Lochne , Benjamin; Wol e , S e anie; Se h, S e an
A icle
AKM e ec s o Ge man labou ma ke da a om 1985 o
2021
Jou nal o Economics and S a is ics
P o ided in Coope a ion wi h:
De G uy e B ill
Sugges ed Ci a ion: Lochne , Benjamin; Wol e , S e anie; Se h, S e an (2024) : AKM e ec s o
Ge man labou ma ke da a om 1985 o 2021, Jou nal o Economics and S a is ics, ISSN
2366-049X, De G uy e Oldenbou g, Be lin, Vol. 244, Iss. 4, pp. 425-431,
h ps://doi.o g/10.1515/jbns -2023-0018
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/333290
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/
Da a Obse e
Benjamin Lochne , S e anie Wol e * and S e an Se h
AKM Effec s o Ge man Labou Ma ke Da a
om 1985 o 2021
h ps://doi.o g/10.1515/jbns -2023-0018
Recei ed Ma ch 27, 2023; accep ed Ma ch 28, 2023
Abs ac : This a icle desc ibes he p ocessing and accessibili y o he pe son and
es ablishmen fixed wage effec s in Ge man adminis a i e da a. These effec s ha e
been es ima ed ollowing he app oach o Abowd, J., K ama z, F., and Ma golis,
D. (1999. High wage wo ke s and high wage fi ms. Econome ica 67: 251–333) and
Ca d, D., Heining, J., and Kline, P. (2013. Wo kplace he e ogenei y and he ise o Wes
Ge man wage inequali y. Q. J. Econ. 128: 967–1015). They can be linked o mos o he
a ailable adminis a i e da ase s p o ided by he Resea ch Da a Cen e (FDZ) o he
Ge man Fede al Employmen Agency a he Ins i u e o Employmen Resea ch
(IAB). They a e a ailable o diffe en ime in e als om 1985 un il 2021. These
effec s ha e been used in nume ous a icles ha deal wi h he con ibu ions o
wo ke s and es ablishmen s o ea nings inequali y.
Keywo ds: wages, wo ke and fi m fixed effec s, AKM, Ge man adminis a i e
mic oda a, labou ma ke da a
JEL Classifica ion: J31, J62, C23
1 In oduc ion
In an influen ial a icle, Ca d e al. (2013) (hence o h CHK) s udy he ole o
es ablishmen -specific wage p emiums o wage inequali y in Ge many. Thei
app oach builds upon a amewo k ini ially p oposed by Abowd e al. (1999)
*Co esponding au ho : S e anie Wol e , Ins i u ü A bei sma k -und Be u s o schung, Nü nbe g,
Ge many, E-mail: [email p o ec ed]
Benjamin Lochne , Ins i u ü A bei sma k -und Be u s o schung, Nü nbe g, Ge many; F ied ich-
Alexande -Uni e si ä E langen-Nü nbe g, Nü nbe g, Ge many; Fo schungsins i u zu Zukun de A bei
GmbH (IZA), Bonn, Ge many; and The Labo and Socio-Economic Resea ch Cen e (LASER), Nu embe g,
Ge many
S e an Se h, Ins i u ü A bei sma k -und Be u s o schung, Nü nbe g, Ge many
Jou nal o Economics and S a is ics 2024; 244(4): 425–431
Open Access. © 2023 he au ho (s), published by De G uy e . This wo k is licensed unde he
C ea i e Commons A ibu ion 4.0 In e na ional License.
(hence o h AKM). I uses ma ched adminis a i e employe –employee da a o
es ima e a s a is ical model o wage de e mina ion, whe e wages a e exp essed as
he sum o wo ke effec s, es ablishmen effec s, ime- a ying co a ia es, and
idiosync a ic e o e ms. Since hen, nume ous a icles
1
ha e used his echnique
o answe many ela ed esea ch ques ions.
The Resea ch Da a Cen e (IAB-FDZ) o he Ge man Fede al Employmen Agency
a he Ins i u e o Employmen Resea ch (IAB) has applied he CHK/AKM app oach
o he uni e se o Ge man employmen da a and es ima ed he pe son and es ab-
lishmen fixed effec s (hence o h AKM effec s). The IAB-FDZ has offe ed hese effec s
o ex e nal da a use s since 2015 and has ecen ly upda ed he p ocedu e and he
co esponding da a files. They a e a ailable o he mos common adminis a i e
da a p oduc s, such as he Sample o In eg a ed Employmen Biog aphies (SIAB; see
F ode mann e al. 2021) o he Linked Employe –Employee Da a om he IAB (LIAB;
see Ru e al. 2021). Cu en ly, he AKM effec s a e a ailable o diffe en ime win-
dows om 1985 un il 2021. Use s wi h an unexpi ed da a usage ag eemen can access
he effec s in addi ion o hei p ima y da a p oduc .
This a icle desc ibes how hese effec s a e es ima ed and p ocessed, how hey
can be linked o he exis ing a ailable da a, and o which esea ch opics he da a
migh be ele an . I is o ganized as ollows. Sec ion 2 explains he da a sou ces.
Sec ion 3 explains he ini ial da a p epa a ion ha has been applied in o de o
es ima e he AKM effec s. Sec ion 4 explains he de ails o he es ima ion p ocedu e.
Sec ion 5 p o ides desc ip i e esul s abou he es ima ed effec s. Sec ion 6 explains
he da a s uc u e and how o access hem. Sec ion 7 summa izes and gi es an
o e iew o po en ial esea ch ques ions.
2 Da a Sou ce
The da a sou ce ha is used o es ima e he AKM effec s is he IAB Employmen
His o y File (BEH) which o igina es om he In eg a ed Employmen Biog aphies
(IEB) which a e p ocessed by he IAB.
The BeH co e s all employees who a e subjec o social secu i y a leas one day
in a gi en yea . Since 1999, he da a also co e employees in ma ginal pa - ime
employmen and unpaid amily wo ke s. Wo ke s’employmen spells a e sup-
plemen ed by es ablishmen in o ma ion om he Es ablishmen His o y Panel
(“Be iebs-His o ik-Panel”, see Ganze e al. 2022), which agg ega es es ablishmen -
le el a iables (wi h June 30 each yea as he eco d day). The IEB uni es da a om
diffe en da a sou ces, each o which may con ain in o ma ion om diffe en
1Up un il Ma ch 2023, acco ding o Google Schola Abowde al. (1999) has been ci ed o e 3600 imes.
426 B. Lochne e al.
adminis a i e p ocedu es. The main da a sou ce is he in eg a ed no ifica ion
p ocedu e o heal h, pension, and unemploymen insu ance.
The BeH co e s he majo i y o he Ge man wo k o ce, only excluding ci il
se an s and he sel -employed.
The main ing edien s ha he CHK/AKM app oach builds upon a e he BeH
in o ma ion on employmen and o al ea nings. Since he iden ifica ion o he effec s
elies on wo ke mobili y, being able o ack wo ke s is c ucial. The BeH offe s an
ideal labo a o y because each wo ke and es ablishmen has dis inc iden ifica ion
numbe s.
3 Da a P epa a ion
The da a p epa a ion mos ly ollows CHK, wi h a ew diffe ences, which we explain
in de ail below. F om he uni e se o BeH employmen spells, we es ic he sample
o ull- ime aged 20 o 60. Fo each o hese wo ke s, we iden i y he main job in a
gi en yea as he job wi h he highes o al ea nings (including any bonus paymen s).
In he BeH da a, ea nings a e igh -censo ed a he con ibu ion assessmen
ceiling.
(“Bei agsbemessungsg enze”).
We define a wage obse a ion as censo ed whene e he epo ed wage is
highe han 98 % o he censo ing h esholds. Following CHK, we fi a se ies o Tobi
eg essions o impu e he igh ail o he wage dis ibu ion. We es ima e he Tobi
eg essions by yea , sex, educa ion, and age g oup.
2
As opposed o CHK, we do no
es ic ou sample o Wes Ge many and males.
The es ima ion is pe o med on he la ges connec ed se o mo e s ha con ains
he g oup o es ablishmen s ha a e connec ed by wo ke mobili y o e each sample
in e al.
Figu e 1 illus a es he concep o he la ges connec ed se wi h ou connec ed
(A, B, C, D) and one unconnec ed es ablishmen (E). We es ima e he AKM model o
fi e non-o e lapping ime in e als: 1985–1992, 1993 –1999, 2000 –2006, 2007 –2013,
and 2014 –2021.
Table 1 shows he numbe o wo ke s and es ablishmen s ha a e in he la ges
connec ed se .
2We also accoun o age, he mean log wage o he wo ke in o he yea s, he ac ion o censo ed
wages in o he yea s, he numbe o ull- ime employees a he cu en es ablishmen and i s squa e,
a dummy o ele en o mo e employees, mean yea s o schooling and ac ion o uni e si y g adua es
a he cu en es ablishmen , mean log wage o co-wo ke s and ac ion o co-wo ke s wi h censo ed
wages, a dummy o indi iduals obse ed only one yea be ween 1985 and 2021, a dummy o
employees o one-wo ke es ablishmen s, and a dummy o Eas Ge many.
AKM Effec s o Ge man Labou Ma ke Da a 427
4 Es ima ion
Pe son and es ablishmen fixed effec s a e es ima ed using he ollowing eg ession
equa ion:
yi =αi+ψJ(i, )+x′
i β+ i .
We es ima e his equa ion sepa a ely o each o he a o emen ioned ime in e -
als. The da a comp ises N*pe son-yea obse a ions on Nwo ke s and J
Figu e 1: Illus a ion o he la ges connec ed se . A, B, C, and D illus a e diffe en es ablishmen s, he
symbols illus a e diffe en wo ke s. The a ows ep esen wo ke ansi ions be ween es ablishmen s.
Sou ce: Own ep esen a ion.
Table :Numbe o wo ke s and es ablishmen s in each ime in e al.
– – – – –
Numbe o wo ke s ,, ,, ,, ,, ,,
Numbe o es ablishmen s ,, ,, ,, ,, ,,
The sample consis s o ull- ime emale and male wo ke s in he connec ed se , aged –, in hei main job. Sou ce:
BeH.
428 B. Lochne e al.
es ablishmen s. J(i, )gi es he iden i y o he unique es ablishmen ha employs
wo ke iin yea .
As in CHK, he daily eal wage yi o indi idual iin yea is assumed o be he sum
o a wo ke componen αi, an es ablishmen componen ψJ(i, ), an index o ime-
a ying obse able cha ac e is ics x′
i β, and an e o componen i .x′
i βincludes an
un es ic ed se o yea dummies as well as quad a ic and cubic e ms in age ully
in e ac ed wi h educa ional a ainmen .
The li e a u e in e p e s he es ablishmen effec ψjas a wage p emium (o
discoun ) ha is paid by es ablishmen j o all o i s employees, while he pe son effec
αi ep esen s ime-in a ian ac o s ha a e a ibu ed o he wo ke .
5 AKM Effec s
Table 2 shows he mean, s anda d de ia ion, and he dis ibu ion o he es ima ed
AKM effec s. The esul s poin o he usual findings ha bo h he pe son effec s and
he es ablishmen effec s become mo e a iable o e ime (CHK). Howe e , in he
mos ecen yea s, his end has s opped (see Lochne e al. 2020).
Table :S a is ics on pe son and es ablishmen effec s o each in e al.
– – – – –
Mean pe son effec . . . . .
SD pe son effec . . . . .
ppe son effec . . . . .
P pe son effec . . . . .
P pe son effec . . . . .
P pe son effec . . . . .
P pe son effec . . . . .
P pe son effec . . . . .
P pe son effec . . . . .
Mean fi m effec −. −. −. −. −.
SD fi m effec . . . . .
pfi m effec −. −. −. −. −.
P fi m effec −. −. −. −. −.
P fi m effec −. −. −. −. −.
P fi m effec . . −. −. −.
P fi m effec . . . . .
P fi m effec . . . . .
P fi m effec . . . . .
The sample consis s o ull- ime emale and male wo ke s in he la ges connec ed se , aged –, in hei main job.
Sou ce: BeH.
AKM Effec s o Ge man Labou Ma ke Da a 429
6 Da a S uc u e and Access
6.1 Da a S uc u e
The AKM effec s a e es ima ed o fi e diffe en non-o e lapping ime in e als. Da a
use s can eadily link he effec s using he p o ided unique pe son and es ablish-
men iden ifie s. The pe son and es ablishmen effec s a e p o ided in sepa a e ime-
specific a iables, i.e. he da a is in wide o ma .
6.2 Da a Access
The es ima ed pe son and es ablishmen effec s a e offe ed by he Resea ch Da a
Cen e o he Fede al Employmen Agency a he Ins i u e o Employmen Resea ch
(IAB-RDC) in addi ion o he SIAB and LIAB da a. The effec s can be accessed du ing
gues isi s o one o he loca ions o he Resea ch Da a Cen e o he IAB-RDC.
Subsequen ly, emo e da a p ocessing is possible.
Ini ially, an applica ion o da a use mus be submi ed o he IAB-RDC. A e
app o al, a con ac needs o be signed be ween he ins i u ion o he esea che and
he IAB. An unexpi ed ag eemen is necessa y o using he es ima ed effec s. De ails
on how o apply o he da a se and he da a p ocessing op ions can be ound on he
websi e o he IAB-RDC (h ps:// dz.iab.de/en/da a-access/). Access o he da a can only
be g an ed once he applican has signed con ac s wi h bo h esea ch da a cen e s.
7 Summa y and Ou look
The AKM app oach allows esea che s o sepa a ely iden i y unobse ed, ime-
in a ian wo ke and es ablishmen he e ogenei y in wages. Using he uni e se o
Ge man employmen da a o ull- ime wo ke s aged 20–60, we ha e applied he AKM
wage model o es ima e pe son and es ablishmen fixed wage effec s. These effec s
ha e been es ima ed o diffe en ime in e als om 1985 o 2021 and can be linked
o he wo kho se da a se s p o ided by he IAB-FDZ.
So a , many a icles ha e used he AKM effec s since he IAB-FDZ s a ed o offe
hem in 2020. The ollowing pa ag aph summa izes a selec i e lis o hem (incom-
ple e and wi hou any specific o de ing):
Bu schek (2022) in es iga es whe he he minimum wage in oduc ion aised
hi ing s anda ds, using wo ke fixed effec s as a p oxy o wo ke p oduc i i y.
Zimme mann (2022) s udies he ela ionship be ween wo ke p oduc i i y and he
430 B. Lochne e al.
sha e o emale manage s. Mülle and Neuschaeffe (2021) s udy he ela ionship
be ween wo ke pa icipa ion, wo ke so ing, and fi m pe o mance and find ha
es ablishmen s wi h wo ks councils employ highe -quali y wo ke s. Niebuh and
Pe e s (2020) s udy wo k o ce composi ion and indi idual wages, using he es i-
ma ed wo ke and es ablishmen fixed effec s as a p oxy o unobse ed ac o s
ha migh influence he en y wages o new hi es. Illing e al. (2021) analyse he
gende wage gap a e job loss and show, among o he hings, ha women end o
wo k a es ablishmen s wi h lowe AKM es ablishmen effec s. Lochne and Me kl
(2022) explo e he ole o gende -specific applica ion beha io as well he selec ion
beha io o fi ms o he gende ea nings gap in Ge many. They find ha women
end o apply less o en a high-wage es ablishmen s, bu i hey do hey ha e abou
he same likelihood o being selec ed by hese fi ms.
As he a o emen ioned examples show, he p o ided AKM effec s offe an ideal
labo a o y o exploi some so o wo ke and es ablishmen he e ogenei y in he
con ex o many diffe en esea ch ques ions.
Re e ences
Abowd, J., K ama z, F., and Ma golis, D. (1999). High wage wo ke s and high wage fi ms. Econome ica
67: 251–333.
Bu schek, S. (2022). Raising he ba : minimum wages and employe s’hi ing s anda ds. Am. Econ. J. Econ.
Pol. 14: 91–124.
Ca d, D., Heining, J., and Kline, P. (2013). Wo kplace he e ogenei y and he ise o Wes Ge man wage
inequali y. Q. J. Econ. 128: 967–1015.
F ode mann, C., Schmucke , A., Se h, S., and Be ge, P.V. (2021). S ichp obe de In eg ie en
A bei sma k biog afien (SIAB) 1975-2019. FDZ-Da en epo 01 (de), Nü nbe g.
Ganze , A., Schmucke , A., S egmaie , J., and S übe , H. (2022). Be iebs-his o ik-panel 1975–2020.
FDZ-Da en epo 03 (de), Nü nbe g.
Illing, H., Schmiede , J.F., and T enkle, S. (2021). The gende gap in ea nings losses a e job displacemen .
Na ional bu eau o economic esea ch wo king pape 29251.
Lochne , B. and Me kl, C. (2022). Gende -specific applica ion beha io , ma ching, and he esidual gende
ea nings gap. IAB Discussion Pape 22, Nü nbe g.
Lochne , B., Se h, S., and Wol e , S. (2020). Decomposing he la ge fi m wage p emium in Ge many. Econ.
Le . 194, h ps://doi.o g/10.1016/j.econle .2020.109368.
Muelle , S. and Neuschaeffe , G. (2021). Wo ke pa icipa ion in decision-making, wo ke so ing, and fi m
pe o mance. Ind. Rela . 60: 436–478.
Niebuh , A. and Pe e s, J.C. (2020). Wo k o ce composi ion and indi idual wages —an employe –
employee da a analysis. B . J. Ind. Rela . 58: 719–742.
Ru , K., Schmid lein, L., Se h, S., S übe , H., and Umkeh e , M. (2021). Linked-employe -employee-da en des
IAB: LIAB Längsschni modell (LIAB LM) 1975 –2019. FDZ-Da en epo 06 (de), Nü nbe g.
Zimme mann, F. (2022). Managing he gende wage gap—how emale manage s influence he gende
wage gap among wo ke s. Eu . Socio Re . 38: 355–370.
AKM Effec s o Ge man Labou Ma ke Da a 431
