The impact of economic policy uncertainty on stock types while considering the economic cycle: A quantile regression approach

Paule-Vianez, Jessica; O den-C uz, Ca men; P ado-Román, Camilo; Ma ínez, Raúl
Gómez
A icle
The impac o economic policy unce ain y on s ock
ypes while conside ing he economic cycle: A quan ile
eg ession app oach
Eu opean Jou nal o Managemen and Business Economics (EJM&BE)
P o ided in Coope a ion wi h:
Eu opean Academy o Managemen and Business Economics (AEDEM), Vigo (Pon e ed a)
Sugges ed Ci a ion: Paule-Vianez, Jessica; O den-C uz, Ca men; P ado-Román, Camilo; Ma ínez, Raúl
Gómez (2025) : The impac o economic policy unce ain y on s ock ypes while conside ing he
economic cycle: A quan ile eg ession app oach, Eu opean Jou nal o Managemen and Business
Economics (EJM&BE), ISSN 2444-8451, Eme ald, Leeds, Vol. 34, Iss. 1, pp. 88-102,
h ps://doi.o g/10.1108/EJMBE-12-2022-0365
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/325587
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
The impac o economic policy
unce ain y on s ock ypes while
conside ing he economic cycle.
A quan ile eg ession app oach
Jessica Paule-Vianez, Ca men O den-C uz, Camilo P ado-Rom
an and
Ra
ul G
omez-Ma 
ınez
Depa men o Business Economics, Rey Juan Ca los Uni e si y–Mad id Campus,
Mad id, Spain
Abs ac
Pu pose –This s udy aims o analyse he e ec s o Economic Policy Unce ain y (EPU) on he e u n o
g ow h/ alue and small/la ge-cap s ocks du ing expansiona y and ecessiona y pe iods ac oss a condi ional
dis ibu ion.
Design/me hodology/app oach –The au ho s selec ed a sample co e ing he pe iod be ween 01/1995–05/
2021. Quan ile eg essions we e applied o he EPU and Russell indices. Business cycles we e es ablished
ollowing he NBER.
Findings –The esul s show ha EPU has a nega i e e ec on s ocks wi h he in ensi y o he e ec
depending on he s ock’s p o ile. Small-cap and g ow h s ocks we e ound o be mos sensi i e o EPU,
especially du ing ecessions. The nega i e e ec is mode a ed by he economic cycle bu is p og essi ely
dilu ed a he lowe ail o he s ock e u n dis ibu ion.
P ac ical implica ions –The indings shed mo e ligh on in es men s a egies o g ow h/ alue in es o s
ha pu sue oppo uni ies a ising om a changing economic cycle.
O iginali y/ alue –This s udy makes he ollowing con ibu ions: (1) explo es he impac o EPU on he
e u n o di e en s ocks ac oss a condi ional dis ibu ion, and (2) p o ides e idence on how he economic cycle
in luences EPU impac on g ow h/ alue s ocks and small/la ge s ocks.
Keywo ds Economic policy unce ain y, S ock ma ke e u ns, Limi ed a bi age, Economic cycles,
Beha iou al inance
Pape ype Resea ch pape
In oduc ion
Following Keynes’(1937) sugges ion ha unce ain y is a undamen al elemen in he
economy, he li e a u e has ocused pa icula ly on he s udy o Economic Policy
Unce ain y (EPU) and i s consequences. This in e es inc eased wi h he unce ain ies
de i ed om he global inancial c isis (2007–2009) ha con ibu ed o a sha p economic
decline, as well as o i s subsequen slow eco e y (In e na ional Mone a y Fund, 2013;
Bake e al., 2016).
EJMBE
34,1
88
© Jessica Paule-Vianez, Ca men O den-C uz, Camilo P ado-Rom
an and Ra
ul G
omez-Ma 
ınez. Published
in Eu opean Jou nal o Managemen and Business Economics. Published by Eme ald Publishing
Limi ed. This a icle is published unde he C ea i e Commons A ibu ion (CC BY 4.0) licence. Anyone
may ep oduce, dis ibu e, ansla e and c ea e de i a i e wo ks o his a icle ( o bo h comme cial and
non-comme cial pu poses), subjec o ull a ibu ion o he o iginal publica ion and au ho s. The ull
e ms o his licence may be seen a h p://c ea i ecommons.o g/licences/by/4.0/legalcode
Funding: This esea ch did no ecei e any speci ic g an om unding agencies in he public,
comme cial, o no - o -p o i sec o s.
Da a a ailabili y s a emen : Da a a ailable on eques om he au ho s
The cu en issue and ull ex a chi e o his jou nal is a ailable on Eme ald Insigh a :
h ps://www.eme ald.com/insigh /2444-8494.h m
Recei ed 6 Decembe 2022
Re ised 26 May 2023
17 July 2023
27 July 2023
9 Augus 2023
Accep ed 11 Augus 2023
Eu opean Jou nal o Managemen
and Business Economics
Vol. 34 No. 1, 2025
pp. 88-102
Eme ald Publishing Limi ed
e-ISSN: 2444-8494
p-ISSN: 2444-8451
DOI 10.1108/EJMBE-12-2022-0365
Changes in exis ing economic policies, o e en he speed o changes ag eed in hese
policies, can in luence in es o s, gene a ing a sen imen o insecu i y in hei expec a ions o
judgemen s abou he alue o asse s (Alc
aza -Blanco e al., 2021). The mos ecen li e a u e
con i ms ha EPU in luences he asse p icing o many ma ke s, including s ocks
(An onakakis e al., 2013;A ou i e al., 2016), bonds (Ioannidis and Ka, 2021;Pham and
Nguyen, 2022) and c yp ocu encies (Cheng and Yen, 2020;Paule-Vianez e al., 2020), among
o he s.
The impac o EPU on s ock ma ke e u ns has been well esea ched, wi h esul s
showing s ong e idence o a nega i e in luence (Bake e al., 2016;A ou i e al., 2016).
Inc eased EPU ampli ies beha iou al biases, wi h mixed esul s being ound on he impac
o EPU on indi idual s ocks (Hu e al., 2018;Luo and Zhang, 2020). Analysing in es o
sen imen , some au ho s show a s onge impac o small-cap and g ow h s ocks
(Lakonishok e al., 1994;Bake and Wu gle , 2006;Waggle and Ag awal, 2015;Wu e al.,
2014;Smales, 2017), while o he s de end he iew ha he impac is s onge o alue
s ocks (Kuma and Lee, 2006;Ba hia and B edin, 2013). Unde unce ain y, Hu e al. (2018)
e eal ha small and g ow h s ocks in China’sA-sha ema ke a emo esensi i e oUS
EPU shocks.
The economic cycle plays a key ole in he impac o EPU on in es o con idence le els
(Ahmad and Sha ma, 2018;Adjei e al., 2022). Policymake s expe ience mo e p essu e o
s imula e he economy du ing economic down u ns, and in es o s a e mo e sensi i e o hei
doub s (Adjei and Adjei, 2017). This is u he e idenced in he in luence o EPU in alue
p emium (B e schge and Lech hale , 2018;Ki by, 2019).
Howe e ,EPUimpac isno hesamein helowe and heuppe quan ileso s ock
e u ns (Kannadhasan and Das, 2020), no in he nonlinea p edic abili y US equi y
p emium models (Beki os e al., 2016). Fo example, Raza e al. (2018) showed ha he
ela ionship be ween equi y p emium and EPU is especially nega i e in he ex eme low
and high ails.
Conside ing ha he impac o EPU on s ock ma ke e u ns does no ha e o be uni o m
and gi en he p eceden s o he ampli ica ion o he beha iou al biases unde unce ain y,
his esea ch goes deepe in o conside ing ela ionships ha p io li e a u e had esea ched
independen ly. The objec i es o his s udy a e he e o e: (1) o explo e whe he he EPU
impac on di e en ypes o s ocks di e s depending on he mos o leas p o i able s ocks,
and (2) o examine how he economic cycle mode a es he in luence o EPU on s ock e u n
conside ing he di e en ypes o s ocks.
Ad ancing on he app oaches unde aken in p e ious li e a u e, we p opose he use o
quan ile eg ession o e alua e EPU impac , as his me hodology o e s a mo e
comp ehensi e dependence s uc u e o he analysis o s ock e u ns unde di e se
ma ke condi ions (Beki os e al., 2016;Kannadhasan and Das, 2020;Jiang e al., 2022).
Linea models based on he condi ional-mean a e insu icien o explain he en i e
condi ional dis ibu ion o he alue p emium (Beki os and Gup a, 2015). We apply
O dina y Leas Squa es (OLS) wi h he e oskedas ici y co ec ion o analyse he obus ness
o he esul s.
This s udy makes a comp ehensi e con ibu ion o he EPU li e a u e ela ed o alue/
g ow h and small/la ge cap s ock e u ns, explo ing he ole he economic cycle plays in his
impac and he e ec unde di e en condi ions o he s ock ma ke . This new e idence o e s
mo e insigh o in es men s s a egies o g ow h/ alue in es o s ha pu sue oppo uni ies
a ising om he changing economic cycle.
The pape is s uc u ed as ollows: Sec ion 2 add esses he heo e ical amewo k in ol ed
in he s udy, and Sec ion 3 explains he da a and a iables used in he s udy. Sec ion 4
explains he me hodology, and in Sec ion 5, he esul s a e p esen ed and discussed. Finally,
Sec ion 6 shows he conclusions.
EPU, s ock
ypes and
economic cycle
89
Theo e ical amewo k
In o ma ion unce ain y is behind se e al indings ha con adic he heo y o equilib ium
in inancial ma ke s (Jiang e al., 2005). Se e al au ho s ha e documen ed how unce ain y
ela ed o social, poli ical o economic condi ions has a conside able in luence on in es o
sen imen (Beugelsdijk and F ijns, 2010;Kuma e al., 2012). B own and Cli (2005) ound ha
in es o sen imen is due o pe sis en and unin o med demand shocks, which leads o a poo
alua ion o p ices in he p esence o limi s o a bi age.
In his pape , we use EPU, unde s anding i as he “non-ze o p obabili y o changes in
exis ing economic policies”(Bake e al., 2016). Unce ain y can inc ease when hose
esponsible o economic policies ail o each imely ag eemen s o change policies
equen ly (Li e al., 2015). E en media specula ions can in luence unce ain y (Adjei and
Adjei, 2017). Since he incep ion o EPU, he e is conside able e idence o i s nega i e impac
on he s ock e u ns using di e en me hodologies and s ock ma ke s (An onakakis e al.,
2013;Kang and Ra i, 2013; be ween o he s).
S udies ocused on he impac o in es o sen imen ha e shown ha s ocks a e a ec ed
by beha iou al biases, especially in s ocks ha a e mo e subjec i e o alue o a e aced by
limi s o a bi age, such as small-cap s ocks (Smales, 2017). In his sense, beha iou al
indica o s such sen imen ha e a g ea explana o y powe (Lemmon and Po niaguina, 2006)
o con idence index ha has a high p edic i e capaci y o small-cap e u ns (Kuma and Lee,
2006). In addi ion, g ow h s ocks o e p ice wi h in es o o e eac ions (Lakonishok e al.,
1994), especially wi h bullish sen imen (Waggle and Ag awal, 2015), making hem mo e
p one o bubbles (Bake and Wu gle , 2006). Ex eme pessimism also a ec s g ow h s ocks
much mo e han alue s ocks (Wu e al., 2014). The e o e, g ow h s ocks a e mo e sensi i e o
changes in in es o con idence.
Though i is con i med ha he impac o in es o sen imen is especially nega i e on
small and g ow h s ocks, he e ec is no so clea unde unce ain y. S udies i s showed
small-cap and alue s ocks as he mos a ec ed nega i ely by EPU (Abou a and A isoy,
2017) and mos p edic able in bea ish ma ke s (Chen e al., 2018). In con as , ecen esea ch
shows small-cap and g ow h s ocks o be mos a ec ed by EPU (Hu e al., 2018;Luo and
Zhang, 2020). These s udies highligh he need o mo e esea ch on he di e en ial ac o
behind hese mixed esul s, leading us o he ollowing hypo hesis:
H1. EPU has a g ea e nega i e impac on g ow h and small-cap s ocks.
The ela ionship be ween s ock e u ns and EPU is no linea , being s onge and mo e
pe sis en du ing pe iods o ex eme ola ili y (A ou i e al., 2016). The quan ile eg ession
me hodology allows us o show di e en asymme ic e ec s. This app oach signi ican ly
enhances ou -o -sample s ock e u n p edic abili y, especially when he ma ke is neu al
(Beki os e al., 2016). Raza e al. (2018) demons a e ha he ela ionship be ween equi y
p emium and EPU is especially nega i e in he ex eme low and high ails. Gi en he p io
li e a u e, we es mo e e idence o he nonlinea ela ionship be ween unce ain y and
indi idual s ock e u ns, p esen ing he ollowing hypo hesis:
H2. EPU has a g ea e impac on he e u ns o s ocks ha a e a he ex emes o he
dis ibu ion.
The economic cycle is o g ea impo ance in he inancial ma ke s as an elemen o sys ema ic
isk in luencing s ock e u ns (Fama and F ench, 1989). Gene ally, he e a e mo e economic
policy adjus men s du ing pe iods o ecession, and in es o s espond mo e o hese changes.
When he economy con ac s, in es o s expec go e nmen al bodies o ake g ea e measu es
han du ing pe iods o expansion. Thespeed a which policies a e implemen ed also in luences
in es men isk pe cep ion (Pas o and Ve onesi, 2013). Thus, he e is a high co ela ion
be ween EPU and he economic cycle (Bake e al., 2016;Adjei and Adjei, 2017).
EJMBE
34,1
90
E idence shows ha a s ock’s p o ile is an impo an elemen o de e mining he impac
o he economic cycle on s ock e u ns. In ega d o alue and g ow h s ocks, Fama and
F ench (1992) led a b oad li e a u e showing how he alue p emium is s a is ically
associa ed wi h mac oeconomic undamen als (Kelly, 2003;A e z e al., 2010; among o he s).
Mo e ecen li e a u e eassesses ha he alue p emium is ela ed o cu en and expec ed
economic g ow h (Lee and Kim, 2017;B e schge and Lech hale , 2018), expec ed business
condi ions (Ki by, 2019) and u u e g ow h consump ion (Roh e al., 2019). The ole o in es o
sen imen is mo e signi ican in he pe iod p eceding he subp ime c isis and du ing he c isis,
ou pe o ming alue s ocks compa ed wi h g ow h s ocks (Ne es e al., 2021). This e idence
could explain how policy make s’in e en ion in imes o ecession, while ying o p e en a
dep ession, makes small-g ow h s ocks ou pe o m small- alue s ocks (Bianchi, 2020).
The size p emium is also demons a ed by an ex ensi e li e a u e (C ain, 2011) e ealing
ha small-caps on a e age ou pe o m la ge-caps o e ime, especially du ing expansions
(Kim and Bu nie, 2002) and a e an economic ough (Swi ze , 2010). This e ec e lec s
he i m’s exposu e o undamen al a iables, bu Van Dijk (2011) also indica ed he
non- a ionali y o alua ion models.
Since in es o sen imen con ibu es o size p emium (Qadan and Aha on, 2019;Song,
2023), EPU impac is especially nega i e o small-cap s ocks (Killins e al., 2022), and gi en
ha size p emium is ela ed o he unce ain y wi h mac oeconomic p oduc ion and
agg ega e consump ion (Scheu le and Sp emann, 2010), we es he ollowing hypo hesis:
H3. EPU has a g ea e nega i e impac on s ock e u ns in imes o ecession, especially
o g ow h and small-cap s ocks.
On he o he hand, as no ed by Beki os e al. (2016), business cycle luc ua ions can cause
di e en EPU impac s on s ock e u ns ac oss quan iles. This may be because his
unce ain y could implici ly inco po a e in o ma ion o some pa s o he e u n dis ibu ion.
This issue has been in es iga ed o he e u n o momen um s a egies bu no o s ocks in
gene al o he di e en s ock ypes. Paule-Vianez e al. (2021) ound ha he momen um e ec
is educed in he p esence o inc eases in EPU, especially in imes o ecession and in he lowe
quan iles o he dis ibu ion. Howe e , in pe iods o expansion, EPU has a posi i e impac on
he uppe quan iles. Conside ing how in es o sen imen impac s s ock e u ns, Bake and
Wu gle (2006) show how he mode a ing e ec o he business cycle g adually loses i s e ec
as he s ock ma ke becomes mo e bea ish. Conside ing he abo e, one would expec EPU o
ha e a g ea e nega i e impac on lowes s ock e u ns in imes o ecession. In con as , EPU
impac should be smalle in imes o expansion and o he bes p o i able s ocks. The e o e,
he las hypo hesis we p opose o es in his s udy is:
H4. The economic cycle mode a es he in luence o EPU on he bes p o i able s ocks.
Da a
In his s udy, we selec ed a sample co e ing he pe iod om Janua y 1995 o May 2021 wi h
mon hly da a.
As a e e ence o EPU, we selec ed he US Mon hly EPU index o Bake e al. (2016) (see
h p://www.policyunce ain y.com/). This index is based on he equency wi h which a icles
in newspape s e e o wo ds such as “economy”o “economic”,“unce ain”o “unce ain y”,
“de ici ”,“Fede al Rese e”,“legisla ion”and “ egula ion”.
To s udy he e ec o unce ain y on in es men e u ns o alue and g ow h s ocks o
highe and lowe capi alisa ion, we selec ed he ollowing indices: Russell 1,000 Value, Russell
1,000 G ow h, Russell 2,000 Value and Russell 2,000 G ow h. Russell 1,000 Value ep esen s
la ge-cap alue s ocks, Russell 1,000 G ow h ep esen s la ge-cap g ow h s ocks, Russell
EPU, s ock
ypes and
economic cycle
91

2,000 Value ep esen s small-cap alue s ocks and Russell 2,000 G ow h ep esen s small-cap
g ow h s ocks.
Following he p oposed objec i es, we also dis inguish he impac o EPU on s ock e u ns
be ween pe iods o ecession and expansion (Table 1).
Economic a iables associa ed wi h s ock ma ke s and poli ical unce ain y we e included
as con ol a iables. The a iables selec ed include US in la ion (A ou i e al., 2016;Chen e al.,
2018), he Indus ial P oduc ion Index (A ou i e al., 2016;Chen e al., 2018), he e m sp ead
be ween he yield o ma u i y o a 10-yea T easu y no e and he 3-mon h T easu y bill
(B ogaa d and De zel, 2015;Adjei and Adjei, 2017), he de aul sp ead be ween yields o BAA-
a ed bonds and AAA- a ed bonds (B ogaa d and De zel, 2015;A ou i e al., 2016;Adjei and
Adjei, 2017) and he US g oss domes ic p oduc index (GDP) (Ku o and S an, 2018).
Table 2 shows he a ge s udy a iables, hei de ini ion and he sou ces om which hey
we e ex ac ed.
Me hodology
To es he p oposed hypo heses, we use quan ile eg ession. This is an ex ensi e o m based
on adi ional eg ession and can b oadly depic a condi ional dis ibu ion (Lee and Chen,
2021). We use quan ile eg ession o e alua e EPU impac , as his app oach o e s a mo e
comp ehensi e dependence s uc u e o he analysis o s ock e u ns unde di e se ma ke
condi ions (Beki os e al., 2016;Kannadhasan and Das, 2020;Jiang e al., 2022). Addi ionally,
Pe iod Economic cycle
Janua y 1995 o Ma ch 2001 Expansion
Ap il 2001 o No embe 2001 Recession
Decembe 2001 o Decembe 2007 Expansion
Janua y 2008 o June 2009 Recession
July 2009 o Feb ua y 2020 Expansion
Ma ch 2020 o Ap il 2020 Recession
May 2020 o May 2021 Expansion
Sou ce(s): Na ional Bu eau o Economic Resea ch (NBER)
Va iable De ini ion Sou ce
RLa ge Value Re u n o Russell 1000 Value index Da as eam da abase
RLa ge−G ow h Re u n o Russell 1000 G ow h index
RSmall−Value Re u n o Russell 2000 Value index
RSmall−G ow h Re u n o Russell 2000 G ow h index
EPU Va ia ion a e o US Economic Policy Unce ain y index Bake e al. (2016)
Cycle Recession (1) o Expansion (2) Na ional Bu eau o Economic
Resea ch (NBER)
De aul _
sp ead
De aul sp ead be ween yields o BAAs- a ed bonds and
AAA- a ed bonds
Fede al Rese e Economic
Da a (FRED) Da abase
In la ion Va ia ion a e o he US Consume P ice index
IPI Va ia ion a e o he Indus ial P oduc ion index
Te m_
sp ead
Te m sp ead be ween he yield o ma u i y o a 10-yea
T easu y no e and he h ee-mon h T easu y bill
GDP Va ia ion a e o US no malised G oss Domes ic P oduc
Sou ce(s): Own elabo a ion
Table 1.
Pe iods o ecession
and expansion in he
sample
Table 2.
Desc ip ion o he
a ge s udy a iables
EJMBE
34,1
92
his me hod’s es ima es a e mo e obus in he p esence o ou lie s, he e oskedas ici y and
skewness han hose o OLS models (Koenke and Hallock, 2001;Koenke , 2005).
The p oposed quan ile eg ession model is he ollowing:
Ri; ¼
α
τ
þβ1;
τ
EPU þβ2;
τ
Cycle þβ3;
τ
De aul sp ead þβ4;
τ
In la ion þβ5;
τ
IPI
þβ6;
τ
Te m sp ead þβ7;
τ
GDP þ
ε
;
τ
;(1)
whe e R
i
is he dependen a iable o model and ep esen s he s ock index i e u n in
mon h ,
α
is he cons an e m, βkis he eg ession coe icien co esponding o each
explana o y a iable k,
τ
he quan ile whose alue will be be ween 0 and 1 ( he quan iles 0.25,
0.5 and 0.75 will be aken in he s udy), and
ε
is he e o e m in mon h .
The p e ious model allows us o e alua e he impac o EPU on he s ock e u ns analysed.
Howe e , o es he ole o he economic cycle in he in luence o EPU on s ock e u ns, he
inclusion o in e ac ion e m be ween EPU and Cycle has been implemen ed.
We un he eg ession in Eq. (1) wi h an addi ional in e ac ion e m gi en by:
Ri; ¼
α
τ
þβ1;
τ
EPU þβ2;
τ
EPU
∙Cycle þβ3;
τ
Cycle þβ4;
τ
De aul sp ead þβ5;
τ
In la ion
þβ6;
τ
IPI þβ7;
τ
Te m sp ead þβ8;
τ
GDP þ
ε
;
τ
;
(2)
To add u he obus ness o he esul s om quan ile eg ession, we p opose o apply linea
eg ession wi h OLS. Gi en he possible he e oskedas ici y p oblem ypical o inancial
se ies, he OLS models a e adjus ed o he e oskedas ici y (whi e c oss-sec ion s anda d
e o s) (Lee and Chen, 2021).
Resul s and discussion
Basic desc ip i e s a is ics
O e he o al sample pe iod, he desc ip i e s a is ics (Table AI) show how R
La ge-G ow h
and R
Small-Value
achie ed a highe a e age e u n (1% e sus 0.9% o R
La ge-Value
and R
Small-
G ow h
). Howe e , when dis inguishing by economic cycle, we ind ha g ow h s ocks had
lowe losses han alue s ocks in ecessions. In pa icula , R
La ge-G ow h
eco ded he smalles
losses (a e age e u n o 0.9%). In con as , la ge cap alue s ocks had he wo s esul s
(R
La ge-Value
:1.8%). Though he a e age e u ns o he di e en ypes o s ocks di e
signi ican ly in ecessions, he di e ences eco ded in expansions a e minimal, wi h R
Small-
G ow h
ob aining he lowes a e age e u n (1.1 e sus 1.2% o he es ). These esul s a e in
line wi h e idence shown by B e schge and Lech hale (2018),Ki by (2019) and Bianchi
(2020). When e alua ing he exis ence o signi ican di e ences in he a e age e u n o hese
ypes o s ocks in ecessions and expansions, only R
La ge-Value
has subs an ial di e ences
depending on he economic cycle, wi h a signi icance le el o 5%.
Al hough in ecessions he dispe sion in all s ocks is highe , we iden i y a clea pa e n: he
e u ns o he smalles capi alisa ion s ocks show g ea e a iabili y, especially hose o
g ow h s ocks. These esul s sugges ha in es men in alue s ocks may be mo e ad isable
in expansions, while g ow h s ocks may pe o m be e in a ecession, despi e aking on
highe isk. In he same ein, Ki by (2019) e idenced a p ocyclical ela ion be ween he
expec ed alue p emium and expec ed business condi ions.
Table AI shows ha EPU has a highe mean alue in ecessions han in expansions
(7.3 s. 1.3%), al hough his di e ence canno be conside ed signi ican . In addi ion, he
dispe sion o EPU is g ea e in ecessions han in expansions. In his sense, he li e a u e
shows how in ecessions, despi e he economic policies ha will be implemen ed, unce ain y
is highe han in expansions (Bake e al., 2016;Adjei and Adjei, 2017).
EPU, s ock
ypes and
economic cycle
93
Rega ding con ol a iables, i should be no ed ha De aul sp ead and Te m sp ead show
signi ican di e ences, wi h a con idence g ea e han 99% in hei mean alue depending on
he economic cycle. These a iables ha e a highe mean alue in ecessions (1.7 and 2.2%
e sus 0.9 and 1.5%, espec i ely). Howe e , in e ms o hei dispe sion, De aul sp ead has
highe a iabili y in ecessions (0.9 s. 0.2%), while Te m sp ead has a highe de ia ion in
expansions (1.1 s. 0.8%).
IPI and GDP show highe alues in expansions (IPI 50.3% and GDP 50.2%) han in
ecessions (IPI 51.4% and GDP 50.5%), and hese di e ences a e signi ican a 99%.
In bo h cases, a iabili y is highe in ecessions.
Finally, i is wo h men ioning ha In la ion does no p esen signi ican di e ences in i s
a e age o he economic cycle, al hough in he pe iod s udied, i has p esen ed a highe
a e age alue in expansions han in ecessions (0.2 e sus 0.1%).
Table AII shows he bi a ia e co ela ions o he a iables used. I should be no ed ha he
co ela ion coe icien be ween all he explana o y a iables among hemsel es and he
dependen a iables is low. Only he co ela ions be ween IPI and GDP (0.637) and De aul
sp ead and Cycle (0.561) a e g ea e han 0.50 bu emain below 0.90, he h eshold
maximum sugges ed by Hai e al. (2010). The e o e, he mul icollinea i y p oblem does no
a ise in his s udy.
A mo e de ailed analysis o he bi a ia e co ela ions be ween he di e en a iables
indica es how he e u ns o he ou ypes o s ocks a e highly co ela ed wi h a signi icance
le el o less han 1%. In pa icula , he co ela ions o R
La ge-Value
yR
Small-Value
(0.859) and
R
Small-Value
yR
Small-G ow h
(0.845) s and ou .
EPU co ela es nega i ely and signi ican ly wi h he e u ns o he ou ypes o s ocks,
whe e he s onges co ela ion is wi h g ow h s ocks, especially small-cap s ocks (0.270).
This esul aligns wi h Hu e al. (2018) and Luo and Zhang (2020). Mo e e idence o his esul
has been ound in he li e a u e ha examines he impac o in es o sen imen on s ocks
depending on hei p o ile (Lakonishok e al., 1994;Bake and Wu gle , 2006;Waggle and
Ag awal, 2015).
Cycle, as expec ed gi en he esul s in Table AI, is posi i ely and signi ican ly co ela ed
wi h s ock e u ns ( ecall Recession 51 and Expansion 52). In ega d o he es o he con ol
a iables, only GDP is posi i ely and signi ican ly co ela ed wi h he e u ns o he ou
ypes o s ocks, especially wi h R
Small-Value
(0.204), while De aul sp ead is nega i ely
and signi ican ly co ela ed only wi h he e u ns o alue s ocks, especially wi h R
La ge-
Value
(0.136).
Resul s o he quan ile-based app oach
Conside ing Eq (1),Table 3 shows he esul s ob ained when analysing he in luence o EPU
on condi ional dis ibu ion o e u ns o alue and g ow h s ocks o highe and lowe
capi alisa ion (PANEL A).
The esul s show how EPU has a nega i e and signi ican impac wi h a signi icance le el
o less han 5% on s ock e u ns, especially small-cap s ocks, and be ween hem, g ow h
s ocks. The e o e, i is e iden ha he mos de e minan cha ac e is ic o measu e he impac
o EPU on s ock e u ns is he la ge o smalle capi alisa ion o he s ocks, wi h smalle
capi alisa ion s ocks being he mos a ec ed. The li e a u e ela ed o he impac o in es o
sen imen on s ock e u ns poin s in he same di ec ion (Lemmon and Po niaguina, 2006;
Kuma and Lee, 2006). Though wi h mo e mino di e ences, ano he de e mining
cha ac e is ic in he in luence o EPU on s ock e u ns is whe he he s ocks a e alue o
g ow h s ocks, whe e he la e a e mos a ec ed by EPU inc eases. In his sense, he e is
e idence ha g ow h s ocks a e mo e sensi i e o in es o sen imen (Lakonishok e al., 1994;
Bake and Wu gle , 2006;Wu e al., 2014). The e o e, we accep H1. I can be con i med ha
EJMBE
34,1
94
Va iable
R
La ge-Value
R
La ge-G ow h
R
Small-Value
R
SmallG ow h
τ
25
τ
50
τ
75
τ
25
τ
50
τ
75
τ
25
τ
50
τ
75
τ
25
τ
50
τ
75
Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue) Coe . (p- alue)
PANEL A: Model 3 es ima es
Cons 0.029 (0.52) 0.004 (0.84) 0.027 (0.35) 0.036 (0.11) 0.022 (0.53) 0.040** (0.03) 0.025 (0.46) 0.024 (0.55) 0.007 (0.78) 0.074 (0.13) 0.001 (0.98) 0.009 (0.86)
EPU 0.056*** (0.01) 0.046*** (0.00) 0.031** (0.03) 0.064*** (0.00) 0.048*** (0.01) 0.043*** (0.00) 0.081*** (0.00) 0.064*** (0.00) 0.036*** (0.00) 0.095*** (0.00) 0.086*** (0.00) 0.055** (0.02)
Cycle 0.028 (0.14) 0.016* (0.07) 0.021* (0.08) 0.020** (0.04) 0.021 (0.16) 0.002 (0.78) 0.006 (0.66) 0.002 (0.89) 0.008 (0.40) 0.035* (0.10) 0.007 (0.74) 0.014 (0.51)
De aul sp ead 3.880*** (0.00) 1.099* (0.08) 2.449*** (0.01) 2.488*** (0.00) 0.163 (0.88) 1.016* (0.06) 5.548*** (0.00) 1.405 (0.24) 1.700** (0.01) 2.546* (0.08) 0.479 (0.74) 1.730 (0.25)
In la ion 1.544 (0.22) 0.572 (0.34) 0.114 (0.89) 0.359 (0.57) 0.503 (0.62) 1.442*** (0.01) 0.034 (0.97) 0.986 (0.39) 1.856*** (0.01) 1.800 (0.19) 0.967 (0.48) 1.510 (0.29)
IPI 1.196** (0.02) 1.095*** (0.00) 0.599* (0.06) 0.667*** (0.01) 0.631 (0.11) 0.927*** (0.00) 1.571*** (0.00) 1.306*** (0.00) 0.768*** (0.00) 1.490*** (0.01) 1.292** (0.02) 0.727 (0.19)
Te m sp ead 0.125 (0.76) 0.045 (0.81) 0.058 (0.82) 0.373* (0.06) 0.158 (0.62) 0.513*** (0.00) 0.198 (0.52) 0.191 (0.60) 0.321 (0.12) 0.262 (0.55) 0.019 (0.97) 0.154 (0.73)
GDP 2.711* (0.06) 2.744*** (0.00) 0.245 (0.79) 2.376*** (0.00) 1.760 (0.13) 3.122*** (0.00) 4.387*** (0.00) 4.776*** (0.00) 1.587** (0.03) 3.037* (0.06) 4.038*** (0.01) 0.775 (0.63)
No Obs 317 317 317 317 317 317 317 317 317 317 317 317
PANEL B: Model 4 es ima es
Cons 0.028 (0.54) 0.001 (0.95) 0.009 (0.78) 0.051*** (0.00) 0.007 (0.81) 0.043** (0.03) 0.025 (0.41) 0.028 (0.37) 0.008 (0.73) 0.076 (0.16) 0.018 (0.65) 0.029 (0.51)
EPU 0.054 (0.53) 0.179*** (0.00) 0.149** (0.02) 0.131*** (0.00) 0.122** (0.02) 0.190*** (0.00) 0.088 (0.13) 0.169*** (0.01) 0.183*** (0.00) 0.081 (0.43) 0.280*** (0.00) 0.293*** (0.00)
EPU ∙Cycle 0.001 (0.98) 0.074*** (0.00) 0.067* (0.05) 0.037** (0.03) 0.041 (0.16) 0.078*** (0.00) 0.006 (0.84) 0.062* (0.06) 0.075*** (0.00) 0.012 (0.83) 0.119*** (0.01) 0.127*** (0.01)
Cycle 0.028 (0.15) 0.016* (0.09) 0.006 (0.66) 0.027*** (0.00) 0.015 (0.21) 0.001 (0.91) 0.007 (0.59) 0.003 (0.80) 0.015 (0.14) 0.036 (0.12) 0.015 (0.38) 0.007 (0.70)
De aul sp ead 3.865*** (0.00) 1.367** (0.03) 1.659* (0.09) 2.442*** (0.00) 0.532 (0.52) 1.008* (0.09) 5.662*** (0.00) 0.437 (0.64) 1.779** (0.01) 2.367 (0.14) 0.652 (0.58) 0.974 (0.46)
In la ion 1.563 (0.22) 0.454 (0.45) 0.390 (0.67) 0.573 (0.20) 0.393 (0.61) 1.132** (0.04) 0.045 (0.96) 1.070 (0.23) 1.772*** (0.01) 1.523 (0.31) 0.900 (0.41) 1.302 (0.30)
IPI 1.172** (0.02) 1.009*** (0.00) 0.718** (0.05) 0.882*** (0.00) 0.533* (0.08) 0.898*** (0.00) 1.712*** (0.00) 1.044*** (0.00) 0.776*** (0.00) 1.610*** (0.01) 0.856** (0.05) 0.529 (0.28)
Te m sp ead 0.114 (0.77) 0.028 (0.88) 0.038 (0.90) 0.431*** (0.00) 0.142 (0.56) 0.560*** (0.00) 0.248 (0.35) 0.215 (0.44) 0.329 (0.12) 0.232 (0.63) 0.224 (0.52) 0.104 (0.79)
GDP 2.714* (0.07) 1.913*** (0.01) 1.159 (0.28) 2.054*** (0.00) 1.234 (0.17) 2.192*** (0.00) 4.470*** (0.00) 3.646*** (0.00) 1.524** (0.05) 4.033** (0.02) 2.304* (0.07) 0.486 (0.74)
No Obs 317 317 317 317 317 317 317 317 317 317 317 317
No e(s): ***, ** and * indica e he signi icance a 1%, 5% and 10% le els, espec i ely
Sou ce(s): Own elabo a ion
Table 3.
Es ima es o he
quan ile eg ession
models
EPU, s ock
ypes and
economic cycle
95
Raza, S.A., Zaighum, I. and Shah, N. (2018), “Economic policy unce ain y, equi y p emium and
dependence be ween hei quan iles: e idence om quan ile-on-quan ile app oach”,Physica A:
S a is ical Mechanics and i s Applica ions, Vol. 492, pp. 2079-2091, doi: 10.1016/j.physa.2017.
11.125.
Roh, T.Y., Lee, C. and Min, B.K. (2019), “Consump ion g ow h p edic abili y and asse p ices”,Jou nal
o Empi ical Finance, Vol. 51, pp. 95-118, doi: 10.1016/j.jemp in.2019.02.001.
Scheu le, P. and Sp emann, K. (2010), “Size, book- o-ma ke , and momen um du ing he business
cycle”,Re iew o Manage ial Science, Vol. 4, pp. 201-215, doi: 10.1007/s11846-010-0041-6.
Smales, L.A. (2017), “The impo ance o ea : in es o sen imen and s ock ma ke e u ns”,Applied
Economics, Vol. 49 No. 34, pp. 3395-3421, doi: 10.1080/00036846.2016.1259754.
Song, G.H. (2023), “Size p emium o size discoun ?–A dynamic capi al mobili y based in e p e a ion”,
S udies in Economics and Finance, Vol. 40 No. 2, pp. 266-285, doi: 10.1108/SEF-04-2022-0211.
Swi ze , L.N. (2010), “The beha iou o small cap s. la ge cap s ocks in ecessions and eco e ies:
empi ical e idence o he Uni ed S a es and Canada”,The No h Ame ican Jou nal o
Economics and Finance, Vol. 21 No. 3, pp. 332-346, doi: 10.1016/j.naje .2010.10.002.
Van Dijk, M.A. (2011), “Is size dead? A e iew o he size e ec in equi y e u ns”,Jou nal o Banking
and Finance, Vol. 35 No. 12, pp. 3263-3274, doi: 10.1016/j.jbank in.2011.05.009.
Waggle, D. and Ag awal, P. (2015), “In es o sen imen and sho - e m e u ns o size-adjus ed alue
and g ow h po olios”,Jou nal o Beha io al Finance, Vol. 16 No. 1, pp. 81-93, doi: 10.1080/
15427560.2015.1000329.
Wu, D.D., Zheng, L. and Olson, D.L. (2014), “A decision suppo app oach o online s ock o um
sen imen analysis”,IEEE T ansac ions on Sys ems Man and Cybe ne ics Sys ems, Vol. 44 No. 8,
pp. 1077-1087, doi: 10.1109/TSMC.2013.2295353.
Supplemen a y ma e ial
The supplemen a y ma e ial o his a icle can be ound online.
Co esponding au ho
Jessica Paule-Vianez can be con ac ed a : [email p o ec ed]
Fo ins uc ions on how o o de ep in s o his a icle, please isi ou websi e:
www.eme aldg ouppublishing.com/licensing/ ep in s.h m
O con ac us o u he de ails: [email p o ec ed]
EJMBE
34,1
102