Adopting restriction: how coalition governments respond to radical right parties

Ellge , Fabio; Klü e , Heike
A icle — Published Ve sion
Adop ing es ic ion: how coali ion go e nmen s espond
o adical igh pa ies
Wes Eu opean Poli ics
P o ided in Coope a ion wi h:
WZB Be lin Social Science Cen e
Sugges ed Ci a ion: Ellge , Fabio; Klü e , Heike (2025) : Adop ing es ic ion: how coali ion
go e nmen s espond o adical igh pa ies, Wes Eu opean Poli ics, ISSN 1743-9655, Taylo &
F ancis, London, Vol. 48, Iss. 6, pp. 1299-1331,
h ps://doi.o g/10.1080/01402382.2024.2341352
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/308005.2
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
Wes eu opean poli ics
2025, Vol. 48, no. 6, 1299–1331
Adop ing es ic ion: how coali ion go e nmen s
espond o adical igh pa ies
Fabio Ellge a and Heike Klü e b
aWZB Be lin social science cen e , Be lin, Ge many; bHumbold -uni e si ä zu Be lin, Be lin,
Ge many
ABSTRACT
Radical igh pa ies (RRPs) ha e gained ep esen a ion in pa liamen s ac oss
Eu ope, bu li le is known abou hei impac on go e nmen policy. E en
hough RRPs end o end up in opposi ion, hey ha e conside able in luence
on policy making wi hin coali ion go e nmen s. One expec a ion is ha coali-
ion go e nmen s a e ough on immig a ion o op imise o e suppo when
being exposed o igh -wing pa ies in pa liamen . Coali ion nego ia ions em-
po a ily educe accoun abili y and allow cabine s o adjus policy posi ions
wi hou bea ing he cos s associa ed wi h oppo unis ic beha iou . This a gu-
men is es ed using no el da a on p e-elec o al policy posi ions and
pos -elec o al immig a ion policies o coali ion cabine s in 24 Eu opean
democ acies om 1980 o 2015. The indings e eal ha go e nmen s shi o
mo e es ic i e immig a ion policies in ace o RRPs. This a icle expands on
p io esea ch on he in luence o he adical igh by demons a ing i s di ec
in luence on coali ion go e nmen s’ join immig a ion policy plans.
KEYWORDS Compa a i e poli ics; coali ion go e nmen ; adical igh pa ies; immig a ion
policy
Do adical igh pa ies in luence coali ion go e nmen policy? Righ -wing
and populis challenge pa ies ha e es ablished hemsel es wi hin a ange
o Eu opean pa y sys ems. While mains eam pa ies su e d ama ic
elec o al losses, adical igh pa ies such as he Ge man Al e na i e ü
Deu schland, he Danish Folkepa i o he Aus ian F eihei liche Pa ei
Ös e eichs a e on he ise. Mo e ecen ly, schola s s a ed o in es iga e
how hese pa ies – albei mos ly being in he opposi ion – ha e a ec ed
he policy posi ions o es ablished pa ies (Abou-Chadi and K ause 2020;
Meguid 2005; Spoon and Klü e 2019) o he a i udes o o e s (Bischo
h ps://doi.o g/10.1080/01402382.2024.2341352
© 2024 he au ho (s). published by in o ma uK limi ed, ading as aylo & F ancis G oup
CONTACT Fabio ellge abio[email p o ec ed]
supplemen al da a o his a icle can be accessed online a h ps://doi.o g/10.1080/01402382.20
24.2341352.
his is an open access a icle dis ibu ed unde he e ms o he c ea i e commons a ibu ion license (h p://
c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any
medium, p o ided he o iginal wo k is p ope ly ci ed. he e ms on which his a icle has been published allow he
pos ing o he accep ed Manusc ip in a eposi o y by he au ho (s) o wi h hei consen .
1300 F. ELLGER AND H. KLÜVER
and Wagne 2019). Rela edly, p io esea ch also shows he impac o
ex eme pa ies on immig a ion policy change (Abou-Chadi 2016b; Folke
2014; Howa d 2010; Schain 2006; Williams 2006). In addi ion, se e al
s udies explo e he di ec and indi ec e ec s o he adical igh on
mains eam pa y posi ions (see Abou-Chadi and K ause 2020; Han 2015;
K ause and Ma sunaga 2023; an Spanje 2010). We con ibu e o his li -
e a u e by conduc ing a no el es o he impac o RRP ep esen a ion on
changes in he join policy agenda o coali ion go e nmen s.
Speci ically, we show ha coali ion go e nmen s shi o mo e es ic-
i e immig a ion policies in hei join ag eemen s whene e adical igh
pa ies a e ep esen ed in pa liamen du ing cabine o ma ion. All pa ies
pay close a en ion o he dis ibu ion o o e p e e ences (Downs 1957).
Once hey ecognise a p e e ence shi among he elec o a e, pa ies adap
hei posi ions in o de o main ain elec o al suppo (Adams e  al. 2004;
Ez ow e  al. 2011). Schola s o pa y beha iou ha e examined his p o-
cess o ‘policy op imisa ion’ and ound a ying s a egies o adjus men .
Adams e al. (2006) show ha mains eam pa ies espond o shi s in he
gene al popula ion while niche pa ies pay special a en ion o p e e ence
changes among hei co e suppo g oups. Klü e and Spoon (2016) sim-
ila ly ind ha mains eam pa ies espond o issue p io i ies o he o e -
all popula ion while niche pa ies mos ly espond o hei own suppo e s.
Bu open policy shi s do come a a cos , as pa ies ace a ange o
es ic ions which cons ain hei beha iou (Meye 2013; Ta i s 2007).
Vo e s, and especially pa isan suppo e s, closely moni o p og amma ic
changes (Cox and McCubbins 1986; Ez ow e  al. 2011) and ha e a desi e
o c edible signals o policy s abili y. A pa y ha changes i s posi ions
egula ly and subs an ially has a ha d ime o con incingly commi o
policy adhe ence in he u u e. Acco dingly, Ta i s (2007) shows ha
oppo unis ic shi s on salien policy issues esul in elec o al penal ies.
Poli ical pa ies he e o e y o s a egically balance egula adap a ion o
p e e ence changes in he elec o a e wi h a oiding o appea as
‘ lip- loppe s’.
Resea ch also shows ha es ablished pa ies pay a en ion o new con-
ende s (Meguid 2005) and he e exis s a g owing in e es in he way
eme ging igh -wing challenge pa ies p o oke mains eam pa y eac-
ions (see Abou-Chadi and K ause 2020; Han 2015). RRPs eme ged in
poli ical a enas h oughou Eu ope and buil hei success on a combina-
ion o an i-eli is he o ic and na ionalis policy posi ions (Ki schel and
McGann 1997; Mudde 2007; Williams 2006). These pa ies ha e also
poli icised new ‘cul u al’ issues (K iesi e  al. 2006), wi h a special empha-
sis on an i-immig a ion policies (Dancygie 2010). When acing success ul
adical igh challenge s, mains eam pa ies o en eac wi h an accom-
moda i e s a egy (Meguid 2005; an Spanje 2010).
WEsT EuROPEAN POLiTiCs 1301
In his a icle, we ocus on he in luence o adical igh pa ies du ing
he o ma ion o mul ipa y coali ions. Cabine pa ies can use con iden-
ial coali ion nego ia ions – a pe iod o low accoun abili y and cla i y o
esponsibili y1 – o op imise hei posi ions on immig a ion wi hou bea -
ing he usual cos s associa ed wi h policy change on salien issues. In he
ace o RRPs ep esen ed in pa liamen , cabine pa ies p opose mo e
es ic i e immig a ion policy posi ions in hei coali ion ag eemen s. We
es ou a gumen based on a unique da ase co e ing 165 coali ion o -
ma ions in 24 Wes and Eas Eu opean pa y sys ems be ween 1980 and
2015. Ou analysis makes use o a no el da a ha combines posi ions
de i ed om coali ion ag eemen s wi h p e-elec o al posi ions om pa y
mani es os (Volkens e  al. 2018). We p o ide mo e in o ma ion on he
o igin o he ag eemen s in Online Appendix Sec ion 1.2.
Ou app oach is unique since we combine pa y posi ions o go e n-
men pa ies ex ac ed om hei p e-elec o al mani es os wi h da a on
policy posi ions o coali ion go e nmen s ha o igina e om hei
pos -elec o al coali ion ag eemen s. The coali ion ag eemen s a e usually
nego ia ed sho ly a e he elec ion and a e published only a ew weeks
a e o e s ha e cas hei ballo , which enables us o ack policy change
o e an excep ionally sho pe iod o ime. We can easonably assume
ha posi ion changes o go e nmen pa ies a e ba ely a ec ed by exoge-
nous, ime- a ying con ounde s. And since he elec ion mani es os and
he coali ion ag eemen s a e analysed using a simila coding p ocedu e,
we can mo eo e assess he magni ude and di ec ion o policy adjus men
du ing coali ion nego ia ions on p ecisely he same ideological scale.
To es ou hypo hesis on immig a ion policy op imisa ion du ing coa-
li ion nego ia ions, we analyse how much he coali ion ag eemen posi ions
(CAPs) de ia e om p e-elec o al expec ed posi ions (EPs) ha we gene a e
on he basis o Gamson (1961). We ind ha RRP p esence in pa liamen ,
a symbol o suppo o ex eme- igh poli ics among he elec o a e, is a
signi ican p edic o o shi s owa ds mo e es ic i e policies. A ange o
addi ional analyses show ha his e ec is no p ima ily d i en by he
size o adical challenge pa ies o hei ini ial en y o pa liamen . We
hen explo e how a ibu es o he incoming go e nmen mode a e he
e ec . While he gene al ideological alignmen o a cabine o he du a-
ion o i s o ma ion does no appea o impac policy change, we ind
ha mino i y cabine s, and cabine s which ha e los elec o al suppo , a e
mo e likely o shi owa ds es ic i e policies. In a inal se ies o obus -
ness es s, we show ha coali ion cabine s exposed o RRPs in pa liamen
a e qui e simila o hose wi hou igh -wing p essu es ac oss a ange o
impo an co a ia es.
O e all, he empi ical analysis p o ides e idence showing ha coali ion
cabine s con on ed wi h RRPs in pa liamen adop signi ican ly mo e
1302 F. ELLGER AND H. KLÜVER
es ic i e immig a ion policy posi ions han hose wi hou simila expo-
su e. Ou indings ha e wo main implica ions. Fi s , we a e going beyond
p io esea ch by highligh ing he impac o RRPs on coali ion cabine s.
In pa icula , we p o ide e idence o he ole o coali ion nego ia ions as
windows o oppo uni y o unpunished policy op imisa ion in ace o
p essu es by he adical igh . Ra ional, o e-seeking cabine pa ies can
accommoda e o mo e es ic i e posi ions while a oiding o appea
oppo unis ic. Second, we show ha he success o adical igh pa ies
does ha e an immedia e e ec on go e nmen policy. Compa ed o posi-
ions d a ed jus a ew weeks o mon hs be o e he elec ions, cabine s
ake on signi ican ly ha she posi ions in hei join ag eemen s when
RRPs a e ep esen ed in pa liamen . O e all, we con ibu e o he li e a-
u e on policy posi ioning in con es ed mul i-pa y sys ems and exis ing
esea ch on consequences o igh -wing pa y success o immig a ion
policy in ad anced democ acies.
Radical igh pa y pe o mance and immig a ion policy
change
In his s udy, we es i he p esence o adical igh pa ies in pa liamen
impac s he policy agendas o coali ion go e nmen s. Mo e speci ically, we
obse e changes in immig a ion policy, a highly salien and con es ed
issue a ea ha is closely linked o adical igh challenge s (Dancygie
2010). Ou a gumen builds on exis ing wo k ha highligh s he in luence
o he adical igh on immig a ion policy. Fo example, Schain (2006)
aces he de elopmen o he ‘F on Na ional’ in F ance and shows how
i s elec o al success causes ealignmen o exis ing pa ies. In a compa a-
i e amewo k, Howa d (2010) showcases ha he s eng h o a - igh
pa ies is he mos ele an p edic o o na u alisa ion p ocesses ac oss
EU membe s. Rela edly, Abou-Chadi (2016b) shows ha he poli icisa ion
and elec o al compe i ion o he immig a ion issue is a pi o al e opoin
o mo e p og essi e immig a ion policies. And in a quasi-expe imen al
s udy on Swedish municipal elec ions, Folke (2014) p o ides a causal
assessmen o he in luence o RRP ep esen a ion on s ic e immig a-
ion policy.
Beyond analysing he impac o RRPs on gene al immig a ion policy,
p io esea ch has also demons a ed hei e ec on o he pa ies. Fo
example, an Spanje (2010), Abou-Chadi (2016a) and Han (2015) show
accommoda i e beha iou and ‘con agion’ e ec s o mains eam pa ies
and pa liamen a ians ac oss es ablished democ acies. In a simila ein,
Abou-Chadi and K ause (2020) implemen a RD-design o p o ide causal
e idence on he impac o RRPs on mains eam pa ies’ immig a ion
posi ions, also inding accommoda ion e ec s among he es ablishmen .

WEsT EuROPEAN POLiTiCs 1303
Going beyond he assessmen o adical pa ies’ impac on hei compe -
i o ’s posi ions, K ause and Ma sunaga (2023) show ha accommoda i e
s a egies o he es ablishmen do no pay o elec o ally. We con ibu e o
his li e a u e by shi ing he a en ion away om indi idual pa y
esponses o eac ions by en i e coali ion cabine s.
Fo sake o compa abili y, ou concep ualisa ion o RRPs ollows es ab-
lished s anda ds se by schola s in he ield. We iden i y RRPs based on
hei c i ical s ance owa ds mains eam immig a ion policy (see
Abou-Chadi and K ause 2020; Roodujin e al. 2019). We concei e o ad-
ical and populis igh -wing pa ies as challenge s o es ablished main-
s eam pa ies because hey h ea en hem elec o ally (Abou-Chadi 2016b;
De V ies and Hobol 2020; Howa d 2010). While mains eam pa ies su -
e d ama ic elec o al losses in ecen yea s, adical and populis igh
pa ies such as he Ge man Al e na i e ü Deu schland, he Danish
Folkepa i o he Aus ian F eihei liche Pa ei Ös e eichs a e on he ise.
Radical igh -wing pa ies exe p essu e on mains eam pa ies by
mobilising on he immig a ion issue and employing an i-es ablishmen
he o ic o unde mine mains eam pa y legi imacy (De V ies and Hobol
2020; Mudde 1999; Schain 2006). One co e claim is a demand o mo e
es ic i e mig a ion laws (B ug & Van Spanje 2009). O en, hese claims
a e combined wi h populis appeals and he usage o acis s e eo ypes
(Mudde 2007). Thus, we a gue ha RRPs exe ‘owne ship’ o e he
immig a ion issue by demanding o ‘go ough’ on immig a ion (see
Abou-Chadi 2016b; Howa d 2010; Schain 2006). Since immig a ion pol-
icy has become inc easingly con en ious and highly poli icised in ecen
elec ions, adical and populis igh -wing pa ies equen ly bene i ed
om a mobilisa ion on his issue (Dancygie 2010; Ki schel and McGann
1997). As challenge s, hey did so by a ac ing o e s om es ablished
pa ies ac oss he adi ional le - igh dimension (McGann and Ki schel
2005; Rydg en 2008).
We he e o e a gue ha success ul adical and populis igh pa ies
pose an elec o al h ea o coali ion go e nmen s. The p esence o RRPs
signals o coali ion pa ies ha a sizable pa o he elec o a e p e e s
mo e es ic i e immig a ion policies. In an e o o appeal o hese
o e s, we expec ha coali ion go e nmen s ha e s ong incen i es o
adop ha she policies on immig a ion. Howe e , one cen al ca ea we
p opose is ha go e nmen s a e no able o eely eposi ion hemsel es
on such a salien issue. Flip- lopping will esul in c edibili y losses and
penalisa ion, especially by dedica ed co e suppo e s (Ez ow e  al.
2011). In he nex sec ion, we lay ou how he blu ed lines o espon-
sibili y du ing con iden ial coali ion nego ia ions can p o ide go e ning
pa ies wi h an oppo uni y o upda e hei policy posi ions wi hou
ha ing o bea he nega i e consequences o posi ion adjus men . A e
1304 F. ELLGER AND H. KLÜVER
de eloping ou a gumen on why mains eam pa ies should make use
o coali ion nego ia ions o upda e hei policy agenda, we conduc an
empi ical es . We p o ide e idence which shows ha cabine s shi
owa ds mo e es ic i e posi ions i hey ha e been exposed o a RRP
in pa liamen .
Policy op imisa ion du ing coali ion nego ia ions
We hypo hesise ha coali ion pa ies will se le wi h mo e es ic i e pol-
icies in hei coali ion ag eemen s compa ed o he policy posi ions hey
p oposed in p e-elec o al campaign mani es os i hey ace RRPs in pa -
liamen . Tha is, because he coali ion o ma ion s age p o ides an ideal
oppo uni y o go e nmen pa ies o adjus hei policy agenda in ligh
o ecen elec ion esul s wi hou bea ing elec o al cos s.
We concei e o policy op imisa ion du ing he coali ion cabine o ma-
ion as a h ee-s age p ocess (see Figu e 1). Fi s , we assume ha pa ies
a e a ional, goal-o ien ed and pu pose ul collec i e ac o s in e es ed in
maximising hei own elec o al suppo (Downs 1957). Following he spa-
ial p oximi y model, we hen assume ha he main ins umen o maxi-
mise elec o al suppo is s a egic posi ion- aking along a se o salien
policy dimensions (S okes 1963).
The spa ial p oximi y model also posi s ha poli ical pa ies ca e ully
analyse he dis ibu ion o o e p e e ences (Enelow and Hinich 1984).
Vo e s a e expec ed o suppo he pa y whose policy p og amme mos
closely aligns wi h hei indi idual p e e ences on he issues hey deem
ele an . Since o e p e e ences change egula ly o e ime (Con e se
1964; Pede sen 1979), pa ies a e incen i ised o cons an ly moni o and
Figu e 1. s a egic policy adjus men o coali ion go e nmen s a ound elec ions.
WEsT EuROPEAN POLiTiCs 1305
espond o shi s in he elec o a e (Adams e  al. 2004; Schumache , de
V ies, and Vis 2013). Thus, we hink o pa y compe i ion as a dynamic
sys em in which pa ies adap hei posi ions o changing o e p e e -
ences (see S imson, Mackuen, and E ikson 1995; So oka and Wlezien
2010; Adams e  al. 2004).
A he same ime, poli ical pa ies a e also held accoun able, especially
by hei co e suppo e s. Ez ow e  al. (2011) ha e con as ed he gene al
elec o a e model, which ocuses on o e maximisa ion among all o e s,
wi h he pa isan cons i uency model, acco ding o which pa ies pay spe-
cial a en ion o he p e e ences o hei closes ollowe s. Poli ical pa ies
ely on a s able and commi ed g oup o ac i is s o he main enance o
local pa y in as uc u e, olun ee ing, and campaigning. Acco ding o
he ‘co e suppo e model’ pa ies he e o e a e mo e esponsi e owa ds
his small g oup and do no isk sca ing hem o (Cox and McCubbins
1986). Since hese co e suppo e s pay special a en ion o policies
(Schumache , de V ies, and Vis 2013), changes o he o icial pa y s ances
on salien ma e s like immig a ion can likely igge a en ion and ange .
Ac i is s, who a e ypically d i en by ideological mo i es, will abandon
o e ly oppo unis ic pa ies (Be nha d , Duggan, and Squin ani 2009).
Thus, open posi ion adjus men can be a cos ly s a egy ha unde mines
elec o al suppo .
Poli ical pa ies need o balance popula demand and co e suppo e
ideology when de ining hei policy posi ions be o e elec ions, as depic ed
in s age one o Figu e 1. They i s d a elec ion mani es os in which
hey lay ou hei ideological s ance. These mani es os a e he mos ency-
clopaedic s a emen s o pa ies’ policy p og ammes and se he guidelines
o he ensuing campaigns, po en ial coali ion nego ia ions and o u u e
legisla i e ac i i y (Budge 1987; Ede , Jenny, and Mülle 2017). Bu pa ies
ha e o c ea e hei mani es os in a con ex o incomple e in o ma ion
abou ac ual o e p e e ences. Guessing how he posi ions ound in he
mani es os will esona e wi h he elec o a e is ha d because he p eceding
elec ion is only a poo indica o o cu en o e p e e ences
(Some -Topcu 2009).
Polls a e used by poli ical eli es o decision making (e.g. Pickup and
Hobol 2015), bu a e a bes a c ude indica o o o e p e e ences. They
deli e an es ima e o o ing in en ions a he ime o hei execu ion ha
comes wi h sampling e o , measu emen e o , and is in luenced by sub-
jec i e weigh ing decisions (see Hudson e  al. 2004). Recen e idence
mo eo e shows ha inc easing elec o al ola ili y in mul i-pa y sys ems
and he eme gence o new poli ical ac o s, such as RRPs, ha e u he
ele a ed ambigui y o e inal elec ion ou comes (Hobol and Tilley 2016;
P osse and Mellon 2018).
1306 F. ELLGER AND H. KLÜVER
I is only h ough elec ion esul s – shown a s age wo o Figu e 1 –
ha pa ies ecei e comple e in o ma ion abou o e p e e ences.
The e o e, we concei e o elec ion ou comes as in o ma ion signals on
o e s’ p e e ed policy posi ions (Abou-Chadi and O lowski 2016; Budge
1994). All poli ical pa ies ecei e di ec eedback on he popula i y o
hei p e-elec o al policy p oposi ions. In an en i onmen wi hou ans-
ac ion cos s, we would expec all pa ies o immedia ely op imise posi-
ions based on hei pe o mance a he ballo . Bu pa ies a e cons ained
since open change o policy posi ions can unde mine hei c edibili y and
suppo .
We a gue ha pa icipa ion in coali ion nego ia ions, he hi d s age
highligh ed in Figu e 1, cons i u e an excep ion om his ule. In mul i-
pa y sys ems, mo e han one pa y ypically o ms a coali ion go e n-
men since no single pa y gains a legisla i e majo i y. Coali ion
go e nmen s a e composed o a leas wo poli ical pa ies ha sha e
execu i e o ices (S øm, Mülle , and Be gman 2008: 6). Go e ning in
coali ions can be in ica e. E en hough coali ion pa ies join o ces o
gain con ol o e execu i e o ices, hey ypically pu sue di e en policy
objec i es. Be o e en e ing go e nmen , coali ion pa ies he e o e engage
in leng hy coali ion nego ia ions ha ypically las a ew weeks o ba gain
o e he alloca ion o minis e ial po olios as well as o e hei join pol-
icy agenda ha is hen w i en down in a public coali ion ag eemen
(Klü e and Bäck 2019; S øm, Mülle , and Be gman 2008). These nego-
ia ions a e o en labelled as a ‘black box’ since he ba gaining p ocess is
highly con iden ial.
We a gue ha hese coali ion nego ia ions p o ide an excellen oppo -
uni y o cabine pa ies o op imise hei posi ions in ace o RRPs.
Cabine pa ies can p o i om being esponsi e by aligning hei p opo-
si ions o p e e ences o he gene al elec o a e and accommoda e o RRP
posi ions wi hou bea ing he cos s associa ed wi h open posi ion swi ch-
ing. This is because coali ion nego ia ions c ea e a con ex o low accoun -
abili y du ing which posi ion swi ching canno be aced back o
indi idual ac o s.
Posi ion swi ching du ing nego ia ions is also appealing as i allows
pa ies o ensh ine hei comp omise in hei join coali ion ag eemen .
The ag eemen s ha e implica ions o ac ual policy making h oughou
he go e nmen ’s e m in o ice. E en hough coali ion ag eemen s a e no
legally binding, hey de ine policy p io i ies o a cabine and cons ain he
beha iou o each coali ion pa y. While coali ion nego ia ions a e sec e ,
he coali ion ag eemen s a e publicly eleased, which enhances he com-
pliance as shi king will lead o public blaming and shaming. Acco dingly,
Thomson (2001), Mou y (2011) and Sche mann and Ennse -Jedenas ik
(2014) show ha elec ion pledges o which coali ion go e nmen s commi
WEsT EuROPEAN POLiTiCs 1313
EP
R
Lsqs
ij
nij
ij
ij j
ni
log log log
()
=+
+







=
= =
∑∑
1 1
05
05
601
.
.∗jj i j
ij ij
ij
qs
qs qs s
++
()
++
()








608 0 5
602 607 0 5
.
.
−∗
log (4)
In Online Appendix Sec ion 6.3, we con i m ha hese expec ed posi-
ions a e highly co ela ed wi h posi ions weigh ed by a cabine ’s po olio
alloca ion (Co ela ion = 0.91; p < 0.001). In an addi ional obus ness
check in Online Appendix Sec ion 6.3, we also eplica e ou main analysis
using hese po olio-weigh ed posi ions. The e, we show ha u ilising
sea sha es ins ead o he po olio dis ibu ion does no change ou
esul s.
Following Gamson’s law (Gamson 1961), we assume ha he expec ed
posi ion (EPs) de i ed om mani es os a e s ong p edic o s o he coali-
ion ag eemen posi ions (CAPs). Bo h measu e he same phenomenon,
immig a ion policy, on exac ly he same scale based on a compa able con-
en analysis o poli ical ex s which a e usually d a ed jus a ew mon hs
apa . Gi en he sho ime span we a e also con iden o ule ou a mul-
i ude o exogenous con ounde s. Bi a ia e eg essions p esen ed in Online
Appendix Sec ion 5.1 p o ide s ong empi ical suppo o ou assump-
ion. The expec ed posi ion on immig a ion policy ha we scale based on
he p e-elec o al mani es os is a majo p edic o o he ac ual immig a ion
posi ion ha coali ions adop in hei coali ion ag eemen .
Ope a ionalisa ion o independen a iables
We ely on wo independen a iables in o de o es ou heo e ical
a gumen ha coali ion go e nmen s adop mo e es ic i e immig a ion
policies when RRPs a e p esen in pa liamen .
Fi s , he main analysis u ilises he dummy a iable RRP Pa liamen ,
which cap u es i a adical igh pa y has gained ep esen a ion a a
gi en elec ion. Ob aining pa liamen a y ep esen a ion cons i u es a s ong
signal aimed a mo e es ablished pa ies (Bischo and Wagne 2019). As a
consequence, mains eam pa ies a e acing RRPs in hei daily pa lia-
men a y wo k. Once elec ed in o he legisla u e, igh -wing challenge s
also ecei e addi ional esou ces and media a en ion which inc eases he
h ea ha hey cons i u e o es ablished pa ies (Abou-Chadi and K ause
2020). The e o e, we expec coali ion go e nmen s o espond by adop ing
mo e es ic i e immig a ion policies. Empi ically, we ind ha igh -wing
challenge s gain ep esen a ion in 61% o all coali ion o ma ion ins ances
obse ed in ou sample. This ensu es ha he e is su icien a ia ion
wi hin he sample, sugges ing ha no e e y cabine o ma ion we obse e
occu s wi h exposu e o a adical pa y in pa liamen . We conside pa ies

1314 F. ELLGER AND H. KLÜVER
ha display s ong na ionalis and an i-immig a ion posi ions as igh -wing
challenge s. The classi ica ion is based on p e alen wo k in he ield,
mos impo an ly by Mudde (2007) and Roodujin e  al. (2019). Fo
Eas e n Eu ope and ecen elec ions, we also consul ed elec ion epo s
and ela ed s udies (e.g. Bus iko a 2014). Online Appendix Sec ion 1.1
lis s all RRPs co e ed in ou analysis.
Addi ionally, we add he second independen a iable RRP Vo e Sha e
o some models. I is a con inuous measu e o he combined o e sha e
o all adical igh pa ies. We u ilise he con inuous measu e o RRP
p e alence o assess i exposu e o la ge igh -wing challenge s has a
mo e p onounced e ec on immig a ion posi ion change. This would
imply ha i is no he me e p esence o challenge s, bu hei elec o al
dominance ha should igge policy op imisa ion du ing nego ia ions.
We do no ha e s ong ideological p io s on he expec ed esul s. Tha is
because Abou-Chadi and K ause (2020) ha e al eady p oduced con incing
causal e idence ha small RRPs, jus abo e he elec o al h eshold, can
in luence posi ions o mains eam pa ies. Ce ainly, a la ge RRP should
signal mo e suppo o an i-immig a ion p oposi ions and hus could
induce la ge changes du ing he nego ia ions. On he o he hand, a la ge
RRP migh educe he unce ain y abou he appeal o an i-immig a ion
p oposi ions among he gene al elec o a e p io o he elec ion day. Tha
is because a igh -wing challenge polling a a ound 15% o mo e o he
popula o e is e y likely o be ep esen ed in pa liamen a e he elec-
ions. I unce ain y abou he appeal o ough posi ions on immig a ion
is low p io o he elec ion, es ablishmen pa ies could al eady accoun
o hem in hei p e-elec o al pa y p og ammes. This could educe hei
po en ial o adap a ion du ing he cabine o ma ion pe iod.
Da a and es ima ion
Ou analysis in es iga es he e ec o adical igh pa ies on coali ion
go e nmen ’s immig a ion policy change du ing hei o ma ion. We aim
o obse e i cabine s ha o m while being exposed o RRPs in pa lia-
men shi o mo e es ic i e immig a ion posi ions han o he coali ion
cabine s.
We he e o e compiled a no el da ase based on p e- and pos -elec o al
posi ions o 165 coali ion cabine s ac oss 24 Wes e n and Eas e n
Eu opean democ acies be ween 1980 and 2015. Fo he selec ion o coun-
ies, we ollowed he es ablished s anda d in coali ion esea ch (see
Ande sson, Be gman, and E sson 2014). The coun ies in ou sample
include Aus ia, Belgium, Bulga ia, Czech Republic, Denma k, Es onia,
Finland, Ge many, G eece, Hunga y, Iceland, I eland, La ia, Li huania,
WEsT EuROPEAN POLiTiCs 1315
Luxembou g, Ne he lands, No way, Poland, Po ugal, Romania, Slo akia,
Slo enia, Sweden, and he Uni ed Kingdom. This coun y sample is cha -
ac e ised by a ia ion in se e al ins i u ional ea u es, which s eng hens
he ex e nal alidi y o ou indings.
Wi h ega d o he collec ion o coali ion ag eemen s, we build on he
wo k by Klü e , Bäck, and K auss (2023). The au ho s ha e compiled an
ex ensi e da ase on he policy con en o coali ion ag eemen s ac oss
Eu ope. C ucially, hey ha e assessed he documen s o cohe ence and
compa abili y. Fo example, he au ho s ensu ed ha each ag eemen ul-
ils ce ain c i e ia (i.e. he discussion o speci ic policy plans o he leg-
isla u e o he public backing by all cabine pa ies) and hei coun y
expe s we e ained o ensu e alidi y and eliabili y. To his da a, we add
he in o ma ion on elec ion esul s and he ep esen a ion o RRPs om
Pa lGo (Dö ing and Manow 2018), as well as p e-elec o al mani es o
posi ions based on he Mani es o P ojec (Budge and Klingemann 2001;
Volkens e  al. 2018).
To es ou hypo hesis on immig a ion policy op imisa ion du ing coa-
li ion nego ia ions, we hen assess how he pos -elec o al coali ion ag ee-
men posi ions de ia e om he p e-elec o al expec ed posi ions ha we
gene a e on he basis o Gamson (1961). By s udying he di e ence
be ween expec ed posi ions based on p e-elec o al elec ion mani es os
wi h ac ual go e nmen posi ions adop ed in coali ion ag eemen s, we
measu e policy change wi hin an ex emely sho ime span. This no
only makes i possible o es ic he in luence o ime- a ying con ound-
e s. The ollowing equa ion ep esen s ou main speci ica ion wi h a di -
e enced ou come a iable:
∆
Posi ion RRPPa liamen
ic c ic ic ,, , ,, ,,
=+ ++αβ β ε
12
X (5)
We a e in e es ed in he change in immig a ion policy posi ion o go -
e nmen i in coun y c a yea as he dependen a iable. This is ep e-
sen ed by ΔPosi ioni,c, , which is he di e ence be ween a go e nmen ’s
p e-elec o al expec ed posi ion (EP) and he pos -elec o al coali ion ag ee-
men posi ion (CAP). No e ha posi i e alues indica e a change owa ds
mo e es ic i e, and nega i e alues indica e a mo e owa ds less es ic-
i e policy posi ions du ing cabine o ma ion. The dependen a iable
hus cap u es he dynamics o he posi ion change a he han he abso-
lu e posi ion a any gi en ime poin .
The key explana o y a iable o in e es is RRP Pa liamen , a dummy
a iable ha indica es he p esence o RRPs in pa liamen . The coe icien
β1 on his a iable cap u es he a e age di e ence in immig a ion policy
posi ion change associa ed wi h he p esence o an RRP in pa liamen ,
ela i e o when an RRP is no p esen , holding o he a iables cons an .
1316 F. ELLGER AND H. KLÜVER
To accoun o in a ian con ounding ac oss coun ies and ime ends,
we add ac, , ep esen ing coun y (c) and cabine o ma ion yea ( ) ixed
e ec s, espec i ely, in ou speci ica ion. These ixed e ec s con ol o
unobse ed he e ogenei y ha may a ec immig a ion policy posi ions,
holding cons an c oss-coun y di e ences and ime ends such as he
gene al poli ical clima e, ins i u ional ac o s, o mac oeconomic condi-
ions. By doing so, we ensu e ha ou coe icien o in e es , β1, cap u es
he wi hin-coun y, wi hin-yea a ia ion a ibu able o he p esence o
adical igh pa ies (RRPs), he eby making ou es ima es mo e obus o
omi ed a iable bias.
In addi ion o he s a e and yea ixed e ec s, we also include Xi,c, , a
ec o o cabine -speci ic con ol a iables. This includes he absolu e
numbe o cabine pa ies (as a measu e o he complexi y o nego ia-
ions), a dummy ha deno es minimum-winning coali ions (which a e he
mos common o m o coali ions, ins ead o mino i y o supe -majo i a ian
cabine s) and he na ional unemploymen a e du ing he yea o cabine
o ma ion as a measu e o gene al economic condi ions. E en hough ou
es ima ion s a egy elies on he analysis o a b oad se o a ailable da a
on coali ion o ma ion in Eu ope, he ime se ies c oss sec ion analysis is
demanding due o he limi ed numbe o coun y-cases and ime poin s.
The esul s he e o e can come wi h a deg ee o s a is ical unce ain y.
Finally, all s anda d e o s a e clus e ed by coun y and desc ip i e s a is-
ics a e p esen ed in Online Appendix Sec ion 2.
Empi ical analysis
In Table 1, we p esen he i s empi ical es o ou a gumen . We un a
se ies o eg ession models ha ely on a di e enced ou come measu e
and include coun y and yea ixed e ec s and clus e ed s anda d e o s
by coun y. The dicho omous RRP Pa liamen dummy ac s as he inde-
penden a iable. Fo he di e enced ou come a iable, posi i e alues
deno e a change owa ds mo e es ic i e policy posi ions, while nega i e
alues indica e a change owa ds mo e p og essi e posi ions on immig a-
ion. Again, hese changes ypically occu o e a ew weeks a ound elec-
ion day. We un sepa a e eg essions o he ull sample o cabine s
(n = 165) and a es ic i e sample (n = 132) ha excludes cabine s which
did no explici ly s a e policy posi ions in ei he pa y mani es os o he
coali ion ag eemen .
Fo all models o Table 1, we obse e ha he en y o adical igh
pa ies in o go e nmen is associa ed wi h a mo e o mo e es ic i e pol-
icy posi ions du ing he coali ion nego ia ion phase. In h ee o ou
models, his associa ion is signi ican abo e con en ional le els o
WEsT EuROPEAN POLiTiCs 1317
s a is ical unce ain y. Impo an ly, he e ec o adical igh pa y p es-
ence in pa liamen on shi s o mo e es ic i e immig a ion policy is
highly obus once we es ic he sample o hose 132 cabine s ha did
co e mig a ion policy in bo h hei elec ion p og ammes and join ag ee-
men s. The co e age o immig a ion by es ablishmen pa ies in hei pol-
icy pla o ms indica es a heigh ened salience o his issue. In e ms o
e ec magni ude, we ind policy shi s o 0.4 o 0.7 poin s on he Kim/
Fo ding scale – wi h a heo e ical maximum o 2 poin s – o be qui e
subs an ial. A maximum change o 2 poin s would cons i u e a change
om he mos ex eme pe missi e p e-elec o al posi ion o he mos
ex eme es ic i e pos -elec o al ag eemen posi ion. Again, we obse e
his change o e he sho ime span be ween d a ing elec o al p o-
g ammes and nego ia ing he join coali ion ag eemen .
In a second s ep, we add RRP Vo e Sha e, he o e sha e o all adical
igh pa ies, as an addi ional independen a iable o es i posi ion
change becomes mo e p onounced once adical challenge s gain la ge
o e sha es. Acco ding o he esul s p esen ed in Table 2, we ind ha
he size o RRPs does yield la ge policy shi s. I any hing, bigge chal-
lenge s appea o lead o ma ginally less es ic i e change. Howe e , mos
es ima es a e no signi ican ly disce nible om ze o.
Why would me e exposu e o RRPs, bu no hei size in luence a cab-
ine ’s immig a ion plans? P o iding conclusi e e idence on his inding is
beyond he scope o ou a icle. Howe e , one plausible explana ion could
be ha es ablishmen pa ies y o ca e owa ds RRP suppo e s as long
as hey emain ma ginal compe i o s. The accommoda i e beha iou o
mains eam cabine s migh be a iable s a egy as long as he adical
compe i o s a e small (Meguid 2005). Once RRPs ha e es ablished hem-
sel es in he pa y sys em and become majo compe i o s o o ices,
Table 1. policy change o ps in pa liamen .
Dependen a iable:
posi ion change (KF) posi ion change (log)
(1) (2) (3) (4)
p pa liamen 0.422* 0.724*** 0.749 1.350**
(0.221) (0.176) (0.614) (0.523)
MWc −0.170 −0.246 −0.307 −0.505
(0.133) (0.145) (0.330) (0.403)
no. cabine pa ies 0.0041 0.061 0.129 0.329
(0.078) (0.081) (0.234) (0.251)
unemploymen −0.024 −0.027 −0.052 −0.080
(0.019) (0.023) (0.058) (0.084)
Fixed e ec s coun y + Yea coun y + Yea coun y + Yea coun y + Yea
obse a ions 165 132 165 132
sample Full es . Full es .
adjus ed 20.319 0.399 0.182 0.232
no e: *p < 0.1; **p < 0.05; ***p < 0.01.
1318 F. ELLGER AND H. KLÜVER
mains eam pa ies could op o di e en ia ion in o de o sha pen hei
own policy p o iles. In line wi h his p oposi ion, Abou-Chadi and K ause
(2020) ha e p esen ed con incing causal e idence on he accommoda i e
beha iou o mains eam pa ies in esponse o small RRPs ha ba ely
en e ed pa liamen a ound he elec o al h eshold.
Cabine cha ac e is ics and he in luence o RRPs
A e showing ha coali ion go e nmen s change hei posi ion on immi-
g a ion policies when acing adical challenge s in pa liamen , we now
explo e wha cabine cha ac e is ics could in luence he e ec . In pa icu-
la , we ocus on ou coali ion a ibu es ha could shape he deg ee o
policy in luence by RRPs.
Fi s , we assess whe he he elec o al pe o mance o a cabine mode -
a es he in luence o RRPs on policy change. One could expec ha cab-
ine s which ha e los o es a e p one o p essu es om he adical igh .
We hus c ea e he bina y a iable Elec o al Loss o indica e i cabine
pa ies ha e on a e age los o es compa ed o he p e ious elec ion. A
alue o one shows ha he cabine has expe ienced an elec o al loss.
Nex , we ocus on he gene al ideology o a cabine as a po en ial mod-
e a o . He e, one could assume ha especially cen e- igh go e nmen s
a e p one o p essu es om adical pa ies because hey a e mo e likely
o lose suppo e s o he a igh . To di e en ia e be ween le and
igh -leaning go e nmen s, we c ea e he bina y a iable Le Cabine . We
measu e cabine ideology based on he weigh ed a e age posi ion on he
gene al le - igh scale (o ‘ ile’-scale) o each cabine pa ies’ mani es o.
Cabine s wi h an a e age ideology sco e lowe han 0 a e coded as being
Table 2. p size and policy change.
Dependen a iable:
posi ion change (KF) posi ion change (log)
(1) (2) (3) (4)
p pa liamen 0.465* 0.919*** 0.784 1.732***
(0.246) (0.184) (0.671) (0.605)
p Vo e sha e −0.006 −0.020** −0.004 −0.039
(0.010) (0.009) (0.023) (0.026)
MWc −0.172 −0.247* −0.309 −0.508
(0.133) (0.138) (0.332) (0.393)
no. cabine pa ies 0.004 0.087 0.129 0.379
(0.079) (0.083) (0.235) (0.259)
unemploymen −0.026 −0.031 −0.053 −0.088
(0.020) (0.026) (0.061) (0.091)
Fixed e ec s coun y + Yea coun y + Yea coun y + Yea coun y + Yea
obse a ions 165 132 165 132
sample Full es . Full es .
adjus ed 20.319 0.435 0.175 0.240
no e: *p < 0.1; **p < 0.05; ***p < 0.01.

WEsT EuROPEAN POLiTiCs 1319
le cabine s. Thi d, we sugges ha he leng h o he coali ion nego ia-
ions hemsel es could impac he in luence o RRPs on policy changes.
We hus ha e esea ched he leng h be ween he elec ion da e and go -
e nmen o ma ion, measu ed in days. We ha e been able o e ie e
in o ma ion on he o ma ion du a ion o 161 ou o 165 cabine s in he
ull sample and 128 o 132 cabine s in he es ic ed sample. In he nex
s ep, we conduc a median spli o c ea e he bina y indica o Long
Nego ia ion. Longe nego ia ion pe iods should exe cabine pa ies wi h
a longe ime pe iod o implici p essu e om he adical igh while
p o iding hem wi h mo e oppo uni ies o shi hei policies. Finally, we
ocus on he coali ion ype i sel . Mino i y cabine s do no possess he
pa liamen a y s eng h o implemen legisla ion alone, hus making hem
ely on pa s o he opposi ion o policy making. This could make hem
mo e p one o shi owa ds es ic i e policies in o de o es ablish sup-
po along he line o immig a ion policy.
We eplica e he p io OLS eg ession models, bu in e ac each o he
bina y cabine cha ac e is ics wi h he RRP Pa liamen dummy a iable. We
measu e policy shi s on he Kim/Fo ding-scale and again include cabine
con ol a iables, coun y and yea ixed e ec s in he models. The s an-
da d e o s a e clus e ed by coun y. The esul s a e p esen ed in Table 3.
The i s wo models pain an inconclusi e pic u e o he mode a ing
e ec o cabine ’s elec o al pe o mance. In he ull sample, he signi ican
in e ac ion e ms sugges ha losing cabine s a e mo e likely o adjus
hei posi ion. Howe e , in he es ic ed sample, he coe icien o he
in e ac ion is smalle and lacks s a is ical signi icance. We in e p e his as
sugges i e, bu no conclusi e e idence o he p oposi ion ha cabine s
ha did lose elec o ally a e mo e p one o accommoda i e s a egies in
ligh o he adical igh . Fo he emaining h ee cabine -speci ic mode -
a ing a iables, he analyses p o ide mo e consis en esul s. Looking a
cabine ideology, he insigni ican in e ac ion e ms showcase ha
le -leaning cabine s a e nei he mo e no less likely o amend hei posi-
ion in he ace o RRPs. The di ec ion o he e ec is nega i e, sugges ing
ha le -leaning cabine s shi owa ds mo e p og essi e posi ions com-
pa ed o igh -leaning coali ions. While he di ec ion o he in e ac ion
e ec aligns wi h ou heo e ical in ui ion, he es ima es a e no s a is i-
cally signi ican .
The same is ue o he indings ega ding go e nmen o ma ion
du a ion. Leng hy nego ia ions in ace o RRPs a e no associa ed wi h
mo e o less in ense policy posi ion changes. The coe icien s in models
i e and six a e close o ze o and lack s a is ical p ecision. Finally, we
obse e posi i e es ima es o he in e ac ion be ween RRP p esence and
he mino i y cabine dummy. Mino i y cabine s, which a e mo e elian
on opposi ion pa y suppo , do shi o signi ican ly mo e es ic i e
1320 F. ELLGER AND H. KLÜVER
Table 3. p in luence by cabine cha ac e is ics.
Dependen a iable:
policy change (KF)
(1) (2) (3) (4) (5) (6) (7) (8)
p pa liamen :los Vo es 0.40*
(0.21)
0.21
(0.24)
p pa liamen :le cabine −0.30
(0.24)
−0.49
(0.32)
p pa liamen :long
Fo ma ion
0.11
(0.19)
−0.11
(0.16)
p pa liamen :Mino i y
cabine
0.55**
(0.20)
1.13**
(0.45)
p pa liamen 0.31 0.67*** 0.57** 0.95*** 0.38 0.81*** 0.42* 0.73***
(0.23) (0.19) (0.26) (0.24) (0.33) (0.25) (0.21) (0.16)
los Vo es −0.12
(0.13)
−0.05
(0.16)
le cabine 0.37**
(0.15)
0.42*
(0.21)
long Fo ma ion −0.19
(0.12)
−0.003
(0.15)
Mino i y cab. −0.74***
(0.22)
−1.55***
(0.54)
MWc −0.14 −0.24 −0.16 −0.24 −0.14 −0.23 −0.44** −0.57**
(0.14) (0.15) (0.13) (0.14) (0.14) (0.15) (0.18) (0.21)
no. cab. pa ies −0.01 0.05 0.01 0.07 0.001 0.04 −0.03 −0.02
(0.08) (0.09) (0.08) (0.08) (0.08) (0.09) (0.09) (0.08)
unemploymen −0.02 −0.03 −0.03 −0.02 −0.02 −0.02 −0.02 −0.03
(0.02) (0.02) (0.02) (0.02) (0.02) (0.02) (0.02) (0.02)
Fixed e ec s coun y + Yea coun y + Yea coun y + Yea coun y + Yea coun y + Yea coun y + Yea coun y + Yea coun y + Yea
obse a ions 165 132 165 132 161 128 165 132
sample Full es . Full es . Full es . Full es .
adjus ed 20.33 0.40 0.34 0.42 0.28 0.36 0.35 0.48
no e: *p < 0.1; **p < 0.05; ***p < 0.01.
WEsT EuROPEAN POLiTiCs 1321
posi ions whene e RRPs a e ep esen ed in pa liamen . This sugges s ha
RRPs a e especially in luen ial o go e nmen policies when he coali ion
cabine i sel has no s ong elec o al manda e.
O e all, he empi ical analyses p o ide ample suppo o ou heo e -
ical a gumen . Coali ions pe cei e he p esence o igh -wing challenge s
as a signal o e usal o hei p io posi ions on immig a ion policy. They
immedia ely eac by adop ing a oughe s ance on immig a ion in com-
pa ison o hose go e nmen s wi hou simila exposu e. In addi ion, we
p esen complemen ing e idence on he ole o RRP size and hei in lu-
ence condi ional on cabine cha ac e is ics. In he auxilia y analyses, we
ind ha la ge RRPs do no ha e s onge in luence on he deg ee o
policy change han smalle RRPs in pa liamen . This aligns wi h p io
esea ch showing ha e en small adical igh challenge pa ies can
in luence hei compe i o s once hey gain ep esen a ion. Wi h ega d o
he cabine cha ac e is ics, we ind some e idence ha hose coali ions
unde elec o al a e swi ching owa ds mo e es ic i e policies. In pa ic-
ula , mino i y coali ion cabine s and cabine s ha los o es show accom-
moda i e beha iou in ace o RRPs.
Fu he obus ness and alidi y checks
In he las empi ical sec ion, we summa ise suppo ing e idence on he ela-
ionship be ween adical challenge s and coali ion go e nmen s’ policy change
on immig a ion du ing nego ia ions. We p obe i cabine s wi h and wi hou
exposu e o RRPs a e compa able wi hin ou sample. We conduc wo es s
showing ha bo h g oups do no di e along ele an cha ac e is ics.
Fi s , Figu e 2 isualises he simila i ies o hei p e-elec o al posi ions.
The plo s depic a coali ion cabine ’s immig a ion posi ion ac oss h ee
ime poin s: The pos -elec o al coali ion ag eemen posi ion CAP, he
p e-elec o al expec ed posi ion EP( 0), and he expec ed posi ion a he
p io elec ion EP( -1). C ucially, we obse e no di e ences ac oss bo h
g oups when mo ing om EP( -1) o EP( 0). Only in he coali ion ag ee-
men posi ions ha a e d a ed a e he elec ions, signi ican di e ences
be ween posi ions aken on by cabine s wi h and wi hou RRP exposu e
occu . This pa e n holds o he Kim-Fo ding scale on he uppe pane,
as well as he log scale on he lowe pane o he igu e.
Second, we explo e i he cabine s di e ge along a se o o he ele an
dimensions. We he e o e conduc ed a ange o - es s o check hei bal-
ance on po en ially ele an co a ia es. Table 4 summa ises he esul s.
Fo each a iable, we showcase he a e age alue o cabine s wi h and
wi hou RRP p esence in pa liamen . The balance column hen indica es
i signi ican di e ences ac oss bo h g oups exis acco ding o a wo- ailed
- es (cu o a p < 0.05).
1322 F. ELLGER AND H. KLÜVER
Figu e 2. immig a ion policy posi ions ac oss cabine s.
no e: he igu e shows ends in immig a ion posi ions o cabine s wi h ps in pa liamen in g een
and cabine s wi hou exposu e o ps in ed. We obse e ha he e a e no signi ican di e ences
ac oss bo h g oups o he p e-elec o al eps a he cu en elec ion ( 0) o a he p eceding elec ion
( -1). signi ican di e ences only eme ge o pos -elec o al coali ion ag eemen posi ions.
WEsT EuROPEAN POLiTiCs 1329
K iesi, Hanspe e , Edga G ande, Romain Lacha , Ma in Dolezal, Simon
Bo nschie , and Timo heos F ey (2006). ‘Globaliza ion and he T ans o ma ion
o he Na ional Poli ical Space: Six Eu opean Coun ies Compa ed’, Eu opean
Jou nal o Poli ical Resea ch, 45:6, 921–56.
Laakso, Ma kku, and Rein Taagepe a (1979). ‘E ec i e Numbe o Pa ies a Measu e
wi h Applica ion o Wes Eu ope’, Compa a i e Poli ical S udies, 12:1, 3–27.
La e , Michael (1998). ‘Models o Go e nmen Fo ma ion’, Annual Re iew o
Poli ical Science, 1:1, 1–25.
Lowe, Will, Kenne h Benoi , Sla a Mikhaylo , and Michael La e (2011). ‘Scaling
Policy P e e ences om Coded Poli ical Tex s’, Legisla i e S udies Qua e ly,
36:1, 123–55.
Ma in, Lanny W., and Geo g Vanbe g (2014). ‘Pa ies and Policymaking in
Mul ipa y Go e nmen s: The Legisla i e Median, Minis e ial Au onomy, and
he Coali ion Comp omise’, Ame ican Jou nal o Poli ical Science, 58:4, 979–96.
McGann, An hony J., and He be Ki schel (2005). ‘The Radical Righ in he
Alps: E olu ion o Suppo o he Swiss SVP and Aus ian FPÖ’, Pa y Poli ics,
11:2, 147–71.
Meguid, Bonnie (2005). ‘Compe i ion be ween Unequals: The Role o Mains eam
Pa y S a egy in Niche Pa y Success’, Ame ican Poli ical Science Re iew, 99:3,
347–59.
Meye , Thomas M. (2013). Cons ain s on Pa y Poli ics. Colches e : ECPR P ess.
Mou y, Ca he ine (2011). ‘Coali ion Ag eemen and Pa y Manda e: How Coali ion
Ag eemen s Cons ain he Minis e s’, Pa y Poli ics, 17:3, 385–404.
Mou y, Ca he ine, and A co Timme mans (2013). ‘In e -Pa y Con lic
Managemen in Coali ion Go e nmen s: Analyzing he Role o Coali ion
Ag eemen s in Belgium, Ge many, I aly and The Ne he lands’, Poli ics and
Go e nance, 1:2, 117–31.
Mudde, Cas (1999). ‘The Single-Issue Pa y Thesis: Ex eme Righ Pa ies and he
Immig a ion Issue’, Wes Eu opean Poli ics, 22:3, 182–97.
Mudde, Cas (2007). Populis Radical Righ Pa ies in Eu ope. Camb idge, UK; New
Yo k: Camb idge Uni e si y P ess.
Mülle , Wol gang C., and Kaa e S om (2003). Coali ion Go e nmen s in Wes e n
Eu ope. Ox o d: Ox o d Uni e si y P ess.
Pede sen, Mogens N. (1979). ‘The Dynamics o Eu opean Pa y Sys ems: Changing
Pa e ns o Elec o al Vola ili y’, Eu opean Jou nal o Poli ical Resea ch, 7:1, 1–26.
Pickup, Ma k, and Sa a B. Hobol (2015). ‘The Condi ionali y o he T ade-o
be ween Go e nmen Responsi eness and E ec i eness: The Impac o Mino i y
S a us and Polls in he Canadian House o Commons’, Elec o al S udies, 40,
517–30.
Powell, G. Bingham (2000). Elec ions as Ins umen s o Democ acy: Majo i a ian
and P opo ional Visions. New Ha en: Yale Uni e si y P ess.
Powell, G. Bingham, and Guy D. Whi en (1993). ‘A C oss-Na ional Analysis o
Economic Vo ing: Taking accoun o he Poli ical Con ex ’, Ame ican Jou nal o
Poli ical Science, 37:2, 391.
P osse , Ch is ophe , and Jona han Mellon (2018). ‘The Twiligh o he Polls? A
Re iew o T ends in Polling Accu acy and he Causes o Polling Misses’,
Go e nmen and Opposi ion, 53:4, 757–90.
Roodujin, Ma hijs, S ijn an Kessel, Ca e ina F oio, And ea Pi o, Sa ah de
Lange, Daphne Halikiopoulou, Paul Lewis, Cas Mudde, and Paul Tagga
(2019). ‘The PopuLis : An O e iew o Populis , Fa Righ , Fa Le and
Eu oscep ic Pa ies in Eu ope.’ h p://www.popu-lis .o g.

1330 F. ELLGER AND H. KLÜVER
Rydg en, Jens (2008). ‘Immig a ion Scep ics, Xenophobes o Racis s? Radical
Righ -Wing Vo ing in Six Wes Eu opean Coun ies’, Eu opean Jou nal o
Poli ical Resea ch, 47:6, 737–65.
Saal eld, Thomas (2008). ‘Ins i u ions, Chance and Choices: The Dynamics o
Cabine Su i al in he Pa liamen a y Democ acies o Wes e n Eu ope
(1945-99),’ in Mülle Wol gang C, S øm Kaa e and Be gman To bjö n (eds.).
Cabine s and Coali ion Ba gaining: The Democ a ic Li e Cycle in Wes e n Eu ope.
Ox o d: Ox o d Uni e si y P ess, 327–68.
Schain, Ma in A (2006). ‘The Ex eme-Righ and Immig a ion Policy-Making:
Measu ing Di ec and Indi ec E ec s’, Wes Eu opean Poli ics, 29:2, 270–89.
Sche mann, Ka in, and Lau enz Ennse -Jedenas ik (2014). ‘Explaining
Coali ion-Ba gaining Ou comes: E idence om Aus ia, 2002-2008’, Pa y
Poli ics, 20:5, 791–801.
Scho ield, No man, and Michael La e (1985). ‘Ba gaining Theo y and Po olio
Payo s in Eu opean Coali ion Go e nmen s 1945-83’, B i ish Jou nal o Poli ical
Science, 15:2, 143–64.
Schumache , Gijs, Ca he ine E. de V ies, and Ba ba a Vis (2013). ‘Why Do Pa ies
Change Posi ion? Pa y O ganiza ion and En i onmen al Incen i es’, Jou nal o
Poli ics, 75:2, 464–77.
Some -Topcu, Zeynep (2009). ‘Timely Decisions: The E ec s o pas Na ional
Elec ions on Pa y Policy Change’, Jou nal o Poli ics, 71:1, 238–48.
So oka, S ua N., and Ch is ophe Wlezien (2010). Deg ees o Democ acy: Poli ics,
Public Opinion, and Policy. Camb idge: Camb idge Uni e si y P ess.
Spoon, Jae-Jae, and Heike Klü e (2019). ‘Pa y Con e gence and Vo e Swi ching:
Explaining Mains eam Pa y Decline ac oss Eu ope’, Eu opean Jou nal o
Poli ical Resea ch, 58:4, 1021–42.
S imson, James A., Michael B. Mackuen, and Robe S. E ikson (1995). ‘Dynamic
Rep esen a ion’, Ame ican Poli ical Science Re iew, 89:3, 543–65.
S okes, Donald E. (1963). ‘Spa ial Models o Pa y Compe i ion’, Ame ican Poli ical
Science Re iew, 57:2, 368–77.
S øm, Kaa e, and Wol gang C. Mülle (1999). ‘The Keys o Toge he ness:
Coali ion Ag eemen s in Pa liamen a y Democ acies’, Jou nal o Legisla i e
S udies, 5:3–4, 255–82.
S øm, Kaa e, Wol gang C. Mülle , and To bjö n Be gman (2008). Cabine s and
Coali ion Ba gaining: The Democ ac ic Li e Cycle in Wes e n Eu ope. Ox o d:
Ox o d Uni e si y P ess.
Ta i s, Ma gi (2007). ‘P inciple s. P agma ism: Policy Shi s and Poli ical
Compe i ion’, Ame ican Jou nal o Poli ical Science, 51:1, 151–65.
Ta i s, Ma gi (2008). ‘The Role o Pa ies’ pas Beha iou in Coali ion Fo ma ion’,
Ame ican Poli ical Science Re iew, 102:4, 495–507.
Thomson, Robe (2001). ‘The P og amme o Policy Linkage: The Ful ilmen o
Elec ion Pledges on Socio-Economic Policy in The Ne he lands, 1986-1998’,
Eu opean Jou nal o Poli ical Resea ch, 40:2, 171–97.
Van Spanje, Joos (2010). ‘Con agious Pa ies: An i-Immig a ion Pa ies and Thei
Impac on O he Pa ies’ Immig a ion S ances in Con empo a y Wes e n
Eu ope’, Pa y Poli ics, 16:5, 563–86.
Volkens, And ea We ne K ause, Pola Lehmann, The es Ma hieß, Nicolas Me z,
S en Regel, and Be nha d Weßels (2018). The Mani es o Da a Collec ion.
Mani es o P ojec (MRG/CMP/MARPOR). Ve sion 2018b.
WEsT EuROPEAN POLiTiCs 1331
Wa wick, Paul V., and James N. D uckman (2001). ‘Po olio Salience and he
P opo ionali y o Payo s in Coali ion Go e nmen s’, B i ish Jou nal o Poli ical
Science, 31:4, 649.
Wa wick, Paul V., and James N. D uckman (2006). ‘The Po olio Alloca ion
Pa adox: An In es iga ion in o he Na u e o a Ve y S ong bu Puzzling
Rela ionship’, Eu opean Jou nal o Poli ical Resea ch, 45:4, 635–65.
Williams, Michelle Hale (2006). The Impac o Radical Righ -Wing Pa ies in Wes
Eu opean Democ acies. New Yo k: Palg a e Macmillan US.