Beckmann, Joscha; Schweicke , Raine ; Jahn, Ma in
A icle — Published Ve sion
Poli ical Business Cycles in Va ie ies o Capi alis ic Sys ems
Economics & Poli ics
P o ided in Coope a ion wi h:
John Wiley & Sons
Sugges ed Ci a ion: Beckmann, Joscha; Schweicke , Raine ; Jahn, Ma in (2025) : Poli ical Business
Cycles in Va ie ies o Capi alis ic Sys ems, Economics & Poli ics, ISSN 1468-0343, Wiley, Hoboken, NJ,
Vol. 37, Iss. 2, pp. 653-668,
h ps://doi.o g/10.1111/ecpo.12331
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323860
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/
Economics & Poli ics
ORIGINAL ARTICLE
Poli ical Business Cycles in Va ie ies o Capi alis ic
Sys ems
Joscha Beckmann
1,2
| Raine Schweicke
2
| Ma in Jahn
1
1
Fe nUni e si ä in Hagen, Facul y o Business Adminis a ion and Economics, Chai o Mac oeconomics, Hagen, Ge many |
2
Kiel Ins i u e o he Wo ld
Economy, Kiel, Ge many
Co espondence: Joscha Beckmann ([email p o ec ed])
Recei ed: 10 Janua y 2024 | Re ised: 4 No embe 2024 | Accep ed: 2 Decembe 2024
Keywo ds: expec a ions | poli ical business cycles | a ie ies o capi alism
ABSTRACT
This s udy con ibu es o he li e a u e on poli ical business cycles by assessing he e ec o elec ions on g ow h expec a ions
based on expe su ey da a ins ead o using ac ual pe o mance da a. We analyze he di e en oles oppo unis ic and pa isan
poli ics play in a ie ies o capi alis ic sys ems as a sou ce o he e ogenei y. Ou esul s show ha expec a ions di e ema kably
be ween Libe al and Coo dina ed Ma ke Economies (LME s. CME) e en independen o elec ion ou comes.
JEL Classi ica ion: P20, P34
1 | In oduc ion
The li e a u e on poli ical business cycles (PBCs) p ominen ly
es ablished by Alesina and co‐au ho s ( o a summa y, see, e.g.,
Alesina, Roubini, and Cohen 1997) emphasizes he ole o ex-
pec a ions. Oppo unis ic heo ies sugges ha bo h igh and
le ‐wing go e nmen s implemen expansiona y policies be o e
elec ions o push economic g ow h. Pa isan heo ies sugges
ha , ollowing elec ions, expansiona y policies a e a he o be
expec ed om le ‐wing go e nmen s, while igh ‐wing go -
e nmen s a e mo e likely o implemen aus e i y policies.
Ye , ecen empi ical e idence is a he mixed conce ning pol-
icies ac ually implemen ed (see, e.g., Imbeau, Pé y, and
Lama i 2001 and Po a ke 2017 on iscal spending and
Beckmann e al. (2017) and Belke and Po a ke (2012)on
mone a y policy). We sugges ha his may be due o missing
impo an aspec s o pa isan poli ics and expec a ion
o ma ion.
Wi h espec o pa isan poli ics, i is a gued ha hey ha e
become less d i en by ideological p og amma ic di e ences
(see, e.g., De Simone and Mou ao 2016; Mai e al. 2014) and a
new clea age independen o he le ‐ igh clea age uns
be ween he suppo o open e sus closed socie ies (see, e.g.,
Enyedi 2008). Howe e , adi ional pa isan ideology may s ill
ma e bu di e in al e na i e economic sys ems. The li e a u e
on a ie ies o capi alism (VoC; Hall and Soskice 2001) dis in-
guishes be ween libe al ma ke economies (LMEs) and co-
o dina ed ma ke economies (CMEs), he la e based mo e on
consensus ins ead o ma ke d i en decisions and, hence, wi h
la ge and mo e ac i e go e nmen s. Consequen ly, le ‐ igh
ideology de e mines de ia ions om coun y g oup‐speci ic
pe cep ions abou economic policy in he subg oups.
Indeed, Shel on (2012) is able o show ha he le ‐ igh clea age
has a di e en meaning o PBCs in LMEs such as he Uni ed
S a es and he Uni ed Kingdom compa ed o CMEs such as
Ge many and Sweden. In he same ein, Beckmann e al. (2020)
we e able o show o EU coun ies ha he LME‐CME di ide is
e en mo e elling abou iscal policy s ance compa ed o bo h
le / igh and eas /wes di ides. Hence, we ollow Shel on and
o he s and hypo hesize ha adi ional pa isan ideology may
s ill ma e o PBCs bu ha his is mode a ed by a e age pe -
cep ion o economic policy p e ailing in he subg oups o eco-
nomic sys ems in which he elec ions ake place.
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s 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.
© 2025 The Au ho (s). Economics & Poli ics published by John Wiley & Sons L d.
653 o 837Economics & Poli ics, 2025; 37:653–668
h ps://doi.o g/10.1111/ecpo.12331
Wi h espec o expec a ion o ma ion, empi ical es s had, wi h
ew excep ions, o ely on assump ions abou p e‐elec ion ex-
pec a ions and ac ual pos elec ion ou comes o mac oeconomic
a iables such as income, unemploymen , o in la ion. This
igno es ha expec a ion o ma ion o ma ke pa icipan s plays a
key ole o he p opaga ion o policy shocks and a e a key ea u e
o economic models. Based on he Lucas (1976)c i iquewhich
emphasized he impo ance o expec a ion building o he e a-
lua ion o policy measu es, a ich li e a u e has analyzed he ex-
pec a ion building mechanism o ma ke pa icipan s, o example
wi h ega d o in o ma ion igidi ies o a ional ina en ion
(Coibion and Go odnichenko 2012,2015). Especially wi h espec
o expe opinions, su ey da a ha einc easinglybecomea ailable
and is widely used in se e al pape s, o example in he con ex o
analyzing unce ain y (Beckmann e al. 2025).
Hence, his s udy explici ly analyzes ma ke expec a ions a ound
elec ions by adop ing he da a se om consensus economics ha
co e s a wide ange o ins i u ions and is a ailable a a mon hly
equency and has been used by se e al au ho s (Is e i and
Mouabbi 2018;Lahi iandZhao2019). This allows o es he
o ma ion o expec a ions in di e en samples o OECD coun ies
o see whe he o no we obse e sys ema ic elec ion e ec s on
GDP dependen o independen o policy change.
The es o hiss udyiso ganizedas ollows.InSec ion2,we
in oduce a gumen s om he VoC li e a u e o show how his
may mode a e elec ions expec a ions. Sec ion 3discusses he da a
and he cons uc ion o mon hly GDP expec a ions o 12 mon hs
ahead. Sec ion 4shows he esul s. A obus sys ema ic e ec o
elec ions on GDP expec a ions only appea s in LME and CME
subsamples, e ealing di e en p e‐and pos elec ion pa e ns e en
i di e ing elec ion ou comes a e neglec ed. LME coun ies e eal
ap e‐elec ion boom and a pos elec ion bus , while he e e se is
ue o CME coun ies. Elec ion ou comes mode a e hese gene al
pa e ns o di e en ex en in bo h sys ems. Sec ion 5concludes.
2 | Li e a u e Re iew
Ou s udy ela es o se e al s ands o he li e a u e. In pa ic-
ula , we con ibu e o he li e a u e on PBC, he li e a u e on
VoC
1
, and empi ical esea ch on expec a ion building.
As ou lined in he in oduc ion, oppo unis ic business cycles
and pa isan business cycles can be dis inguished. Fo mal
models o he oppo unis ic business cycle go back o he pio-
nee ing wo k o No dhaus (1975) while he idea ha poli ical
pa ies ha e di e en p e e ences o e in la ion and
unemploymen goes back Hibbs (1977). The li e a u e has
assessed PBCs om a ious pe spec i es (see, e.g., he su eys
p o ided by D azen 2000 and Dubois 2016).
One s and o he empi ical li e a u e, ollowing Pe sson and
Tabellini (2005), has es ablished di e ences in PBC in p esi-
den ial sys ems and pa liamen a y sys ems wi h majo i y o ing
on he one hand and pa liamen a y sys ems wi h p opo ional
o ing on he o he hand (see also Blume e al. 2009). S onge
accoun abili y o a p esiden o a single‐pa y go e nmen is
likely o p oduce s onge oppo unis ic business cycles c ea ing
a boom o expec a ions in he p e‐elec ion pe iod. Howe e ,
empi ical e idence is a he mixed and B ende and D azen
(2005) a gue ha inding oppo unis ic PBC is due o he
inclusion o “new”democ acies, while es ic ing he sample o
coun ies o “old”democ acies shown insigni ican esul s
because expe ienced o e s end o punish p e‐elec o al
manipula ion. In addi ion, Canes‐W one and Pa k (2012)
a gue ha e en a e e se elec o al cycle is possible i p i a e
in es o s pos pone in es men due o p e‐elec ion unce ain y,
which hey assume o inc ease wi h elec o al compe i i eness.
In ou con ex , i is impo an o no e ha majo i y o ing is
ypical o libe al ma ke economies (LMEs) wi h, in con as o
coo dina ed ma ke economies (CMEs), a gene al p e e ence o
compe i i e o ganiza ion o bo h economic and poli ical sys-
ems (Lijpha 1999). Hence, oppo unis ic business cycles, i
p esen in es ablished democ acies a all, should be mo e likely in
LME coun ies wi h possible e e sals mo e likely due o unce -
ain y abou elec ion ou comes.
Empi ical wo k on pa isan PBCs ocusing on expec a ions o
he pos elec ion pe iod depending on elec ion ou comes is
based on he wo k ini ia ed by Alesina and co‐au ho s (see, e.g.,
Alesina, Roubini, and Cohen 1997) and mos ly suppo s he
s anda d expec a ion ha le ‐wing go e nmen s a e expec ed o
ollow expansiona y policies p oducing highe (medium‐ e m)
g ow h, lowe unemploymen , and highe in la ion. Alesina,
Roubini, and Cohen (1997, Ch. 6) did no conside su p ise
e ec s explici ly, ha is, implici ly assuming only changes in
go e nmen as a su p ise a he han each elec ion as c ea ing
unce ain y (e en in he case o no change). Mo e ecen wo k,
o example, by Be lemann and Ma kwa d (2006,2007) and
Shel on (2012), explici ly conside measu es o su p ise e ec s
d i ing he signi icance o pa isan e ec s.
Independen o conside ing explici measu es o su p ise, wo
pape s e eal di e ences be ween coun ies o ganized as LMEs
e sus CME. Based on consensus da a o ecas s, Shel on (2012)
shows ha he expec ed expansiona y e ec in case o a le win is
only p esen in LME coun ies, while a nega i e g ow h e ec
appea s in CME coun ies. He a gues ha ideological o ien a ions
ha e di e en meanings in bo h ypes o capi alis sys ems and,
hence, c ea e di e en expec a ions. Mo e speci ically, in CME “…
conse a i e iscal policy is pe cei ed o be s ic ly supe io …,
deli e ing highe ou pu g ow h…” (Shel on 2012, p. 237).
This may explain a “ ea o a le win”in CME coun ies leading o
e e sed expec a ions abou pa isan p e e ences. In addi ion,
ac ually e e sed p e e ences a e obse ed bo h wi h espec o le
and w igh pa ies. Non adi ional pa isan beha io is wi nessed
by, o example, le ‐wing labo ma ke e o ms and iscal con-
se a ism in Ge many ollowing he Ha z‐ e o ms and expan-
siona y policies by populis igh ‐wing go e nmen s in Hunga y
and Poland ( o a discussion, see, e.g., Beckmann e al. 2020).
2
Howe e , go e nmen ac i i y, ha is, iscal spending and
egula ion, is gene ally highe in CME coun ies, ha is, a le
win in a low‐ac i i y LME en i onmen may well be less ex-
pansiona y compa ed o a igh win in a high‐ac i i y CME
en i onmen . In bo h cases, he sys ems con e ge owa d he
(in e na ional) a e age. This is in line wi h he esul s p o ided
654 o 837 Economics & Poli ics, 2025
by Beckmann e al. (2020) on he di ide be ween LME and CME
coun ies conce ning pa isan and o he d i e s o ealized is-
cal policy in EU coun ies. The di ide be ween LME and CME
subsamples clea ly domina es ideological e ec s indica ing ha
he assumed le ‐ igh clea age conce ning go e nmen ac i i y
has become less elling o e ecen decades. A leas , i is o be
ound wi hin LME and CME subsamples o speci ic cases only.
In addi ion, Po a ke (2012) e eals ha he expansiona y le
win e ec is only signi ican in wo‐pa y sys ems. As a gued
abo e his is cha ac e is ic o LME coun ies wi h majo i y
o ing in ei he p esiden ial o pa liamen a y sys ems, ha is,
wi h single‐pa y go e nmen s. The majo a gumen made by
Po a ke is ha single‐pa y go e nmen s ha e he poli ical
manda e o ac ually implemen policies wi hou comp omising
in coali ion nego ia ions. Taking his a gumen o g an ed, i
also applies o he p e‐elec ion oppo unis ic policy choice.
Incumben single‐pa y go e nmen s a e mo e likely o be held
accoun able o a p e‐elec ion boom in expec a ions. This would
sugges ha bo h p e‐elec ion booms in gene al and pos elec ion
booms in he case o a le win is mo e likely in LME coun ies.
Mos empi ical e idence p oduced so a su e ed om he ac
ha li le is known abou ac ually ealized expec a ions o ma ke
pa icipan s a ound elec ions and ha hypo heses abou ex-
pec a ions ha e been es ed by ac ual pe o mance. As men-
ioned abo e, he excep ion is Shel on (2012) implemen ing
expec a ions de i ed om consensus economics and also dis in-
guishing be ween a ie ies o capi alis sys ems and di e en
ou comes. The desc ip i e pa e n o cycles in ou pu g ow h
o ecas s o he pe iod 9 mon hs be o e and 9 mon hs a e
elec ions (Shel on 2012, Figu e 1) e eals clea di e ences
be ween le and igh wins in LME and CME coun ies. In
addi ion, cyclical pa e n o p e‐and pos elec ion expec a ions in
he exis ing li e a u e ha e sugges ed ha oppo unis ic and
pa isan aspec s ha e o be in e p e ed in an in eg a ed app oach.
Expec a ion building has a ac ed a lo o a en ion ecen ly and he
da a by consensus economics has been widely used ecen ly.
Biljanowska, G igoli, and Hengge (2021)conside policy
unce ain y spillo e s om he Uni ed S a es o Canada and he
Uni ed Kingdom and be ween Eu ope, China, and he Uni ed
S a es espec i ely based in da a om consensus economics and
Beckmann and Geige (2024) assess how unce ain y is ansmi ed
ia expec a ion shocks. Do e n, F i sche, and Scalacek (2012) ely
on Consensus da a o ocus on disag eemen among p o essionals.
3 | Da a and Me hods
3.1 | Subsampling and Da a Cons uc ion
3.1.1 | Subsampling
Because we assume ha bo h sys em and ou come ma e , we
a oid he use o dummy a iables. Ou empi ical in es iga ion
will adop a subsample app oach sepa a ing he OECD wo ld
in o mo e libe al and mo e coo dina ed ma ke economies (LME
s. CME) o an empi ical es o he signi icance o mon hly
de ia ions om end o ecas s. The sample o OECD coun ies is
de e mined by he a ailabili y o o ecas da a by consensus
economics.
3
Fo de ining LME and CME subsamples, we ely on
empi ical e idence p o ided by he clus e analysis based on
mac oeconomic policy and pe o mance a iables (Ahlbo n,
Ah ens, and Schweicke 2016).
4
The subsample o he
coo dina ed‐ma ke economies (CME) consis s o Ge many,
Japan, F ance, I aly, No way, and Sweden. The subsample o
libe al‐ma ke economies (LME) include he Uni ed S a es,
Canada, he Uni ed Kingdom, New Zealand and Swi ze land.
5
The subsample app oach also allows o sub‐samples wi h
espec o he ou come o elec ions bu di e en o o he ap-
p oaches also allows o conside a Scena io o “same di ec ion”
explici ly. This scena io is no mally ea ed as he benchma k
scena io o he analysis o “ igh ”and “le .”Basically, “same
di ec ion”is ea ed implici ly as he no‐su p ise scena io by
Alesina, Roubini, and Cohen (1997). Di e en o he li e a u e
we also allow o a subsample showing he e ec s i he di ec-
ion o ideology is owa d “Cen e ,”which is howe e only
ele an o CME coun ies wi h coali ion go e nmen s.
FIGURE 1 | Seasonal componen s. The e ec s show he decomposi ion assuming an addi i e se ies. [Colo igu e can be iewed a
wileyonlinelib a y.com]
655 o 837
3.1.2 | Expec a ion Fo ma ion
Because we use expe opinions, we assume ha he e is no
sys ema ic su p ise in o ecas s conce ning he ideological o i-
en a ion o incoming go e nmen s. As a consequence,
insigni ican esul s may be due o a iola ion o his assump-
ion. This assump ion is plausible gi en ha o ecas e use all
a ailable in o ma ion in an e icien way. I is also impo an o
keep in mind ha o ecas s a e o en no e ised due o a ional
ina en ion which e lec s he ac ha aqui ing in o ma ion
comes wi h subs an ial cos s o indi iduals. Hence, any e i-
sion in o ecas s a ound elec ions would e lec s ong e idence
ha he in o ma ion in elec ions is conside ed ele an
in o ma ion.
3.1.3 | G ow h Expec a ions
The e is no unique measu e o expec a ions. In a nu shell, su ey
and ma ke ‐based measu es can be dis inguished. Su ey da a
a e widely used o analyze expec a ion building and unce ain y
among ma ke pa icipan s. Da a om consensus economics has
been adop ed by se e al s udies and au ho s o unde s and how
expec a ions a e o med and how hey adjus o new in o ma ion.
The e is moun ain e idence ha expec a ions o en adjus wi h a
delay o new in o ma ion since in o mal ic ions make i di i-
cul o cos ly o p ocess new in o ma ion. Assessing whe he
expec a ion change be o e o a e an elec ion is impo an since
i allows o a di ec es o bo h a ional and PBCs. So a , e i-
dence on expec a ions in he con ex o elec ions is sca ce. Ex-
pec a ion da a on annual GDP g ow h is cons uc ed on he basis
o mon hly da a om Consensus Economics The numbe o
o ecas e s a ies be ween 15 and 40 depending on he a iable
unde in es iga ion. The ac ha he name o con ibu ing
p o essional o ecas e is published inc eases he c edibili y o
o ecas s due o epu a ion e ec s.
6
The o ecas s p o ided by consensus economics a e ixed e en
o ecas s, ha is expec a ions a e p o ided o he cu en and he
nex yea a each poin in ime. gij,deno es he expec ed g ow h
a e in pe iod
j
o pe iod
i
. This implies ha disag eemen abou
he cu en yea na u ally dec eases o e ime, ha is he
unce ain y abou his yea 's in la ion o GDP g ow h is o ex-
ample, much lowe in No embe han in Janua y. We he e o e
adop he app oach sugges ed by Pa on and Timme mann
(2010), which has also been applied by Do e n (2015) o ans-
o m ixed e en in o ixed ho izon o ecas s ia weigh ed a e -
aging. See Knüppel and Vladu (2016) o an al e na i e way o
ans o ming ixed e en in o ixed ho izon o ecas s by choosing
a di e en weigh ing. The in ui i e idea is o use he weigh ed
a e age o ixed e en o ecas s o he cu en and he nex yea
wi h he weigh o he o me (la e ) linea ly dec easing
(inc easing) as ime e ol es based on he ollowing o mula
gwg wg
ˆ=ˆ+(1−)ˆ
,
,−12 1,0 2,1
(1)
whe e
g
ˆ
,−1
2
deno es he app oxima ed ixed ho izon g ow h
a e o ecas while g
ˆ1,
0
and g
ˆ2,
1
gi e he ixed e en o ecas s o
he cu en and he nex yea and
w
deno es he ad hoc weigh
∕ (24 −)1
2
.
The mon hly da a‐se p o ides a o ecas alue o he end o
he espec i eyea ,
x
m
,
, and he o ecas alue o he end o
he nex yea xm ,+1
. Thus, he e a e 24 o ecas alues
a ailable o each yea and wi h each mon h app oaching he
end o he p ojec ion ho izon, he o ecas alues a e ge ing
mo e p ecise and a e less unce ain, which is shown by lesse
disag eemen be ween he o ecas e s alongside a dec easing
p ojec ion ho izon. In addi ion o he dec easing a iabili y
wi hin he o ecas se ies, he noncon inui y o he da a
poses ano he issues as he e is a jump in alues be ween
Decembe o Janua y, whe e he o ecas alue o xm ,+1 u ns
in o
x
m
,
and a new, independen , xm ,+1is en e s he da a‐se .
To ge a ound hese issues we combined he wo o ecas
alues o c ea e a weigh ed a e age be ween o he wo
which esul s in he p ojec ed amoun o o al GDP g ow h in
he nex 12 mon hs, x12m;.
7
Wi h his ans o ma ion owa d
a con inuous 12 mon h ahead o ecas alue, we pa ly
accoun o he he e ogenei y o he da a, as adjus men s o
he o ecas s owa ds he end o he yea , which ha e usually
been smalle , may appea wi h he same likelihood as in
p e ious mon hs.
The x12m‐se ies sha e se e al p ope ies such as a seasonal
componen , con ibu ing posi i ely in he beginning o he
beginning o he second hal o he yea (Figu e 1) and a ange
o GDP g ow h be ween −5% and 5% wi hin he 12‐mon h
pe iod. Wi h an a e age alue o 2.324%, GDP g ow h o ecas
we e he highes in LME coun ies. G ow h‐ o ecas s in LME
coun ies ha e been signi ican ly lowe , 1.797% and mo e
ola ile (Tables A1 and A2). Pe iods o nega i e g ow h can be
mainly a ibu ed o he inancial c isis and o ha e emained
an excep ion apa om Spain and I aly which expe ienced
se e al pe iods o nega i e g ow h du ing he eu o‐c isis a e
he inancial c isis un il 2015.
Wi hin he LME subsample he se ies show s ong c oss‐
coun y co ela ions (Figu e 2). While his is he case o a
subg oup o CME coun ies as well (Ge many, Spain, F ance,
and I aly), his is less he case o he Asian a ian (Japan) and
he No dic a ian (No way, Sweden). Hence, expec a ions
abou business cycles e lec he esul s om clus e analysis on
ins i u ional design ha LME coun ies a e much mo e
homogeneous compa ed o CME coun ies.
3.1.4 | Elec ions
Global da a on elec ions is aken om he Da abase o Poli ical
Ins i u ions (DPI2020) (Sca ascini, C uz, and Kee e 2021). The
da a se co e s a wide a ie y o elec ions, which on he na ional
le el a e di ided in o legisla i e and execu i e elec ions. The da a
is summa ized in Table 1(de ails a e p o ided in Table A3).
Howe e , he conside a ion o one o ano he depends as “ he
main”elec ion a iable depends on he poli ical sys em. The
au ho s mainly di ided hem in o p esiden ial and pa liamen a y
sys ems. Fo he p esiden ial sys ems, he execu i e elec ions a e
o be conside ed and o he pa liamen a y sys ems he legisla i e
elec ions a e o be conside ed. We label hese elec ions as Gen-
e al Elec ions.
8
Since he s a o he panel in 1989 up o 2018,
F ance has held 6 Gene al Elec ions which ep esen s he lowes
alue o he sample as gene al elec ions a e held e e y 5 yea s.
656 o 837 Economics & Poli ics, 2025
Japan wi h 10 Gene al Elec ions, a a equency o he 3 yea s,
ep esen s he Coun y wi h he highes numbe o gene al
elec ions. These Gene al Elec ions can be a ilia ed wi h a change
in go e nmen ideology as well.
3.1.5 | Ideology
The DPI da a se also allows o c ea e subsamples indica ing
whe he he go e nmen s ideology has changed due o an
elec ion and i , o which di ec ion: same, igh , cen e , and le .
This a iable ep esen s he o e all change which does no
include he magni ude o he change, ha is, a change owa ds
he le could o igina e om a p e ious cen is o a igh wing
go e nmen . Howe e , his is only ele an o he CME g oup
because in he wo‐pa y wo ld o LME coun ies he choice is
be ween wha is (sys em‐dependen ) le o igh . Anyhow, in
some cases he pa y ideology could no be de e mined and he
da a is ma ked as NA. No ably, he conside ed sample o
coun ies shows a e y s able pa e n, whe e o he majo i y o
he elec ions e ealed no change in ideology occu ed. This is
an addi ional jus i ica ion o conside his ca ego y explici ly
and no only as a benchma k assuming no su p ise. Reg ession
analysis will ha e o show whe he we de ec in e p e able
cycles in his ca ego y as well.
3.2 | Me hods
A s anda d way o es o he exis ence o a poli ical cycle is o
un an au o eg ession o he a iable i sel , a small se o eco-
nomic a iables, and poli ical dummies (D azen 2000).
FIGURE 2 | C oss‐coun y co ela ion. The igu e shows he co ela ion o he
x12m
‐se ies wi hin he wo subse s. Da a a e aken om Con-
sensus Economics. [Colo igu e can be iewed a wileyonlinelib a y.com]
TABLE 1 | Elec ions by coun y.
Coun y AE GE Righ Le Cen e Same
Aus alia 9 9 2 1 0 6
Canada 8 8 1 2 0 5
F ance 6 6 2 2 1 1
Ge many 8 8 1 1 0 6
I aly 8 8 3 1 1 2
Japan 10 10 0 1 0 8
New Zealand 9 9 1 2 0 6
No way 7 7 2 1 0 4
Spain 8 8 2 1 0 5
Sweden 7 7 2 2 0 3
Swi ze land 7 7 0 1 0 4
Uni ed Kingdom 7 7 1 1 0 5
Uni ed S a es 14 7 2 2 0 3
No e: Desc i i e s a is ics by coun y. Includes all OECD 1990 coun ies elec ion winne in Japan in 2009 has no been classi ied and he e o e no di ec ion has been
assigned. Same o I aly in 2013 and Swi ze land in 2003 and 2015.
657 o 837
In his spi i , ou modeling app oach is based on he ollowing
equa ion:
xx12 = 12 + Elec ion + Op ional Con ols
.
i i
i i ,,−1
=−9
=+9
,,
On he le ‐hand side o he equa ion, we ha e he GDP o ecas
in 12 mon hs (x12i
,) and on he igh ‐hand side he i s lag o
he dependen a iable, he espec i e elec ion a iables as
dummy a iables, eaching om 9 mon hs be o e an elec ion
un il 9 mon hs a e an elec ion. We also p o ide obus ness
check by including he e ec i e eal exchange a e as a ca ch‐all
con ol a iable. I seems o be jus i ied no o conside o he
con ols as in he li e a u e using ac ual policy ou comes, o
which he pa in luenced by expec a ions would ha e o be
de e mined.
Wi h espec o he lag‐s uc u e o he endogenous a iable he
na u e o he ime‐se ies p oblem a hand poses a ade‐o because
he ime se ies could be easily modeled comple ely by including
addi ional lags lea ing no oom o an explana ion by he
hypo hesis discussed abo e. We a gue ha sho ‐ e m expec a ions
a eadequa e o hebaselinemodelbecause hey e lec hedi ec
esponse o elec ion e en s. Al e na i ely, we also conside lags o
one qua e and 1 yea espec i ely in obus ness checks. The
co esponding esul s a e p o ided in Appendix A.
Seasonali y na u ally a ises since 12‐mon h expec a ions a e
s ongly co ela ed wi h he p e ious mon h. Using he lag o
expec a ions o e he nex 12 mon hs on he igh ‐hand side ac-
coun s o he esul ing au oco ela ion which a ises om he
cons uc ion o he da a based on a e ages o g ow h expec a ions
o hisyea andnex yea .Thes ongo e lapp esul sins ong
co ela ions be ween bo h se ies which is p opaga ed by he pa -
e n ha expec a ions also equen ly emain unchanged. Hence,
signi icance o any elec ions dummy cons i u es a s ong esul .
Ou sample pe iod consis s o da a om 1989:01 un il 2019:12
on a mon hly basis.
Table 2p o ides s anda d panel es s o he da a unde
in es iga ion which sugges ha ixed e ec s e lec a mo e
adequa e speci ica ion compa ed o andom e ec s. Howe e ,
ou main app oach is based on a Panel AR me hodology
because his allows us o accoun o au oco ela ion in expe
o ecas s which a ises om he na u e o ixed e en o ecas s.
4 | Resul s
The esul s o ou AR(1) panel based on he P ais‐Wins en
easible gene alized leas squa es (FGLS). a e p esen ed
g aphically in Figu e 3, which shows he poin es ima es o he
elec ion lag dummies om
=−9
o
=+9
highligh ing he
signi ican esul s a , a leas he 10% le el. The co esponding
es ima es a e p o ided in Appendix A.
As a i s impo an esul , he g aph o gene al elec ions, ha
is, no dis inguishing be ween elec ion ou comes, e eals clea ly
dis inc pa e ns o he LME and CME subg oups esul ing in
insigni ican esul s o he OECD sample coun ies as a g oup.
Wi h espec o oppo unis ic cycles, his explains insigni ican
esul s in he li e a u e bu is also he con i ma ion o he LME
g oup—as sugges ed by Pe sson and Tabellini —as well as he
e e sal o oppo unis ic cycle based on unce ain y—as sug-
ges ed by Canes‐W one and Pa k— o he CME g oup. Com-
bining ou indings wi h hese hypo heses, he expe
expec a ions seem o ollow he idea ha LME single‐pa y
go e nmen o p esiden s espec i ely ind i much mo e
a ac i e o c ea e boom expec a ions because o he likelihood
o eap he bene i s in e ms o elec o al success, while his is
much less clea in CME wi h unce ain coali ion o ma ion.
This migh , a he same ime, explain a highe le el o unce -
ain y domina ing he p e‐elec ion phase in CME. The p e‐
elec ion pa o he cycle ends wi h elec ions upcoming wi h
bo h boom expec a ions in LME and bus expec a ions in CME
ending close o end expec a ions. The pos elec ion cycle
s a ing wi h elec ion mon h shows a clea e e sal wi h abo e‐
end expec a ions in CME and below‐ end expec a ions in
LME. S ill looking a all elec ions independen o ou come his
is a he signi ican o he case o LME s a ing in he second
qua e a e elec ion mon h. Taking he cyclical pa e n o
g an ed, his i s nicely wi h he unce ain y hypo hesis in he
case o CME. I unce ain y abou elec ion ou come has mean
lowe expec a ions o 12‐mon h GDP g ow h, his should c e-
a e a posi i e cycle a e unce ain y is elimina ed. A he same
ime, he nega i e phase o he cycle in LMEs can be explained
by ela i e aus e i y (expec a ions) a e he p e‐elec ion boom.
I , o example, iscal o mone a y ha e been ( hough o be)
TABLE 2 | S anda d es s o panel da a.
Me hod S a is ic p‐ alue Al e na i e
F es o indi idual e ec s 1.594 0.086 Signi ican e ec s
Lag ange mul iplie es — ime e ec s (B eusch‐Pagan) o unbalanced
panels
5
, 110.97
4
0 Signi ican e ec s
Lag ange mul iplie es —(Honda) o unbalanced panels 0.434 0.332 Signi ican e ec s
Lag ange mul iplie es — ime e ec s (Honda) o unbalanced panels 71.491 0 Signi ican e ec s
B eusch‐Pagan LM es o c oss‐sec ional dependence in panels
5
, 942.703
0 C oss‐sec ional dependence
Pesa an CD es o c oss‐sec ional dependence in panels 74.455 0 C oss‐sec ional dependence
Hausman es 10.406 0.960 One model is inconsis en
No e: Desc ip i e s a is ics by coun y. Includes all OECD 1990 coun ies.
658 o 837 Economics & Poli ics, 2025
expansiona y mac oeconomic balance o he cycle equi es
some conse a ism. The o he g aphs in Figu e 3help o explain
o which ex end his o e all pic u e is d i en by pa isan e ec s.
As can be seen, he boom expec a ions in LME coun ies in he
p e‐elec ion phase a e o be obse ed in all possible ou come
scena ios and wi hin simila mon hs in he wo p e‐elec ion
qua e s. This u he suppo s he Pe sson‐Tabellini‐hypo hesis
abou p onounced oppo unis ic cycles in sinle‐pa y go e n-
men coun ies. Di e en o ha , in CME he nega i e de ia ion
om end expec a ions be o e elec ion is only signi ican in he
case o no pos elec ion change in go e nmen ideology. As a -
gued abo e, we assume ha he e is no sys ema ic bias in expe
o ecas s o elec ion ou comes. Gi en his assump ion, he
unce ain y a gumen seem o be especially ele an i no clea
change owa ds ex eme ideological posi ions is expec ed o
happen. Gi en ha he numbe o pa ies o ming coali ions
inc eases o mino i y go e nmen s a e common, his seems o
be plausible. Conce ning he pos elec ion phase o he cycle, he
pic u e is di e en o he case o LME coun ies. He e only no
change p oduces a nega i e impac on expec a ions, which
sugges s ha he expec a ion o an aus e i y phase ollowing a
boom phase is especially ele an i he incumben p esiden /
go e nmen s ays in powe . A new incoming go e nmen may
eel less obliged o co ec o expansiona y p e‐elec ion policies
o hose he o me go e nmen could be blamed. The scena io
o same ideological o ien a ion also p oduces he mos p o-
nounced esul o he CME subsample. A e no‐change is
ac ually he esul , simila cons ella ions o coali ion pa ne s
may ag ee on expansiona y packages o close coali ion ag ee-
men s. All in all, he esul s p esen ed in Figu e 3suppo bo h
he hypo hesis abou elec o al sys ems and abou elec ion
unce ain y i mode a ed by he LME/CME‐di ide. The e is also
some suppo o he hypo hesis ha PBCs a e mo e likely and/
o mo e p onounced in an LME en i onmen . Beyond ha ,
howe e , he e is no indica ion ha ei he le o igh wins i
connec ed o a change a e hough o make an expansiona y
scena io mo e likely. The no‐change scena io p oduces he mos
clea ‐cu esul s o bo h LME and CME coun ies. Some
obus ness checks o he esul s including all elec ions a e
p o ided in Tables A4 and A5.TableA4 e eals ha , neglec ing
he lagged endogenous a iable as an explana ion o cu en
expec a ions and including o he e ec i e eal exchange a e
does no change he esul s p esen ed in Figu e 2.InTableA5,
di e en lag s uc u es o he endogenous a iable as an ex-
plana ion o cu en expec a ions a e conside ed wi h a 1‐yea
lag (
M
1
2
) and lags o one qua e and 1 yea (
M
12,3
). While, again
he inclusion o he eal exchange a e does no ha e an e ec on
he esul s. Assuming longe lags o he endogenous a iable
does con i m he gene al esul o con as ing boom and bus
cycles in LME and CME subsamples. Looking a mo e long‐ e m
de ia ions shi s he cycle wi h he pos elec ion cycle al eady
se ing in in he las qua e be o e elec ions. Wi h espec o
expec a ion o ma ion his is plausible gi en ha o ecas s a e
capable o inco po a ing ele an in o ma ion a he quickly.
FIGURE 3 | Panel AR esul s. [Colo igu e can be iewed a wileyonlinelib a y.com]
659 o 837
5 | Conclusion
This s udy has con ibu ed o he li e a u e on PBCs by
assessing he e ec o elec ions on g ow h expec a ions based on
expe su ey da a ins ead o using ac ual pe o mance da a.
Mon hly expe o ecas s ha e been used o cons uc a mon hly
da a se o o ecas s o annual GDP g ow h, ha is, 12 mon hs
ahead o o ecas mon hs. Fo a sample o 13 OECD coun ies,
o which da a we e a ailable o he pe iod om 1989, we ha e
analyzed he p e‐and pos elec ion de ia ions o end o ecas s
om 9 mon hs be o e o 9 mon hs a e elec ions.
Ou esul s based on a Panel AR app oach, ha is, AR(1) panel
model which allows o speci ic au o‐co ela ion as well as
panel he e oskedas ici y and/o con empo aneous co ela ion
based on he P ais‐Wins en easible gene alized leas squa es
(FGLS), shows ha signi ican esul s a e e ealed wi hin sub-
samples o LMEs and CMEs as sugges ed by Shel on (2012).
Howe e , ou esul s on bo h p e‐and pos elec ion phases o
PBCs e eal opposi e signs wi h boom expec a ions be o e and
bus expec a ions a e elec ions in LMEs, while he opposi e is
ue o CMEs. This esul is independen o elec ion ou comes
and suppo s he Pe sson‐Tabellini (2005) hypo hesis on di -
e ences be ween majo i y o ing (as in LME) and p opo ional
o ing (as in CME) explaining he LME cycle as well as he
Canes‐W ong/Pa k (2012) hypo hesis on elec ion unce ain y
d i ing CME oppo unis ic cycles. Hence, oppo unis ic cycles
a e no es ic ed o new democ acies as sugges ed by B ende
and D azen (2005) bu o LMEs.
Pos elec ion booms as assumed in he li e a u e due o le wins
in p esiden ial sys ems (LMEs) and ound by Po a ke (2012)
and due o a igh ‐win ( elie om le ‐win) as ound by Shel on
(2012) o CMEs a e no con i med by ou esul s. P e‐elec ion
booms a e e ealed o LMEs in all cases, ha is, changing
ideology o le , igh , and s aying he same. Fo CMEs, p e‐
elec ion bus and pos elec ion boom a e d i en mainly o he
same di ec ion scena io, in which, a guably, bo h unce ain y
be o e elec ion and elie a e elec ion is highe compa ed o
LMEs. In he scena io o a change o ideology o a le ‐leaning
go e nmen we ind no expansiona y pos elec ion e ec on
expec a ions a all.
O e all, we show ha allowing o he e ogenei y o OECD
coun ies wi h espec o hei ins i u ional design, wi h LMEs
p e e ing compe i i e solu ions o bo h he poli ical and he
economic sys em, is in o ma i e o analyzing hypo hesis aised
in he li e a u e. This is in line wi h indings ha especially
d i e s o iscal policy a e di e en in bo h sys ems wi h he
LME/CME di ide domina ing ideology e ec s. A he same
ime, heo izing on his di ide s ill emains eclec ic and would
bene i om a consis en amewo k o analysis and empi ical
es ing. F om a gene al poin o iew, ou esul s illus a e ha
expec a ion building ake in o accoun poli ical decisions
making and he poli ical en i onmen , illus a ing he need o
clea policy communica ion.
Acknowledgmen s
Open Access unding enabled and o ganized by P ojek DEAL.
Da a A ailabili y S a emen
The da a ha suppo he indings o his s udy a e a ailable on eques
om he co esponding au ho . The da a a e no publicly a ailable due
o p i acy o e hical es ic ions.
Endno es
1
We use he e m Va ie ies o Capi alism because his is well es ab-
lished in he li e a u e ollowing he seminal pape by Hall and
Soskice (2001), which ocused on inno a ion and p oduc ion la gely
neglec ing he ole o go e nmen . Following Esping‐Ande sen (1990)
and o he s, a consis en implemen a ion o economic sys ems implies
ha he di ide be ween LME and CME coun ies is also o be ound
wi h espec o go e nmen ac i i y b oadly de ined. Fo a summa y
and empi ical e alua ion see, o example, Sch öde (2013).
2
Wi h espec o Eas e n Eu opean coun ies, Ta i s and Le ki (2009)
ha e a gued such e e sed pa isan p e e ences o le ‐wing pa ies
in hese coun ies. Howe e , his has been shown o be a empo a y
e ec mos likely due o EU enla gemen condi ionali y (see
Beckmann e al. 2020 o a discussion o e e sed p e e ences).
3
Widening he app oach o non‐OECD coun ies has been im-
plemen ed bu did no show signi ican esul s, nei he wi h egional
no wi h sys em subsampling. Ob iously, he PBC app oach de e-
loped o OECD coun ies is no app op ia e o o he ins i u ional
en i onmen s.
4
Clus e analysis is he wo kho se echnique o empi ical analyses in
he VoC li e a u e because i maximizes he homogenei y o sub-
g oups o coun ies acco ding o a heo e ically es ablished se o
indica o s. Ahlbo n e al. eplica e ea lie clus e esul s based on
mic o indica o s by using agg ega e mac o a iables on ele an
go e nmen policy a iables (o e all go e nmen size, ans e
spending, and egula ion) and pe o mance a iables (inno a i e
capaci y, iscal s abili y, and income equali y) likely o be ele an o
expe ma ke obse e s as well. Hence, his app oach is closes o ou
analysis. Howe e , he e is also a conside able consensus in he VoC
li e a u e wi h espec o he alloca ion o ou adi ional Eu opean
OECD coun ies in o LME and CME subsamples.
5
Di e en o Shel on (2012), we do no conside so‐called Medi e a-
nean ma ke economies (MME). As shown by clus e analysis his is a
subca ego y o CME sys ems. In he same ein, allowing o sub‐
clus e s, Japan ep esen s an Asian model and No way and Sweden
belong o a No dic Clus e (see, e.g., Sch öde 2013). In gene al, his
e eals ha he CME clus e is signi ican ly mo e he e ogenous
compa ed o he LME clus e . This is also alid o di e ences in
iscal policy (see Beckmann e al. 2020 on his poin ).
6
See h p://www.consensuseconomics.com/ o u he de ails.
7
The ollowing o mula is applied: xαxαx12 = × + (1 −)×
mm m ,,+
1
);
whe e
∕
α
m=(13−)1
2
.
8
We also conside ed a a iable o Any Elec ions. Howe e , he only
coun y in which legisla i e and execu i e elec ions a e held inde-
penden ly a e he Uni ed S a es. In con as o he emainde o he
sample, legisla i e elec ions in he Uni ed S a es a e held indepen-
den ly o m he execu i e ones esul ing in a highe numbe o any
elec ions. In o he cases, such as F ance, bo h elec ions a e held a he
same ime. In addi ion, using Any Elec ions did no p oduce di e en
esul s.
Re e ences
Ahlbo n, M., J. Ah ens, and R. Schweicke . 2016. “La ge‐Scale T an-
si ion o Economic Sys ems—Do CEECs Con e ge Towa d Wes e n
P o o ypes?”Compa a i e Economic S udies 58, no. 3: 430–454. h ps://
doi.o g/10.1057/s41294-016-0009-x.
Alesina, A., N. Roubini, and G. D. Cohen. 1997. Poli ical Cycles and he
Mac oeconomy. Camb idge: MIT P ess.
660 o 837 Economics & Poli ics, 2025
TABLE A5 | Robus ness baseline esul s.
OEC LME CME OEC LME CME OEC LME CME OEC LME CME
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
lag(GE, −9:9)‐9−0.026 −0.014 −0.038 −0.020 0.002 −0.040 −0.026 −0.014 −0.038 −0.020 0.001 −0.040
(0.029) (0.022) (0.053) (0.056) (0.026) (0.105) (0.030) (0.022) (0.053) (0.057) (0.026) (0.105)
lag(GE, −9:9)‐8−0.046 −0.062 −0.032 −0.066 −0.050 −0.081 −0.047 −0.062 −0.033 −0.067 −0.050 −0.082
(0.029) (0.045) (0.035) (0.061) (0.054) (0.106) (0.029) (0.045) (0.035) (0.061) (0.054) (0.106)
lag(GE, −9:9)‐7 0.010 0.041*−0.020 −0.060 −0.023 −0.095 0.010 0.041*−0.020 −0.061 −0.023 −0.096
(0.017) (0.024) (0.018) (0.056) (0.066) (0.087) (0.017) (0.024) (0.018) (0.056) (0.066) (0.087)
lag(GE, −9:9)‐6−0.006 0.039*** −0.049*−0.041 0.029 −0.106*−0.006 0.039*** −0.049*−0.042 0.029 −0.107
(0.019) (0.012) (0.026) (0.050) (0.066) (0.064) (0.019) (0.012) (0.026) (0.050) (0.065) (0.065)
lag(GE, −9:9)‐5 0.013 0.040 −0.011 0.018 0.122** −0.077 0.013 0.040 −0.011 0.018 0.122** −0.077
(0.021) (0.026) (0.029) (0.049) (0.053) (0.060) (0.021) (0.025) (0.030) (0.049) (0.052) (0.061)
lag(GE, −9:9)‐4 0.021 0.015 0.026 0.028 0.100*−0.035 0.021 0.015 0.026 0.028 0.100*−0.035
(0.019) (0.030) (0.023) (0.045) (0.055) (0.060) (0.019) (0.030) (0.023) (0.045) (0.055) (0.061)
lag(GE, −9:9)‐3 0.021 −0.019 0.059 0.054 0.045 0.067 0.021 −0.019 0.058 0.054 0.046 0.066
(0.023) (0.014) (0.036) (0.041) (0.051) (0.058) (0.023) (0.014) (0.036) (0.041) (0.051) (0.058)
lag(GE, −9:9)‐2 0.029 −0.008*0.065*0.071 0.003 0.141*0.029 −0.008 0.065*0.072 0.003 0.143*
(0.021) (0.005) (0.033) (0.051) (0.040) (0.078) (0.021) (0.005) (0.033) (0.052) (0.040) (0.079)
lag(GE, −9:9)‐1 0.008 −0.064** 0.079*0.058 −0.073*0.192*0.009 −0.063** 0.080*0.059 −0.073*0.195*
(0.033) (0.029) (0.042) (0.069) (0.038) (0.101) (0.033) (0.029) (0.041) (0.069) (0.038) (0.101)
lag(GE, −9:9)0 −0.014 −0.054** 0.024 0.024 −0.108** 0.158*−0.014 −0.054** 0.025 0.026 −0.108** 0.161*
(0.022) (0.024) (0.030) (0.065) (0.053) (0.086) (0.022) (0.024) (0.030) (0.066) (0.054) (0.086)
lag(GE, −9:9)1 −0.008 −0.056** 0.037 −0.009 −0.154** 0.136 −0.008 −0.056** 0.038 −0.008 −0.154** 0.139
(0.028) (0.025) (0.042) (0.074) (0.073) (0.095) (0.028) (0.025) (0.042) (0.075) (0.074) (0.095)
lag(GE, −9:9)2 −0.022 −0.053** 0.008 −0.039 −0.147** 0.069 −0.021 −0.052** 0.009 −0.036 −0.146** 0.073
(0.020) (0.025) (0.026) (0.064) (0.072) (0.085) (0.020) (0.026) (0.026) (0.064) (0.072) (0.084)
lag(GE, −9:9)3 −0.046*−0.076*** −0.019 −0.071 −0.173*** 0.029 −0.045*−0.075*** −0.018 −0.069 −0.171*** 0.032
(0.025) (0.023) (0.041) (0.065) (0.063) (0.094) (0.025) (0.023) (0.041) (0.065) (0.063) (0.094)
lag(GE, −9:9)4 0.010 −0.026 0.045 −0.051 −0.145** 0.040 0.011 −0.026 0.046 −0.049 −0.144** 0.043
(0.027) (0.020) (0.045) (0.063) (0.060) (0.096) (0.027) (0.021) (0.044) (0.063) (0.061) (0.095)
lag(GE, −9:9)5 0.001 −0.007 0.008 −0.027 −0.102** 0.042 0.001 −0.007 0.008 −0.026 −0.102** 0.045
(Con inues)
667 o 837
TABLE A5 | (Con inued)
OEC LME CME OEC LME CME OEC LME CME OEC LME CME
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
(0.017) (0.023) (0.024) (0.055) (0.046) (0.087) (0.017) (0.024) (0.024) (0.056) (0.046) (0.086)
lag(GE, −9:9)6 0.011 0.018 0.002 0.025 −0.016 0.061 0.011 0.018 0.003 0.026 −0.016 0.063
(0.021) (0.014) (0.039) (0.052) (0.035) (0.091) (0.021) (0.015) (0.038) (0.051) (0.035) (0.090)
lag(GE, −9:9)7 −0.003 −0.016 0.008 0.013 −0.006 0.027 −0.002 −0.016 0.009 0.015 −0.006 0.030
(0.021) (0.017) (0.035) (0.055) (0.038) (0.099) (0.021) (0.017) (0.035) (0.055) (0.039) (0.099)
lag(GE, −9:9)8 0.008 0.024*** −0.009 0.016 0.022 0.005 0.009 0.024*** −0.008 0.018 0.022 0.008
(0.015) (0.007) (0.026) (0.054) (0.029) (0.099) (0.015) (0.007) (0.026) (0.054) (0.030) (0.099)
lag(GE, −9:9)9 −0.025 0.005 −0.055*−0.016 0.011 −0.047 −0.025 0.006 −0.054*−0.015 0.011 −0.043
(0.019) (0.005) (0.031) (0.047) (0.024) (0.084) (0.019) (0.005) (0.031) (0.047) (0.024) (0.084)
lag(EREER_IX) −0.001 −0.001*−0.001 −0.002*−0.001 −0.003
(0.0004) (0.0003) (0.001) (0.001) (0.001) (0.002)
lag(M_12) 0.979*** 0.975*** 0.982*** 0.979*** 0.975*** 0.981***
(0.003) (0.003) (0.003) (0.003) (0.004) (0.003)
lag(M_12, 3) 0.885*** 0.858*** 0.899*** 0.884*** 0.858*** 0.897***
(0.015) (0.012) (0.018) (0.014) (0.013) (0.017)
Obse a ions 4211 1936 2275 4211 1936 2275 4211 1936 2275 4211 1936 2275
R
2
0.965 0.953 0.971 0.797 0.745 0.826 0.965 0.953 0.971 0.797 0.745 0.827
Adjus ed R
2
0.965 0.952 0.971 0.796 0.741 0.824 0.965 0.953 0.971 0.796 0.741 0.824
No e: This able p o ides he obus baseline esul s wi h an al e na i e lag s uc u e.
***p< 0.01; **p< 0.05; *p< 0.1.
668 o 837 Economics & Poli ics, 2025
