Hoang, Khanh; Hoang, Huy Vie ; Anh, Dao Le T ang
A icle
The wa che s on he wall: In e -Ko ean geopoli ical isk
and i m isk- aking
BRQ Business Resea ch Qua e ly
P o ided in Coope a ion wi h:
Asociación Cien í ica de Economía y Di ección de Emp esas (ACEDE), Mad id
Sugges ed Ci a ion: Hoang, Khanh; Hoang, Huy Vie ; Anh, Dao Le T ang (2025) : The wa che s on he
wall: In e -Ko ean geopoli ical isk and i m isk- aking, BRQ Business Resea ch Qua e ly, ISSN
2340-9444, Sage Publishing, London, Vol. 28, Iss. 1, pp. 288-303,
h ps://doi.o g/10.1177/23409444231184479
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/327069
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-nc/4.0/
h ps://doi.o g/10.1177/23409444231184479
Business Resea ch Qua e ly
2025, Vol. 28(1) 288 –303
© The Au ho (s) 2023
A icle euse guidelines:
sagepub.com/jou nals-pe missions
DOI: 10.1177/23409444231184479
jou nals.sagepub.com/home/b q
C ea i e Commons Non Comme cial CC BY-NC: This a icle is dis ibu ed unde he e ms o he C ea i e Commons
A ibu ion-NonComme cial 4.0 License (h ps://c ea i ecommons.o g/licenses/by-nc/4.0/) which pe mi s non-comme cial use,
ep oduc ion and dis ibu ion o he wo k wi hou u he pe mission p o ided he o iginal wo k is a ibu ed as speci ied on he SAGE and
Open Access page (h ps://uk.sagepub.com/abou us/openaccess.h m).
In oduc ion
How does co po a e isk- aking eac o inc easing geopo-
li ical isk and h ea s o wa ? P e ious s udies show ha
co po a e isk- aking is sensi i e o changes in he mac oe-
conomic and poli ical ins i u ions (Boubak i e al., 2013;
T an, 2019). Howe e , li le is known abou how geopoli i-
cal isk a ec s co po a e isk- aking beha io . As unce -
ain y in geopoli ics leads o changes in he business
en i onmen and a ec s co po a e ope a ions (Julio & Yook,
2012; Le & T an, 2021), i ms may ha e o conside unde -
aking ewe isky ope a ions i he poli ical isk is high.
Using a sample o Sou h Ko ean i ms ha ope a e
unde a high deg ee o geopoli ical isk, his s udy seeks o
e eal how In e -Ko ean geopoli ical isk in luences co -
po a e isk- aking. The unique geopoli ical con ex o he
Ko ean Peninsula p esen s an in e es ing se ing o s udy
he impac o geopoli ical isk on co po a e isk- aking. In
p inciple, No h Ko ea and Sou h Ko ea ha e been a wa
since 1950. The wo pa s o Ko ea only p ese e a so-
called “peace” by an a mis ice in 1953, which only se ed
he pu pose un il he Second Ko ean Wa 1966–1969.
Since he 1990s, he e ha e been inc easing conce ns om
Sou h Ko ea and he in e na ional communi y as No h
Ko ea has been de eloping nuclea weapons. F om 2006
o da e, he e ha e been six obse ed nuclea es s con-
duc ed in No h Ko ea’s e i o y, leading o heigh ened
geopoli ical ensions in he peninsula and he Asia-Paci ic
egion. As he h ea s o wa and nuclea isk become mo e
isible, i is a compelling con ex o s udy how co po a es
beha e unde ex eme unce ain y.
We ollow Jung e al. (2021) o measu e In e -Ko ean
geopoli ical isk using a news-based app oach. Because
geopoli ical isk is unobse able and canno be quan i ied
in con en ional ways, Jung e al. (2021) measu e his ac-
o by coun ing he equency o news a icles wi h key-
wo d combina ions ep esen ing In e -Ko ean geopoli ical
ensions and mili a y e en s, ha is, he isk associa ed
wi h geopoli ics be ween No h and Sou h Ko ea. Figu e 1
The wa che s on he wall: In e -Ko ean
geopoli ical isk and i m isk- aking
Khanh Hoang1, Huy Vie Hoang1 and Dao Le T ang Anh2
Abs ac
This a icle in es iga es he ela ionship be ween In e -Ko ean geopoli ical isk (IKGPR) and Sou h Ko ean i ms’ isk-
aking in he con ex o highly poli ical ensions in he Ko ean peninsula. Using a ecen ly de eloped news-based index
o geopoli ical isk, ou empi ical analysis shows an in e se U-shaped ela ionship be ween IKGPR and i m isk- aking
in Sou h Ko ea, implying ha Sou h Ko ean i ms’ isk- aking inc eases wi h IKGPR when IKGPR is low bu becomes
p uden when IKGPR is high. Such an e ec o IKGPR on i m isk- aking is weake o i ms ha su i e wa ime (1900–
1969) and is s onge o i ms ha we e es ablished a e he Second Ko ean Wa (1966–1969), sugges ing he ole o
co po a e esilience o geopoli ical isk in mo i a ing di e ences in co po a e beha io s. Fu he analyses indica e ha
Sou h Ko ean i ms end o ake less isk when he nuclea isk is highe and when hey a e mo e echnology-o ien ed.
JEL CLASSIFICATION: G30, P00, O31
Keywo ds
Co po a e go e nance, co po a e isk- aking, geopoli ical isk, in e -Ko ea, nuclea h ea s, R&D in ensi y
1
Uni e si y o Economics Ho Chi Minh Ci y (UEH), Ho Chi Minh Ci y,
Vie nam
2RMIT Uni e si y Vie nam, Hanoi, Vie nam
Co esponding au ho :
Dao Le T ang Anh, RMIT Uni e si y Vie nam, Handi Resco Building,
521 Kim Ma, Ngoc Khanh, Ba Dinh, Hanoi, 100000, Vie nam.
Email: [email p o ec ed]
1184479BRQ0010.1177/23409444231184479Business Resea ch Qua e lyHoang e al.
esea ch-a icle2023
Regula Pape
Hoang e al. 289
shows he main de elopmen s o In e -Ko ean geopoli ics
and he mo emen o he Geopoli ical Risk Index p oposed
by Jung e al. (2021).
Using a mul iple ixed e ec model and in ensi e con-
ols o i m-le el, indus y-le el, and mac o-le el con-
ounding ac o s du ing he 2002–2019 pe iod, we ind
ha , in gene al, Sou h Ko ean i ms’ isk- aking inc eases
wi h In e -Ko ean geopoli ical isk when In e -Ko ean
geopoli ical isk is low; howe e , his ela ionship e e ses
o o m a U-shaped cu e when In e -Ko ean geopoli ical
isk is high. The empi ical inding emains quali a i ely
unchanged a e a ba e y o sensi i i y es s. In e es ingly,
analysis shows ha such an e ec is mo e p onounced in
i ms es ablished a e he Second Ko ean Wa (1966–
1969) while emaining weake in hei coun e pa s.
Fu he analyses demons a e he a ia ions o he impac
du ing No h Ko ea’s nuclea es e en s and among i m
g oups o di e en echnology in ensi ies and chaebols.
The indings a e no el and no documen ed elsewhe e in
he li e a u e.
Ou s udy con ibu es o he li e a u e in se e al ways.
Fi s , we show a nonlinea ela ionship be ween In e -
Ko ean geopoli ical isk and Sou h Ko ean i ms’ isk-
aking, indica ing he asymme y in co po a e beha io
ac oss di e en le els o geopoli ical isk. Ou inding
well suppo s he no ion ha i ms a ionally ake isk and
can dis inguish be ween aking isk and gambling (Ma ch
& Shapi a, 1987). The e o e, despi e i ms aking isk
when he isk is low, hey become p uden when he le el
o geopoli ical isk exceeds a ce ain h eshold. F om
he e, we indica e ha he di e ences in he esea ch
designs (linea s nonlinea model se ing) may e en ually
lead o con adic o y indings on he ela ionship be ween
geopoli ical isk and co po a e isk- aking. This is an
impo an ema k ha p o ides implica ions o u u e
s udies in his ield.
Second, we sugges he ole o co po a e esilience o
geopoli ical isk in mo i a ing di e ences in co po a e
beha io s unde unce ain y. In he con ex o he Ko ean
Peninsula, i ms ha su i e wa ime may de elop a ce -
ain deg ee o esilience o he poli ical ension be ween
he Sou h and he No h. In con as , i ms es ablished
a e he Second Ko ean Wa (e.g., i ms wi h less esil-
ience o poli ical ension) exhibi s onge nega i e eac-
ions o un a o able de elopmen s in In e -Ko ean
poli ics. This inding is in ui i e as Sou h Ko ean i ms
es ablished and su i ed he In e -Ko ean wa pe iods
(1950–1953 and 1966–1969) ha e expe ience dealing
wi h ex eme unce ain y du ing he wa , hus ha ing be -
e co po a e esilience o such ex aneous shocks. As a
esul , nega i e de elopmen s in In e -Ko ean geopoli i-
cal ensions do no a ec hei beha io as much as i ms
ha did no expe ience wa .
Thi d, we p o ide addi ional e idence o how he
impac a ies in he c oss-sec ions and he ime dimension,
hus adding new unde s andings o he new ound
200201
200207
200301
200307
200401
200407
200501
200507
200601
200607
200701
200707
200801
200807
200901
200907
201001
201007
201101
201107
201201
201207
201301
201307
201401
201407
201501
201507
201601
201607
201701
201707
201801
201807
201901
201907
0
50
100
150
200
250
300
350
In e -
Ko ea
Second
nuclea
Fih nuclea es
Kaesong Indus ial Complex shu dow
In e connenal ballisc
missile launching
Th ea s o "Ch is mas gi"
Fi s nuclea es
In e -Ko ean
summi
A misce enounciaon
Figu e 1. De elopmen s in In e -Ko ean geopoli ics and he IKGPR index du ing he 2002–2019 pe iod.
290 Business Resea ch Qua e ly 28(1)
ela ionship. The inding sugges s he ole o esea ch and
de elopmen (R&D) in ensi y in co po a e ope a ions
unde ex eme geopoli ical isk, whe e echnologically
in ensi e i ms ake less isk han hei coun e pa s when
he h ea s om he No h d aw nea . As in es men s in
R&D ac i i ies a e usually ega ded as isky ope a ions
(Xu e al., 2019), i ms wi h mo e R&D in ensi y may ha e
less incen i e o ake mo e isk in hei ope a ions unde
inc eased geopoli ical isk ela i e o less R&D-in ensi e
i ms. The inding p o ides impo an implica ions o he
decision-making o R&D-in ensi e i ms unde ex eme-
isk si ua ions.
The es o he a icle p oceeds as ollows. Sec ion
“Li e a u e backg ound and hypo hesis de elopmen ”
e iews ela ed li e a u e and de elops hypo heses. Sec ion
“Me hodology and da a” p esen s he esea ch me hodol-
ogy and da a used in his s udy. Sec ion “Empi ical esul s
and discussion” epo s and discusses empi ical esul s.
Sec ion “Conclusion, limi a ions, and u u e esea ch”
concludes his a icle.
Li e a u e backg ound and hypo hesis
de elopmen
Some pa icula a eas a e mo e complica ed han o he s
due o hei his o ical e en s ha mo i a e an in ense
a mosphe e in he a eas. One o he mos unp edic able
egions is he Ko ean Peninsula, whe e wo coun ies o
No h Ko ea and Sou h Ko ea co-exis . The ension in he
In e -Ko ea ela ionship, as elabo a ed by Jung e al. (2021,
p. 7), is due o hei “in e na ional sanc ions agains No h
Ko ea, bila e al and mul ila e al alks o seek econcilia-
ion, and, inally, economic coope a ion be ween Sou h
and No h Ko ea.” The igh ening economic and inancial
sanc ions om he in e na ional o ganiza ions and he
Uni ed S a es h ea en No h Ko ea’s economy and a
some poin s igge agg essi e eac ions om No h
Ko ea, which, in u n, ele a e ension on he geopoli ical
landscape o he Ko ean Peninsula. Al hough he ela ion-
ship be ween No h and Sou h Ko ea has imp o ed
ecen ly, ma ked by he his o ical handshake o P esiden
Kim Jong Un and P esiden Moon Jae In a he sha ed bo -
de o he wo coun ies, hei s ain is s ill he e o sol e.
Di e en om commonly used p oxies o unce ain y
such as (economic) policy unce ain y and poli ical unce -
ain y, geopoli ical isk is o e looked in i m-le el
esea ch. The sca ci y o esea ch on In e -Ko ean geopo-
li ical isk is p obably pa ly because o a lack o p ope
measu emen s o his e m alongside he complexi y o
he bila e al ela ionship. To e ec i ely g asp he idea o
geopoli ical isk, i is impo an o unde s and ha isk is
associa ed wi h a known ou come, while unce ain y e e s
o he sho age o eliable in o ma ion o come up wi h an
e ec i e p edic ion o he ou come (Kob in, 1979). Ma ch
and Shapi a (1987, p.1404) make a poin ha manage s
“make a sha p dis inc ion be ween aking isks and gam-
bling.” Acco ding o P a (1964) and o he ea ly ea -
men s, i an indi idual s ands be ween wo op ions wi h
simila expec ed alues, in which one’s ou come is known
o ce ain and he o he is a gamble, he pe son will lean
owa d he o me op ion. I i ms may conside ading o
isks o po en ial u u e p o i s when he ou come is
known, hey end o be mo e p uden wi h business deci-
sions when hey canno an icipa e he consequence. Hence,
i ms’ eac ions o isk and unce ain y may no be iden i-
cal. Despi e ha isk and unce ain y di e in na u e, he
an agonis ic his o y o No h and Sou h Ko ea implies ha
geopoli ical isk may b ing o h geopoli ical unce ain y
(i.e., an inc easing p obabili y o an a med con lic leading
o unp edic able business p ospec s) when poli ically isky
e en s incessan ly ake place. The e o e, ising In e -
Ko ean geopoli ical isk also heigh ens unce ain y h ea s
in his speci ic con ex .
Despi e co po a e esponses o In e -Ko ean geopoli i-
cal swings being unde -s udied, se e al pape s on s ock
pe o mance in he ace o In e -Ko ean geopoli ical luc-
ua ions ha e yielded dissimila esul s. Jung e al. (2021)
use hei sel -cons uc ed In e -Ko ean geopoli ical isk
index, based on machine lea ning echniques, o e eal an
in e se associa ion be ween geopoli ical isk and s ock
e u ns o Sou h Ko ean i ms. In con as , Pyo (2021)
poin s ou ha he ma ke eac s a o ably when he geo-
poli ical si ua ion in he Peninsula u ns milde ; howe e ,
he ma ke shows limi ed a en ion o nega i e In e -Ko ea
e en s. Mo eo e , how his ype o isk impac s Sou h
Ko ea’s mac oeconomic indica o s also emains inconclu-
si e due o con adic o y empi ical indings (Ha e al.,
2022; Pyo, 2021). The di e gence in empi ical indings
may s em om di e en measu emen cons uc ions; how-
e e , ano he possible scena io is he empi ical con lic s
could be an indica ion o a ia ions in he impac o In e -
Ko ean geopoli ical isk a di e en isk le els, which has
no been apped in o by exis ing s udies.
Gi en he looming isk o geopoli ics and wa in he
Ko ean Peninsula o almos a cen u y, he la e explana-
ion is no i a ional. Al hough his inhe en isk may
somehow al e he business p ac ice o Sou h Ko ean
i ms, i is well no ed ha mos cu en ly ope a ing i ms
we e es ablished a e he Second Wo ld Wa ; he e o e,
hey a e p obably bo n wi h a ole ance o In e -Ko ean
geopoli ical swing. In o he wo ds, Sou h Ko ean i ms
may no s ongly eac o mino luc ua ions in geopoli ical
isk unless when i conside ably ele a es. In ha ligh ,
Sou h Ko ean i ms likely pu sue some isky p ojec s un il
he isk le el eaches a h eshold. Theo e ically, Kahneman
and T e sky (1979) indica e ha when an agen ’s u ili y is
based on gains o losses a he han inal income le els,
he e is a endency owa d isk-seeking beha io in he
domain o mino losses. Adding o ha , F ench and Sichel
(1993) sugges ha because heigh ened unce ain y is
Hoang e al. 291
o en a esul o nega i e shocks, he ad e se e ec o
unce ain y ends o domina e when unce ain y is high.
Con e sely, he e is oom o a posi i e e ec when unce -
ain y is low. These heo e ical iewpoin s imply ha he
associa ion be ween co po a e isk- aking and geopoli ical
isk is no necessa ily linea .
Whe eas no p io esea ch di ec ly del es in o he link-
age be ween co po a e isk- aking and In e -Ko ean geo-
poli ical isk, cu en s udies p o ide some hin s a a
nonlinea impac o unce ain y om he in es men and
cash-holding channels. Gi en ha In e -Ko ean geopoli i-
cal isk se es as a sou ce o unce ain y in he Ko ean
Peninsula, i is logical o associa e In e -Ko ean geopoli i-
cal isk wi h unce ain y when s udying Ko ean i ms’
isk- aking. The e m isk- aking is highly linked wi h
in es men , e e ing o a i m’s in es men decision o
ade o an icipa ed cash lows o unce ain isks
(Acha ya e al., 2011). Sa ka (2000), al hough ag ees wi h
he alidi y o he eal op ions heo y, also poin s ou ha
unce ain y does no always inc ease he alue o he
op ion o wai . Ins ead, highe unce ain y may also b ing
up he p obabili y o in es men ; howe e , his posi i e
e ec seems o be es ic ed only wi hin a low-unce ain y
domain and e e ses when unce ain y exceeds a ce ain
h eshold. The e o e, acco ding o Sa ka (2000), he ela-
ionship be ween in es men and unce ain y should ol-
low an in e ed U cu e. The nonlinea e ec o unce ain y
on in es men is also shown in Bo and Lensin (2005), an
empi ical s udy on a sample o Du ch non- inancial i ms.
Pa icula ly, hey ind ha when he unce ain y is low
(high), an inc ease in unce ain y leads o highe (lowe )
in es men s. F om he cash-holding pe spec i e, Su e al.
(2020) e eal ha when unce ain y eme ges, i ms end o
ade o hei cash holdings o seize new in es men
oppo uni ies o boos hei ea nings. Howe e , once
unce ain y is clea ly p esen , a isk-a e sion a i ude p e-
ails as con inuously heigh ened unce ain y makes i
challenging o co po a e manage s o assess he compa-
ny’s abili y o handle he isks, leading o a mo e conse a-
i e in es men app oach and an inc ease in he amoun o
cash kep o p ecau iona y easons (Ko cha in &
Maneenop, 2020; Lee & Wang, 2021). These indings
in o m ha i ms lean owa d isky decisions when unce -
ain y ises as long as i has no eached a ce ain
h eshold.
Fi ms’ abili y o ake isks du ing escala ed unce ain y
is also es ained by he capi al ma ke . As unce ain y
heigh ens, banks will be mo e p uden in lending deci-
sions, leading o a sho age o cash supply limi ing i ms’
a ailable unds o in es men (De Nicolò e al., 2010; Hu
& Gong, 2018). In ha con ex , banks’ p ecau iona y poli-
cies when geopoli ical unce ain y u ns isible cons ain
i ms’ inancing op ions (B ogaa d & De zel, 2015; Zhang
e al., 2015). The inc easing isk o inancing g adually
decima es he expec ed bene i om in es men s, making
in es men op ions unjus i iable and encou aging manag-
e s o adop a mo e conse a i e app oach (Be nanke,
1983; Go mley & Ma sa, 2016; Gulen & Ion, 2016;
Shapi a, 1995). This exogenous impac incen i izes i ms
o minimize hei exposu e o unce ain y in he ea o he
unan icipa ed ad e se e ec caused by g owing unce -
ain y. Co po a e isk- aking, hence, dec eases i unce -
ain y is con inuously magni ied as a esul o geopoli ical
swing. Fi ms’ isk- aking, howe e , may inc ease in he
sho un because he bene i gained om in es men s ill
ou weighs i s cos and banks do no necessa ily adop
s ingen measu es agains unno iceable unce ain y.
The di e en co po a e isk- aking beha io du ing di -
e en pe iods o unce ain y indica es ha he ela ionship
be ween i ms’ isk- aking and unce ain y is likely o be
nonlinea . Since In e -Ko ean geopoli ical isk ac s as a
o m o unce ain y imposed on he Sou h Ko ean econ-
omy, we conjec u e ha Sou h Ko ean i ms demons a e
di e en isk- aking a i udes ac oss escala ing geopoli i-
cal ension and calm pe iods. Following his conjec u e,
we p opose he esea ch hypo hesis as ollows:
Hypo hesis 1: The ela ionship be ween In e -Ko ean
geopoli ical isk and i m isk- aking is in e ed
U-shaped.
Me hodology and da a
Va iable measu emen s
In e -Ko ean geopoli ical isk. We use he In e -Ko ean geo-
poli ical isk index p oposed by Jung e al. (2021) as ou
a iable o in e es in his s udy. Acco ding o Jung e al.
(2021), he geopoli ical isk in ol es ou key d i e s ha
link o each o he and c ea e a geopoli ical backg ound o
he Ko ean Peninsula. The d i e s include he con lic
be ween he Sou h and No h Ko ean mili a y, in e na-
ional emba goes on No h Ko ea, bila e al and mul ila -
e al con e sa ions o econcile, and coope a ion be ween
Sou h and No h Ko ea’s economies. Jung e al. (2021)
cons uc he In e -Ko ean geopoli ical isk index based on
he news a icles o 18 b oadcas e s and newspape s o
BigKinds, a news-analyzing i m o med by he Ko ea
P ess Founda ion. The opics o he news ocus on eco-
nomics, in e na ional ela ionships, and poli ics. Then, a
lis o keywo ds on headlines o con en s is de e mined.
F om he sea ch o keywo ds, Jung e al. (2021) calcu-
la e he news equency in each ca ego y (mili a y issues,
sanc ions, con e sa ion o econcilia ion, and economic
coope a ion) and di ide i in o wo g oups: posi i e and
nega i e news. Posi i e and nega i e news indica e he
news ha epo s he ension dec ease and inc ease
be ween Sou h and No h Ko ea, espec i ely. Le Nj be
he o al numbe o news om media j a ime . The ela-
i e equency
()
Xj o nega i e news
()
,
Nnega j and
292 Business Resea ch Qua e ly 28(1)
posi i e news
()
,
N
posi j o e he o al numbe s o news
Nj is compu ed as ollows
X
NN
N
j
nega j posi j
j
,,
Jung e al. (2021) hen ans o m Xj o a posi i e numbe
as ollows
XX
j j
05 01
2
..
Fo each media sou ce, Jung e al. (2021) s anda dize
Xj
by s dj ( he ime se ies wi hin newspape s anda d de ia-
ion o
Xj ) o ge a se ies Yj
YX
s d
j
j
j
=
Yj is a e aged ac oss K media sou ces as ollows
YKY
j
N
j
1
1
Finally, Y is no malized o ge he In e -Ko ean geopoli i-
cal isk index IKGPR wi h he mean alue o Y
()
Y
IKGPR
Y
Y
=100
The highe he alue o IKGPR, he highe he geopoli ical
isk be ween No h and Sou h Ko ea is.
Co po a e isk- aking. Following he commonly accep ed
p ac ice in he p e ious li e a u e (Kim e al., 1993; W igh
e al., 2007; Yung & Chen, 2018), we use he olling s and-
a d de ia ion o e u n-on-asse s a io (SDROA) as a meas-
u emen o co po a e isk- aking. SDROA is calcula ed
o e a olling window o h ee con inuous yea s. Fo
obus ness check, we compu e an al e na i e measu e o
isk- aking: he h ee-yea olling s anda d de ia ion o
e u n-on-common equi y a io (SDROE). The highe he
alue o SDROA and SDROE, he iskie p ojec s he i m
akes on and he mo e ola ile i s income is.
Using an al e na i e app oach o measu e i m isk-
aking (Ma ins, 2020), we p oxy i m isk- aking by he
absolu e de ia ion om expec ed pe o mance o each
yea . Ma ins (2020) models he expec ed i m pe o -
mance using he ollowing equa ion
ROASIZELEVERAGE
CASH INVESTMENT
GR
i i i
i i
,, ,
,,
OOWTH FIXED ASSETS
i i
ii
,,
,
_
(1)
whe e ROAi , is he e u n-on- o al asse s a io, SIZEi , is
he na u al loga i hm o he book alue o o al asse s,
LEVERAGEi , is he o al deb o o al asse s a io,
CASHi , s ands o cash and cash equi alen s scaled by
ne o al asse s, INVESTMENTi , is he a io o he capi al
expendi u e on he beginning balance o o al asse s,
GROWTHi , is he annual changes in sales scaled by
1-yea lagged sales, FIXED ASSETSi
_, is he angible
asse s scaled by o al asse s a io,
δ
i is he i m- ixed
e ec , all numbe s a e om i m i du ing yea . The abso-
lu e alue o he esidual
ε
i , is de ined as he de ia ion o
ac ual ROA om he expec ed ROA, hus se ing as he
al e na i e measu e o i m isk- aking (ABS_ROA_RES).
The highe he alue o ABS_ROA_RES, he mo e isk-
aking he i m is. Simila app oaches a e adop ed in he
li e a u e (Adams e al., 2005; Cheng, 2008; Nakano &
Nguyen, 2012).
Con ol a iables. Based on he p e ious li e a u e (T an,
2019; Yung & Chen, 2018), we adop coun y-le el,
indus y-le el, i m-le el, and co po a e go e nance con-
ols o cap u e he speci ic cha ac e is ics o each i m in
each yea o ou sample. Rega ding coun y-le el con-
ols, we employ Sou h Ko ea’s Economic Policy Unce -
ain y Index (Bake e al., 2016) and he g oss domes ic
p oduc (GDP) g ow h a e (in pe cen age) o Sou h
Ko ea. The coun y-le el con ols indica e he mili a y
issue, economic, and poli ical backg ound o Sou h Ko ea.
In e ms o indus y-le el con ols, he s anda d de ia ion
o o al sales o i ms in he same indus y and he s and-
a d de ia ion o ROA o i ms in he same indus y a e
included. Fo he i m-le el con ols, we selec popula -
used con ol a iables in he li e a u e, including i m
size, i m age, le e age, sales g ow h, cash a io, ixed
asse s a io, cash low, Z-sco e, ma ke - o-book a io, and
annual s ock e u ns ollowing Yung and Chen (2018),
T an (2019), and Koi ala e al. (2020).
Empi ical model
We use he ollowing empi ical model o in es iga e he
impac o In e -Ko ean geopoli ical isk on i m
isk- aking
RISK TAKING IKGPRSQ
IKGPRCONTROL
ij
i
i i
__
,
,
,
(2)
Hoang e al. 293
whe e
RISK TAKINGij
_
, is he measu e o i m i’s isk-
aking du ing pe iod ; IKGPR and
β
IKGPRSQ
_ a e
he In e -Ko ean Geopoli ical Risk Index and i s squa e
du ing pe iod , espec i ely; Σ
CONTRO
Li , is he ec o
o con ol a iables a i m-le el, indus y-le el, and
mac o-le el du ing pe iod ;
δ
i and
θ
a e he indus y-
and yea - ixed e ec s, espec i ely;
ε
i , is he e o e m
o he model. When es ima ing he model, we clus e
s anda d e o s by i m o alle ia e he po en ial impac o
he e oskedas ici y and au oco ela ion. In gene al, he use
o a a iable and i s squa ed e m in he same eg ession
model does no a ec hei p alues (Alisson, 2012), and
he in e ences can be sa ely in e p e ed as long as he
Table 1. Va iable desc ip ions.
Va iable Desc ip ion Sou ces
ROA Ne income scaled by a e age o al asse s (in %) Bloombe g
ROE Ne income scaled by a e age common equi y Bloombe g
SDROA 3-yea olling s anda d de ia ion o ROA ollowing Yung and Chen (2018) Bloombe g
SDROE 3-yea olling s anda d de ia ion o ROE ollowing Yung and Chen (2018) Bloombe g
ABS_ROA_RES The on-yea isk- aking measu e which equals he absolu e alue o he
de ia ion om he expec ed pe o mance o i m ollowing Ma ins (2020)
IKGPR The annualized In e -Ko ean Geopoli ical Risk Index calcula ed by aking
he mean o mon hly In e -Ko ean Geopoli ical Risk Index du ing a yea ,
hen scaled by 100
Jung e al. (2021)
IKGPR_SQ The squa e o IKGPR Jung e al. (2021)
MA3_IKGPR 3-yea mo ing a e age o IKGPR Jung e al. (2021)
MA3_IKGPR_SQ The squa e o MA3_IKGPR
IW_IKGPR Inc easing-weigh ed IKGPR calcula ed by assigning inc easing weighs o
1–12 o mon hs o he yea om Janua y o Decembe , espec i ely,
hen aking he weigh ed a e age and scaled by 100
Jung e al. (2021)
IW_IKGPR_SQ The squa e o IW_IKGPR
DW_IKGPR Dec easing-weigh ed IKGPR calcula ed by assigning dec easing weighs o
12–1 o mon hs o he yea om Janua y o Decembe , espec i ely,
hen aking he weigh ed a e age and scaled by 100
Jung e al. (2021)
DW_IKGPR_SQ The squa e o DW_IKGPR
Res_IKGPR The esiduals om he eg ession o IKGPR on he Geopoli ical isk
(GPR) index o China and Russia (Calda a and Iaco iello, 2022). China’s
and Russia’s GPR indexes a e scaled by 100
Jung e al. (2021), Calda a
and Iaco iello (2022)
Res_IKGPR_SQ The squa e o Res_IKGPR
SIZE Na u al loga i hm o o al asse s Bloombe g
LEVERAGE To al deb scaled by o al asse s Bloombe g
GROWTH G ow h in sales scaled by 1-yea lagged sales Bloombe g
CASH Cash and cash equi alen s scaled by o al asse s Bloombe g
FIXED_ASSETS P ope y, plan , and equipmen scaled by o al asse s Bloombe g
CASHFLOW Ne ope a ing cash low scaled by o al asse s Bloombe g
ZSCORE Al man’s Z-sco e o he i m Bloombe g
MTB Ma ke - o-book a io Bloombe g
SRETURN Annual s ock e u ns o he i m Bloombe g
FIRM_AGE Age o he i m ( om inco po a ion o da e) Bloombe g
SD_SALES_IND S anda d de ia ion o o al sales o i ms in he same indus y Bloombe g
SD_ROA_IND S anda d de ia ion o ROA o i ms in he same indus y Bloombe g
KEPU Sou h Ko ean’s Economic Policy Unce ain y Index Bake e al. (2016)
GDP GDP g ow h o Sou h Ko ea (in %) Wo ld Bank
NUCLEAR The numbe o No h Ko ea’s nuclea es s pe yea
GDP: g oss domes ic p oduc .
es ima es a e s a is ically signi ican o i he uppe and
lowe limi s o he con idence in e al a e in hei heo e i-
cal ange (O’b ien, 2007).
Following he hypo heses de elopmen , we expec he
coe icien
β
o IKGPR_SQ o be nega i e and signi i-
can . Va iable desc ip ions a e p esen ed in Table 1.
Da a and sample
We collec da a om se e al sou ces. Financial da a o
Sou h Ko ean i ms a e om he Bloombe g da abase. We
e ie e mac oeconomic da a om Wo ld Bank open da a-
base; he economic policy unce ain y index o Sou h
294 Business Resea ch Qua e ly 28(1)
Ko ea (Bake e al., 2016) is om he websi e www.poli-
cyunce ain y.com. Fo independen a iable, we adop
he In e -Ko ean Geopoli ical Risk Index p oposed by
Jung e al. (2021). Ou ini ial sample includes all i ms
lis ed on Ko ea S ock Exchange om 2002 o 2019. We
sc een he sample and exclude all inancial and u ili y
i ms because he isk- aking na u e is di e en om ha
o non- inancial i ms. A e excluding missing da a, he
inal sample consis s o 1,668 Sou h Ko ean i ms wi h a
o al o 15,412 i m-yea obse a ions. All con inuous
a iables a e winso ized a he 1s and he 99 h pe cen iles
o alle ia e he po en ial impac s o ou lie s on he ou -
comes o da a analysis.
Empi ical esul s and discussion
Desc ip i e s a is ics
The a iables’ desc ip i e s a is ics a e displayed in Table
2. The a e age co po a e isk- aking le el (SDROA) o
he lis ed Ko ean i ms is 4.429, wi h a s anda d de ia-
ion o 6.109. The a iable IKGPR du ing he sample
pe iod anges om 0.527 o 2.389, wi h a mean o 1.257.
Figu e 1 shows ha he In e -Ko ean geopoli ical isk
index eaches new peaks a he announcemen o impo -
an nega i e news, such as he i s and second nuclea
es s, he Yeonpyeong Island a ack. The index d ops o
low le el when posi i e news, o example, In e -Ko ean
summi and In e -Ko ean and No h Ko ea-Uni ed S a es
summi s, a e in o med.
Rega ding he con ol a iables, an a e age Ko ean-
lis ed i m has le e age and sales g ow h o 23.2% and
9%, espec i ely. The a e age ixed asse s a io and cash
a io a e 32.8% and 10.5%, co espondingly. The e a e
bo h young i ms and ma u e i ms in he sample, wi h
an a e age i m age (FIRM_AGE) o 30.3 yea s, whe e
he younges i ms a e only 1 yea old and he oldes
i ms a e 121 yea s old. In e ms o ma ke pe o mance,
Ko ean-lis ed i ms’ ma ke - o-book a io and annual
s ock e u ns a e a e aged a 1.5% and 15.8%, espec-
i ely. Fo he co po a e go e nance ac o , he s a is ics
show ha a majo i y (81.5%) o he Chie Execu i e
O ice is also he chai o Ko ean i ms’ boa d o di ec-
o s. Besides, as shown in Table 2, du ing he s udy
pe iod (2002–2019), he a e age GDP g ow h o Ko ea
is 3.428%. The a e age numbe o No h Ko ea’s nuclea
es s is 0.419 imes pe yea .
Table 8 in he Appendix shows he pai wise co ela-
ion ma ix o ou eg ession model’s a iables. Since all
co ela ion alues a e smalle han 0.5, he e a e no no a-
ble co ela ions among independen a iables in he
eg ession model.
Table 2. Summa y s a is ics.
Va iable Obs M SD Minimum Maximum
SDROA 15,412 4.429 6.109 0.000 40.366
SDROE 15,371 10.628 16.912 0.000 110.458
ABS_ROA_RES 15,412 4.206 5.253 0.000 53.863
IKGPR_SQ 15,412 1.854 1.598 0.277 5.708
IKGPR 15,412 1.257 0.523 0.527 2.389
MA3_IKGPR 15,412 121.914 32.927 73.250 177.188
IW_IKGPR 15,412 124.158 52.281 50.974 234.000
DW_IKGPR 15,412 127.284 53.300 54.359 243.833
Res_IKGPR 15,412 11.539 39.539 −46.222 94.430
SIZE 15,412 5.480 1.566 1.756 10.119
ROA 15,412 1.156 9.925 −52.101 27.224
LEVERAGE 15,412 0.232 0.175 0.000 0.696
GROWTH 15,412 0.090 0.355 −0.645 2.514
CASH 15,412 0.105 0.128 0.001 0.945
FIXED_ASSETS 15,412 0.328 0.185 0.003 0.793
CASHFLOW 15,412 −0.004 0.090 −0.384 0.242
ZSCORE 15,412 3.511 6.176 −12.981 457.813
MTB 15,412 1.500 1.656 0.120 11.439
SRETURN 15,412 0.158 0.635 −0.776 3.295
FIRM_AGE 15,412 30.308 16.537 1.000 121.000
SD_SALES_IND 15,412 3,143.717 5,861.372 6.653 52,813.613
SD_ROA_IND 15,412 9.230 4.259 0.035 35.688
KEPU 15,412 144.888 43.217 68.64 257.362
GDP 15,412 3.428 1.412 0.793 7.725
NUCLEAR 15,412 0.419 0.644 0.000 2.000
Hoang e al. 295
Baseline esul s and discussion
The baseline eg ession esul s o Model 2 a e p esen ed
in Table 3. As shown in Table 3, he coe icien o IKGPR
is posi i ely and signi ican ly associa ed wi h co po a e
isk- aking a 1% le el in all h ee eg ession analyses
( educed- o m, ull model, and s and-alone e ec o
Table 3. Baseline eg ession.
Va iables (1) (2) (3)
Reduced- o m model Full model S and-alone e ec o IKGPR
SDROA SDROA SDROA
IKGPR_SQ −10.958***
(2.615)
−1.341***
(0.013)
IKGPR 16.299***
(4.416)
4.384***
(0.054)
0.230***
(0.057)
SIZE −0.507***
(0.065)
−0.507***
(0.066)
ROA −0.166***
(0.017)
−0.166***
(0.017)
LEVERAGE 0.820
(0.546)
0.820
(0.543)
GROWTH 0.584**
(0.204)
0.584**
(0.205)
CASH 2.058**
(0.786)
2.058**
(0.789)
FIXED_ASSETS −3.001***
(0.615)
−3.001***
(0.619)
CASHFLOWS −3.651**
(1.525)
−3.651**
(1.519)
ZSCORE −0.032**
(0.012)
−0.032**
(0.012)
MTB 0.611***
(0.085)
0.611***
(0.084)
SRETURN −0.445***
(0.131)
−0.445***
(0.132)
FIRM_AGE −0.008 −0.008
(0.005) (0.006)
SD_SALES_IND 0.000
(0.000)
0.000
(0.000)
SD_ROA_IND 0.063***
(0.016)
0.063***
(0.016)
KEPU −0.019***
(0.001)
−0.010***
(0.001)
GDP 0.048
(0.043)
0.198***
(0.043)
Cons an 3.953***
(0.126)
7.007***
(0.603)
7.403***
(0.588)
Indus y- ixed e ec Yes Yes Yes
Yea - ixed e ec Yes Yes Yes
Obse a ions 17,350 15,412 15,412
Adjus ed R2.084 .258 .258
This able epo s he eg ession esul s o ou baseline eg ession. The dependen a iable is SDROA. Column 1 epo s he eg ession esul o he
educed- o m model, Column 2 epo s he eg ession esul s o he ull model, while Column 3 shows he es ima ion o he s andalone e ec o
IKGPR on SDROA a e con olling o he a iables. All a iable de ini ions a e p esen ed in Table 1. S anda d e o s a e clus e ed by i m. Indus y-
and yea - ixed e ec s a e included. Numbe s in pa en heses a e s anda d e o s.
***, **, and * deno e signi icance le els o 1%, 5%, and 10%, espec i ely.
IKGPR). This esul sugges s ha co po a e isk- aking o
Ko ean-lis ed i ms gene ally inc eases wi h In e -Ko ean
geopoli ical isk. Howe e , he isk- aking a i ude could
be e e ed when he geopoli ical isk ele a es o a ce ain
le el as documen ed by Bo and Lensin (2005).
Fu he mo e, he esul s show ha he coe icien o
IKGPR_SQ is nega i ely and signi ican ly ela ed o
302 Business Resea ch Qua e ly 28(1)
Ma ins, H. C. (2020). The B azilian bank up cy law e o m,
co po a e owne ship concen a ion, and isk- aking.
Manage ial and Decision Economics, 41, 562–573.
Nakano, M., & Nguyen, P. (2012). Boa d size and co po-
a e isk- aking: Fu he e idence om Japan. Co po a e
Go e nance: An In e na ional Re iew, 20(4), 369–387.
O’b ien, R. M. (2007). A cau ion ega ding ules o humb
o a iance in la ion ac o s. Quali y & Quan i y, 41, 673–
690.
Pandi , S., Wasley, C. E., & Zach, T. (2011). The e ec o
esea ch and de elopmen (R&D) inpu s and ou pu s
on he ela ion be ween he unce ain y o u u e ope -
a ing pe o mance and R&D expendi u es. Jou nal o
Accoun ing, Audi ing & Finance, 26(1), 121–144. h ps://doi.
o g/10.1177/0148558X11400583
P a , J. W. (1964). Risk a e sion in he small and in he la ge.
Econome ica, 32, 122–136. h ps://doi.o g/10.1016/B978-
0-12-214850-7.50010-3
Pyo, D. J. (2021). Does geopoli ical isk ma e ? E idence om
Sou h Ko ea. De ence and Peace Economics, 32(1), 87–106.
h ps://doi.o g/10.1080/10242694.2020.1829937
Sa ka , S. (2000). On he in es men –unce ain y ela ionship in
a eal op ions model. Jou nal o Economic Dynamics and
Con ol, 24(2), 219–225. h ps://doi.o g/10.1016/S0165-
1889(99)00005-6
Shapi a, Z. (1995). Risk aking: A manage ial pe spec i e.
Russell Sage Founda ion.
Su, X., Zhou, S., Xue, R., & Tian, J. (2020). Does economic
policy unce ain y aise co po a e p ecau iona y cash
holdings? E idence om China. Accoun ing & Finance,
60(5), 4567–4592. h ps://doi.o g/10.1111/ac i.12674
T an, T. Q. (2019). Economic policy unce ain y and co -
po a e isk- aking: In e na ional e idence. Jou nal o
Mul ina ional Financial Managemen , 52, 100605. h ps://
doi.o g/10.1016/j.mul in.2019.100605
Tübbicke, S. (2022). En opy balancing o con inuous ea men .
Jou nal o Econome ic Me hods, 11(1), 71–89.
Vo, L. V., & Le, H. T. T. (2017). S a egic g ow h op ion,
unce ain y, and R&D in es men . In e na ional Re iew
o Financial Analysis, 51, 16–24. h ps://doi.o g/10.1016/j.
i a.2017.03.002
W igh , P., K oll, M., K ug, J. A., & Pe us, M. (2007). In luences
o op managemen eam incen i es on i m isk aking.
S a egic Managemen Jou nal, 28(1), 81–89. h ps://doi.
o g/10.1002/smj.548
Xu, D., Zhou, K. Z., & Du, F. (2019). De ian e sus aspi a ional isk
aking: The e ec s o pe o mance eedback on b ibe y expend-
i u e and R&D in ensi y. Academy o Managemen Jou nal,
62(4), 1226–1251. h ps://doi.o g/10.5465/amj.2016.0749
Yeoh, S. B., & Hooy, C. W. (2022). Gene a ion e ec s and
manage ial isk aking. Jou nal o Business Resea ch, 139,
918–934. h ps://doi.o g/10.1016/j.jbus es.2021.09.063
Yung, K., & Chen, C. (2018). Manage ial abili y and i m
isk- aking beha iou . Re iew o Quan i a i e Finance
and Accoun ing, 51, 1005–1032. h ps://doi.o g/10.1007/
s11156-017-0695-0
Zhang, G., Han, J., Pan, Z., & Huang, H. (2015). Economic
policy unce ain y and capi al s uc u e choice: E idence
om China. Economic Sys ems, 39(3), 439–457. h ps://doi.
o g/10.1016/j.ecosys.2015.06.003
Zhao, Q., & Pe ci al, D. (2017). En opy balancing is doubly
obus . Jou nal o Causal In e ence, 5(1), 20160010.
Hoang e al. 303
Appendix 1
Table 8. Co ela ion ma ix.
Va iables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19)
(1) SDROA 1.000
(2) SDROE 0.758 1.000
(3) IKGPR_SQ 0.041 0.022 1.000
(4) IKGPR 0.037 0.020 0.983 1.000
(5) MA3_IKGPR 0.037 0.002 0.622 0.585 1.000
(6) IW_IKGPR 0.034 0.017 0.974 0.990 0.591 1.000
(7) DW_IKGPR 0.039 0.023 0.972 0.990 0.567 0.960 1.000
(8) RES_IKGPR 0.026 0.013 0.938 0.949 0.588 0.966 0.913 1.000
(9) SIZE −0.265 −0.152 −0.035 −0.036 −0.050 −0.033 −0.038 −0.023 1.000
(10) ROA −0.381 −0.377 −0.017 −0.015 −0.052 −0.012 −0.017 −0.011 0.223 1.000
(11) LEVERAGE 0.071 0.246 −0.059 −0.058 −0.114 −0.060 −0.054 −0.061 0.166 −0.312 1.000
(12) GROWTH 0.017 0.004 −0.020 −0.031 −0.058 −0.024 −0.037 −0.027 0.020 0.124 −0.011 1.000
(13) CASH 0.115 0.011 0.092 0.093 0.171 0.095 0.088 0.090 −0.189 0.023 −0.305 0.015 1.000
(14) FIXED_ASSETS −0.151 −0.049 −0.057 −0.059 −0.109 −0.060 −0.058 −0.065 0.195 0.016 0.381 −0.047 −0.340 1.000
(15) CASHFLOW −0.206 −0.195 0.011 0.019 −0.015 0.016 0.023 0.004 0.127 0.436 −0.295 0.010 0.132 −0.102 1.000
(16) ZSCORE −0.003 −0.079 0.066 0.065 0.099 0.064 0.065 0.060 −0.124 0.146 −0.323 0.036 0.172 −0.189 0.104 1.000
(17) MTB 0.258 0.280 0.087 0.088 0.129 0.086 0.087 0.081 −0.185 −0.186 0.002 0.078 0.178 −0.174 −0.123 0.315 1.000
(18) SRETURN −0.049 −0.034 −0.021 −0.012 −0.133 −0.025 0.002 −0.024 −0.014 0.153 −0.035 0.129 0.021 0.006 0.089 0.112 0.214 1.000
(19) FIRM_AGE −0.161 −0.097 −0.013 −0.017 −0.022 −0.015 −0.019 −0.012 0.319 0.051 0.061 −0.055 −0.182 0.177 0.030 −0.110 −0.212 −0.003 1.000
This able epo s he pai wise co ela ion ma ix o In e -Ko ean geopoli ical isk measu es and i m-le el a iables used in he baseline model. Bold numbe s ep esen he co ela ion coe icien s ha a e signi ican a 10% le el o
s onge .
