Mielca z, Paweł; S uciński, Ad ian; Osiichuk, Dmy o; Jin, Mickael
A icle
The impac o he Russo-Uk ainian wa on he
in e na ional so e eign deb ma ke : The ole o ade
channel
Con empo a y Economics
P o ided in Coope a ion wi h:
VIZJA Uni e si y, Wa saw
Sugges ed Ci a ion: Mielca z, Paweł; S uciński, Ad ian; Osiichuk, Dmy o; Jin, Mickael (2024) : The
impac o he Russo-Uk ainian wa on he in e na ional so e eign deb ma ke : The ole o ade
channel, Con empo a y Economics, ISSN 2300-8814, Uni e si y o Economics and Human Sciences
in Wa saw, Wa saw, Vol. 18, Iss. 3, pp. 352-364,
h ps://doi.o g/10.5709/ce.1897-9254.542
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/312958
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/
www.ce. izja.pl
352
This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License.
The pape quan i ies he impac o he ou b eak o he Russo-Uk ainian wa on so e eign bond yields o
58 coun ies. Ou indings based on e en s udy me hodology and mul i a ia e c oss-sec ional eg ession
analysis highligh he salien ole o ade channel in shaping he ma ke s’ eac ion o he hos ili ies. The up ick
in yields was highe in coun ies wi h a g ea e sha e o ag icul u al aw ma e ials impo s om Uk aine. No
o he pe sis en c oss-geog aphy di e ences in ma ke eac ions we e ound. The s eng h o he eac ion
was con ingen on he indi idual coun ies’ mac oeconomic condi ions. The pos -e en cumula i e abno mal
e u ns exhibi ed a posi i e associa i e link wi h he baseline in la ion in he s udied coun ies poin ing o
he in la iona y p essu e po ended by he ou b eak o wa . No s a is ically signi ican links be ween he ex
an e si ua ion on he labo ma ke and he subsequen bond ma ke eac ion we e obse ed. A he ou se o
hos ili ies, he dynamics o yields we e d i en by he pe cei ed likelihood o escala ion, while a la e s ages,
he ade channel appea s o ha e shaped he ma ke esponse. The pape p o ides insigh s in o he ac o s
shaping he spillo e e ec s o he wa on ixed-income ma ke s and quan i ies he speed o ma ke adjus -
men in esponse o an ex e nal shock.
1. In oduc ion1. In oduc ion
The Russian in asion o Uk aine which s a ed on
Feb ua y 24, 2022, has had p o ound consequences o
he global economy. The sanc ions imposed on Russia
by he in e na ional communi y and he blockade o
a signi ican pa o ma i ime ade in commodi ies
caused an inc ease in global in la iona y p essu e. The
hos ili ies also con ibu ed o an ele a ed sys emic
isk ac oss a numbe o economies which we e pe -
cei ed as ulne able o he po en ially dis up i e im-
pac o he wa .
Since bo h Uk aine and Russia a e among he
wo ld's op expo e s o ag icul u al commodi ies,
and Russia is one o he mos impo an expo e s o
ene gy commodi ies, he dis up i e impac o he wa
on he global economy has been p ima ily ope a ing
h ough he ade channel (Fang & Shao, 2022). As he
global dependence on ossil uels emains high wi h
al e na i e ene gy sou ces expe iencing apid g ow h
om a low base (Boubake e al., 2022), he supply
bo lenecks in he commodi y ma ke s ha e had a
p o ound impac on he implemen a ion and p io i-
ies o iscal and mone a y policies.
The Impac o he Russo-Uk ainian Wa on he
In e na ional So e eign Deb Ma ke : The Role o
T ade Channel
ABSTRACT
F10, F34, F40.
KEY WORDS:
JEL Classi ica ion:
so e eign deb , in e na ional ade, ixed-income analysis, sys emic isks.
1
Depa men o Finance, Kozminski Uni e si y, 03-301, 57/59 Jagiellonska S ., Wa saw, Poland
2
Depa men o Accoun ing, Kozminski Uni e si y, 03-301, 57/59 Jagiellonska S ., Wa saw, Poland
3
Una ilia ed, Wa saw, Poland
Co espondence conce ning his a icle should be add essed o:
Dmy o Osiichuk,
Depa men o Finance, Kozminski Uni e si y,
03-301, 57/59 Jagiellonska S ., Wa saw, Poland
.
E-mail: [email protected]
Paweł Mielca z1 , Ad ian S uciński2 , Dmy o Osiichuk1 , and Mickael Jin3
P ima y submission: 29.08.2023 | Final accep ance: 03.02.2024
353
Paweł Mielca z , Ad ian S uciński, Dmy o Osiichuk, Mickael Jin
10.5709/ce.1897-9254.542DOI: CONTEMPORARY ECONOMICS
Vol. 18 Issue 3 352-3642024
P io empi ical s udies ha e shown ha he Russo-
Uk ainian wa has exe cised a p o ound impac on he
global capi al ma ke s (Boubake e al., 2022; Bassene
e al., 2022; Ahmad e al., 2022), commodi y ma -
ke s (Fang & Shao, 2022; Cui e al., 2023), and e en
al e na i e in es men s (Goodell e al., 2022). The
con lic has impac ed he decision-making pa e ns o
in e na ional in es o s (Pa el e al., 2022; Glambosky
& Pe e bu gsky, 2022; She h & Uslay, 2023) and p e-
cipi a ed a e-e alua ion o sys emic isks accompany-
ing he c oss-bo de lows o capi al. The coun ies
which ei he bo de Russia o a e hea ily dependen
on Russia o Uk aine in hei supply o ag icul u al o
ene gy commodi ies ha e expe ienced an inc ease in
he pe cei ed le el o sys emic isk. The la e ca ies
po en ially dis up i e consequences o hei domes-
ic economic policies.
The p esen s udy seeks o in es iga e he impac
o he Russo-Uk ainian wa on coun y-speci ic isks
h ough he obse a ion o wa -associa ed cumula i e
abno mal e u ns o so e eign bonds.
By ocusing on he de e minan s shaping he so -
e eign deb ma ke eac ion o he ou b eak o hos-
ili ies, we a emp o elucida e he ole o he ade
channel in he p opaga ion o sys emic isk. Relying
on a c oss-sec ional analysis o cumula i e abno mal
e u ns on so e eign bonds issued by 58 coun ies, we
e idence ha coun ies wi h a highe dependence on
Uk aine in hei supply o ag icul u al commodi ies
expe ienced a ela i ely s onge up ick in so e eign
bond yield. A he same ime, he s eng h o he ma -
ke eac ion is shown o be con ingen on he indi-
idual coun ies’ mac oeconomic condi ions such as
he baseline le el o in la ion and he gene al le el o
unemploymen . Fu he mo e, we no e ha du ing he
ea ly s ages o he con lic , he obse ed dynamics o
he bond yields we e s ongly skewed by he pe cei ed
likelihood o a po en ial con lic escala ion beyond he
Uk ainian bo de .
2. Li e a u e Re iew2. Li e a u e Re iew
Mili a y con lic s exe cise a p o ound impac
on inancial ma ke s and ca y epe cussions o
in es o s ac oss all asse classes (Izzeldin e al., 2023;
Pandey e al., 2023). The spillo e e ec s ope a e
ac oss a numbe o ansmission mechanisms
(Qu eshi e al., 2022; Tae a e al., 2023). To s a wi h,
mili a y hos ili ies may cause ma ke panic and a
‘ ligh o sa e y’ (Ahmed, 2023) esponse on he pa
o in es o s. As a esul o an inc ease o expec ed isk
p emia (G ana e al., 2023), ma ke alua ions may
umble, while he asse s pe cei ed as less exposed
o he possible epe cussions o he con lic o likely
o bene i he e om may expe ience an addi ional
in low o capi al (Singh e al., 2022). Wi hin he
coun ies bo de ing ju isdic ions in ol ed in he
con lic , expec a ions o possible secu i y p oblems,
p opaga ion o he con lic , and bo de acciden s
may also p ecipi a e he ealloca ion o capi al
owa ds a sa e en i onmen (Kuma i e al., 2023).
The o ecas ing windows sh ink, and in es o s may
be inclined o pos pone long- e m capi al alloca ion
decisions.
The p esen s udy a emp s o quan i y he sho -
e m easu y bonds ma ke esponse o he ou b eak
o wa and in es iga es he ac o s, which a e
associa ed wi h he magni ude o yield shi s. The eby,
we y o cla i y, which ac o s may be guiding capi al
ealloca ion and pinpoin he de e minan s o isk
p emia unde condi ions o an open mili a y con lic .
The e a e se e al di ec quan i iable channels,
h ough which he impac o he hos ili ies ope a es
ac oss he inancial ma ke s. The mos impo an one,
h ough which he wa ’s e ec s spill o e bo de s,
is he ade channel (Es ada & Kou onas, 2022;
S einbach, 2023). La ge-scale con lic s a e always
associa ed wi h a decline in in e na ional ade. As
a esul , he magni ude o possible epe cussions o
he coun ies no di ec ly in ol ed in he con lic
depends on he composi ion o he dis up ed ade
lows (Ma kus, 2022). Commodi y ma ke s appea
o be he mos ulne able o he dis up i e impac
o wa s (Chish i e al., 2023; Cui e al., 2023). Supply
con ac ions, which canno be compensa ed wi hou
a signi ican ime lag, exe cise upwa d p essu e on
p ice, which in u n ansla es in o a cascade o
adjus men s ac oss he supply chains. A e a ce ain
pe iod o ime, hese spillo e e ec s become he
p incipal engines o in la iona y p essu e p omp ing
an adjus men o he mac oeconomic policies o he
a ec ed coun ies.
Bo h Russia and Uk aine a e sys emically
c ucial expo e s o commodi ies. Russia is a majo
p oduce o oil and na u al gas, as well as a numbe
www.ce. izja.pl
354
The Impac o he Russo-Uk ainian Wa on he In e na ional So e eign Deb Ma ke : The Role o T ade Channel
This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License.
o ag icul u al commodi ies. Uk aine p o ides o a
signi ican po ion o ag icul u al commodi y ma ke s
(Abay e al., 2023). Bo h ene gy and ag icul u al
commodi ies ha e expe ienced a signi ican upsho
in spo and u u es p ices ollowing he ou b eak o
hos ili ies. O e all, du ing he ini ial mon hs o he
wa , whea p ices inc eased by abou 35%, and na u al
gas p ices inc eased by abou 20% om an al eady
high base. O e ime, hese upsho s pa ially subsided.
E en ually, he pe u ba ions on commodi ies ma ke s
ickled down h ough he supply chains (Uma e al.,
2022) in o in la ion igu es. As a esul , ollowing a
s a k inc ease in in la ion p ojec ions a e a ce ain
ime lag, cen al banks ac oss nume ous coun ies
ha e ini ia ed a wa e o mone a y policy igh ening
aimed a s i ling he upsu ge o in la ion (Chibane
& Kuhana han, 2023). The esul ing al e a ions in
mone a y policies combined wi h a change o in es o
sen imen and consume expec a ions ha e led o an
adjus men on he inancial ma ke s.
The p esen s udy ocuses only on he sho - e m
ma ke eac ion o he ou b eak o hos ili ies. Wi hin
he s udied ime ame, only in es o sen imen
shapes he di ec ion o eadjus men s ac oss majo
asse classes. P ice luc ua ions a e p ima ily d i en
by ealloca ion o capi al guided by he shi s in he
pe cei ed le els o isk. The p esen s udy a emp s
o elucida e, which undamen al ac o s shaped hose
expec a ions, i.e., which ex an e ac o s co ela e
wi h he magni ude o sho - e m p ice shi s on he
selec ed inancial ma ke s.
The ex an empi ical li e a u e delinea es se e al
ac o s associa ed wi h he magni ude o he easu y
ma ke eac ion o he ou b eak o mili a y con lic .
To s a wi h, he scale o he con lic appea s o be
s ongly associa ed wi h in es o s’ esponse. Bo h
he size o economies in ol ed in he con lic and
he p obabili y o con lic escala ion appea o be
d i ing he dynamics o easu y yields in bo h
pa icipa ing and non-pa icipa ing coun ies. The
magni ude o he esponse is documen ed o be la ge
in coun ies, which a e pe cei ed likely o be angled
in he con lic , su e colla e al damage, expe ience
spillo e e ec s such as e ugee c ises, inc eased a ms
a ic, illici ade (Tong, 2024). The geog aphical
p oximi y o he bellige en coun ies is also e idence
o be an impo an ac o (Sil a e al., 2023). P ices
o easu ies a e also likely o dec ease i ma ke s
an icipa e a subs an ial expansion in go e nmen
expendi u es in esponse o he con lic ega dless o
whe e he addi ional inancing is likely o be di ec ed.
The p esen s udy analyzes wo o he po en ial
de e minan s o he easu y ma ke eac ion in
esponse o he mili a y con lic . The i s one is he
an icipa ed impac o he con lic on in e na ional
ade lows, pa icula ly o hose coun ies, which
a e likely o be di ec ly o indi ec ly a ec ed by
he ou b eak o hos ili ies. The second ac o is
mac oeconomic policy mix and mac oeconomic
se ings p e ailing in he a ec ed coun ies p io o
he wa 's ou b eak.
P io s udies in es iga ed he impac o ade low
dis up ions on easu y yields. The e is a s a is ically
signi ican obse able pa e n, whe eby bond yields
may inc ease in esponse o he an icipa ed o ac ual
dis up ion o ade lows (Chen e al., 2023). The
magni ude o he yields inc ease is con ingen upon
a numbe o quan i iable ac o s. To s a wi h, yields
may eac mo e s ongly i he dis up ed ade lows
a e comp ised o commodi ies, in which he a ec ed
coun y is de icien (Nazlioglu e al., 2020). This is
because he likelihood o sho ages is associa ed wi h
an an icipa ion o inc eased in la ion, which in u n
is likely o igge a esponse on he pa o mone a y
au ho i ies in he o m o inc eased in e es a es
(Abbas & Lan, 2020). The la e di ec ly ansla e
in o bond alua ions. The second ac o which may
be signi ican ly associa ed wi h he magni ude o he
ma ke eac ion is he pe cei ed leng h o he con lic .
I he dis up ion o ade is likely o be sho - e m, he
empo a y supply-side sho ages may be compensa ed
by o he ma ke pa icipan s o by a ge ed sho - e m
a ioning policies. Howe e , long- e m dis up ions
a e likely o be associa ed wi h signi ican adjus men
cos s and ma ke ic ions (Fe guson & Ubila a, 2022;
Zhou e al., 2023), all o which is likely o push up
in la iona y p essu e. The hi d ac o , which may be
shaping he scale o easu y yield shi s ollowing he
ou b eak o wa , is he pe cei ed impac o he con lic
on he dynamics o GDP g ow h (Liadze e al., 2023).
The moun ing unce ain y (He, 2023) usually esul s in
a educ ion o agg ega e consump ion and in es men
expendi u es, which may esul in dep essed
economic ac i i y. Whe eas s agna ion coupled wi h
355
Paweł Mielca z , Ad ian S uciński, Dmy o Osiichuk, Mickael Jin
10.5709/ce.1897-9254.542DOI: CONTEMPORARY ECONOMICS
Vol. 18 Issue 3 352-3642024
de la ion may be ackled wi h a ge ed iscal s imulus
and, whene e possible, mone a y easing, s ag la ion
is likely o esul in mone a y policy igh ening and
concomi an inc ease in easu y yields. The esul ing
ne impac on easu y ma ke depends on he
combina ion o ou pu and p ice dynamics.
While i seems clea ha he scale o easu y
ma ke eac ion o ex e nal shocks may be con ingen
on he ini ial mac oeconomic condi ions in he
a ec ed coun y (Moench & Soo i-Sia ash, 2022),
he ac ual empi ical e idence in ha ega d appea s
sca ce. P io s udies sugges ha easu y ma ke s o
coun ies wi h weake economies cha ac e ized by
ele a ed le els o cyclical o s uc u al unemploymen
a e mo e ulne able o ex e nal shocks and
expe ience subs an ial luc ua ions unde un o eseen
ci cums ances. The magni ude o his condi ional
ela ionship wi h key mac oeconomic a iables,
howe e , emains unquan i ied. Ele a ed in la ion,
be i he esul o supply-side shocks o o issues wi h
he independence o he cen al bank (Zhang e al.,
2023), is also con ibu ing o he ulne abili y o he
easu y ma ke . Unde uns able mac oeconomic
condi ions, easu y yields ha e been shown o
expe ience signi ican shi s in esponse o signaling
e en s (Ozbekle e al., 2021), including policy
changes o eshu les o senio public o icials.
The quan i a i e impac o an ou b eak o mili a y
con lic has, howe e , no been a subjec o empi ical
in es iga ion. The p esen pape a emp s o ill in his
gap by s udying how easu y bond yields eac o an
ou b eak o wa unde he e ogeneous mac oeconomic
condi ions wi hin he a ec ed coun ies.
3. Me hodology 3. Me hodology
In o de o quan i y he impac o he ou b eak
o he Russo-Uk ainian wa on he so e eign
deb ma ke , we ely on e en s udy me hodol-
ogy (B ad o d & Da id-Robison, 1997; Guidolin
& La Fe a a, 2010; Leigh e al., 2003). A de ailed
p esen a ion o he me hodology can be ound in
MacKinlay (1997) and Campbell e al. (1997). The
p incipal objec i e o an e en s udy consis s in
measu ing he isk-adjus ed e u n on an unde ly-
ing secu i y du ing a speci ic e en window chosen
by an in es iga o .
In his s udy, di e en e en windows we e used
o assess whe he he choice o a pa icula ime
in e al a ec s he esul s ob ained in he empi i-
cal model. Day 0 ( he e en da e) was assumed o
be he day o he Russia's in asion o Uk aine, i.e.
Feb ua y 24, 2022. The longes e en window ad-
op ed was [-40; 40] and he sho es was [-1; 1]. The
longe e en window, he e o e, co e s he pe iod
om Decembe 30, 2021 o Ap il 21, 2022, while
he in e al om Janua y 5, 2021 (day -297) o De-
cembe 29, 2021 (day -41) was adop ed as he es i-
ma ion window. The es ima ion window, he e o e,
includes 257 daily obse a ions.
In line wi h e en s udy me hodology, we i s
calcula ed he ac ual daily changes in yields (YTM)
on 10-yea go e nmen bonds (Ac R
ij
) o he se-
lec ed coun ies:
(1)
whe e 10
y
YTM
ij
is 10-yea YTM on easu y bonds
o coun y j on day i, ex ac ed om he da abase
Cbonds. Due o lack o da a o some coun ies o
p olonged pe iods wi hou any ansac ions in go e n-
men bonds, some ins umen s we e excluded om
he s udy. Finally, da a o 58 coun ies we e used in
he cumula i e a e age abno mal e u ns (CAAR) cal-
cula ion.
In he nex s ep, abno mal changes in yields (AR
ij
)
on each day o he e en window, de ined as he di -
e ence be ween he ac ual change (Ac R
ij
) and he ex-
pec ed change (A R
j
), we e es ima ed.
(2)
whe e A R
j
is a e age ac ual daily change o 10y
YTM on bonds o coun y j obse ed du ing he es i-
ma ion window:
(3)
whe e
0
ma ks he beginning o he es ima ion win-
dow (in his s udy,
0
=-297), while
1
'-1 is he day p e-
ceding he ea lies e en window (in his s udy,
1
'-1=
-41). In o de o educe he impac o ou lie s, he da a
we e winso ized a 2,5 h and 97,5 h pe cen ile.
The cumula i e abno mal e u ns (CAR) on he
bonds o coun y j o e he in e al [
1
;
2
] we e es i-
www.ce. izja.pl
356
The Impac o he Russo-Uk ainian Wa on he In e na ional So e eign Deb Ma ke : The Role o T ade Channel
This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License.
ma ed as ollows:
(4)
The cumula i e a e age abno mal e u ns (CAAR)
we e hen compu ed as ollows:
(5)
whe e N is he o al numbe o ins umen s consid-
e ed in he e en s udy.
In o de o e i y he s a is ical signi icance o he
CAAR, a pa ame ic es ounded on he a iance es-
ima ed on he basis o he c oss sec ion o cumula i e
abno mal e u ns, was pe o med:
(6)
(7)
The s udy elies on he analysis o changes in bond
yields (YTM) a he han o daily changes in bond
p ices due o he ollowing ac o s:
he i s objec i e o he s udy is o de e mine
how he ou b eak o wa in Uk aine has a ec ed he
cos o inancing o indi idual coun ies, a he han
how he e en unde s udy has a ec ed sho - e m
yields ea ned by go e nmen bondholde s,
he second, and key, objec i e o he s udy is o
assess whe he he deg ee o a coun y's economic
linkage wi h Uk aine has an impac on he absolu e
change in he cos o unding. We a e, he e o e, in-
e es ed in whe he he magni ude o he absolu e
change in he cos o long- e m deb depends on a
coun y's economic ela ionship wi h Uk aine.
The second impo an me hodological choice
made in he s udy is he use o absolu e changes in
bond yields ins ead o ela i e changes. This choice
is dic a ed by he need o elimina e he isk o a -
ibu ing disp opo iona ely la ge ela i e changes o
coun ies wi h high c edi a ings. Fo example, i one
compa es wo coun ies whe e one is a ed AAA and
he o he BB, o he i s coun y, e en small absolu e
changes in he yield a e will esul in e y la ge ela-
i e ola ili y, while o he second coun y, simila
absolu e shi s in yield will esul in compa a i ely
low ela i e changes. Since ma ke ola ili y a imes
such as he ou b eak o wa is ypically signi ican , an
analysis o ela i e changes in yields could esul in
a ibu ing much la ge inc eases in he cos o deb
o coun ies wi h high c edi a ings. Hence, in his
s udy we ocus on analysing he absolu e changes in
coun ies' long- un unding cos s esul ing om he
ou b eak o wa in Uk aine.
A he second s age o empi ical analysis, we a -
emp o es ablish which ac o s exe cise a signi i-
can impac on he magni ude and di ec ion o he
so e eign deb ma ke eac ion o he ou b eak o
wa . To ha end, we un a se o linea eg ession
models, whe e he explana o y a iable is he cumu-
la i e changes in bond yields (CAR[
1
;
2
],j). The se o
explana o y a iables includes p oxies o a coun y's
size and p incipal mac oeconomic pa ame e s, such
as he le el o go e nmen deb , he baseline in la ion
a e, he unemploymen a e, he budge de ici . The
p incipal expe imen al a iables subjec o analysis
ela e o he subjec coun ies’ ulne abili y o he sys-
emic isks c ea ed by he ou b eak o hos ili ies and
p opaga ing h ough he ade channel. In pa icula ,
we in oduced a iables encoding he geog aphical
p oximi y o he s udied coun ies o Uk aine and
Russia (Uk Bo de , RusBo de ), hei dependence on
he supply o commodi ies and o he p oduc s om
Uk aine (Ag _impo %20), and o he ele an geo-
poli ical ac o s, which in luence he pe cei ed le el
o sys emic isk o a gi en coun y. The las ca ego y
o a iables includes bina y eg esso s which encode
he coun y’s being a membe o he Eu opean Union
(EU) and NATO. Da a on GDP, in la ion, budge bal-
ance and unemploymen we e ex ac ed om he da-
abases Cbonds and Ou Wo ld in Da a, while da a
on he scale o ade ope a ions wi h Uk aine we e
ob ained om he Wo ld Bank da abase.
The ull lis o a iables used in he eg ession
analysis is p esen ed in Table 1. The summa y s a is-
ics a e p esen ed in Table 2. The baseline eg ession
model es ed in his s udy has he ollowing speci ica-
ion:
(8)
357
Paweł Mielca z , Ad ian S uciński, Dmy o Osiichuk, Mickael Jin
10.5709/ce.1897-9254.542DOI: CONTEMPORARY ECONOMICS
Vol. 18 Issue 3 352-3642024
4. Empi ical Findings4. Empi ical Findings
Table 3 and Figu e 1 show he a e age abno mal
changes (AAR) in bond yields o each day o he
e en window. Jus be o e he ou b eak o wa , we
obse e a s a is ically signi ican inc ease in bond
yields. I is wo h no ing, howe e , ha he e was
an obse able dec ease in bond yields immedia ely
a e he ou b eak o wa in Uk aine ( he a e age
yield changes on days = 2 and = 3 a e nega i e
and s a is ically signi ican ). On subsequen days,
howe e , bond yields s a ed o inc ease ma kedly.
Table 4 shows he CAARs es ima ed o di e en
e en windows, which con i m ou ini ial indings.
The ou b eak o hos ili ies was associa ed wi h a
s a is ically signi ican inc ease in so e eign bond
yields ac oss he analyzed coun ies.
The esul s o mul i a ia e econome ic analysis
o he de e minan s o cumula i e abno mal e-
u ns on go e nmen bonds a e p esen ed in Tables
5 and 6. In he eg ession equa ions summa ized in
Table 1
De ini ions o Explana o y Va iables Used in a Linea Reg ession Model
Va iable name De ini ion
lnGDP20 Na u al loga i hm o GDP (in USD bn.) o 2020
Deb _ o_GDP20 Deb o GDP in 2020 (in %)
In la ion20 In la ion in 2020 (in %)
Budge _balance_ o_
GDP20
Budge balance (su plus o de ici ) o GDP in 2020
Unemploymen 20 Unemploymen in 2020 (in %)
Ag _impo %20 The a io o he alue o impo s o ag icul u al aw ma e ials om Uk aine o he coun y's
o al impo s o ag icul u al aw ma e ials in 2020
NATO Bina y a iable, whe e 1 indica es a NATO coun y
Uk Bo de Bina y a iable, whe e 1 indica es a coun y bo de ing Uk aine
RusBo de Bina y a iable, whe e 1 indica es a coun y bo de ing Russia
EU Bina y a iable, whe e 1 indica es a membe coun y o he Eu opean Union
De eloped Bina y a iable, whe e 1 indica es a de eloped economy pe he UN classi ica ion
Table 2
Desc ip i e S a is ics o Explana o y Va iables
Va iable Mean S . de ia ion Minimum Maximum
lnGDP20 6.0066 1.5087 2.3571 9.9472
Deb _ o_GDP20 78.516 44.722 24.8 266.2
In la ion20 1.8773 2.9304 -2.5403 13.246
Budge _balance_ o_GDP20 -7.483 3.5862 -14.9 4
Unemploymen 20 6.9521 5.4974 0.14 33.3
Ag _impo %20 0.011061 0.023175 0 0.12202
NATO 0.37931 0.48945 0 1
Uk Bo de 0.068966 0.25561 0 1
RusBo de 0.051724 0.2234 0 1
EU 0.37931 0.48945 0 1
De eloped 0.53448 0.50317 0 1
www.ce. izja.pl
358
The Impac o he Russo-Uk ainian Wa on he In e na ional So e eign Deb Ma ke : The Role o T ade Channel
This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License.
Table 3
A e age Abno mal Bond Yield Changes Wi hin an E en Window
Day AAR Tes
s a is ic
Day AAR Tes
s a is ic
Day AAR Tes
s a is ic
-40 0.00 -0.603 -13 0.02*** 3.990 14 0.00 0.407
-39 0.00 -0.658 -12 0.03*** 4.635 15 -0.01** -2.211
-38 0.03*** 4.605 -11 -0.03*** -3.780 16 -0.01** -1.042
-37 0.03*** 4.518 -10 0.03*** 4.534 17 0.05*** 6.133
-36 0.00 0.320 -9 0.01** 0.891 18 0.06*** 7.919
-35 0.03*** 7.428 -8 0.01** 1.689 19 0.01 0.560
-34 0.02*** 4.750 -7 0.01** 1.913 20 0.05*** 5.641
-33 0.02*** 3.925 -6 -0.01** -2.050 21 0.03*** 4.991
-32 0.00** -0.883 -5 -0.02*** -5.255 22 -0.01** -0.731
-31 -0.03*** -4.909 -4 -0.02*** -3.192 23 -0.03** -2.188
-30 -0.03*** -4.604 -3 0.00 0.086 24 -0.02** -2.455
-29 0.02*** 4.045 -2 0.02*** 3.927 25 -0.06*** -5.301
-28 0.03*** 5.748 -1 0.02*** 3.033 26 0.02** 1.562
-27 0.02** 2.017 0 0.01 0.584 27 -0.02** -2.198
-26 0.00 -0.081 1 0.00 -0.141 28 0.05*** 6.617
-25 -0.04*** -4.499 2 -0.02** -1.700 29 0.07*** 6.009
-24 -0.03*** -5.905 3 -0.08*** -6.040 30 0.02*** 3.768
-23 -0.01** -1.112 4 0.04*** 4.192 31 0.05*** 5.148
-22 0.01** 1.055 5 0.01** 0.927 32 0.05*** 7.081
-21 0.01** 1.929 6 0.01 0.547 33 -0.02** -1.556
-20 0.02*** 3.504 7 0.05*** 3.287 34 -0.01** -1.876
-19 0.00 -0.249 8 0.08*** 5.860 35 0.02** 1.921
-18 0.01** 2.537 9 0.03** 2.327 36 0.01*** 3.455
-17 -0.01** -1.753 10 0.05*** 5.372 37 0.01*** 2.862
-16 0.00 -0.371 11 0.00 -0.013 38 0.04*** 4.788
-15 0.03*** 3.640 12 0.07*** 7.881 39 -0.03*** -3.935
-14 0.04*** 4.059 13 -0.02*** -3.482 40 0.02*** 3.280
***, **, * indica es s a is ical signi icance a he 1%, 5%, 10% le el espec i ely. AAR – a e age abno mal change in bond
yields.
Table 5, he explained a iable is he cumula i e ab-
no mal e u n (CAR) obse ed o e e en windows
p eceding he ou b eak o hos ili ies. In Table 6, we
p esen he indings o eg ession models ea u -
ing he cumula i e abno mal e u ns ollowing he
commencemen o he wa as he explained a i-
able. All models exhibi sa is ac o y econome ic
p ope ies (F-s a is ic) and can se e as a basis o
alid s a is ical in e ence. I is wo h no ing ha we
an eg essions o a numbe o di e en e en win-
dows. Since he esul s we e quali a i ely simila
ac oss he majo i y o e en windows, only selec ed
esul s we e included in he inal e sion o he pa-
pe wi h he emainde a ailable o e iew.
We documen a signi ican ole o he ade chan-
nel in shaping he ma ke eac ion o he ou b eak
o wa . The a iables measu ing he s udied coun-
ies’ dependence on he supply o commodi ies
om Uk aine exhibi a pe sis en ly signi ican posi-
i e associa i e link wi h he cumula i e abno mal
359
Paweł Mielca z , Ad ian S uciński, Dmy o Osiichuk, Mickael Jin
10.5709/ce.1897-9254.542DOI: CONTEMPORARY ECONOMICS
Vol. 18 Issue 3 352-3642024
Table 4
CAARs o Va ious E en Windows.
E en window CAAR Tes s a is ic E en window CAAR Tes s a is ic
[-40; 40] 0.795*** 10.260 [-1; 30] 0.453*** 10.437
[-40; 30] 0.653*** 10.328 [-1; 40] 0.595*** 13.824
[-40; 20] 0.593*** 10.574 [-5; 10] 0.164*** 3.903
[-40; 10] 0.388*** 7.129 [-5; 20] 0.369*** 6.822
[-40; 5] 0.175*** 3.305 [-5; 30] 0.429*** 10.056
[-40; 1] 0.23*** 4.307 [-5; 40] 0.571*** 13.124
[-30; 30] 0.557*** 9.696 [-10; 20] 0.422*** 7.813
[-30; 20] 0.497*** 9.952 [-10; 30] 0.481*** 10.970
[-30; 10] 0.292*** 6.242 [-10; 40] 0.624*** 13.378
[-30; 5] 0.079** 1.769 [-20; 30] 0.587*** 11.995
[-30; 1] 0.134*** 2.919 [-20; 40] 0.73*** 12.899
[-20; 20] 0.528*** 10.677 [-30; 40] 0.699*** 10.092
[-20; 10] 0.322*** 7.943 [-40; -1] 0.223*** 3.718
[-20; 5] 0.109*** 3.689 [-40; -5] 0.2*** 3.314
[-20; 1] 0.164*** 5.222 [-40; -10] 0.206*** 3.548
[-10; 10] 0.216*** 5.064 [-30; -1] 0.127*** 2.460
[-10; 5] 0.003 0.133 [-30; -5] 0.104** 2.004
[-10; 1] 0.058*** 3.310 [-30; -10] 0.11** 2.240
[-5; 5] -0.049** -1.890 [-20; -1] 0.157*** 4.441
[-5; 1] 0.006 0.354 [-20; -5] 0.134*** 3.823
[-1; 1] 0.03** 1.770 [-20; -10] 0.14*** 4.196
[-1; 5] -0.025 -0.995 [-10; -1] 0.051*** 3.404
[-1; 10] 0.188*** 4.580 [-10; -5] 0.028** 2.025
[-1; 20] 0.393*** 7.139
***, **, * indica es s a is ical signi icance a he 1%, 5%, 10% le el espec i ely.
Figu e 14
Daily AARs O e he Obse a ion Window.
