou s and wo king condi ions. Scandina ian Jou nal o Public
Heal h 41(5), 470-478. h ps://doi.o g/10.1177/1403494813482400
Raba é, S. & Rochu , J. (2020): Employmen and subs i u ion e ec s o aising
he s a u o y e i emen age in F ance. Jou nal o Pension Economics & Fi-
nance 19, 293–308. h ps://doi.o g/10.1017/S1474747218000392
Rob oek, S., Rongen, A., A s, C., O en, F., Bu do , A. & Schu ing, M.
(2015): Educa ional inequali ies in exi om paid employmen among Du ch
wo ke s: The in luence o heal h, li es yle and wo k. PLoS ONE 10(8), 1–12.
h ps://doi.o g/10.1371/jou nal.pone.0134867
Soosaa , O., Puu , A. & Leppik, L. (2021): Does aising he pension age p olong
wo king li e? E idence om pension age e o m in Es onia. Jou nal o Pen-
sion Economics & Finance 20(2), 317–335.
h ps://doi.o g/10.1017/S1474747220000244
S aubli, S. (2011): The impac o s ic e c i e ia o disabili y insu ance on labo
o ce pa icipa ion. Jou nal o Public Economics 95(9-10), 1223-1235.
S aubli, S. & Zweimülle , J. (2013): Does aising he ea ly e i emen age inc ease
employmen o olde wo ke s? Jou nal o Public Economics 108, 17–32.
h ps://doi.o g/10.1016/j.jpubeco.2013.09.003
Re o ming he Pension Sys em in Ge many –
An Empi ical S udy o Re o m A e sions
Ma hias Die meie / Ru h Ma ia Schüle
Con ac In o ma ion
D . Ru h Ma ia Schüle
Economis o Social Secu i y Sys ems and Income and Weal h Dis ibu ion a he
Ge man Economic Ins i u e in Cologne
E-Mail: schuele @iwkoeln.de
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
62
Abs ac
The Ge man pension sys em is unde eno mous p essu e o e o m due o de-
mog aphic change. Despi e conside able conce n, no e o m o insu ance sys em
mechanisms enjoys majo i y suppo among he Ge man popula ion. In a ac o ial
su ey expe imen conduc ed as pa o a popula ion su ey by he Ge man Eco-
nomic Ins i u e (IW) 2023, con ibu ion a e, pension le el and e i emen age a e
examined in ela ion o each o he , and po en ial e o m adjus men s a e simu-
la ed. E en he explici p esen a ion o e o m scena ios is unable o o e come
he exis ing a e sion o e o m. The s a us quo ecei es majo suppo , bu in
heo y i would only be inancially sus ainable by inc easing ax subsidies, which
would place a conside able bu den on he s a e budge . In a di ec compa ison o
e o m op ions, an inc ease in con ibu ion a es is opposed he leas . Pension
cu s a e seen as he mos pain ul. Delaying e i emen by one yea is iewed as
nega i ely as an inc ease in he con ibu ion a e o a ound h ee pe cen age
poin s o a educ ion in pension le els o a ound 4 pe cen .
The lowes accep ance o e o ms was o be ound among hose whom hey
would leas a ec : he o e -50-yea -olds. Among hose unde he age o 50, who
a e mos conce ned abou hei pensions, he ejec ion o all e o m op ions is
conside ably weake . The ac ha he isible p essu e on he pension sys em is
pushing younge people in pa icula o be mo e lexible should se e as an a -
gumen o a ge he o e -50s and highligh he consequences o inac ion o so-
cie y as a whole. Failu e o ac would ei he equi e a g ea e commi men o oc-
cupa ional and p i a e pension p o ision o , while gua an eeing a cons an le el
o pensions, lead o an immense bu den on he s a e budge . Delaying he e o m
u he isks ein o cing he public’s expec a ion o a cons an le el o pensions
wi hou hei ha ing o con ibu e, educing he scope o manoeu e and in ensi-
ying he deba e on pensions.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
63
Table o Con en s
Abs ac .............................................................................................................. 62
Table o Con en s .............................................................................................. 63
Lis o Figu es .................................................................................................... 63
Lis o Tables ..................................................................................................... 63
3.1 Ageing and pension insu ance .................................................................. 64
3.2 Pension e o m p e e ences acco ding o a popula ion su ey ................. 64
3.3 Fac o ial su ey expe imen ....................................................................... 67
3.4 A e sion o imposi ions in he pension deba e .......................................... 70
3.5 S ong s a us quo bias ............................................................................... 72
3.6 Ine i able cu s ............................................................................................ 74
Re e ences ........................................................................................................ 77
Lis o Figu es
Figu e 1: Ranking o dimensions o pension e o m in Ge many,
pe cen age o esponden s ......................................................... 66
Figu e 2: Example o a igne e ............................................................... 68
Figu e 3: Coe icien plo s o e o m sa is ac ion ..................................... 73
Lis o Tables
Table 1: Dimensions and le els o he ac o ial su ey expe imen
on pension e o ms ..................................................................... 69
Table 2: Ranking o e o m op ions conside ing he Ge man pension
sys em ........................................................................................ 70
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
64
3.1 Ageing and pension insu ance
The s a u o y pension insu ance sys em in Ge many is o ganised as a pay-as-
you-go sys em. This means ha oday’s con ibu o s inance he pensions o cu -
en pensione s. Howe e , due o he ageing o Ge man socie y, he e a e in-
c easingly ewe con ibu o s compa ed o he numbe o pensione s. This is e-
lec ed in he de elopmen o he old-age dependency a io. In 2021, he e we e
32 people aged o e 67 o e e y 100 people in he ypical wo king age g oup o
20-67 (S a is isches Bundesam , 2023). In 2035, he old-age dependency a io
will exceed 40 o he i s ime. This means ha he pension sys em’s expendi u e
will no be co e ed by i s e enue unde cu en condi ions.
In he Ge man pay-as-you-go sys em, h ee inhe en mechanisms a e designed
o balance he pension sys em’s e enues and expendi u es:
• Inc easing he con ibu ion a e can gene a e mo e e enue.
• Lowe ing he pension le el can educe pension insu ance expendi u e.
• Raising he e i emen age means ha people e i e la e and he e o e e-
cei e a pension o a sho e pe iod and con ibu e o a longe pe iod as hey
emain in he labou o ce longe .
Howe e , all hese inhe en e o m op ions a e no suppo ed by he majo i y o
he Ge man popula ion (Fo schungsg uppe Wahlen, 2018; Boe i e al., 2002).
The ie ce deba e on pension e o m and he s ong p o es s agains he ecen
inc ease in he s a u o y e i emen age om 62 o 64 in F ance illus a e he so-
cial ensions inhe en in such e o ms. To iden i y a e o m op ion o he Ge man
pension sys em ha has a ealis ic chance o being implemen ed wi hin he Ge -
man poli ical sys em, i is impo an o unde s and he a i ude o he Ge man
popula ion owa ds hese h ee mechanisms. This in ol es explici ly explaining
o esponden s he in e ela ionship be ween hese h ee mechanisms and he
consequences o inac ion, and mo i a ing hem o weigh up he h ee op ions.
This s udy analyses a ac o ial su ey expe imen conduc ed as pa o a popula-
ion su ey by he Ge man Economic Ins i u e (IW) in 2023 ( o in o ma ion on he
su ey and he sample, see Die meie e al., 2023).
3.2 Pension e o m p e e ences acco ding o a popula ion su ey
As pa o a popula ion su ey by he Ge man Economic Ins i u e 2023, espond-
en s we e asked in which o he h ee dimensions a change would be mos o
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
65
leas accep able om a social pe spec i e. Responden s we e asked o ank he
h ee mechanisms – highe con ibu ion a e, highe e i emen age and lowe
pension le el – acco ding o hei pe cei ed “accep abili y”. In he in oduc ion o
he expe imen , esponden s we e in o med abou he inancing di icul ies o he
pension sys em in he con ex o demog aphic change. Via an online access
panel p o ided by Bilendi & espondi, be ween 27 Feb ua y and 6 Ma ch 2023,
he su ey ecei ed 4,899 alid esponses om indi iduals aged 18 and o e on
hei p e e ences ega ding pension insu ance. The su ey was ep esen a i e
and s a i ied by gende , age (c oss- abula ed), place o esidence by ede al
s a e and ne mon hly household income (Die meie e al., 2023). The numbe o
esponden s who iden i ied hemsel es as pensione s in he s a u o y pension
insu ance sys em was 28.2 pe cen and 3.1 pe cen in ci il se ice espec i ely.
The pe cen age o pensione s was hus sligh ly lowe han he pe cen age o peo-
ple in Ge many who ecei e a pension om he s a u o y pension insu ance sys-
em, which is 30 pe cen . The numbe o pensione s was 2 pe cen .
O e all, abou 35 pe cen o he adul Ge man popula ion exp essed g ea con-
ce n o e hei own e i emen p o ision. The desc ip i e analyses (Figu e 1) also
show a clea hie a chy in e ms o which dimensions o he de e io a ion in he
pension insu ance si ua ion a e pe cei ed as pa icula ly pain ul. 52.7 pe cen o
esponden s in Ge many said ha a lowe pension le el would be he leas ac-
cep able. 54.8 pe cen o esponden s conside ed a highe con ibu ion a e o
be he mos accep able. Acco ding o he ex en o which i would a ec hem,
59.1 pe cen o pensione s said a pension cu would be he leas accep able.
The pe cen age o hose who ind a highe e i emen age mos accep able is
highes among pensione s no a ec ed by his e o m op ion, a 35 pe cen . In-
e es ingly, he iews o hose o e he age o 50 who do no iden i y hemsel es
as pensione s a e only sligh ly di e en . Howe e , hose unde he age o 50 ha e
a e y di e en pe spec i e. They a e mo e likely o accep a lowe pension le el.
They a e mos opposed o a highe e i emen age. An ini ial desc ip i e inding
sugges s ha p e e ences a y acco ding o gene a ional a ilia ion and a e hus
pa ly ela ed o indi idual conce ns.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
72
3.5 S ong s a us quo bias
A ac o ial su ey expe imen allows a di ec compa ison o di e en dimensions
h ough he app o al o e o m scena ios, as well as a dis inc ion o e ec s be-
ween di e en gene a ions. I is ue ha esul s o da e ha e indica ed a clea
p io i isa ion – inc easing con ibu ions is he leas pain ul; cu ing pensions is
pa icula ly unpleasan . Howe e , hey ha e no aken in o accoun ha he le els
o he dimensions a e exp essed in di e en uni s – such as inc easing he e i e-
men age in wo-yea inc emen s and educing he pension le el in 10 pe cen
inc emen s.
A eg ession wi h e o m consen as he dependen a iable and he h ee dimen-
sions o e i emen age, con ibu ion a e and pension le el as independen a i-
ables p o ides an insigh in o hei ela i e impo ance. The mos a ou able le el
o each dimension is de ined as he e e ence ca ego y and compa ed o he o he
wo le els. As a la ge pa o he a iance can be explained by di e en esponse
pa e ns among esponden s, a mul ile el model is es ima ed. The mul ile el
models a e speci ied sepa a ely o pensione s, he g oup o non-pensione s o e
he age o 50 and hose unde he age o 50.
Figu e 3 shows he coe icien plo s o he eg essions. O e all, inc easing he
con ibu ion a e and e i emen a 67 mee wi h he leas esis ance. Pension cu s
and e i emen a 69 mee wi h he s onges esis ance. On he ag eemen scale
om 1 o 7, a 20 pe cen educ ion in pensions leads o he s onges ejec ion
in all h ee g oups, which is s a is ically signi ican . While he app o al o e o m
dec eases by mo e han 1.3 scale poin s o esponden s o e he age o 50 in
he case o a 20 pe cen educ ion in pensions, he e ec o hose unde he age
o 50 is s a is ically signi ican ly lowe , a 1.17 scale poin s. I is no ewo hy ha
aising he e i emen age o 69 is he second mos ejec ed op ion among he
wo g oups o e he age o 50, e en hough hey would no be a ec ed o only
pa ially a ec ed (non-pensione s o e he age o 50). O he h ee g oups, he
non-pensione s who ha e no ye eached e i emen age a e mos opposed o
e i emen a 67 – o his g oup he s anda d e i emen age is al eady a leas
66. An inc ease in he con ibu ion a e o 22 pe cen is he leas opposed o all
age g oups. Despi e he g ea e impac on he younge age coho , an inc ease
in he con ibu ion a e o 25 pe cen also mee s wi h he leas esis ance.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
73
Figu e 3: Coe icien plo s o e o m sa is ac ion; Mul ile el andom e ec s
(GLS) and clus e ed esponden -le el s anda d e o s
A mul ile el model con ols he hie a chical da a s uc u e o a da ase – in his case, he
nine igne es a ed by he 4,899 esponden s. Such a model was speci ied because, a
38 pe cen , a signi ican p opo ion o he a iance can be explained by di e en esponse
pa e ns be ween esponden s (in a-class co ela ion coe icien : 0.38). Only 62 pe cen
o he a iance can be explained wi hin each esponden ’s nine answe s. Basis: 4,899
obse a ions om a popula ion su ey by he Ge man Economic Ins i u e in Feb ua y and
Ma ch 2023.
Sou ce: Ge man Economic Ins i u e
Compa isons o he e ec sizes show ha a e i emen age o 69 would ha e a
simila ly nega i e impac on hose unde he age o 50 as a 10 pe cen educ ion
in hei pension. Fo olde gene a ions, howe e , a e i emen age o 69 would be
signi ican ly wo se. Assuming a linea ela ionship, an inc ease in he e i emen
age by one yea would be ejec ed by all h ee demog aphic g oups in a simila
way as an inc ease in he con ibu ion a e o a ound 3 pe cen age poin s o a
educ ion in pensions o a ound 4 pe cen . This ela ionship emains consis en ,
wi h small a ia ions, e en when con olling o socioeconomic cha ac e is ics
(educa ion le el, income, gende , child en, age) and he esiden ial loca ion o he
esponden s (Eas s. Wes ).
-1,50 -1,25 -1,00 -0,75 -0,50 -0,25 0,00
Re i emen age o 67 ( e .: 65)
Re i emen age o 69 ( e .: 65)
Con ibu ion a 22% ( e .: 18.6%)
Con ibu ion a 25% ( e .: 18.6%)
Pensions 10% lowe ( e .: same)
Pensions 20% lowe ( e .: same)
Pensione s Non-Pensione s abo e 50 yea s Unde 50 yea s
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
74
3.6 Ine i able cu s
People unde he age o 50 a e he mos conce ned abou hei pensions and a
he same ime he leas esis an o pension e o m. I is he e o e impo an o
de e mine how o use he window o oppo uni y o e o m c ea ed by ising e-
i emen conce ns among younge people be o e demog aphic ends shi ma-
jo i ies u he owa ds he olde popula ion.
The majo i y o Ge mans emain opposed o any changes ha would make he
pension sys em mo e sus ainable. The desi e o main ain he s a us quo is deeply
ing ained. Responden s a e mos willing o accep a highe con ibu ion a e,
while a he same ime s essing he impo ance o main aining a cons an pen-
sion le el. Speci ically, delaying e i emen by one yea is abou as pain ul as
inc easing he con ibu ion a e by abou 3 pe cen age poin s o educing he pen-
sion le el by abou 4 pe cen .
This is a e y uncom o able si ua ion o policymake s. While demog aphic p es-
su es undoub edly equi e a high p ice o be paid o sa is y he desi e o high
pension paymen s, any adjus men wi hin he insu ance logic o he s a u o y pen-
sion sys em would equi e he unco e ed expenses o be inanced om he ed-
e al budge o o he sou ces. The cu en s udy shows ha a e o m would alien-
a e a ious popula ion g oups. Howe e , he leas objec ionable pa h o inc eas-
ing he con ibu ion a e may u n ou o be misguided o se e al easons.
Fi s ly, i is unclea whe he people ejec an abs ac inc ease in he con ibu ion
a e simply because i is less angible han a la e e i emen age o a pension
cu . Once he inancial bu den becomes angible, ejec ion o he e o m may ol-
low. The e is e idence ha people who can p ecisely es ima e he con ibu ion
a e a e much mo e likely o ejec i s inc ease.
Secondly, highe con ibu ion a es also a ec companies, some o which a e cu -
en ly ope a ing a capaci y. The esul ing highe labou cos s o employe s could
lead o nega i e employmen e ec s and educed in es men ac i i y by compa-
nies. Fo example, Beznoska e al. (2017) show ha an equal di ision o addi-
ional con ibu ions o s a u o y heal h insu ance be ween employees and em-
ploye s would lead o a decline in g oss domes ic p oduc and an inc ease in he
unemploymen a e.
Finally, demog aphic ageing will equi e no only an inc ease in he con ibu ion
a e o s a u o y pension insu ance, bu also an inc ease in con ibu ions o heal h
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
75
and long- e m ca e insu ance. The implica ions o hese indi ec e ec s a e p ob-
ably no aken in o accoun by mos people in hei conside a ions, bu hey hold
conside able po en ial o disappoin men .
To some ex en , he en i lemen men ali y o he Ge man popula ion has been
ein o ced by poli ical p omises. The in oduc ion o non-deduc ible ea ly e i e-
men o pa icula ly long-se ing employees, he so-called “ e i emen a 63”, as
well as he mo he s’ pension and he o en- oiced demands o ixing he pension
le el and he con ibu ion a e – and hei ac i e implemen a ion h ough he dou-
ble s abilisa ion lines – ga e he imp ession ha he s a us quo could be main-
ained o e en imp o ed wi hou addi ional inancial bu dens. This is no he case.
I con ibu ion a es and e i emen ages a e no adjus ed, he Ge man popula-
ion’s p io i y wish o a cons an le el o pensions o he cu en gene a ion o
pensione s will lead o eno mous bu dens. I is necessa y o explain o he public
ha his would esul in educed public se ices o highe axes o he popula ion
as a whole, possibly a he expense o he ne income de elopmen o younge
coho s. In addi ion, he pension gua an ee ensu es ha pensions will no be cu .
The p io i y wish o he Ge man popula ion is al eady ensh ined in cu en law. I
is he e o e impo an o emphasise and cla i y ha dec easing he s a is ical pen-
sion le el does no mean ha he pensions paid will dec ease. The s a is ical
pension le el simply compa es he s anda d pension, i.e., he pension o a hypo-
he ical pensione who has ecei ed he a e age wage o all pensione s o 45
yea s and paid he co esponding con ibu ions, o he cu en a e age wage o
all pensione s. The s a is ical pension le el hus only se es o compa e he pe -
o mance o he pension sys em o e ime.
To p omo e unde s anding o he undamen al e o ms o he Ge man pension
sys em, mo e emphasis should be placed on he mul iple challenges o demo-
g aphic change and he consequences o indi iduals should be illus a ed wi h
examples. This equi es bo h clea ly iden i ying he indi idual and collec i e bu -
dens o e o m and highligh ing he ade-o s be ween di e en mechanisms in
he pension sys em. In pa icula , he oppo uni y cos s o inac ion should be
clea ly highligh ed (Ens e e al., 2009). Main aining he s a us quo would equi e
highe ax subsidies, limi ing oppo uni ies o in es men and o he go e nmen
unc ions. Wi h ega d o he e i emen age, o example, i should be con eyed
ha la e e i emen bene i s each indi idual in he o m o a highe pension (Pim-
pe z, 2023). Finally, pension paymen s depend on he amoun and du a ion o
con ibu ions.
The cu en analyses sugges ha he key o pension e o m wi h majo i y suppo
may lie in olde people aking he p e e ences o younge people se iously. A e
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
76
all, he la e a e pa icula ly a ec ed bu a e ne e heless mo e willing o accep
he cu s.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
77
Re e ences
Auspu g, K., Hinz, T. & Liebig, S. (2009): Komplexi ä on Vigne en, Le ne ek e
und Plausibili ä im Fak o iellen Su ey, in: Me hoden, Da en, Analysen
(mda), 3. Jg., N . 1, S. 59-96.
Beznoska, M., Kole , G. & Pimpe z, J. (2017): Mak oökonomische E ek e eine
pa i ä ischen Bei ags inanzie ung: Eine Analyse ak uelle Re o m o -
schläge ü die Gese zliche K anken- und die soziale P lege e siche ung,
IW-Policy Pape , N . 15, Köln.
Boe i, T., Boe sch-Supan, A. & Tabellini, G. (2002): Pension Re o ms and he
Opinions o Eu opean Ci izens, in: Ame ican Economic Re iew, 92. Jg., N .
2, S. 396-401.
DFPA – Deu sche Finanz P esse Agen u (2020): S udie: Deu sche um die 40 –
Gene a ion Ren enangs , 15.7.2020, h ps://www.d pa.in o/mae k e-s u-
dien-news/s udie-deu sche-um-die-40-gene a ion- en enangs .h ml
[20.7.2023].
Die meie , M., Engle , J.F. & Schä e , H. (2023): Zu iel ode zu wenig Re o m?.
Die ö en liche Ve handlung des Bü ge geldes, in: IW-T ends, 50. Jg., N . 1,
S. 101-124.
Ens e, D.H., Ha e kamp, A. & Fe chenhaue , D. (2009): Un e schiede im Denken
zwischen Ökonomen und Laien – E klä ungsansä ze zu Ve besse ung de
wi scha spoli ischen Be a ung, in: Pe spek i en de Wi scha spoli ik, 10.
Jg., N . 1, S. 60-78.
Fo schungsg uppe Wahlen (2018): Poli ba ome e Augus II 2018,
h ps://www. o schungsg uppe.de/Um agen/Poli ba ome e /A chi /Poli ba-
ome e _2018/Augus _II_2018/ [14.7.2023].
Ge man Economic Ins i u e (2023): Popula ion Su ey by he Ge man Economic
Ins i u e (IW) 2023, Köln.
Häuse mann, S, Ku e , T. & T abe , D. (2019): The Poli ics o T ade-O s:
S udying he Dynamics o Wel a e S a e Re o m wi h Conjoin Expe imen s,
in: Compa a i e Poli ical S udies, 52. Jg., N . 7, S. 1059-1095.
Pimpe z, J. (2023): Länge a bei en lohn sich im Al e , IW-Ku zbe ich , N . 65,
Köln.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
78
S a is isches Bundesam (2023): E gebnisse de 15. koo dinie en Be ölke ungs-
o ausbe echnung, En wicklung de Be ölke ungszahl bis 2070 nach aus-
gewähl en Va ian en de 15. Koo dinie en Be ölke ungs o ausbe echnung,
Wiesbaden.
Non inancial De ined Re u n (NDR) Pension
F amewo k and a New Pe spec i e on Pension
Sys em Sus ainabili y
Ma in D ees
Con ac In o ma ion
Ma in D ees
PhD s uden in ma hema ics a he Uni e si y o Bonn
E-Mail: mail@md ees.de
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
80
Abs ac
We p opose Non inancial De ined Re u n (NDR) as a no el abs ac amewo k
o pension sys ems, building upon he Non inancial De ined Con ibu ions
(NDC) scheme. NDC has been implemen ed in se e al coun ies, including
Sweden. I sha es key ea u es wi h he Ge man pension sys em. NDR empha-
sizes a di ec link be ween con ibu ions and bene i s, wi h each uni o con ibu-
ion co esponding o one expec ed uni o bene i . Pa icipan s accumula e
index poin s acco ding o speci ic con ibu ion ules. These index poin s a e
hen con e ed in o annual pensions ia an insu ance mechanism. Balancing
ules manage he di e ence be ween con ibu ion e enues and pension ex-
penses.
NDR-GDP ep esen s he special case whe e he indexa ion me hod is ied o
GDP, balanced h ough go e nmen ans e s by de aul . We u he in oduce
he concep o del a-sus ainabili y. A pension sys em is called del a-sus ainable
i he sum o un unded liabili ies ela i e o GDP emains cons an o e ime,
and go e nmen ans e s educe liabili ies by he amoun o he ans e . NDR-
GDP is del a-sus ainable, jus i ying he balancing ia go e nmen ans e s.
The NDR amewo k and he concep o del a-sus ainabili y o e esh pe spec-
i es on pension sys em sus ainabili y, wi h NDR-GDP p esen ing a no el ap-
p oach o add ess demog aphic challenges in pension e o ms.
Acknowledgemen s
I would like o hank Meike Neuwohne , Be Rü up, Ch is ina Wilke, Ch is ian
Zimpelmann and many o he s o ui ul discussions and help ul commen s.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
81
Table o Con en s
Abs ac .............................................................................................................. 80
Lis o Figu es .................................................................................................... 82
4.1 In oduc ion ................................................................................................ 83
4.2 Simpli ied desc ip ion o NDC .................................................................... 85
4.3 Non inancial De ined Re u n amewo k .................................................... 88
4.3.1 Index poin s ................................................................................... 88
4.3.2 Sepa a ing he insu ance mechanism .......................................... 89
4.3.3 Desc ip ion o NDR ....................................................................... 90
4.3.4 Equi alence p inciple e sus pa icipa ion equi alence ............... 91
4.4 NDR-GDP .................................................................................................. 92
4.4.1 Desc ip ion .................................................................................... 92
4.4.2 NDR-GDP and NDC wi hin he con ex o NDR ............................ 93
4.5 Del a-sus ainabili y ..................................................................................... 95
4.5.1 De ini ion ....................................................................................... 95
4.5.2 NDR-GDP is del a-sus ainable ..................................................... 95
4.5.3 Compa ison wi h simila no ions o pension
sys em sus ainabili y ..................................................................... 96
4.6 High-le el e o m s a egy: In a ian s and lexibili y .................................. 97
4.7 C ucial di e ence o NDC and NDR-GDP ................................................. 98
4.8 Conclusion and ou look ............................................................................. 99
Re e ences ...................................................................................................... 100
Appendix 4.1:Te minology o Non inancial De ined Re u n amewo k .......... 102
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
88
Fo his pape , a basic unde s anding o he concep o u u e con ibu ions, as
illus a ed by he abo e example, is su icien . The key poin is ha u u e con-
ibu ions a e concep ually in eg a ed in o he conside a ion o inancial balance.
Se e g en and Mikula (2005) p o ide a comp ehensi e de ini ion o he u no e
du a ion and he u u e con ibu ions.
Ra e o e u n and au oma ic balancing mechanism (ABM)
By le e aging he concep s o liabili ies and asse s, an au oma ic balancing
mechanism (ABM) can be implemen ed wi hin an NDC scheme. When asse s
and liabili ies a e unequal, adjus men s o he a e o e u n can es o e equilib-
ium.
In heo y, one could di ec ly manipula e he alue o liabili ies o main ain pa i y
wi h asse s, he eby implying a con inuous change o alues in indi idual ac-
coun s. Howe e , in p ac ice, an a p io i choice o he a e o e u n is made —
o ins ance, ied o changes in he wage sum. I he dispa i ies be ween liabili-
ies and asse s become signi ican , he ABM in e enes o adjus he a e o
e u n acco dingly.
In his pape , NDC e e s o a scheme whe e he a e o e u n is de e mined by
inancial balance, simila ly o wha Palme (2005) e e s o as gene ic NDC.
4.3 Non inancial De ined Re u n amewo k
Wi hin his sec ion, we p esen he no el abs ac pension amewo k Non inan-
cial De ined Re u n (NDR). Simila o he NDC scheme, he NDR amewo k
employs he concep o no ional indi idual accoun s. Howe e , unlike he NDC
scheme, i does no equi e ixed con ibu ion a es o a a e o e u n induced
by inancial balance; ins ead, i only necessi a es he exis ence o wha we call
con ibu ion ules and any a e o e u n. Be o e del ing in o he in icacies o
he NDR amewo k, we will i s in oduce wo undamen al componen s o ab-
s ac ion: he concep o index poin s and he sepa a ion o he insu ance
mechanism.
4.3.1 Index poin s
Ins ead o ep esen ing a no ional alue di ec ly on indi idual accoun s, he
amewo k u ilizes index poin s. These index poin s possess a ixed alue a any
gi en ime. Ra he han applying he a e o e u n di ec ly o he mone a y alue
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
89
wi hin indi idual accoun s, i is applied o he alue o an index poin . While his
app oach is equi alen in ou come, he numbe o index poin s emains cons an
when he a e o e u n is applied. Consequen ly, his me hod cap u es wha
emains cons an when no con ibu ions a e made.
Subsequen ly, we will desc ibe con ibu ions no ed on indi idual accoun s as
he p ocess o buying index poin s.
4.3.2 Sepa a ing he insu ance mechanism
In a pension sys em, a key unc ion is o insu e indi iduals agains he isk o
ou li ing hei sa ings by p o iding li elong income upon e i emen . I is wo h
no ing ha he e exis s he possibili y o sepa a e he accumula ion o pension
claims om his insu ance aspec . Gi en ha insu ance na u ally en ails he
possibili y o gene a ing su pluses o de ici s, he idea is o seg ega e his as-
pec om he b oade pension sys em. Ra he han allowing o he con e sion
o alue om he indi idual accoun in o a li elong annui y wi hin he pension
sys em, a dedica ed pension insu ance mechanism is in oduced.
The pension insu ance is esponsible o disbu sing li elong annui ies, while any
su pluses o de ici s wi hin he insu ance mechanism a e in e nally managed.
Fo a isual ep esen a ion o hese mechanics, e e o Figu e 2. The admin-
is a ion handles he buying and selling o index poin s, a p ocess ha may po-
en ially esul in imbalances, a opic we will explo e u he .
Figu e 2: Sepa a ing he insu ance mechanism
Sou ce: Own compu a ion
Adminis a ion
Buy index poin s
Pa icipan s Pension insu ance
Sell index poin s
Index poin s o
insu ance benefi
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
90
4.3.3 Desc ip ion o NDR
The NDR amewo k4 pu s a s ong emphasis on he con ibu ion-bene i link,
ensu ing ha e e y uni o con ibu ion esul s in one uni o pension insu ance
bene i . These uni s a e ep esen ed by index poin s, he alue o which chang-
es acco ding o an indexa ion me hod.
Con ibu ion ules delinea e how pa icipan s con ibu e o he sys em, in ol ing
he pu chase o index poin s a hei cu en alue, which a e hen eco ded on
indi idual accoun s.
Th ough an insu ance mechanism, pa icipan s ha e he op ion o exchange
index poin s o pension bene i s, such as a li elong annui y. Subsequen ly, he
pension insu ance sells index poin s o co e he cos o hese bene i s.
Simila o he NDC scheme, he e enue om con ibu ions is no in es ed in
inancial ma ke asse s bu is u ilized o pay pensions o o he pa icipan s,
albei indi ec ly h ough he pension insu ance.
The o al e enue om con ibu ions may di e om he expenses o he pen-
sion insu ance, esul ing in a disc epancy be ween he numbe s o index poin s
bough and sold, espec i ely. Balancing ules a e employed o add ess such
si ua ions.
Examples o balancing ules include he es ablishmen o a ese e und, he
o mula ion o con ibu ion ules and indexa ion me hods, and he p o ision o
go e nmen ans e s.
In summa y, he NDR amewo k comp ises he ollowing componen s:
• Con ibu ion ules
• Indexa ion me hod
• Insu ance mechanism
• Balancing ules
I is impo an o no e ha he NDR amewo k emains abs ac , as i , o exam-
ple, does no p esc ibe speci ic con ibu ion ules. Ra he , i manda es ha any
conc e e pension sys em ollowing his amewo k mus de ine i s con ibu ion
ules.
4 A discussion on he e minology o he NDR amewo k can be ound in Appendix
4.1.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
91
La e on, we will del e in o how NDC aligns wi h he NDR amewo k. Addi ion-
ally, we will in oduce NDR-GDP, which also con o ms o he amewo k bu
implemen s di e en choices han NDC.
4.3.4 Equi alence p inciple e sus pa icipa ion equi alence
The NDR amewo k ope a es on he co e p inciple ha one index poin o con-
ibu ion co esponds o one uni o (expec ed) pension bene i , a p inciple
e med he equi alence p inciple. In con as , he Ge man pension sys em elies
on he p inciple o (pa icipa ion equi alence), whe e e e y uni o con ibu ion a
he same poin in ime mus lead o he same expec ed pension claim.
To illus a e he concep o pa icipa ion equi alence wi hin he NDR amewo k,
we u ilize he no ions o index poin s and he sepa a ed insu ance mechanism.
Pa icipa ion equi alence en ails ha a ime 𝑡𝑡, a con ibu ion o one index poin
leads o a pension bene i o 𝑥𝑥𝑡𝑡 index poin s, whe e 𝑥𝑥𝑡𝑡 is equal o all pa ici-
pan s. Unlike he equi alence p inciple, 𝑥𝑥𝑡𝑡 may a y o e ime, and i may di e
om 1. I is impo an o no e ha he equi alence p inciple di ec ly implies pa -
icipa ion equi alence.
The call o he equi alence p inciple ep esen s a s onge claim han ad oca -
ing o pa icipa ion equi alence. While he o me demands an exac co e-
spondence be ween con ibu ions and bene i s, he la e only equi es p opo -
ionali y, wi h he ac o possibly a ying o e ime. Howe e , s a emen s abou
he implica ions o pa icipa ion equi alence a e a s onge claim han hose
abou he equi alence p inciple. This is because any s a emen implied by pa -
icipa ion equi alence is also co e ed by he mo e speci ic equi alence p inci-
ple.
The possibili y o changing 𝑥𝑥𝑡𝑡 o e ime can o cou se ha e e ec s on in e gen-
e a ional edis ibu ion. Fu he mo e, e en i 𝑥𝑥𝑡𝑡 emains cons an bu di e s om
1, he e a e e ec s on in agene a ional edis ibu ion. Fo ins ance, i 𝑥𝑥𝑡𝑡= 1.2
o all 𝑡𝑡, he o al a e o e u n pe yea is highe o con ibu ions held o sho -
e pe iods because he bonus o 20 pe cen is sp ead o e a sho e pe iod o
ime. This edis ibu es income in agene a ionally om ea ly o la e ea ne s.
Addi ionally, one can a gue ha a p opo ional ac o di e en om 1 a ec s
edis ibu ion, i espec i e o his empo al pe spec i e. In he special case
whe e 𝑥𝑥𝑡𝑡= 0 o all 𝑡𝑡, he e a e no pension claims a all and hose wi h la ge
con ibu ions pay mo e bu ecei e no hing in e u n. Con e sely, i 𝑥𝑥𝑡𝑡=1000
o all 𝑡𝑡, doubling he con ibu ion has a signi ican ly smalle impac han dou-
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
92
bling he bene i . These e ec s a e much less p onounced bu s ill p esen i 𝑥𝑥𝑡𝑡
is close o 1 bu di e en .
While bo h he equi alence p inciple and pa icipa ion equi alence ha e implica-
ions o edis ibu ion, he dynamic alua ion o e ime and di e en li e expec-
ancies complica e asse ions abou edis ibu ion unde ei he p inciple.
4.4 NDR-GDP
In his sec ion, we will desc ibe NDR-GDP, a pension scheme ha adhe es o
he NDR amewo k, and compa e i wi h he NDC scheme.
4.4.1 Desc ip ion
NDR-GDP adop s GDP as he indexa ion me hod, whe e he alue o index
poin s is de e mined by a cons an ac o imes he cu en GDP. Unlike NDC,
NDR-GDP o e s comple e lexibili y in con ibu ion ules and imposes no e-
s ic ions on he insu ance mechanism. This lexibili y ex ends o he abili y o
change con ibu ion ules, making NDR-GDP pa icula ly adap able o e ol ing
ci cums ances. Balancing wi hin NDR-GDP is achie ed h ough go e nmen
ans e s, whe e he go e nmen buys and sells index poin s o ensu e he o al
numbe o bough and sold index poin s emains equal (see Figu e 3). The ad-
minis a ion hus ne e holds index poin s. An ini ial amoun o index poin s is
gi en by exis ing pension claims.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
93
Figu e 3: Schema ic o NDR-GDP
Sou ce: Own compu a ion
Fo simplici y, we assume ha he go e nmen always holds a nonnega i e
balance o index poin s. This can be p ac ically achie ed by combining he sys-
em wi h a unded sys em o u ilizing o ced payou mechanisms.
A undamen al cha ac e is ic o NDR-GDP is ha he o al sum o index poin s
ac oss pa icipan s, he pension insu ance, and he go e nmen emains con-
s an o e ime. This p ope y ensu es ha he numbe o index poin s held by
he go e nmen is bounded by he ini ial amoun in he sys em, limi ing he o al
ans e s ela i e o GDP.
The jus i ica ion o go e nmen ans e s is p o ided by del a-sus ainabili y,
which we will in oduce in Sec ion 5.
4.4.2 NDR-GDP and NDC wi hin he con ex o NDR
Figu e 4 illus a es he di e en choices made by NDC and NDR-GDP o he
co e componen s o he NDR amewo k.
Adminis a ion
Buy index poin s
Pa icipan s Pension insu ance
Sell index poin s
Index poin s o
insu ance benefi
Go e nmen
Buy/sell index poin s
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
94
Figu e 4: Choices o NDC and NDR-GDP
Sou ce: Own compu a ion
NDC ypically employs a ixed a e o wages as i s con ibu ion ule, ensu ing
ha con ibu ions a e se o allow o he compu a ion o he con ibu ion asse .
In con as , NDR-GDP o e s lexibili y in con ibu ion ules.
Rega ding indexa ion, NDC elies on he au oma ic balancing mechanism, while
NDR-GDP op s o GDP-based indexa ion. Long- e m balancing in NDC is
achie ed h ough he au oma ic balancing mechanism, while sho e - e m bal-
ancing is o en acili a ed by a ese e und. Con e sely, NDR-GDP elies on
go e nmen ans e s o balancing.
In many desc ip ions o NDC, he insu ance mechanism is po ayed as an in e-
g a ed componen , whe eas in NDR-GDP, i is en i ely sepa a ed and accoun -
ed o independen ly. I is possible o also sepa a e he insu ance aspec in
NDC schemes.
This compa ison highligh s he echnical dispa i ies be ween NDC and NDR-
GDP. In Sec ion 7, we will del e in o he c i ical concep ual di e ence be ween
NDC and NDR-GDP, pa icula ly ocusing on he no ion o pension sys em sus-
ainabili y.
NDR
NDR-GDP
NDC
Flexible
GDP
Go e nmen ans e s
Sepa a ed
Fixed a es o wages
Induced by ABM
Rese e und, ABM
In eg a ed
Con ibu ion ules
Indexa ion
Balancing
Insu ance
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
95
4.5 Del a-sus ainabili y
In his sec ion, we in oduce he cen al concep o his pape : del a-
sus ainabili y. We begin by p esen ing i s de ini ion. Subsequen ly, we demon-
s a e ha NDR-GDP adhe es o he p inciple o del a-sus ainabili y. Fu he -
mo e, we unde ake a compa a i e analysis, con as ing del a-sus ainabili y wi h
o he no ions o pension sys em sus ainabili y, in pa icula wi h he concep o
inancial balance p e alen in NDC schemes.
4.5.1 De ini ion
A pension sys em is e med del a-sus ainable i he sum o un unded liabili ies5
ela i e o an economic indica o does no inc ease o e ime. Addi ionally, any
ans e o alue 𝑥𝑥 in o he sys em mus lead o a dec ease in he sum o un-
unded liabili ies by 𝑥𝑥. Simila ly, in he case o a nega i e 𝑥𝑥, an inc ease by |𝑥𝑥| is
pe mi ed.
When he economic indica o is no explici ly designa ed, i de aul s o GDP,
gi en i s p e alence as he p ima y indica o o ela i e public deb .
Del a-sus ainabili y does no p o ide a snapsho o he cu en s a us bu in-
s ead ocuses on how he pension sys em’s s a us e ol es o e ime.
4.5.2 NDR-GDP is del a-sus ainable
We will demons a e ha NDR-GDP, wi h an insu ance mechanism ha is sepa-
a ely accoun ed o , adhe es o del a-sus ainabili y.
Recall ha he o al sum o index poin s held by he pa icipan s, he pension
insu ance, and he go e nmen emains cons an .6 Wi hou any ans e s, as
index poin s a e indexed ia GDP, he o al alue o index poin s o pa icipan s
and he pension insu ance ela i e o GDP emains cons an . These index
poin s ep esen he un unded liabili ies o he sys em. Thus, wi hou ans e s,
he sum o un unded liabili ies ela i e o GDP emains cons an .
When he go e nmen makes a ans e o alue 𝑥𝑥, i pu chases an amoun o
index poin s o o al alue 𝑥𝑥. As he o al sum o index poin s is cons an , he
5 These a e liabili ies no backed by inancial asse s, o example index poin s ha a e
no held by he go e nmen .
6 The adminis a ion does ne e hold index poin s in NDR-GDP.
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
96
numbe o index poin s held by pa icipan s and he pension insu ance dec eas-
es by ha amoun . Consequen ly, he alue o un unded liabili ies dec eases by
𝑥𝑥.
In summa y, NDR-GDP ul ills bo h condi ions o del a-sus ainabili y. No e ha
NDR, when employing an a bi a y indexa ion me hod alongside go e nmen
ans e s as a balancing ule, main ains del a-sus ainabili y ela i e o he cho-
sen indexa ion me hod. Consequen ly, NDR-GDP holds a dis inc posi ion, wi h
GDP assuming a undamen al ole in assessing ela i e deb le els.
4.5.3 Compa ison wi h simila no ions o pension sys em sus ainabili y
Va ious no ions o pension sys em sus ainabili y exis in he li e a u e, each
o e ing a unique pe spec i e. We will ocus on a compa ison o del a-
sus ainabili y wi h he closes ones sugges ed be o e and o inancial balance.
De esa-Ca pio and De esa-Ca pio (2010) p o ide a comp ehensi e o e iew
o no ions o pension sys em sus ainabili y o ypical PAYG pension sys ems
ha a e no NDC. An impo an concep ela ed o del a-sus ainabili y men ioned
in hei pape is he sum o un unded liabili ies, o en e e ed o as implici deb .
Fo ins ance, He d and an den Noo d (1993) ha e compu ed implici deb o
se e al la ge economies.
Howe e , De esa-Ca pio and De esa-Ca pio (2010) no e ha solely measu ing
implici deb does no p o ide a comp ehensi e de ini ion o pension sys em
sus ainabili y, as i emains unclea wha he maximum le el o implici deb
should be. To add ess his, hey in oduce he concep s o ac ua ial imbalance
and uni a y pension cos . These concep s compa e he o al bene i s ecei ed
by a g oup o pa icipan s o hei o al con ibu ions. This idea esembles he
no ion o del a-sus ainabili y in ha he sus ainabili y o he unning sys em is
conside ed. Ne e heless, hei app oach assumes a closed sys em and does
no di ec ly accoun o changes in un unded liabili ies.
Holzmann e al. (2004) emphasize he impo ance o measu ing implici deb
and sugges assessing pension e o ms by conside ing explici deb and implici
deb join ly, bu do no p o ide an explici ool o do so. The no ion o del a-
sus ainabili y cons i u es a way o implemen his sugges ion.
Financial balance, as de ined wi hin NDC schemes, is a c i ical me ic used o
e alua e pension sys em sus ainabili y. I e lec s he equilib ium be ween u u e
con ibu ions and pension bene i s. Mo e p ecisely, he no ion o inancial bal-
ance sugges s ha he sys em, unde p ede e mined con ibu ion a es, can
KCV Sch i en eihe, Vol. 8, Wilke: Pension Re o ms: No dic Coun ies and Ge many
97
po en ially ope a e au onomously wi hou necessi a ing ans e s. In essence,
achie ing inancial balance indica es ha con ibu ions a e p ojec ed o su i-
cien ly co e pension bene i s o e he long e m.
Financial balance, howe e , does no imply ha ha ing a pension sys em in
inancial balance absol es he s a e om o e all liabili ies. A s a e lacking such
a pension sys em could in oduce one and e ec i ely alloca e he esul ing un-
unded pension claims o i sel . Hence, ha ing he pension sys em in inancial
balance a leas ep esen s an “oppo uni y liabili y”. Indeed, he liabili y can also
be unde s ood as he obliga ion o main ain he se con ibu ion a es.
Addi ionally, assuming no unded ese es, p opo ionally inc easing pension
bene i s and con ibu ion a es does no al e whe he an NDC scheme is in
inancial balance. This scena io e ec i ely gene a es addi ional pension claims
wi hou inc easing inancial asse s.
In con as , del a-sus ainabili y o e s a di e en pe spec i e. I does no ea
he pension sys em as a closed en i y and explici ly pe mi s go e nmen ans-
e s. Del a-sus ainabili y conside s he holis ic liabili ies o he pension sys em
and ocuses solely on changes o e ime. When ansi ioning om a legacy
pension sys em o a new one, del a-sus ainabili y does no hold he new sys em
accoun able o exis ing liabili ies. Ins ead, i e alua es whe he he new sys em
in oduces u he liabili ies o e ime.
In summa y, inancial balance and del a-sus ainabili y add ess dis inc aspec s
o pension sys em sus ainabili y. While inancial balance pe ains o he sel -
su iciency o he sys em gi en p ede e mined con ibu ion a es, del a-
sus ainabili y o e s a b oade e alua ion, conside ing he sys em’s e olu ion
o e ime and i s impac on o e all liabili ies.
4.6 High-le el e o m s a egy: In a ian s and lexibili y
In his sec ion, we explo e he o e a ching e o m s a egy implied by he NDR-
GDP scheme. As p e iously men ioned, he implemen a ion o NDR-GDP in-
ol es con e ing exis ing pension claims in o index poin s. Howe e , he spe-
ci ics o his p ocess a e complex and con ingen upon he nuances o he cu -
en pension sys em. De e mining exis ing pension claims in ol es poli ical con-
side a ions, such as whe he hey should be compu ed based on pas con ibu-
ions o p omised bene i s.
Volumes Published in he KCV Sch i en eihe:
Volume 1 (2020)
P. an de Meulen / T. Ch is iaans / M. D ewes / S. F ohwe k / M. Göke / P. Hen-
necke / T. Hol o / K. Obe mann / L. Rebeggiani / A. Spe mann / C. Thiel-
sche / A.-K. Voi / C. B. Wilke / M. Wohlmann
Was kos e uns Co ona? Volkswi scha liche Auswi kungen eine globalen
Pandemie
ISBN (P in ) 978-3-89275-168-7 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-169-4 ISSN (eBook) 2701-9411
Volume 2 (2021)
A. M. P asuhn / C. B. Wilke
Re o mop ion Bü ge e siche ung? Eine Nu zwe analyse o dem Hin e g und
ak uelle und kün ige He aus o de ungen des deu schen K anken e siche-
ungssys ems
ISBN (P in ) 978-3-89275-172-4 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-173-1 ISSN (eBook) 2701-9411
Volume 3 (2021)
C. F i sche
On Mig a ion and Immig a ion Policy in Eu ope
ISBN (P in ) 978-3-89275-220-2 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-221-9 ISSN (eBook) 2701-9411
Volume 4 (2022)
L. Rebeggiani / C. Roppel / F. Sch ickel
Gib es in Deu schland zu iele K ankenkassen?
Eine Analyse de We bewe bsin ensi ä in de gese zlichen K anken e siche-
ung Deu schlands
ISBN (P in ) 978-3-89275-254-7 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-255-4 ISSN (eBook) 2701-9411
Volume 5 (2022)
L. Rebeggiani / M. D ewes
We bewe b und We bewe bspoli ik im Spo : Bookle zu 25. Jah es agung des
A bei sk eises Spo ökonomie
ISBN (P in ) 978-3-89275-272-1 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-273-8 ISSN (eBook) 2701-9411
Volume 6 (2022)
E. G oßpie sch
Challenges o Nu sing Schools due o Digi alisa ion
ISBN (P in ) 978-3-89275-296-7 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-297-4 ISSN (eBook) 2701-9411
Volume 7 (2023)
S. Bä / C. Benkel / T. Bezold / P. Biebl / M. B eue / O. Budzinski / S. Cha a h /
F. Daumann / S. Fauls ich / A. Fedde sen / B. F ick / F. Gassmann / J. Hell / L.
M. Hoang / I. Hugae s / A. Klad oba / T. Könecke / P. Kunz-Kal enhäuse / M.
Ku scheid / K. Lind / L. Mo sie e / G. Nu e / L. Rebeggiani / B. Römmel / H.
Schunk / V. Spindle / A. S ickel / L. Thieme / G. T osien
We bewe b und We bewe bspoli ik im Spo : Sammelband zu 25. Jah es a-
gung des A bei sk eises Spo ökonomie
ISBN (P in ) 978-3-89275-364-3 ISSN (P in ) 2701-9403
ISBN (eBook) 978-3-89275-365-0 ISSN (eBook) 2701-9411
Meh als 50.000 S udie ende, 25 Fo schungsein ich ungen und 500 Ve ö en lichungen
im Jah – dami zähl die FOM zu den g öß en und o schungss ä ks en Hochschulen Eu opas.
Ini iie du ch die gemeinnü zige S i ung BildungsCen um de Wi scha olg sie einem
kla en Bildungsau ag: Die FOM e möglich Be u s ä igen, Auszubildenden, Abi u ien en und
in e na ional S udie enden ein quali a i hochwe iges und i nanziell agba es Hochschuls u-
dium. Als gemeinnü zige Hochschule is die FOM nich gewinno ien ie , sonde n ein es ie
säm liche Gewinne – un e ande em in die Leh e und Fo schung.
Die FOM is s aa lich ane kann und bie e meh als 50 akk edi ie e Bachelo - und Mas e -
S udiengänge an – im Campus-S udium an 35 Hochschulzen en ode im einziga igen Digi alen
Li e-S udium gesende aus den High ech-S udios de FOM.
Leh ende und S udie ende o schen an de FOM in einem g oßen Fo schungsbe eich aus
hochschuleigenen Ins i u en und Kompe enzCen en. Do we den anwendungso ien ie e
Lösungen ü be iebliche und gesellscha liche P oblems ellungen gene ie . Ak uelle
Fo schungse gebnisse l ießen unmi elba in die Leh e ein und kommen so den Un e nehmen
und de Wi scha insgesam zugu e.
Zudem ö de die FOM g enzübe sch ei ende P ojek e und Pa ne scha en im eu opäischen
und in e na ionalen Fo schungs aum. Du ch Publika ionen, übe Fach agungen, wissen-
scha liche Kon e enzen und Vo agsak i i ä en wi d de T ans e de Fo schungs- und
En wicklungse gebnisse in Wissenscha und Wi scha siche ges ell .
Fo schungss a k und p axisnah
FOM.
Die Hochschule
besonde en
Fo ma s
Alle Ins i u e und Kompe enzCen en un e
om.de/ o schung
FOM Hochschulzen um
Düsseldo
Mi und 50.000 S udie enden is die FOM eine de
g öß en Hochschulen Eu opas und üh sei 1993
S udiengänge ü Be u s ä ige du ch, die einen s aa -
lich und in e na ional ane kann en Hochschulab-
schluss (Bachelo /Mas e ) e langen wollen.
Die FOM is de anwendungso ien ie en Fo schung
e p lich e und e olg das Ziel, adap ions ähige
Lösungen ü be iebliche bzw. wi scha snahe ode
gesellscha liche P oblems ellungen zu gene ie en.
Dabei spiel die Ve zahnung on Fo schung und
Leh e eine g oße Rolle: Kong uen zu den Mas e -
p og ammen sind Ins i u e und Kompe enzCen en
geg ünde wo den. Sie geben de Hochschule ein
achliches P o il und e ö nen sowohl Wissenscha -
le innen und Wissenscha le n als auch engagie en
S udie enden die Gelegenhei , sich ak i in den Fo -
schungsdisku s einzub ingen.
Wei e e In o ma ionen inden Sie un e om.de
Das KCV Kompe enzCen um ü angewand e
Volkswi scha sleh e deck ein b ei es Spek um
on de mik o- zu mak oökonomischen sowie
on de heo e ischen zu empi ischen Fo schung
ab. Besonde e inhal liche Schwe punk e liegen in
den g oßen He aus o de ungen unse e Zei : dem
demog a ischen Wandel, de Globalisie ung, dem
Klimawandel sowie de Digi alisie ung.
Neben diese inhal lichen Aus ich ung übe nimm
das KCV eine Que schni s unk ion im Be eich de
Me hodik und Da a Li e acy an de FOM Hoch-
schule, indem äche übe g ei end e ne z es,
analy isches Denken ge ö de und sowohl ü die
Fo schung als auch ü die Leh e nu zba gemach
wi d.
Wei e e In o ma ionen inden Sie un e om-kc .de
FOM Hochschule KCV
KCV Kompe enzCen um
ü angewand e Volkswi scha sleh e
de FOM Hochschule ü Oekonomie & Managemen
ISSN (P in ) 2701-9403
ISSN (eBook) 2701-9411ISBN (eBook) 978-3-89275-387-2
ISBN (P in ) 978-3-89275-386-5
De Wissenscha sblog
de FOM Hochschule
De Wissenscha sblog de FOM Hochschule bie e Einblicke in die iel äl igen
Themen, zu denen an de FOM ge o sch wi d: om-blog.de
