D eusch, Dennis; Reichling, Pe e
A icle
The pe o mance o me ging coope a i e banks in
Ge many
Ge man Economic Re iew (GER)
P o ided in Coope a ion wi h:
Ve ein ü Socialpoli ik / Ge man Economic Associa ion
Sugges ed Ci a ion: D eusch, Dennis; Reichling, Pe e (2025) : The pe o mance o me ging
coope a i e banks in Ge many, Ge man Economic Re iew (GER), ISSN 1468-0475, De G uy e ,
Be lin, Vol. 26, Iss. 3, pp. 193-227,
h ps://doi.o g/10.1515/ge -2024-0087
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/331954
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/
ge 2025; 26(3): 193–227
Dennis D eusch* and Pe e Reichling
The Pe o mance o Me ging Coope a i e
Banks in Ge many
h ps://doi.o g/10.1515/ge -2024-0087
Recei ed Augus 6, 2024; accep ed Feb ua y 3, 2025; published online Feb ua y 17, 2025
Abs ac :Mo i a ed by he ecen inc ease in bank me ge s, his pape examines
he pe o mance o Ge man coope a i e banks ha me ged be ween 2014 and 2019.
We a e pa icula ly in e es ed in whe he ele a ed me ge a es a e due o bank
ine iciencies o o challenging policy measu es such as low- o -long in e es a es.
The esul s indica e ha banks ha pe o m ela i ely wo se be o e and du ing
he low in e es en i onmen exhibi a g ea e p obabili y o becoming a a ge du -
ing his pe iod. Consolida ion gene ally occu s among low pe o ming banks whe e
la ge and well-capi alized banks me ge wi h hei small and ine icien pee s. Ul i-
ma ely, ou esul s a ibu e he inc eased numbe o me ge s o ine iciencies in
he banking indus y, as banks ha exi ed he ma ke we e ine icien p io o he
ad e se low in e es a e en i onmen .
Keywo ds: banks; me ge s; egula ion; low in e es en i onmen ; e iciency
JEL Classi ica ion: G21; G34
1 In oduc ion
O e he pas decade, a significan numbe o Ge man banks ha e exi ed he ma -
ke h ough me ge s, which many banks see as he esul o excessi e egula ion
and low- o -long in e es a es. The sh inking banking landscape hus aises po en-
ial conce ns ha di icul policy measu es ha e d i en ex an e good banks ou o
*Co esponding au ho : Dennis D eusch, O o- on-Gue icke Uni e si y Magdebu g,
Uni e si ä spla z 2, 39106 Magdebu g, Ge many, E-mail: [email p o ec ed].
h ps://o cid.o g/0009-0000-7295-6982
Pe e Reichling, O o- on-Gue icke Uni e si y Magdebu g, Uni e si ä spla z 2, 39106 Magdebu g,
Ge many, E-mail: [email p o ec ed]
Open Access. ©2025 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.
194 —D. D eusch and P. Reichling
he ma ke . This pape examines his issue by es ima ing he de e minan s o bank
me ge s be o e and du ing he low in e es a e en i onmen , he eby iden i ying
he pe o mance cha ac e is ics o me ging banks in wo ime pe iods. This p o-
idesin o ma ionno onlyonwhe he me gingbankswe egoodo badpe o me s
ex an e, bu also on he p ima y cause o he me ge wa e: Me ge s a e likely o
be d i en by bank ine iciency i a ge banks pe o m ela i ely poo ly be o e
and du ing he egula o y and in e es a e changes and e en ually exi he ma -
ke du ing his ad e se pe iod. Con e sely, inc eased me ge ac i i y may eflec
challenging policies i a ge banks ini ially pe o m well o simila ly o hei non-
me gingpee s,bu unde pe o min he cou se o he low in e es a een i onmen
and e en ually lea e he ma ke .
Gi en he ecen inc ease in he numbe o me ge s, i seems in e es ing o ask
how ad e se ope a ing condi ions a e shaping and impac ing he banking indus-
y. The pas decade p o ided o a leas wo c ucial de elopmen s ha , possibly
each indi idually, bu pa icula ly when iewed oge he , esul ed in o a di icul
su ounding. Fi s and o emos , he low- o -long in e es a e policy in he a e -
ma h o he 2007–2009 financial c isis g adually educed banks’ p ofi abili y and
ne in e es ma gin and e ec i ely de e io a ed hei financial leeway (e.g. Bo io,
Gambaco a, and Ho mann 2017;Busch e al. 2022;Claessens, Coleman, and Don-
nelly 2018;Genay and Podjasek 2014). Exis ing wo k sugges s ha ex ended pe iods
o low o nega i e policy a es induce banks o adap . Banks inc ease hei isk-
aking by lowe ing hei loan s anda ds (Maddaloni and Peyd ó 2011) and aising
he po ion o isk asse s (Delis and Kou e as 2011;Heide , Saidi, and Schepens 2019).
Fu he mo e, banks cu hei lending ac i i y (Heide , Saidi, and Schepens 2019),
adjus hei unding s uc u e, and exchange in e es -gene a ing engagemen s wi h
ee- ela ed and ading ac i i ies (B ei, Bo io, and Gambaco a 2020). Impo an ly,
he e ec s a e mo e p onounced o small banks wi h highe deposi sha es (e.g.
Claessens, Coleman, and Donnelly 2018;Genay and Podjasek 2014;Heide , Saidi,
and Schepens 2019;Ke bl and Sigmund 2016;Meye 2018). Besides igge ing low
policy a es, he financial c isis also spu ed au ho i ies o ho oughly igh en and
expand exis ing banking egula ions. Conside ing he shee complexi y and impac
on he indus y, he Basel III e o m he eby ha dly compa es o p io ame-
wo ks. In ac , e o ms en ail subs an ial cos s o banks o es ablish pe missible
isk-managemen sys ems and comply wi h no el egula o y s anda ds and me -
ics (e.g. Bonne and Eij inge 2015;Die ich, Hess, and Wanzen ied 2014;Hando
2014;King 2013). Since egula o y bu den can mani es in a ious aspec s, such as
inc eased spending on adminis a ion and s a aining, howe e , no eliable quan-
ifica ion exis s wi h espec o banks’ incu ed cos o egula ion (Coch ane 2014;
Hoskins and Labon e 2015). The li e a u e shows ha banks anspose and espond
o egula o y claims o highe capi al and liquidi y equi emen s by cu ing hei
The Pe o mance o Ge man Coope a i e Banks —195
lending (De Nicolò, Gamba, and Lucche a 2014;Mésonnie and Monks 2015)and
isk-weigh ed asse s (G opp e al. 2019) and by employing di e en s a egies based
on hei p ofi abili y (And le e al. 2017;Cohen and Sca igna 2016).
Taken oge he , banks ha e aced a di icul su ounding o mo e han a
decade,e okingchanges in asse composi ion and undings uc u e.Gi en he ela-
i ely sha p decline in he numbe o independen ly ope a ing EU banks, howe e ,
ad e se condi ions no only seem o ha e spa ked no able adap a ions in finan-
cial p ofiles. They likely also con ibu ed o an accele a ed consolida ion p ocess in
which a subs an ial numbe o banks le he ma ke h ough me ge s.1The unde -
lying easoning is in ui i e. S enuous pe iods equi e an e ec i e alloca ion o
esou ces o banks o emain compe i i e. Banks ailing o cope wi h de imen-
al ci cums ances a e hen ei he o ced o wind up o me ge wi h o he banks.
Wha emains unclea , howe e , is whe he ad e se condi ions induced ex an e
ine icien banks o exi he ma ke in he spi i o Schumpe e (1939) (he ea e
he ‘e iciency- iew’), o whe he hey encou aged e en ini ially well-pe o ming
banks o lea e he ma ke by nega i ely a ec ing hei lending and deposi busi-
ness. The uncha ed issue o ele a ed bank exi s s eps in o he line o ques ions
ela ed o he e ec s o ul a-loose mone a y policies and s ingen egula ions,
and is o pa amoun impo ance o comp ehending de elopmen s in he banking
landscape. I ecen policy measu es p ima ily u n small, albei ini ially well pe -
o ming, banks in o me ge a ge s, hen his should be o in e es o au ho i ies.
Likewise, i is impo an o know i consolida ion occu s among low pe o ming
banks, which could help educe o e -capaci ies, enhance p ofi abili y, and imp o e
esilience agains shocks (ECB 2019). Gi en hese un esol ed ques ions su ounding
he ecen me ge ac i i y, his pape examines he ela i e pe o mance and o he
key cha ac e is ics o banks ha exi ed he ma ke h ough me ge s in ecen yea s.
We do so by conside ing he me ge dynamics in he Ge man coope a i e bank-
ing indus y be ween 2014 and 2019. We selec he Ge man banking ma ke because
i is one o he la ges in he EU, while he coope a i e sec o o e s an abundan
numbe o independen ly bu simila ly ope a ing banks. This na u ally mi iga es
conce ns abou po en ial di e ences in unobse ed ime-in a ian bank cha ac e -
is ics (Mood 2010), and simul aneously sa isfies c ucial homogenei y assump ions
o well-es ablished e iciency measu emen echniques (Dysone al. 2001). The cho-
sen ime span eflec s ou in en ion o in es iga e c i ical pe iods. On he one hand,
1Acco ding o he Eu opean Banking Fede a ion, he numbe o EU banks be ween 2013 and
2019 declined by 23 %, which is oughly wice he decline om 2007 o 2013 (10 %). Simila
findings a e ob ained o he numbe o MFIs in he Eu o a ea: h ps://www.ecb.eu opa.eu/
s a s/ecb_s a is ics/escb/h ml/ able.en.h ml?id=JDF_MFI_MFI_LIST.
196 —D. D eusch and P. Reichling
he Capi al Requi emen s Regula ion en e s legal o ce in 2014, g adually impos-
ing s ic e s anda ds on banks’ capi al and liquidi y managemen . On he o he
hand, s ipula ed efinancing a es and s anding acili ies con inue hei declining
pa h, u he educing banks’ in e es expenses bu also hei o e all p ofi abili y.
Fo ou empi ical analyses, we d aw on he ich li e a u e on bank me ge de e mi-
nan s. As such, we fi s es ima e a se ies o s anda d mul inomial logis ic models o
he pe iod 2014 o 2019, linking a ious pe o mance measu es o he p obabili y o
becoming a a ge o an acqui e . This allows impo an insigh s in o consolida ing
banks’ pe o mance cha ac e is ics a ha junc u e. Subsequen ly, we conside an
ea lie ime pe iod om 2010 o 2012, when in e es a es p e ail a common le -
els and no el egula ions s ill emain o be pou ed in o binding law. We ei e a e
ou analyses in simila ashion bu his ime ela ing banks’ a e age pe o mance
o e he pe iod 2010 o 2012 o he p obabili y o becoming a u u e a ge a some
poin be ween 2014 and 2019. The idea is o es whe he pe o mance, as measu ed
be o e he low in e es en i onmen , significan ly de e mines he p obabili y o
becoming a u u e a ge . I so, we can ule ou he no ion ha he a ge s we e
ini ially well-pe o ming banks ha became unde pe o ming due o he ad e se
policy ac ions and e en ually disappea ed om he ma ke as a esul . We p o ide
wo main findings. Fi s , banks ha pe o m ela i ely wo se be o e and be ween
2014 and 2019 a e mo e likely o exi he ma ke du ing his pe iod. Con olling o
bank size and ela ed me ge de e minan s, he p obabili y o becoming a a ge
inc eases wi h a highe cos -income a io, lowe g ow h a es, and poo e asse
managemen . Second, depending on he pe o mance measu e o choice, acqui e s
pe o m ei he simila o wo se, bu in no case be e , han he e e ence g oup wi h
no me ge occu ence. Consolida ion hus ends o occu among low pe o ming
banks, whe e la ge and well-capi alized banks ake o e hei small and ine icien
pee s. Based on he findings in his pape , we conclude ha he ecen me ge wa e
p ima ily a ises om bank ine iciencies: Banks ha pe o m ela i ely wo se o e
he 2010–2012 pe iod a e mo e likely o become a ge s in he low in e es a e
en i onmen .
In addi ion o s udies analyzing he e ec s o banking egula ions and low-
o -long in e es a e policies, ou pape ela es o se e al o he s eams o li e -
a u e, including esea ch on bank me ge mo i es and de e minan s (e.g. Beccalli
and F an z 2013;Foca elli, Pane a, and Salleo 2002;Godda d, McKillop, and Wil-
son 2009;Hadlock, Hous on, and Ryngae 1999;Hannan and Pillo 2009;Hannan
and Rhoades 1987;He nando, Nie o, and Wall 2009;Huh ilainen, Saas amoinen, and
Suhonen 2022;Koe e e al. 2007;Lanine and Vande Venne 2007;Moo e 1996,1997;
Pasiou as, Tanna, and Gaganis 2011;Wheelock and Wilson 2000;Wo hing on 2004),
bank dis ess and ailu e (e.g. Be ge and Bouwman 2013;Be ge , Imbie owicz, and
Rauch 2016;Cole and Gun he 1995;Cole and Whi e 2012;DeYoung and To na 2013;
The Pe o mance o Ge man Coope a i e Banks —197
Es ella, Pa k, and Pe is iani 2000;Wheelock and Wilson 2000), and he cleansing
e ec o c ises (Spoke iciu e, Keasey, and Vallascas 2019). Ou pape closely con-
nec s o Spoke iciu e, Keasey, and Vallascas (2019), which seems o be he only
pe cei able wo k in es iga ing he cleansing e ec s o financial c ises wi hin he
banking indus y. The wo k p edic s he p obabili y o bank ailu e o acquisi ion
by in e ac ing c ises wi h banks’ cos e iciency. They find ha he sa ings and loan
c isis in he mid 1980s and ea ly 1990s inc eases he exi p obabili y o less e i-
cien US comme cial banks mo e han o he g oup o e icien banks. Howe e ,
he 2007–2009 financial c isis escala es he exi p obabili y ega dless o banks’ cos
e iciency. The wo k hus p o ides mixed e idence on he e iciency- iew ha c ises
encou age less e icien banks o d op ou o he ma ke . Ou pape di e s om he
exis ing li e a u e in ha we examine he link be ween ecen me ge ac i i y and
(mone a y) policy measu es. By showing ha consolida ion occu s among low pe -
o ming banks, we p o ide an in e es ing a enue o esea ch analyzing he belie
ha “in sys ems wi h many weak-pe o ming small banks, consolida ion wi hin
hei domes ic sys em could imp o e pe o mance” ECB (2019), p. 107. Rela ed, we
also ex en he li e a u e on bank me ge p edic abili y.
2 Me ge s Made in Ge many
The Ge man coope a i e banking indus y has expe ienced an ex ao dina y
inc ease o me ge s o e he pas yea s. This is illus a ed in Figu e 1(a),which
depic s he annual numbe o banks exi ing he ma ke along wi h he ma ginal
lending acili y as a measu e o he ECB’s in e es a e policy. Impo an ly, me ge s
accele a e a e 2011, and peak in 2017 when 57 a ge s a e aken o e by 40 acqui -
ing banks. These numbe s can be pu in o pe spec i e by compa ing he ecen
me ge ac i i y wi h he si ua ion a ound he 2007–2009 financial c isis. Al hough
he c isis is known o i s subs an ial impac on he banking indus y, ela i ely ew
banks engaged in o me ge s du ing ha pe iod. Conside ing he en i e ime span
o ou sample om 2013 o 2020, me ge dynamics induce an o e all decline in he
quan i y o independen coope a i e banks by nea ly 25 %.2
The d i ing mo i e behind hese ele a ed me ge a es could a guably be
linked o he low in e es en i onmen , which exposes banks o his o ically low
ea nings. Such linkage is suppo ed by he simple ac ha me ge s occu in a
ime when in e es a es concu en ly exhibi a declining pa h. Lowe ma ke a es,
2Bundesbank da a sugges s 1,065 banks by he end o 2013, and 804 banks by he end o 2020. In
his espec , no e ha supe iso y agencies posses no di ec manda e o en o ce me ge s. Gi en
numbe s hence comp ise me ge s made on a olun a y basis.
198 —D. D eusch and P. Reichling
Figu e 1: The e olu ion o coope a i e bank me ge s in Ge many. Figu e (a) shows ecen
de elopmen s in Ge man coope a i e bank me ge ac i i y along wi h he lending acili y.
Figu e (b) depic s he a e age ne in e es ma gin and capi aliza ion o his indus y. Shaded a eas
espec i ely indica e he 2007–2009 inancial c isis and he 2014–2019 low-in e es en i onmen as
conside ed in ou analysis. The da a a e aken om he Bundesbank: “Banks ellens a is ik” and
“Zei eihen-Da enbanken”.
in u n, come along wi h a educ ion o banks’ ne in e es ma gin, as shown in
Figu e 1(b) by he solid line. In his ega d, banks migh me ge o encoun e hei
dec easing ma gins by ealizing economies o scale o scope (Amel e al. 2004). How-
e e , low in e es a es may no be he only mo i e. The ecen me ge wa e also
alls in a pe iod in which au ho i ies depa owa ds a s ic e banking egula ion.
Ex ensi e egula ion in e ms o he Basel III amewo k could hen impede g ow h
and bu den banks wi h cos s (e.g. Die ich, Hess, and Wanzen ied 2014;Mésonnie
and Monks 2015), al hough hese cos s may be abso bed, a leas o some ex en , by
he compound s uc u es o he Ge man coope a i e sys em. While he quan ifica-
ion o such egula o y cos s is di icul , he impac o g ea e (capi al) equi emen s
can be shown clea ly by an upwa d sloping capi al a io, as depic ed in Figu e 1(b)
by he dashed line. Thus, banks migh also me ge o comply wi h egula ions a
ewe cos s, o ins ance by p ofi ing om a di e sifica ion o hei loan po o-
lio (Amel e al. 2004). Apa om egula o y mo i es, me ge s can also be based
on managemen es uc u ing goals and he es ablishmen o a new co po a e cul-
u e (Gindele e al. 2019). In hei ecen s udy, Gindele e al. (2019) explici ly s a e
ha me ge s among coope a i e banks a e p ima ily d i en by s uc u al changes
in managemen , a he han a emp s o imp o e p ofi abili y. Al hough he anal-
ysis is limi ed o banks in he Ge man s a e o Baden-Wü embe g be ween 2009
and 2016, hese managemen mo i es make i gene ally unclea whe he he a ge
banks a e unde pe o ming o dis essed.
The Pe o mance o Ge man Coope a i e Banks —199
As o he cu en wa e o me ge s, low in e es a es seem o be he domi-
nan mo i e. D awing on banks’ websi es, p ess eleases, local newspape a icles
ha comp ise in e iews wi h bank execu i es, and ela ed sou ces, banks mainly
a ibu e hei decision o me ge o he low in e es en i onmen and an ex ensi e
se o egula ions, bu also o a cos ly layou o a digi al in as uc u e. Mo e specifi-
cally, among he 228 a ge s subjec o ou empi ical analysis, 211 banks jus i y hei
me ge by e e ing o a combina ion o a leas wo ou o hese h ee elemen s. Fo
he emaining 17 banks, we find no in o ma ion. An exempla y s a emen eflec ing
key me ge mo i es is aken om a p ess elease by he Volksbank Un e e Saa eG
and he Ve einig e Volksbank eG Saa louis – Losheim am See – Sulzbach/Saa :
As o he main easons o he me ge , Soes e names he exube an egula o y equi emen s,
he al e ed cus ome beha io due o an inc easing digi aliza ion, and he impac o he
ongoing low in e es en i onmen , which causes declines in ea nings yea a e yea . (P ess
elease, accessed on he 11.01.2023 a 11:30)
Ano he example conce ns he me ge ac i i y be ween he VR-Bank Huns ück-
Mosel eG and he Ve einig e Volksbank Rai eisenbank eG:
The me ge is mainly aimed a mee ing he challenges o he low-in e es policy and he
inc easing supe iso y egula ion. (Websi e, accessed on he 11.01.2023 a 11:30)
Since mos s a emen s esemble each o he o a g ea ex en , bo h examples can be
aken as ep esen a i e o he gene al easoning in he popula ion. While i could
be ue ha some banks mask o he decisi e me ge in en ions behind s a emen s
such as he wo p esen ed, he disad an ageous na u e o he low in e es en i on-
men is e iden . E en i he e a e cases whe e unobse ed ac o s ul ima ely seal
he me ge , such as excessi e losses leading o dis essed me ge s, i seems likely
ha such ac o s a e again closely ela ed o he s a e o p ofi abili y, o o he cos s
o egula ion and digi aliza ion, al hough he la e may again be abso bed by he
p e ailing ne wo k s uc u es in he coope a i e sec o .
Gi en he high numbe o me ge s, pai ed wi h he ac ha mos banks
a ibu e hei me ge decision o he fie ce ope a ing si ua ion, i appea s na u-
al o ask i such en i onmen encou ages a he low pe o ming banks o exi
he ma ke . To ob ain a fi s , mo i a ing glimpse on he di e ences be ween a -
ge s and non-me ging banks, Table 1 con as s bo h g oups wi h ega d o c ucial
bank (pe o mance) cha ac e is ics. Beginning wi h Panel A, which conside s all
bank-yea obse a ions om ou p ima y da a as discussed in he subsequen
sec ion, we obse e a significan di e ence in o al asse s. Ta ge banks a e on
a e age only hal he size o non-me ging banks. Mo eo e , non-me ging banks
exhibi g ea e g ow h a es, a e seemingly be e capi alized, p ofi om signifi-
can ly lowe NPL- a ios, and achie e mo e a o able cos - o-income balances. We
200 —D. D eusch and P. Reichling
Table 1: Pe o mance di e ences. This able examines pe o mance di e ences be ween a ge
banks and non-me ging banks. Panel A conside s he ull sample om 2014 o 2019, o which we
obse e 599 non-me ging banks o e 6 yea s yielding 3,594 bank-yea obse a ions. Panel B
addi ionally conside s he 2014–2017 sub-sample comp ising 599 non-me ging banks. All a iables
a e epo ed in pe cen , excep asse s and GRP pe capi a, which a e epo ed in millions and
housands o Eu os, espec i ely. S.E. e e s o he s anda d e o o he mean. The i s p- alue e e s
o a wo-sample - es , which es s he H0 ha he di e ence in means equals ze o. The second
p- alue e e s o a Wilcoxon ank sum es and is epo ed o add ess conce ns abou da a no mali y.
Panel A: Full sample
Ta ge Non-me ging
NMean S.E. NMean S.E. Di e ence p- alue Wp- alue
Asse s 837 389 12.91 3,594 785 33.70 396 0.00 0.00
Asse g ow h 609 3.73 0.144 2,995 4.69 0.076 0.96 0.00 0.00
Loan g ow h 609 4.20 0.211 2,995 5.67 0.099 1.47 0.00 0.00
Equi y sha e 837 9.14 0.069 3,594 9.46 0.036 0.32 0.00 0.00
Tie 1 a io 812 14.6 0.141 3,557 15.1 0.067 0.50 0.00 0.00
NPL a io 758 1.79 0.061 3,403 1.55 0.025 −0.24 0.00 0.00
Re u n on asse s 837 0.27 0.006 3,594 0.28 0.003 0.01 0.11 0.72
Re u n on equi y 837 2.94 0.056 3,594 2.94 0.029 0.00 0.96 0.26
Cos -income a io 837 70.0 0.360 3,594 67.0 0.332 −3.00 0.00 0.00
GRP pe capi a 833 34.7 0.439 3,582 35.4 0.220 0.70 0.16 0.14
Panel B: Sub-sample –
Ta ge Non-me ging
NMean S.E. NMean S.E. Di e ence p- alue Wp- alue
Asse s 228 402 21.58 2,396 749 38.97 347 0.01 0.05
Asse g ow h 171 3.56 0.271 1,797 4.43 0.101 0.87 0.01 0.05
Loan g ow h 171 4.11 0.370 1,797 5.46 0.131 1.35 0.00 0.01
Equi y sha e 228 9.08 0.134 2,396 9.25 0.045 0.17 0.25 0.16
Tie 1 a io 228 14.3 0.246 2,361 14.6 0.083 0.30 0.23 0.24
NPL a io 213 1.92 0.142 2,259 1.73 0.032 −0.19 0.09 0.09
Re u n on asse s 228 0.25 0.009 2,396 0.29 0.004 0.04 0.00 0.00
Re u n on equi y 228 2.82 0.091 2,396 3.20 0.036 0.38 0.00 0.01
Cos -income a io 228 71.0 0.710 2,396 67.0 0.441 −4.00 0.01 0.00
GRP pe capi a 224 35.2 0.897 2,388 34.4 0.260 −0.80 0.36 0.23
nei he obse e di e ences in he p ofi abili y, no in he egional economic ou pu
as p oxied by he G oss Regional P oduc pe capi a. Since Figu e 1(b) e eals ends
in banks’ capi aliza ion and ne in e es ma gin, howe e , i is likely ha compa -
isons among ela ed a iables a e flawed. This is because non-me ging banks, by
defini ion, emain h oughou he sample pe iod and a e hus mo e exposed o
The Pe o mance o Ge man Coope a i e Banks —207
The esul ing e iciency alues – o b e i y, adjus ed alues – a e hence adjus ed
o ele an en i onmen al ac o s. Since an inc easing 𝜃indica es dec easing bank
ine iciency, we expec a nega i e ela ionship be ween bo h e iciency measu es
and he p obabili y o becoming a a ge .8
Apa om DEA, we also analyze six con en ional accoun ing-based a ios. As
o he fi s fi e measu es, we conside he cos - o-income a io, he non-in e es
expenses o asse a io, he e u n on equi y and asse s, and loan g ow h. These
me ics a e selec ed due o hei equen use in ela ed wo k (e.g. Lanine and Van-
de Venne 2007), and ecogni ion as pe o mance measu es by supe iso s (e.g.
ECB 2010). The six h measu e is a liquidi y indica o , defined as he sum o cash and
cen al bank holdings as a sha e o o al asse s. Al hough p opo ionally la ge s akes
o hese liquid asse s seem o appea desi able a fi s sigh , hey migh indica e idle
esou ces wi h poo e u ns (Koe e e al. 2007). This could be pa icula ly ue
when conside ing he low in e es sphe e wi h ze o o nega i e ma ginal deposi
acili ies. We hus iew he liquidi y sha e as an indica ion o how banks pe o m
ega ding hei asse managemen . In his espec , we expec he liquidi y sha e o
be posi i ely associa ed wi h he p obabili y o becoming a a ge bank.
4.2 Con ol Va iables
Conside ing he findings o Koe e e al. (2007),He nando, Nie o, and Wall (2009),
Pasiou as, Tanna, and Gaganis (2011) and Beccalli and F an z (2013), amongs o h-
e s, he banking li e a u e has con e ged o a se o bank-specific ac o s ha a e
eliably associa ed wi h bank me ge s. To make su e ha hese common de e -
minan s o bank me ge s a e no d i ing ou pe o mance es ima es, we include
co esponding ac o s as con ols in he eg essions. Mo e p ecisely, we ollow He -
nando, Nie o, and Wall (2009) and Pasiou as, Tanna, and Gaganis (2011) and include
he na u al log o o al asse s as a measu e o bank size. Fu he mo e, we include
he pe cen age change in o al asse s as a p oxy o g ow h p ospec s as well as he
a io o equi y o o al asse s as a p oxy o capi aliza ion. Mo eo e , we conside
cus ome loans as a sha e o o al asse s o con ol o di e ences in specializa ion
and asse di e sifica ion (e.g. Beccalli and F an z 2013;Huh ilainen, Saas amoinen,
and Suhonen 2022). To ensu e obus ness, we u he ake in o accoun al e na i e
wi h la ge sha es o elde people a e a ec ed de imen ally, which could be due o educed loan
demand and g ea e cos s o pe sonal ad iso y (Con ad, Neube ge , and T igo 2009).
8Rela ed wo k o en conside s p ofi and cos e iciency, which, howe e , equi e da a on uni
p ices and cos s. He e, we ins ead ocus on he echnical-physical aspec s o in e media y e i-
ciency as we lack eliable in o ma ion on espec i e a iables such as dep ecia ion and numbe
o employees (Coope , Sei o d, and Tone 2007).
208 —D. D eusch and P. Reichling
me ics such as he Tie 1 Capi al a io (ins ead o he equi y sha e) and he sha e
o secu i ies in o al asse s (ins ead o he loan sha e) as in Koe e e al. (2007).
Because bank cha ac e is ics may depend on ex e nal condi ions, we also
include economic and demog aphic in o ma ion o he coun y in which each bank
ope a es. We accoun o he possibili y ha banks in economically mo e a o -
able en i onmen s could p ofi in e ms o asse g ow h and loan quali y. In his
espec ,weaddi ionallyinclude heunemploymen a e op oxy egionaleconomic
s eng h. We also suspec ha u ban banks migh be la ge and could benefi om a
p oduc i i y p emium is-à- is banks in u al a eas (Ande sson, Bu gess, and Lane
2007). The e o e, we conside he na u al log o he popula ion densi y as a measu e
o he u baniza ion ex en . Fo obus ness, we check i a coun y’s coope a i e bank
densi y and ma ke concen a ion influence ou esul s. As such, we accoun o
he a iables bank densi y, p oxied by he numbe o banks ela i e o he coun y’s
popula ion size, and HHI, which is he no malized He findahl–Hi schman Index
measu ing cus ome loan ma ke concen a ion wi hin each coun y.
5 Resul s
This sec ion add esses ou empi ical esul s. We begin by es ima ing a ia ions o
equa ion (1) o he pe iod 2014 o 2019 o unde s and i bank pe o mance sig-
nifican ly associa es wi h he p obabili y o u ning in o a a ge o an acqui e .
Subsequen ly, ou ocus shi s o an ea lie poin in ime, when in e es a es p e-
ail a somewha common le els, and new egula ions jus begin o unpack hei
influence on he indus y. The idea is o analyze how a ge banks pe o m p io o
low ma ke a es and fie ce egula ions. This could show ha ad e se condi ions
a e no he p ima y cause o a ge banks’ low pe o mance. Ins ead, ini ially low
pe o ming banks may ha e a g ea e p obabili y o exi he ma ke la e on.
5.1 Resul s o he Mul inomial Model
To examine how pe o mance ela es o he likelihood o a bank being a a ge o
an acqui e , we conduc a se ies o mul inomial logis ic eg essions. Es ima ions
a e based on a sample o 201 a ge s, 169 acqui e s, and 599 non-me ging banks.
No e ha he es ima ions equi e independence among all bank-yea obse a ions.
Due o he panel s uc u e, howe e , his is unlikely o be he case, since banks
ha con inue o exis in some yea canno ha e become a ge s in −1, and ice
e sa (Shumway 2001). We ackle his issue by employing s anda d e o s ha a e
clus e ed a he bank le el. This lea es es ima ed coe icien s unchanged bu o e s
obus in e ence by allowing dependence wi hin clus e s. We epo ou esul s in
The Pe o mance o Ge man Coope a i e Banks —209
Table 2. Recall ha we discuss he coe icien s in e ms o RRR, which cap u e he
change in he ela i e p obabili y gi en a uni change in he p edic o a iable.
Fo example, he coe icien on he liquidi y sha e o a ound 0.092 indica es ha
a one pe cen age poin inc ease in his a iable inc eases he p obabili y o being
in he a ge g oup ela i e o he p obabili y o being in he e e ence g oup by
[(e0.092 −1) ×100 %]≈10 %.
Two main findings a e appa en . Fi s , passing h ough a ge g oup esul s
ac oss columns (T), we no ice ha a ge s pe o m wo se han he e e ence g oup
on nea ly all pe o mance measu es. S a ing wi h column (1), which cap u es
banks’ unadjus ed e iciency as in e media ies, we obse e a significan and neg-
a i e coe icien . Consis en wi h ou expec a ion, inc easing bank e iciency is
associa ed wi h a dec easing p obabili y o becoming a a ge . The esul pa icu-
la ly implies ha a ge banks a e cha ac e ized by ela i ely low e iciency alues,
indica ing hei endency o ely on g ea e inpu olumes compa ed o hei non-
me ging pee s. This finding emains in column (2) when we epea he es ima ion
bu use adjus ed alues which accoun o di e ences in ope a ing en i onmen s.
Simply pu , a ge banks end o d aw on ela i ely g ea e inpu quan i ies e en
when conside ing hei ope a ion in po en ially disad an ageous loca ions.
Tu ning o ou se o accoun ing-based pe o mance measu es, we documen
a significan , albei economically small, e ec o he cos -income a io in column
(3). The p obabili y o becoming a a ge inc eases as banks bea la ge cos s pe
income uni . This again shows ha he a ge s unde conside a ion a e no pa -
icula ly well pe o ming banks bu ins ead end o exhibi ela i ely g ea e cos -
income a ios. Simila ly, column (4) implies ha a ge banks a he su e om
a g ea e sha e o non-in e es expenses. This indica es ela i ely la ge spending
on s a and adminis a ion o manage one asse uni . Con inuing wi h column (5),
we obse e a posi i e ela ionship be ween banks’ liquidi y sha e and he likeli-
hood o being a a ge , which echoes he findings o Koe e e al. (2007).Recall
ha we expec his ou come since excessi e cash posi ions and cen al bank hold-
ings o e poo e u ns, especially du ing pe iods o low ma ke in e es a es. The
p ofi abili y measu e in column (6) u ns ou insignifican a e con olling o com-
mon de e minan s o bank me ge s. The ou come does no change i we ei e a e
he exe cise bu use he e u n on asse s ins ead (un epo ed). Finally, column (7)
indica es a nega i e ela ionship be ween loan g ow h and he p obabili y o being
a a ge such ha banks wi h g ea e g ow h a es, any hing else equal, a e less
likely o engage in o me ge s. No e ha we a e awa e o po en ial mul icollinea i y
issues be ween loan and asse g ow h. We accep his possibili y o main ain model
consis ency.
Second, we find mixed e idence o he acqui e g oup (A). On he one hand,
columns(1),(2), (4), and (7) imply ha acqui e s ope a eine icien ly ega ding hei
210 —D. D eusch and P. Reichling
Table 2: Mul inomial logi esul s. This able explo es how a ious pe o mance measu es ela e o he p obabili y o a bank becoming a a ge o an acqui e .
Es ima ions a e based on a sample o 201 a ge s, 169 acqui e s, and 599 non-me ging banks o e he pe iod 2015 o 2019. Me ge ac i i ies and bank-yea
obse a ions in 2014 a e omi ed due o he lagged na u e o asse g ow h. All es ima ions conside yea e ec s as shown a he bo om o he able. S anda d
e o s a e clus e ed a he bank le el. P- alues a e epo ed in pa en heses.
(1) (2) (3) (4) (5) (6) (7)
TATATATATATATA
E iciency −.∗∗ −.∗∗∗
(.) (.)
E iciency adjus ed −.∗∗∗ −.∗∗∗
(.) (.)
Cos -income a io .∗∗ .
(.) (.)
Expense sha e .∗∗∗ .∗∗
(.) (.)
Liquidi y sha e .∗∗ −.
(.) (.)
Re u n on equi y −. −.
(.) (.)
Loan g ow h −.∗∗∗ −.∗∗∗
(.) (.)
ln(asse s) −.∗∗∗ .∗∗∗ −.∗∗∗ .∗∗∗ −.∗∗∗ .∗∗∗ −.∗∗∗ .∗∗∗ −.∗∗∗ .∗∗∗ −.∗∗∗ .∗∗∗ −.∗∗∗ .∗∗∗
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.)
Asse g ow h −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗∗ −.∗∗ .
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.)
Equi y sha e −.∗.∗∗ −.∗.∗∗ −.∗∗ .∗∗ −.∗∗ .∗−.∗∗ .∗∗ −.∗∗ .∗∗ −.∗∗ .∗∗
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.)
Loan sha e −.∗. −.∗∗ . −.∗∗ . −.∗∗ . −.∗∗ . −.∗. −. .
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.)
Unemploymen a e −. −. −. −. −. −. −.∗−.∗−. −. −. −. −. −.
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.)
ln(popula ion densi y) .∗−.∗∗ . −.∗∗∗ . −.∗∗ .∗∗ −.∗∗ . −.∗∗∗ . −.∗∗∗ . −.∗∗∗
(.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.) (.)
Yea e ec s Yes Yes Yes Yes Yes Yes Yes
N, , , , , , ,
Pseudo R2. . . . . . .
Wald 𝜒2. . . . . . .
Pseudo loglikelihood −,. −,. −,. −,. −,. −,. −,.
∗∗∗,∗∗ and ∗Respec i ely deno e s a is ical signi icance a he 1 %, he 5 % and he 10 % le el.
The Pe o mance o Ge man Coope a i e Banks —211
inpu -ou pu alloca ion, expense sha e, and loan g ow h, espec i ely. On he o he
hand, columns (3), (5), and (6) sugges ha acqui e s’ cos -income a io, liquidi y
sha e, and p ofi abili y do no significan ly di e om hose o he e e ence g oup.
Thus, depending on he indica o o choice, acqui e s ei he pe o m wo se o he
same, bu in no case be e han he e e ence g oup.
The emaining se o con ol a iables is no o key in e es o ou pu poses.
Ne e heless, i may be eassu ing o no e ha he obse ed e ec s on all bank-
specific ac o s a e gene ally consis en wi h hose ound in ela ed wo k. In pa -
icula , he esul s uni o mly indica e no able di e ences in bank size and cap-
i aliza ion. Acqui e s ( a ge s) a e ela i ely la ge (smalle ) and be e (wo se)
capi alized, which aligns wi h he findings o Koe e e al. (2007) and Beh and
Heid (2011). Las ly, we s ess he obus ness o hese esul s ac oss a ious model
specifica ions. In un epo ed es ima ions, he esul s no only hold when di e en
co a ia es a e used, such as he Tie 1 Capi al a io o he secu i y sha e. They also
emainla gelyunchangedwhenes ima ionsinclude mo e han one designa ed pe -
o mance measu e a a ime. Fo example, he cos -income e ec s do no change
significan ly when he liquidi y sha e is added o he es ima ions. Fu he mo e, he
esul s emain when we addi ionally include he me ge e en s in 2014, which a e
omi ed he e due o he lagged na u e o asse g ow h. This is shown in column
(2) o Table A3, which achie es an ex ended sample by neglec ing he asse g ow h
a iable.9Ano he conce n ha o en goes unno iced in allied wo k is he po en-
ial in e dependence be ween me ge s and egional ma ke concen a ion. In his
ein, we suspec ha me ge s could mo e likely occu in coun ies wi h a g ea e
coope a i e bank densi y and a mo e dispe sed ma ke s uc u e. To con ol o
his possibili y, columns (3) and (4) o Table A3 espec i ely include he a iable
bank densi y, which is he numbe o banks in a coun y ela i e o he coun y’s
popula ion, and he HHI, which is he no malized He findahl–Hi schman index
ha measu es he concen a ion o he ma ke o consume loans. Ou findings
hold up agains bo h specifica ions. Consis en wi h ou expec a ions, highe bank
densi y and dispe sed local ma ke powe seem o encou age me ge s, bu he pe -
o mance aspec p e ails. Finally, we acknowledge ha Tobi eg essions may yield
biased second-s age e iciency measu es (Sima and Wilson 2007). The e o e, col-
umn (5) uses bias-co ec ed e iciency as he dependen a iable, which is based
on he p ocedu e p oposed by Sima and Wilson (2007). The p e iously obse ed
esul s emain.
Wha we ake away om hese eg essions is ha consolida ing banks a he
pe o m wo se han hei non-me ging pee s, bu nuances exis among a ge s and
9We ocus on adjus ed e iciency o b e i y.
212 —D. D eusch and P. Reichling
acqui e s. While a ge s clea ly exhibi financially weak p ofiles, pa icula ly om
a cos pe spec i e, acqui e s do benefi om g ea e size, s onge capi aliza ion,
and some pe o mance ou comes ha a e indis inguishable om he pee g oup.
The esul s so a p o ide impo an insigh s in o he cha ac e is ics o consolida -
ing banks, bu hey a e no su icien o eliably e i y he e iciency- iew in he
Ge man banking ma ke . This is because we ha e o conside he possibili y ha
he a ge banks pe o med well p io o he low in e es a e e a, bu de e io a ed
mainlydue o head e seen i onmen .Suchapossibili yissuppo edby heobse -
a ion ha he a ge s a e o en small, egional banks. These banks a e ypically
mo e elian on deposi and lending ac i i ies and may be mo e exposed o changes
in he in e es a e en i onmen (e.g. Claessens, Coleman, and Donnelly 2018;Genay
and Podjasek 2014). I so, hen he sole eliance on p e ious analyses would likely
yield allacious conclusions, and impugn he e iciency- iew. The e o e, he nex
sec ion complemen s he p e ious analyses by examining bank pe o mance p io
o he low-in e es pe iod and be o e he egula o y in e en ions ook legal e ec
in 2014.
5.2 Resul s o he Logi Model
In o de o unde s and i a ge s pe o m ela i ely wo se be o e ou ini ial in es i-
ga ion pe iod, we d aw on ou seconda y da a se . Fo each o he 218 u u e a ge s
and 883 o he coope a i e banks in his sample, we conside a e aged alues o e
hepe iod2010 o2012.
We spli he analysis in o wo pa s. Fi s , we examine whe he a ge s a e
unde pe o ming, as in ou p ima y analysis. In his ein, we fi s es ima e
equa ion (2), ela ing he p obabili y o a bank being acqui ed wi hin he pe iod
2014 o 2019 o he a e age bank pe o mance o e he yea s 2010–2012. The goal
o his p ocedu e is o show ha he low in e es sphe e does no u n ex an e well
pe o ming banks in o a ge s. Ins ead, banks ha ope a e compa a i ely wo se
be ween 2010 and 2012 a e mo e likely o become a ge s la e on. Second, we assign
banks o di e en g oups based on hei size and pe o mance, and epea p e i-
ous es ima ions using g oup indica o s. The aim is o unde s and, in pa icula , how
pe o mance ela es o he su i al chances o small banks. F om an e iciency pe -
spec i e, we expec small and well pe o ming banks o benefi in e ms o lowe
exi p obabili ies compa ed o hei unde pe o ming pee s. Simila ly, he su i al
p obabili ies o la ge and well-pe o ming banks should be significan ly highe
han o hei coho s o simila size.
Be o e p oceeding wi h he analysis, finally no e ha we es ic he anal-
ysis in his pa o ou se o accoun ing-based pe o mance indica o s. This is
no by choice bu a he due o he ac ha uni measu emen s di e o some
The Pe o mance o Ge man Coope a i e Banks —213
obse a ions, which se e ely dis o s all DEA e iciency alues. We also do no co e
he non-in e es expense sha e he e as he da a-se a hand lacks in o ma ion on
adminis a i e expenses.
5.2.1 Ta ge Bank Pe o mance Be o e he Low-Fo -Long In e es Ra e E a
We begin by es ima ing equa ion (2) o see i ex an e pe o mance significan ly
de e mines u u e a ge banks. Mo e p ecisely, we conside a bina y logis ic model
which links he p obabili y o a bank becoming a a ge a some poin be ween 2014
and 2019 o he a e age pe o mance o e he pe iod 2010 o 2012. In case ha es i-
ma ed coe icien s esemble he alues om ea lie eg essions, he low in e es
en i onmen and ela ed ac o s a e unlikely o be he cause o a ge banks’ low
pe o mance in he mo e ecen yea s. In addi ion o he logis ic eg essions, we
also apply a linea p obabili y model (LPM). Al hough LPMs su e om commonly
known, undesi able p ope ies, we find ha hey p o ide a con enien and a o d-
able way o u he ein o ce ou esul s. In his ein, we eg ess he bina y a iable
Fu u eTa ge on he same ec o x, again comp ising one pe o mance measu e a
a ime.
We epo ou esul s in Table 3. The fi s fi e columns, (1) o (5), show he
ou come o he logis ic model. Recall ha he coe icien s o hese eg essions mea-
su e he change in log-odds o a uni -change in he p edic o a iable. Applying
he ans o ma ion yields [(e0.060 −1) ×100 %]≈6.2 %, such ha a one pe cen age
poin inc ease in he cos -income a io inc eases he odds o becoming a a ge by
abou 6.2 %. Pu di e en ly, banks wi h a one pe cen age poin highe cos -income
a io a e 1.062 imes mo e likely o become a a ge han banks wi hou his addi-
ional inc ease. Beginning wi h column (1), we obse e a significan and posi i e
e ec o he cos -income a io. This implies ha a ge banks exhibi ela i ely
g ea e cos -income balances be o e he eme gence o he ze o in e es en i on-
men . The finding aligns wi h he ea lie obse a ion om Table 2 and pa icula ly
sugges s ha he ze o in e es sphe e does no u n well ope a ing banks in o a -
ge s. Ins ead, banks wi h ex an e g ea e cos -income a ios mo e likely become
a ge s la e on du ing ope a ionally disad an ageous pe iods. Con inuing wi h col-
umn (2), we find a posi i e and significan associa ion be ween banks’ liquidi y
sha e and hei p obabili y o u ning in o a u u e a ge . This again ma ches ou
p e ious obse a ions such ha banks wi h p opo ionally la ge holdings o idle
asse s end o be mo e likely u u e a ge s. In con as o ou p ima y analysis,
columns (3) and (4) documen a nega i e and significan e ec o bo h p ofi abili y
measu es. Ini ially mo e p ofi able banks a e less likely o be a a ge in he u u e.
Tu ning o column (5), we find a nega i e and s a is ically significan e ec o loan
g ow h. G ea e loan g ow h educes he p obabili y o a u u e ma ke exi . This
214 —D. D eusch and P. Reichling
Table 3: Bina y logi esul s. This able analyzes how pe o mance o e he pe iod 2010 o 2012 in luences he p obabili y o becoming a u u e a ge be ween
2014 and 2019. The i s i e columns, (1) o (5), show he ou come o he bina y logis ic model. Columns (6) o (10) show he esul s o he linea p obabili y
model. Es ima ions a e based on a sample o 218 u u e a ge s and 838 o he banks assumed o ha e no engaged in o me ge s du ing he 2010–2012 pe iod.
The numbe o obse a ion a ies depending on da a a ailabili y. We do no equi e in o ma ion on all a iables o p e en a po en ial sample selec ion bias.
Robus Hube -Whi e s anda d e o s a e employed. P- alues a e epo ed in pa en heses.
Logi LPM
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
FT FT FT FT FT FT FT FT FT FT
Cos -income a io 0.060∗∗∗ 0.00874∗∗∗
(0.000) (0.000)
Liquidi y sha e 0.023∗∗ 0.00419∗∗
(0.024) (0.021)
Re u n on equi y −0.086∗∗ −0.0113∗∗
(0.039) (0.027)
Re u n on asse s −1.221∗∗ −0.158∗∗
(0.041) (0.025)
Loan g ow h −0.074∗∗∗ −0.0114∗∗∗
(0.008) (0.007)
ln(asse s) −0.078∗−0.105∗∗ −0.098∗∗ −0.100∗∗ −0.097∗∗ −0.0118∗−0.0158∗∗ −0.0146∗∗ −0.0148∗∗ −0.0145∗∗
(0.079) (0.029) (0.041) (0.038) (0.043) (0.066) (0.015) (0.024) (0.022) (0.026)
Asse g ow h −0.081∗∗ −0.123∗∗∗ −0.111∗∗∗ −0.111∗∗∗ −0.076∗−0.0124∗∗∗ −0.0181∗∗∗ −0.0163∗∗∗ −0.0165∗∗∗ −0.0101∗
(0.017) (0.000) (0.001) (0.001) (0.051) (0.009) (0.000) (0.001) (0.001) (0.073)
Equi y sha e 0.080∗0.029 0.029 0.081∗0.022 0.0119∗0.00533 0.00512 0.0124∗0.00416
(0.068) (0.502) (0.494) (0.086) (0.602) (0.084) (0.441) (0.458) (0.099) (0.542)
The Pe o mance o Ge man Coope a i e Banks —215
Table 3: (con inued)
Logi LPM
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
FT FT FT FT FT FT FT FT FT FT
Loan sha e 0.028∗∗ 0.015 0.022∗0.022∗0.020 0.00347∗∗ 0.00206 0.00291∗0.00280∗0.00258
(0.031) (0.194) (0.070) (0.075) (0.105) (0.031) (0.200) (0.068) (0.078) (0.104)
Unemploymen a e −0.078∗∗ −0.060∗−0.070∗∗ −0.070∗∗ −0.044 −0.0112∗∗ −0.00927∗−0.00981∗−0.00959∗−0.00657
(0.025) (0.074) (0.047) (0.046) (0.180) (0.021) (0.062) (0.054) (0.056) (0.176)
ln(popula ion densi y) 0.031 −0.006 −0.002 −0.001 −0.014 0.00776 −0.000424 −0.000425 −0.000939 −0.00166
(0.737) (0.944) (0.978) (0.987) (0.870) (0.562) (0.975) (0.975) (0.944) (0.900)
N960 961 960 960 961 960 961 960 960 961
R20.063 0.037 0.038 0.038 0.040 0.060 0.036 0.036 0.036 0.038
Wald 𝜒252.01 31.84 27.04 27.10 31.74
Pseudo loglikelihood −466.61 −479.61 −478.93 −477.75 −478.05
∗∗∗,∗∗ and ∗Respec i ely deno e s a is ical signi icance a he 1 %, he 5 % and he 10 % le el.
216 —D. D eusch and P. Reichling
esul again implies ha a ge banks exhibi ela i ely poo g ow h a es be o e
he eme gence o de imen al condi ions.
Apa om hese pe o mance measu es, we make an in e es ing disco e y
conce ning he equi y and loan sha e. In con as o p e ious es ima ions, bo h
me ics now indica e a s a is ically ma ginal bu uni o mly posi i e e ec . Be e
capi alized banks and banks holding p opo ionally mo e loans in hei asse s bea
a g ea e p obabili y o becoming a u u e a ge . Since small, adi ional coope -
a i e banks ma ch his desc ip ion ema kably well, we see he ou come s ongly
in line wi h he li e a u e finding mo e p onounced in e es a e policy e ec s o
smallbanks.In pa icula ,Heide ,Saidi, and Schepens(2019),p.3741,find ha “High-
deposi banks a e also smalle , ha e highe equi y a ios (6.2 % s. 5.0 %), highe
loans- o-asse s a ios” and ha he in oduc ion o nega i e policy a es leads “ o
mo e isk- aking and less lending by eu o-a ea banks wi h a g ea e eliance on
deposi unding”.
Rela ed, Claessens, Coleman, and Donnelly (2018), p. 8, find ha “small banks
ha e g ea e di icul y main aining hei NIMs in a low in e es a e en i onmen ”.
Ou es ima ion esul s complemen hese findings by showing ha banks, ini ially
ulfilling he adi ional ole o well-capi alized and loan-o ien ed in e media ies,
also en ail an inc eased p obabili y o exi ing he ma ke in he cou se o a las -
ing low-in e es en i onmen . In his espec , he low- o -long in e es a e policy
could ha e influenced small banks mo e d as ically. Impo an ly, howe e , his does
no change he no ion ha a ge s al eady pe o m ela i ely wo se be o e such
ad e se en i onmen . The subsequen sec ion u he disen angles pe o mance
om bank size and analyzes di e ences in exi p obabili ies be ween small and
la ge banks.
P oceeding wi h columns (6) o (10), which show he es ima ion esul s o he
LPM, we e i y all p e ious findings. Fo ins ance, he coe icien on he cos -income
a io indica es ha an inc ease o he a e age cos -income a io by one pe cen age
poin is associa ed wi h an inc ease in he p obabili y o becoming a u u e a ge by
a ound one pe cen age poin , holding o he ac o s fixed. Banks wi h highe cos -
income a ios a e he e o e mo e likely o become a ge s la e on.
Concluding, we find ha a ge s pe o m wo se be o e he low- o -long in e -
es a e en i onmen and he Basel III implemen a ion in 2014. This is pa icu-
la ly e iden om Table 3, which assigns compa a i ely low pe o ming banks
a g ea e p obabili y o becoming a u u e a ge . Ou analysis hus a o s he
e iciency- iew, bu so a has allen sho o compa ing he su i al chances ac oss
banks wi h di e en sizes and pe o mance le els. We add ess his issue in he
ollowing.
The Pe o mance o Ge man Coope a i e Banks —223
Table A3: (con inued)
(1) (2) (3) (4) (5)
TA TA TA TA TA
ln(popula ion densi y) 0.0885 −0.307∗∗∗ 0.134 −0.295∗∗∗ 0.0828 −0.204∗∗ 0.0942 −0.315∗∗∗ 0.136 −0.219∗∗
(0.343) (0.001) (0.118) (0.001) (0.399) (0.041) (0.312) (0.001) (0.129) (0.012)
Yea e ec s Yes Yes Yes Yes Yes
N4,449 5,445 4,449 4,449 4,449
Pseudo R20.060 0.052 0.064 0.062 0.056
Wald 𝜒2194.77 206.18 209.71 193.35 190.10
Pseudo loglikelihood −1,385.69 −1,605.17 −1,379.44 −1,383.46 −1,391.75
∗∗∗,∗∗ and ∗Respec i ely deno e s a is ical signi icance a he 1 %, he 5 % and he 10 % le el.
224 —D. D eusch and P. Reichling
Re e ences
Amel, D., C. Ba nes, F. Pane a, and C. Salleo. 2004. “Consolida ion and E iciency in he Financial
Sec o : A Re iew o he In e na ional E idence.” Jou nal o Banking & Finance 28 (10): 2493−519..
Ande sson, F., S. Bu gess, and J. Lane. 2007. “Ci ies, Ma ching and he P oduc i i y Gains o
Agglome a ion.” Jou nal o U ban Economics 61 (1): 112−28..
And le, M., J. B ada, V. Tomšík, and J. Vlček. 2017. Banks’ Adjus men o Basel Iii Re o m: A Bank-Le el
Pe spec i e o Eme ging Eu ope. Wo king Pape 17/24. Washing on: In e na ional Mone a y Fund.
Banke , R., and R. Mo ey. 1986. “E iciency Analysis o Exogenously Fixed Inpu s and Ou pu s.”
Ope a ions Resea ch 34 (4): 513−21..
Banke , R., A. Cha nes, and W. Coope . 1984. “Some Models o Es ima ing Technical and Scale
Ine iciencies in Da a En elopmen Analysis.” Managemen Science 30 (9): 1078−92..
Ba h, J., C. Lin, Y. Ma, J. Seade, and F. Song. 2013. “Do Bank Regula ion, Supe ision and Moni o ing
Enhance o Impede Bank E iciency?” Jou nal o Banking & Finance 37 (8): 2879−92..
Beccalli, E., and P. F an z. 2013. “The De e minan s o Me ge s and Acquisi ions in Banking.” Jou nal o
Financial Se ices Resea ch 43: 265−91..
Beh , A., and F. Heid. 2011. “The Success o Bank Me ge s Re isi ed. An Assessmen Based on a
Ma ching S a egy.” Jou nal o Empi ical Finance 18 (1): 117−35..
Be ge , A., and C. Bouwman. 2013. “How Does Capi al A ec Bank Pe o mance du ing Financial
C ises?” Jou nal o Financial Economics 109 (1): 146−76..
Be ge , A., and D. Humph ey. 1997. “E iciency o Financial Ins i u ions: In e na ional Su ey and
Di ec ions o Fu u e Resea ch.” Eu opean Jou nal o Ope a ional Resea ch 98 (2): 175−212..
Be ge , A., B. Imbie owicz, and C. Rauch. 2016. “The Roles o Co po a e Go e nance in Bank Failu es
du ing he Recen Financial C isis.” Jou nal o Money, C edi , and Banking 48 (4): 729−70..
Bonne , C., and S. Eij inge . 2015. “The Impac o Liquidi y Regula ion on Bank In e media ion.”
Re iew o Finance 20 (5): 1945−79..
Bo io, C., L. Gambaco a, and B. Ho mann. 2017. “The In luence o Mone a y Policy on Bank
P o i abili y.” In e na ional Finance 20 (1): 48−63..
B ei, M., C. Bo io, and L. Gambaco a. 2020. “Bank In e media ion Ac i i y in a Low-In e es -Ra e
En i onmen .” Economic No es 49 (2): e12164..
Busch, R., H. Li ke, C. Memmel, and S. Niede aue . 2022. “Ge man Banks’ Beha io in he Low
In e es Ra e En i onmen .” Financial Ma ke s and Po olio Managemen 36 (3): 267−96..
Cha nes, A., W. Coope , and E. Rhodes. 1978. “Measu ing he E iciency o Decision Making Uni s.”
Eu opean Jou nal o Ope a ional Resea ch 2(6):429−44..
Cho a eas, G., C. Gi a done, and A. Ven ou i. 2012. “Bank Supe ision, Regula ion, and E iciency:
E idence om he Eu opean Union.” Jou nal o Financial S abili y 8(4):292
−302..
Claessens, S., N. Coleman, and M. Donnelly. 2018. ““Low-Fo -Long” In e es Ra es and Banks’ In e es
Ma gins and P o i abili y: C oss-Coun y E idence.” Jou nal o Financial In e media ion 35: 1−16..
Coch ane, J. H. 2014. “Challenges o Cos -Bene i Analysis o Financial Regula ion.” The Jou nal o
Legal S udies 43 (S2): S63−S105..
Cohen, B., and M. Sca igna. 2016. “Banks and Capi al Requi emen s: Channels o Adjus men .” Jou nal
o Banking & Finance 69: S56−S69..
Cole, R., and J. W. Gun he . 1995. “Sepa a ing he Likelihood and Timing o Bank Failu e.” Jou nal o
Banking & Finance 19 (6): 1073−89..
Cole, R., and L. Whi e. 2012. “Déjà Vu All o e Again: The Causes o US Comme cial Bank Failu es This
Time A ound.” Jou nal o Financial Se ices Resea ch 42: 5−29..
The Pe o mance o Ge man Coope a i e Banks —225
Con ad, A., D. Neube ge , and G., L. T igo. 2009. The Impac o Regional and Demog aphic Fac o s on he
E iciency o Ge man Sa ings Banks. Wo king Pape 111. Ros ock: Uni e si y o Ros ock, Ins i u e o
Economics.
Coope , W., L. Sei o d, and K. Tone. 2007. Da a En elopmen Analysis: A Comp ehensi e Tex wi h Models,
Applica ions, Re e ences and DEA-Sol e So wa e, 2nd ed. New Yo k: Sp inge .
De Nicolò, G., A. Gamba, and M. Lucche a. 2014. “Mic op uden ial Regula ion in a Dynamic Model o
Banking.” Re iew o Financial S udies 27 (7): 2097−138..
Delis, M., and G. Kou e as. 2011. “In e es Ra es and Bank Risk-Taking.” Jou nal o Banking & Finance 35
(4): 840−55..
DeYoung, R., and G. To na. 2013. “Non adi ional Banking Ac i i ies and Bank Failu es du ing he
Financial C isis.” Jou nal o Financial In e media ion 22 (3): 397−421..
Die ich, A., K. Hess, and G. Wanzen ied. 2014. “The Good and Bad News abou he New Liquidi y
Rules o Basel III in Wes e n Eu opean Coun ies.” Jou nal o Banking & Finance 44: 13−25..
Domb e , A. 2016. A e The e Too Many Banks? The Banking Sec o a e he Financial C isis. Vienna: Speech
gi en by And eas Domb e a he Gene al Mee ing o he Aus ian Socie y o Bank Resea ch.
D ake, L., M. J. Hall, and R. Simpe . 2009. “Bank Modelling Me hodologies: A Compa a i e
Non-pa ame ic Analysis o E iciency in he Japanese Banking Sec o .” Jou nal o In e na ional
Financial Ma ke s, Ins i u ions and Money 19 (1): 1−15..
Dyson, R., R. Allen, A. Camanho, V. Podino ski, C. Sa ico, and E. Shale. 2001. “Pi alls and P o ocols in
DEA.” Eu opean Jou nal o Ope a ional Resea ch 132 (2): 245−59..
ECB. 2010. Beyond ROE −How o Measu e Bank Pe o mance. Appendix o he Repo on eu Banking
S uc u es. F ank u am Main: Eu opean Cen al Bank.
ECB. 2019. Financial S abili y Re iew, No embe 2019. Financial S abili y Re iew. F ank u am Main:
Eu opean Cen al Bank.
Es ella, A., S. Pa k, and S. Pe is iani. 2000. “Capi al Ra ios as P edic o s o Bank Failu e.” Fede al
Rese e Bank o New Yo k Economic Policy Re iew 6(2):33
−52.
Fa ell, M. J. 1957. “The Measu emen o P oduc i e E iciency.” Jou nal o he Royal S a is ical Socie y:
Se ies A 120 (3): 253−81..
Fio en ino, E., A. Ka mann, and M. Koe e . 2006. The Cos E iciency o Ge man Banks: A Compa ison o
SFA and DEA. Discussion Pape 10/2006. F ank u am Main: Deu sche Bundesbank.
Foca elli, D., F. Pane a, and C. Salleo. 2002. “Why Do Banks Me ge?” Jou nal o Money, C edi , and
Banking 23 (4): 1047−66..
F ied, H., S. Schmid , and S. Yaisawa ng. 1999. “Inco po a ing he Ope a ing En i onmen in o a
Nonpa ame ic Measu e o Technical E iciency.” Jou nal o P oduc i i y Analysis 12: 249−67. .
Genay, H., and R. Podjasek. 2014. Wha is he Impac o a Low In e es Ra e En i onmen on Bank
p o i abili y? Chicago Fed Le e 324. Chicago: Fede al Rese e Bank o Chicago.
Gindele, N., R. Doluschi z, L. Sau e , and P. Mülle . 2019. “Fusionen on Volks-und Rai eisenbanken:
E gebnisse eine empi ischen Analyse 2017 in Baden-Wü embe g.” Zei sch i ü das Gesam e
Genossenscha swesen 69 (3): 162−78..
Godda d, J., D. McKillop, and J. O. Wilson. 2009. “Which C edi Unions A e Acqui ed?” Jou nal o
Financial Se ices Resea ch 36: 231−52..
G opp, R., T. Mosk, S. Ongena, and C. Wix. 2019. “Banks Response o Highe Capi al Requi emen s:
E idence om a Quasi-Na u al Expe imen .” Re iew o Financial S udies 32 (1): 266−99..
Hadlock, C., J. Hous on, and M. Ryngae . 1999. “The Role o Manage ial Incen i es in Bank
Acquisi ions.” Jou nal o Banking & Finance 23 (2−4): 221−49..
Hando , W. 2014. “The Cos o Bank Liquidi y.” Jou nal o Banking Regula ion 15: 1−13..
226 —D. D eusch and P. Reichling
Hannan, T. H., and S. J. Pillo . 2009. “Acquisi ion Ta ge s and Mo i es in he Banking Indus y.” Jou nal
o Money, C edi , and Banking 41 (6): 1167−87..
Hannan, T. H., and S. A. Rhoades. 1987. “Acquisi ion Ta ge s and Mo i es: The Case o he Banking
Indus y.” The Re iew o Economics and S a is ics 69 (1): 67−74..
Heide , F., F. Saidi, and G. Schepens. 2019. “Li e below Ze o: Bank Lending unde Nega i e Policy
Ra es.” Re iew o Financial S udies 32 (10): 3728−61..
He nando, I., M. Nie o, and L. Wall. 2009. “De e minan s o Domes ic and C oss-Bo de Bank
Acquisi ions in he Eu opean Union.” Jou nal o Banking & Finance 33 (6): 1022−32..
Hoskins, S., and M. Labon e. 2015. An Analysis o he Regula o y Bu den on Small Banks. Repo P epa ed
o he Membe s and Commi ees o Cong ess. Washing on: Cong essional Resea ch Se ice.
Hosme , D. W., and S. Lemeshow. 2000. Applied Logis ic Reg ession. New Yo k: Wiley.
Huh ilainen, M., J. Saas amoinen, and N. Suhonen. 2022. “De e minan s o Me ge s and Acquisi ions
Among Finnish Coope a i e and Sa ings Banks.” Jou nal o Banking Regula ion 23 (3): 339−49..
Jensen, M., and R. Ruback. 1983. “The Ma ke o Co po a e Con ol: The Scien i ic E idence.” Jou nal
o Financial Economics 11 (1−4): 5−50..
Ke bl, S., and M. Sigmund. 2016. F om Low o Nega i e Ra es: An Asymme ic Dilemma. Financial S abili y
Repo 32. Vienna: Aus ian Cen al Bank.
King, M. R. 2013. “The Basel III Ne S able Funding Ra io and Bank Ne In e es Ma gins.” Jou nal o
Banking & Finance 37 (11): 4144−56..
Koe e , M., T. Nes mann, S. S olz, and M. Wedow. 2006. “S ill O e banked and Unp o i able? Two
Decades o Ge man Banking.” C edi and Capi al Ma ke s −K edi und Kapi al 39 (4): 497−511..
Koe e , M., J. Bos, F. Heid, J. Kola i, C. Kool, and D. Po a h. 2007. “Accoun ing o Dis ess in Bank
Me ge s.” Jou nal o Banking & Finance 31 (10): 3200−17..
Lang, G., and P. Welzel. 1996. “E iciency and Technical P og ess in Banking Empi ical Resul s o a
Panel o Ge man Coope a i e Banks.” Jou nal o Banking & Finance 20 (6): 1003−23..
Lanine, G., and R. Vande Venne . 2007. “Mic oeconomic De e minan s o Acquisi ions o Eas e n
Eu opean Banks by Wes e n Eu opean Banks.” The Economics o T ansi ion 15 (2): 285−308..
Maddaloni, A., and J.-L. Peyd ó. 2011. “Bank Risk-Taking, Secu i iza ion, Supe ision, and Low In e es
Ra es: E idence om he Eu o-A ea and he U.S. Lending S anda ds.” Re iew o Financial S udies
24 (6): 2121−65..
Manne, H. 1965. “Me ge s and he Ma ke o Co po a e Con ol.” Jou nal o Poli ical Economy 73 (2):
110−20..
Mésonnie , J.-S., and A. Monks. 2015. “Did he EBA Capi al Exe cise Cause a C edi C unch in he Eu o
A ea?” In e na ional Jou nal o Cen al Banking 11 (3): 75−117.
Meye , E. 2018. “The New Economic Scena io and i s Impac on he Coope a i e Banking Business
Model.” In New Coope a i e Banking in Eu ope: S a egies o Adap ing he Business Model Pos
C isis, edi ed by M. Miglio elli, 29−45. Cham: Sp inge In e na ional Publishing.
Mood, C. 2010. “Logis ic Reg ession: Why We Canno Do wha We Think We Can Do, and wha We Can
Do abou i .” Eu opean Sociological Re iew 26 (1): 67−82..
Moo e, R. R. 1996. “Banking’s Me ge Fe o : Su i al o he Fi es .” In Fede al Rese e Bank o Dallas
Financial Indus y S udies,9
−15.
Moo e, R. R. 1997. Bank Acquisi ion De e minan s: Implica ions o Small Business C edi . Wo king Pape .
Dallas: Fede al Rese e Bank o Dallas.
Palepu, K. G. 1986. “P edic ing Takeo e Ta ge s: A Me hodological and Empi ical Analysis.” Jou nal o
Accoun ing and Economics 8(1):3−35..
The Pe o mance o Ge man Coope a i e Banks —227
Pasiou as, F., S. Tanna, and C. Gaganis. 2011. “Wha D i es Acquisi ions in he EU Banking Indus y?
The Role o Bank Regula ion and Supe ision F amewo k, Bank Speci ic and Ma ke Speci ic
Fac o s.” Financial Ma ke s, Ins i u ions & Ins umen s 20 (2): 29−77..
Reichling, P., and G. Schulze. 2018. “Regional Di e ences in he E iciency o Ge man Sa ings Banks.”
C edi and Capi al Ma ke s −K edi und Kapi al 51 (3): 445−64..
Schumpe e , J. 1939. Business Cycles: A Theo e ical, His o ical, and S a is ical Analysis o he Capi alis
P ocess. New Yo k: McG aw-Hill.
Sealey, C. W., and J. T. Lindley. 1977. “Inpu s, Ou pu s, and a Theo y o P oduc ion and Cos a
Deposi o y Financial Ins i u ions.” The Jou nal o Finance 32 (4): 1251−66..
Shumway, T. 2001. “Fo ecas ing Bank up cy Mo e Accu a ely: A Simple Haza d Model.” Jou nal o
Business 74 (1): 101−24..
Sima , L., and P. W. Wilson. 2007. “Es ima ion and In e ence in Two-S age, Semi-pa ame ic Models o
P oduc ion P ocesses.” Jou nal o Econome ics 136 (1): 31−64..
Spoke iciu e, L., K. Keasey, and F. Vallascas. 2019. “Do Financial C ises Cleanse he Banking Indus y?
E idence om US Comme cial Bank Exi s.” Jou nal o Banking & Finance 99: 222−36..
Tische , J. 2018. Quan i a i e Easing, Po olio Rebalancing and C edi G ow h: Mic o E idence om
Ge many. Discussion Pape 20/2018. F ank u am Main: Deu sche Bundesbank.
Wheelock, D. C., and P. W. Wilson. 2000. “Why Do Banks Disappea ? The De e minan s o US Bank
Failu es and Acquisi ions.” The Re iew o Economics and S a is ics 82 (1): 127−38..
Wo hing on, A. C. 2004. “De e minan s o Me ge and Acquisi ion Ac i i y in Aus alian Coope a i e
Deposi -Taking Ins i u ions.” Jou nal o Business Resea ch 57 (1): 47−57..
