Muelle , Ma ius; Swoboda, Be nha d
A icle — Published Ve sion
De e minan s o Wholly Owned Fo eign Di ec
In es men s in E-Comme ce Fi ms: A Hie a chical Coun y-
and Fi m-Le el Analysis
Managemen In e na ional Re iew
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Muelle , Ma ius; Swoboda, Be nha d (2025) : De e minan s o Wholly Owned
Fo eign Di ec In es men s in E-Comme ce Fi ms: A Hie a chical Coun y- and Fi m-Le el Analysis,
Managemen In e na ional Re iew, ISSN 1861-8901, Sp inge , Be lin, Heidelbe g, Vol. 65, Iss. 4, pp.
699-736,
h ps://doi.o g/10.1007/s11575-025-00575-7
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/330650
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/
Vol.:(0123456789)
Managemen In e na ional Re iew (2025) 65:699–736
h ps://doi.o g/10.1007/s11575-025-00575-7
RESEARCH ARTICLE
De e minan s o Wholly Owned Fo eign Di ec
In es men s inE‑Comme ce Fi ms: AHie a chical Coun y‑
andFi m‑Le el Analysis
Ma iusMuelle 1· Be nha dSwoboda1
Recei ed: 6 Augus 2024 / Re ised: 11 Feb ua y 2025 / Accep ed: 14 Ma ch 2025 /
Published online: 15 May 2025
© The Au ho (s) 2025
Abs ac
Despi e hei digi al na u e, leading and globally expanding e-comme ce i ms
es ablish a physical p esence in selec coun ies o e ime o implemen ull local
ope a ions and managemen . Howe e , we know su p isingly li le abou he d i -
e s o such wholly owned o eign di ec in es men decisions in ce ain coun ies,
while o he coun ies a e se ed i ually. This s udy p oposes a amewo k o he
speci ic, business model- ela ed i m- and hos -coun y an eceden s o e-comme ce
i ms’ wholly owned o eign di ec in es men . We employ no el da a conce ning
1,726 ope a ion modes chosen by he 241 leading e-comme ce i ms in Eu ope om
2010–2020 and apply mul ile el modeling. In con as o manu ac u ing i ms, he
insigh s e eal how e-comme ce i ms le e age p e iously unknown online expe-
ience capabili ies o in o m o eign di ec in es men decisions, challenging he
cu en ly ques ioned ole o expe ience in in e na ional business esea ch on digi-
al i ms. Mo eo e , business model- ela ed coun y-le el an eceden s, such as
hos coun y logis ics pe o mance, he in e ne use popula ion, o he ule o law,
also explain such decisions di e en ly. These indings con ibu e o he eme g-
ing esea ch on digi al i ms’ in e na ionaliza ion and ha e di ec implica ions o
e-comme ce manage s in e es ed in he d i e s o o eign di ec in es men bu also
o manu ac u e s. The la e inc easingly use e-comme ce i ms o hei own in e -
na ional expansion, whe eas policymake s aim o a ac g owing e-comme ce i ms
o es ablish hei own p esence in hos coun ies.
Keywo ds E-comme ce i ms· Fo eign di ec in es men · Logis ics pe o mance·
In e ne use· Rule o law· In e na ional online expe ience· Mul ile el modeling
The au ho s hank h ee e iewe s o hei e y help ul andcons uc i e commen s. A p elimina y
e sion o he manusc ip was awa ded he bes con e ence pape in he In e na ionaliza ion P ocess
o SMEs and In e na ional En ep eneu ship T ack a he Annual Con e ence o he Eu opean
In e na ional Business Academy, 2023 in Lisbon.
Ex ended au ho in o ma ion a ailable on he las page o he a icle
700
M.Muelle , B.Swoboda
1 In oduc ion
Dynamic digi aliza ion has spa ked a deba e on i s impac on IB esea ch and
heo y (e.g., S allkamp e al., 2023; Yang e al., 2025). While mos i ms a e
a ec ed, i ual expansion s a egies a e pa icula ly e iden among e-comme ce
i ms (ECFs), i.e., i ms selling physical goods ia own online shops o ma ke -
places o consume s ( eaching 23% o global e ail sales in 2027; ITA, 2024).
They mos ly in e na ionalize i ually, ha e immedia e access o ma ke s a low
cos , and also o e new digi al expansion oppo uni ies o manu ac u e s (e.g.,
Gong e al., 2024). Howe e , leading ECFs add selec i e physical p esences o e
ime. Fo example, se ing 150 coun ies i ually, Amazon (2024), is p esen in
21 coun ies wi h wholly owned o eign di ec in es men s (FDI, i.e., websi es
wi h a local domain, language, cu ency, and in es men s in own physical asse s,
such as wa ehouses and subsidia ies). Alibaba (2024) has FDI in 11 coun ies;
Raku en (2024) has FDI in ou coun ies. Such selec i e choices o e ECFs local
ad an ages such as pene a ion o ma ke s h ough adap ed o e s and same-day-
deli e y se ices, local images, o managemen (e.g., Ba sakis e al., 2023a; Hua
& Wu, 2024). IB esea ch assumes ha digi aliza ion educes he impo ance o
i ms’ expe ience o coun y ba ie s and acili a es apid i ual in e na ionaliza-
ion (e.g., Schu e al., 2016). Howe e , his nascen esea ch hasha dly add essed
ECFs’ FDI, which is also impo an o digi al i ms. This s udy p o ides a i s
heo y-based unde s anding o ECFs’ FDI and helps es ablish an empi ical base-
line o esea ch on digi al i ms’ in e na ionaliza ion (S allkamp e al., 2023;
Ve beke e al., 2018). ECFs’ speci ic digi al and physical business model a ec s
ECFs’ FDI di e en ly han he p oduc ion-based FDI o manu ac u e s does (e.g.,
B ou he s e al., 2022). Thus, we seek no el insigh s by linking ECFs’ i ual
and physical business model elemen s, i.e., ye unknown speci ic online lea ning
capabili ies and hos coun y ba ie s in FDI. Ex ending e ined in e na ionaliza-
ion p ocess models, we explo e why ECFs pu sue FDI despi e i ual op ions o
low coun y ba ie s and whe he i ual p esence p o ides su icien lea ning o
make FDI decisions (e.g., Swoboda & Sinning, 2022).
Empi ical s udies ha e mainly add essed FDI among manu ac u e s (e.g.,
me a-analyses by Cue o‐Cazu a e al., 2023; Wan e al., 2023) and se ice o
IT i ms. The la e , o example, highligh adi ional an eceden s o FDI, such
as ins i u ional dis ance (Pico -Coupey e al., 2014; Sánchez Peinado & Pla Ba -
be , 2006), hos coun y o FDI expe ience (E ans e al., 2008; Pla-Ba be e al.,
2014), and i m size o ma ke a ac i eness (Czinko a e al., 2009; Ekeledo &
Si akuma , 2004a). Fo digi al i ms, especially ECFs, only a ew s udies exis
on a ious decisions. Examples include he oles o dis ances o capabili ies in
ECFs’ in e na ionaliza ion speed (Luo e al., 2005; Schu e al., 2016) o o dis-
ances in ma ke selec ion (Schu & Mo sche , 2017). Fo en y choice, dis ances
ha e been shown o be ele an (Lee e al., 2023; Ro hae mel e al., 2006), along-
side epu a ion o geog aphic scope (Ko ha e al., 2001). Coun y-le el an eced-
en s we e add essed by h ee s udies on speed (e.g., legal ac o s, in e ne use;
Luo e al., 2005), ma ke selec ion (e.g., ule o law (ROL), ma ke a ac i eness;
701
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
Schu & Mo sche , 2017), and ma ke en y (cul u e; Ro hae mel e al., 2006).
Vi ual p esence, i.e., do -com s. local domains, was shown o be in luenced by
he en ep eneu ial o ien a ion o small expo ing manu ac u e s only (Ipsmille
e al., 2022); FDI s. o he modes we e linked o he ins i u ional dis ances o
digi al i ms only (S allkamp e al., 2023). No s udy on ECFs has conside ed i -
ual s. physical p esence. Concep ual pape s e e o new heo e ical an eced-
en s and i ual modes (B ou he s e al., 2022), whe eas o he pape s ques ion he
no el y o bo h o ECFs (Henna , 2022).
Despi e his con o e sy, a su p ising esea ch gap exis s in he li e a u e on
digi al i ms’ FDI. While ea lie s udies expec ed he in e ne o eplace physical
p esence (e.g., Sinko ics e al., 2013), schola s ha e ecen ly a gued ha ECFs and
pla o ms can ob ia e ma ke -seeking FDI (e.g.,S allkamp e al., 2023; Yang e al.,
2025). Howe e , p ac ical e idence e eals ha ECFs es ablish selec ed physical
p esences o e ime, which ewe schola s ha e no ed, leading o calls o close
examina ion o ECFs’ local asse decisions owing o hei g owing ele ance and
speci ic hyb id (digi al and physical) business model (e.g., Ba sakis e al., 2023a;
Swoboda & Sinning, 2022). We add ess his gap by analyzing an impo an esea ch
ques ion: Whe he and how s ongly do speci ic i m- and hos coun y-le el ac o s
d i e ECFs’ choice o wholly owned FDI? In heo izing espec i e e ec s, we add o
mode choice esea ch o digi al i ms and o e wo con ibu ions o li e a u e.
Fi s , in ligh o ECFs’ speci ic business model, we expand heo e ical discussions
o physical ma ke p esence in he eme ging esea ch on digi al i ms. Thei ma ke -
seeking FDI di e s om manu ac u e s’ FDI ocus on wholly owned p oduc ion
subsidia ies ab oad o he online p esence o small manu ac u e s o en s udied in
IB (e.g.,Ipsmille e al., 2022; Tols oy e al., 2023), as hose lacking online knowl-
edge o comme ce skills also u n o ECFs ab oad (Gong e al., 2024). Al hough
ECFs can ship om ab oad, only a physically p esen business can acili a e locally
adap ed goods, p ices o se ices, managemen , o ela ionships wi h cus ome s,
au ho i ies, o supplie s (Ba sakis e al., 2023a; Muelle & Swoboda, 2025). As
ECFs physically mo e goods, hey also di e om IT i ms such as Google o O a-
cle and i-business i ms such as Ne lix o eBay (Lee e al., 2023; Paul & Gup a,
2014). Such i ms’ o e s and cus ome in e ac ions a e usually digi al wi h di e en
FDI equi emen s ha suppo a scaling o pla o ms ia da a cen e s o cloud in a-
s uc u e (B ou he s e al., 2022). ECFs’ speci ic hyb id business model equi es an
analysis o speci ic FDI decisions. This calls o de eloping heo e ical a ionales
on he an eceden s o ECFs’ decisions o add a physical p esence in a coun y o no
(Meye e al., 2023).
Second, we con ibu e no el insigh s o he li e a u e on digi al in e na ionaliza-
ion conce ning he join ole o i m- and coun y-le el an eceden s o FDI (which
is gene ally a ely done in FDI esea ch, e.g., Wan e al., 2023). We add ess esea ch
calls o cla i y o wha ex en ECFs’ FDI depends on speci ic online capabili ies and
ex e nal ba ie s (e.g., Meye e al., 2023; Tols oy e al., 2021). In pa icula , a he
i m le el, we know ha dis ances a ec a ious decisions o ECFs, whe eas he ole
o expe ience in digi al i ms’ in e na ionaliza ion has been deba ed (e.g.,Gab iels-
son e al., 2022; Pezde ka & Sinko ics, 2011). We challenge skep ics by highligh -
ing ECFs’ speci ic online capabili ies and pa h-dependen i ual lea ning o e ime
702
M.Muelle , B.Swoboda
(e.g., in gene al and in hos coun ies; Swoboda & Sinning, 2022). We newly heo-
ize ha , unlike o line i ms, ECFs gain expe ience digi ally om success ul o e s,
p omo ions, o local online in e ac ions wi h consume s ha allow hem o exploi
FDI oppo uni ies, o example. We p o ide a i s unde s anding o such unknown
online lea ning mechanisms o a physical p esence, as esea ch on his opic has
been limi ed, despi e he g ow h o digi al businesses (Gong e al., 2024). A he
coun y le el, speci ic an eceden s o ECFs’ business model ab oad seem ob ious:
hos coun ies’ logis ics, in e ne use s, o ROL. We heo ize ha hese ma ch he
logic o speci ic ba ie s o loca ion ad an ages in digi al ma ke s o ECFs’ hyb id
business model (e.g., Dunning & Wymbs, 2001; in di e ence o hose mos o en
s udied o manu ac u e s, Pezde ka & Sinko ics, 2011). Only mac oeconomic s ud-
ies ha e discussed speci ic logis ics o in e ne ba ie s, wi h con adic o y indings
o expo s. FDI lows. Finally, we also espond o calls o adop ing hie a chical
modeling when analyzing i m- and coun y-le el d i e s o digi al i ms’ in e na-
ionaliza ion (e.g., Su ana e al., 2024).
The emainde o his s udy p oceeds as ollows. D awing on concep ualiza ions
and heo y, we de i e and es hypo heses based on 1,726 mode decisions o e 11
yea s co esponding o 241 ECFs o igina ing in 22 home coun ies. A e he esul s
a e p esen ed, we p o ide implica ions and di ec ions o u he esea ch.
2 Concep ualiza ion, Theo y, andHypo heses
2.1 Ope a ion Modes o ECFs
Ope a ion modes ep esen go e nance s uc u es ha allow i ms o implemen
hei business model in hos coun ies di e en ly (e.g., B ou he s e al., 2022). We
concep ualize ECFs’ decisions ega ding i ual s. physical p esence o se e al
easons ha a e speci ic o ECFs. The business model and in e na ionaliza ion o
ECFs a e based on a i ual p esence wi h digi al lea ning bu equi e physical dis-
ibu ion, such as “las -mile” logis ics and locally adap ed o e s (e.g., Meye e al.,
2023). While IB esea ch has discussed many d i e s o manu ac u e s’ FDI, we
e hink his decision o digi al i ms and ad ance heo y on how ECFs’ ope a ion
mode decisions a e made. ECFs may simpli y (mode) decisions o p ac ical deci-
sions (Pan & Tse, 2000). The la e , we a gue, ul ima ely ocus on whe he a physi-
cal p esence should be added and de ine whe he he business model can be ully
ans e ed ab oad.
Vi ual p esence among in e na ional ECFs e e s o coun y-speci ic websi es
(e.g., do -uk domain, language, cu ency) whe e goods a e shipped om ab oad
(whe eas do -com domains a e no hos coun y speci ic; Hua & Wu, 2024; Ipsmille
e al., 2022; Swoboda & Sinning, 2022). This also applies o ECFs’ coope a ion wi h
con ac ual o local se ice p o ide s ha a e pa ly s a egically con olled (e.g.,
ne wo ks), which emain expo -based (Henna , 2022). Join en u es wi h sha ed
ope a ional con ol a e a e and a ypical (e.g., B ou he s e al., 2022; Muelle &
Swoboda, 2025, and in ou da a). Physical p esence e e s o wholly owned FDI ha
is speci ic o ECFs’ business needs, i.e., local domains and long- e m in es men s
703
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
in physical asse s ha acili a e he local eplica ion o he business model (e.g.,
S allkamp e al., 2023). Speci ic examples include wholly owned wa ehouses ha
ocus on las -mile deli e y and ma ke ing subsidia ies ha acili a e local cus ome
ela ionship managemen and o e s (e.g., asso men s, p ices, cus ome se ice,
Tols oy e al., 2021). Howe e , es ablishing such a physical p esence depends on
local ma ke knowledge and speci ic in as uc u e (e.g., Ahi e al., 2023).
2.2 Theo y andECFs’ Speci ic An eceden s
The only s udy on digi al i ms’ FDI heo e ically explains he ole o dis ance
o FDI on he basis o he ans e abili y o complemen a y esou ces ab oad
(S allkamp e al., 2023). We ex end his iew and e e o capabili y and ede ined
in e na ionaliza ion p ocess easoning (majo heo ies in esea ch on he in e na ion-
aliza ion o digi al i ms, e.g., Cahen & Bo ini, 2020; Vahlne & Johanson, 2017;
Yang e al., 2025). We heo ize ha ECFs can scale and ans e online capabili ies
and le e age digi al lea ning as a pa h-dependen p ocess in FDI (Schu & Mo sche ,
2017). ECFs’ expe iences om i ual p esence, which a e dis inc om adi ional
lea ning h ough he physical p esence o manu ac u e s, o example (e.g., Co iello
e al., 2017), de e mine hei decisions o add physical p esence in selec ed coun ies
o e ime. We make bo h heo ies applicable o explaining ECFs’ FDI decisions,
o e ing no el a ionales o acili a ing digi al capabili y ans e and oppo uni y
ecogni ion ia i ual p esence and ocused exploi a ion wi h FDI (e.g.,Swoboda
& Sinning, 2022; Vahlne & Johanson, 2017). In pa icula , we o e a ionales on
he ela i e alue and limi s o di e en ypes o online expe ience capabili ies and
cla i y he ques ionable ole o expe ience in digi al i ms (Pezde ka & Sinko ics,
2011). As bo h heo ies only implici ly conside coun y ac o s, we complemen a-
ily heo ize business model-speci ic a ionales on ba ie s and loca ion ad an ages
o ECFs (Dunning & Wymbs, 2001). This suppo s a join iew o he i m- and
coun y-le el an eceden s o FDI. We a gue ha a lack o speci ic logis ic and in e -
ne in as uc u e o ROL in a coun y may be a ba ie o FDI (i.e., o ce i ual
p esence, Meye e al., 2023), whe eas hei p esence o e s speci ic loca ion ad an-
ages o ECFs’ business model.
In line wi h capabili y and p ocess heo ies, we concep ualize i ual loca ion-
and nonloca ion-bound online knowledge (e.g.,Cahen & Bo ini, 2020; Vahlne &
Johanson, 2017). Nonloca ion-bound online knowledge pe ains o ECFs’ lea ned
ou ines in dealing wi h o eign cus ome s i ually o exploi FDI oppo uni ies in
o eign ma ke s o e ime (oppo uni y iden i ica ion is likely simpli ied i ually;
Tang & Gude gan, 2018). Such knowledge also pe ains o p io FDI online expe i-
ence, which may a ec subsequen FDI in o he coun ies (Schwens e al., 2018). I
can be ans e ed as an online capabili y ac oss o eign ma ke s i ually. Loca ion-
bound knowledge pe ains o ECFs’ insigh s in o a speci ic coun y’s needs om
i ual p esence o e ime (no ans e able o o he ma ke s; He e al., 2013). How-
e e , ECFs may ely oo much on online capabili ies and un he isk o alling in o
a “ i uali y ap” (Sinko ics e al., 2013). We hus iden i y knowledge based on o
704
M.Muelle , B.Swoboda
lea ning ia i ual p esence as an impo an bu con o e sial basis o FDI among
digi al i ms.
We concep ualize ECFs’ speci ic hos coun y in as uc u e and ROL as speci ic
ba ie s o ECFs’ hyb id business model and sou ces o speci ic loca ion ad an ages
(Dunning & Wymbs, 2001; Meye e al., 2023). While many ex e nal an eceden s
a ec manu ac u ing and se ice i ms’ FDI (e.g., Wan e al., 2023), ob ious local
ba ie s o ad an ages o digi al i ms’ o ECFs’ FDI emain la gely unexplo ed
(e.g., Cumming e al., 2023). Coun ies’ logis ics in as uc u e, commonly e e ed
o as “logis ics pe o mance”, and in e ne use s o in as uc u e, o example, ha e
been linked o bo h FDI and expo s by mac oeconomic s udies only (Halaszo ich
& Kin a, 2020; La i e al., 2018). Despi e he high logis ics cos s o ECFs (o e
25% o o al cos s; Reu e s, 2023; Janje ic & Winkenbach, 2020) o hei eliance
on su icien in e ne use s in a coun y, he c i ical ole o hos coun y logis ics pe -
o mance o in e ne use s o ECFs needs heo iza ion. Simila ly, he c i ical ole
o ROL in digi al business models has been highligh ed (e.g., Schu & Mo sche ,
2017), bu i s ole in mode choices has no . Add essing his heo e ically, we p opose
ha ECFs may ace s ong local ba ie s and cos s o FDI i speci ic e-comme ce
in as uc u e is lacking, while i s p esence could be a speci ic loca ion ad an age.
Fo example, a coun y’s low logis ics pe o mance can be a ba ie o ECFs’ abil-
i y o ensu e quick deli e y om wa ehouses, whe eas high pe o mance can make
deli e y mo e e icien . A lack o in e ne use s can be a ba ie o eaching cus om-
e s, while he opposi e o e s loca ion ad an ages wi h a la en b oad cus ome base
(e.g., Meye e al., 2023). ECFs may also ace local legal ba ie s and compliance-
ela ed isks o inc eased coo dina ion cos s due o a lack o ROL (e.g., Luo e al.,
2005).
Addi ionally, a sys ema ic li e a u e e iew suppo s ou choices o an eced-
en s (see Figu e1).[1] We s udy expe ience as an impo an an eceden acco ding
o o line mode esea ch, whe eas he ole o (online) expe ience is unknown and
ques ioned o online businesses (e.g., Sinko ics e al., 2013). A he i m le el,
some an eceden s o ECFs’ in e na ionaliza ion (e.g., mainly dis ances o ma ke
capabili ies) ha e been explo ed, bu no ypes o online expe ience. A he coun y
le el, ECFs’ business model-speci ic an eceden s ha e mos ly no been conside ed
Fi m Le elCoun
y
Le el
Indus y In e na ional
Expe ience
Hos Coun y
Expe ience
FDI
Expe ience
O he
(see me a-analyses)
Logis icsIn e ne Rule o LawO he
(see me a-analyses)
O line
Manu ac-
u ing
Dow & La -
imo, 2009; He
nández &
Nie o, 2015;
Tang & Gude
-
g
an, 2018, e c.
He e al., 2013;
Wan e al., 2023;
e c.
Cui e al., 2013;
Schwens e al.,
2018; e c.
Beugelsdijk e al., 201
8
Klie e al., 2017;Wan
e al., 2023;e c.
–– Bailey, 2018;
S oian &
Filippaios,
2008, e c.
Cue o‐Cazu a e al.
,
2023; Kos o a e al.,
2020;Mo sche e al.,
2010; e c.
Se ices/
IT
Czinko a e al.,
2009; Ekeledo
& Si akuma ,
2004a; E ans
e al., 2008
Pla-Ba be e al., 2014 Pico -Coupey e al.,
2014; Sánchez Peinado
& Pla Ba be , 2006;
e c.
–––Pico -Coupey e al.,
2014; Sánchez
Peinado & Pla Ba -
be , 2006; e c.
Online
Manu ac-
u ing
– – Ipsmille e al., 2022
–
–– –
Se ices/
IT1– – – S allkamp e al., 2023
–
–– –
N
o e
:
I alics
=S udies on bina y mode choice, e.g., FDI s. o he modes. E c.=Fu he s udies no lis ed he e.1 Including i-business and ECFs.
Fig. 1 Li e a u e e iew on he an eceden s o ope a ion mode decisions
705
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
empi ically. This e iew highligh s he lack o knowledge and join heo e ical con-
side a ion o i m- and coun y-le el an eceden s o digi al i ms’ mode decisions.
Nex , we de i e hypo heses o each ela ionship based on heo e ical a ionales
and cla i y ou cu en empi ical knowledge i s . We con ol and es adi ional
an eceden s in s abili y checks (e.g., i m size o age and ma ke a ac i eness ac-
o s, such as size, g ow h, popula ion, o de elopmen ).
2.3 Hypo heses Rega ding Online Expe ience Fac o s
As men ioned, we a gue ha speci ic online capabili ies eme ge om ECFs’ pa h-
dependen i ual lea ning p ocesses in o eign ma ke s ha a e used o make deci-
sions on physical p esence (e.g., Swoboda & Sinning, 2022). ECFs can ans e gen-
e al, coun y-speci ic, o FDI- ela ed online expe ience capabili ies di e en ly based
on i ual lea ning o enhance hei abili y o ecognize and pa icula ly exploi
selec ed FDI oppo uni ies (e.g., Vahlne & Johanson, 2017).
In e na ional online expe ience e e s o ECFs’ deg ee o gene al nonloca ion-
bound knowledge ega ding i ual ope a ions in o eign ma ke s, which is acqui ed
ia local domains o e ime (as an empi ically es ablished p oxy; e.g., Wan e al.,
2023). Empi ical s udies ha e o en iden i ied in e na ional expe iences as an eced-
en s o manu ac u e s’ FDI (e.g., inconclusi ely, Dow & La imo, 2009, He nández
& Nie o, 2015, Tang & Gude gan, 2018; see Figu e1) and ha o se ice and IT
i ms (e.g.,Czinko a e al., 2009; Ekeledo & Si akuma , 2004a). Howe e , we a gue
ha i ually gained in e na ional expe ience di e s om he expe ience o man-
u ac u ing o se ice i ms (e.g., Pezde ka & Sinko ics, 2011). Online expe ience
pe ains o ECFs’ digi al capabili ies, which enable hem o p o ide o e s and se -
ices o in e ac wi h au ho i ies in di e en coun ies o e ime and can be used o
exploi FDI oppo uni ies ab oad. While o line expe ience may be used o p omo e
i ual expansion in o he coun ies (as indica ed o local domains o small manu-
ac u ing i ms; Ipsmille e al., 2022), i is unclea whe he online expe ience is
su icien o deciding on he physical p esence o ECFs (e.g., Co iello e al., 2017),
as we a gue.
We heo e ically a gue ha high in e na ional online expe ience acili a es lea n-
ing and he ecogni ion o ma ke oppo uni ies o FDI. ECFs accumula e knowl-
edge and ans e hei online capabili ies ab oad in e ms o digi ally success ul
o e s, se ices such as p oduc s o p omo ions, o ela ions o o eign au ho i ies
(e.g.,Schu & Mo sche , 2017; Vahlne & Johanson, 2017). O e ime, ECFs ec-
ognize sales oppo uni ies mo e easily in ce ain g owing e-comme ce ma ke s and
especially exploi hem wi h FDI and local supply chain p ocesses such as logis ics
and pu chasing (e.g., Swoboda & Sinning, 2022). This may be less isky e en i
physical p ocesses a e no es ablished ab oad. Fo example, wi h nea ly 20 yea s o
in e na ional online expe ience, Amazon ecognized oppo uni ies o FDI in Aus-
alia and Singapo e in 2017. This was based on he i m’s digi al capabili ies om
a long i ual p esence ab oad and lea ned po en ial o es ablish supply chain p o-
cesses in hese g owing ma ke s (Pash, 2017). We hus a gue ha ECFs wi h high
706
M.Muelle , B.Swoboda
in e na ional online expe ience ha e a be e sense o whe e o exploi FDI and may
be less eluc an o in es in new physical and adap ed i ual p ocesses.
In he case o a low in e na ional online expe ience, ew i ual capabili ies o
ans e o ecognize oppo uni ies ab oad exis (e.g., Tols oy e al., 2021). Fi ms
ha e less knowledge o how o p omo e digi al o e s o which p oduc s a e suc-
cess ul ab oad o how o deal wi h au ho i ies in di e en coun ies. Such ECFs’
abili y o ecognize in es men oppo uni ies is limi ed. ECFs a e especially limi ed
in exploi ing ma ke oppo uni ies in a o eign coun y (e.g., de eloping new supply
chain p ocesses alongside adap ed ma ke o e s, S allkamp e al., 2023). Thei isks
o FDI a e g ea e and missed oppo uni ies may hinde success (e.g., Swoboda &
Sinning, 2022). ECFs may e en unde es ima e such expe ience, which could cause
hei business models o ail ab oad (e.g., Pezde ka & Sinko ics, 2011). We p opose
he ollowing:
Hypo hesis 1 (H1) In e na ional online expe ience has a posi i e e ec on an ECF’s
p opensi y o choose wholly owned FDI.
Hos coun y online expe ience is acqui ed h ough i ual p esence o e ime
and pe ains o he deg ee o a i m’s loca ion-bound knowledge ega ding i ual
ope a ions in a pa icula o eign ma ke . Empi ical s udies ha e o en shown he
ole o hos coun y expe ience o manu ac u e s’ FDI (e.g., He e al., 2013; see
Figu e1) bu only once o se ice i ms (Pla-Ba be e al., 2014). These s udies
ha e a gued ha i ms wi h such expe ience iden i y mo e local oppo uni ies, o e -
come local obs acles, and inc ease hei con ol in coun ies (e.g., Wan e al., 2023).
In con as , o digi al i ms, we only know ha such ma ke (bu no online) knowl-
edge a ec s ma ke selec ion (Schu & Mo sche , 2017). We a gue ha ECFs wi h
hos coun y online expe ience can ans e hei capabili ies and knowledge o local
cus ome s o business p ac ices mo e easily, especially o exploi FDI oppo uni ies
in a coun y (e.g., Ipsmille e al., 2022). This loca ion-bound knowledge is likely a
majo online capabili y o exploi FDI oppo uni ies in a speci ic coun y (e.g., Col-
on e al., 2010).
We a gue ha high hos coun y online expe ience p o ides unique knowledge
om ECFs’ coun y-speci ic i ual in e ac ions o e ime. This includes especially
knowledge o local cus ome needs and p e e ences bu also local business p ac ices,
compe i o s, o u he condi ions. ECFs wi h such knowledge de elop coun y-spe-
ci ic online capabili ies and can mo e easily ecognize local ma ke oppo uni ies
and a sui able ime o exploi hem wi h FDI in a coun y. Fo example, a coun y
may exhibi s ong local e-comme ce g ow h bu also speci ic paymen o deli e y
condi ions. Exploi ing such local oppo uni ies equi es adap ed i ual o e s o cus-
ome needs and especially suppo i e physical in es men s (e.g., locally adap ed
deli e y, e u ns, p oduc s, p ices, o en e p ise esou ce planning sys ems; Janje-
ic & Winkenbach, 2020). This may help es ablish he ECF physically as a local
business. Fo example, Amazon en e ed B azil in 2012 bu chose wholly owned
FDI only in 2019. The i m’s main easons included di e en axa ion ules ac oss
egions (e.g., speci ic axa ion in each ede al geog aphical uni ) and complica ed
713
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
he home ma ke o he acqui ed company was coun ed (no addi ional coun ies).
We had o exclude 61 ECFs wi h in es men s only p io o 2010, mos ly adi ional
b ick-and-mo a ood e aile s, and ashion e icals en e ing Eas e n Eu opean and
Asian ma ke s since he 1990s, and o 45 ECFs, we did no ind in o ma ion on he
ope a ion mode a all o could no econs uc i m-le el in o ma ion o e ime (one
yea be o e FDI). The sample included 244 i ms wi h 1,806 ope a ion mode deci-
sions in 169 hos coun ies.
To include coun y-le el independen a iables (e.g., ollowing Kos o a e al.,
2020), we had o ob ain addi ional da a om he Wo ld Bank (2022a, 2022b, 2022c,
an es ablished sou ce in IB, e.g., Kos o a e al., 2020) and he only one a ailable
p o iding hese da a o e ime, which we ma ched wi h he ime-lagged mode deci-
sion yea s. We had o exclude 80 ope a ion mode decisions in 25 e y small hos
coun ies o which da a on he Wo ld Go e nance Indica o s (WGIs) and Wo ld
De elopmen Indica o s o on he Logis ics Pe o mance Index we e no a ailable
(19 o 61 decisions and 4 o 21 coun ies, espec i ely). Addi ional i m-le el a ia-
bles, such as in e na ional expe ience, i m size, o pla o m business, we e ob ained
om Digi al Comme ce 360 wi h a one-yea ime lag.
Table 1 Sample dis ibu ion Home coun ies Numbe o ECFs Numbe o en ies
Aus ia 1 8
Belgium 1 1
China 7 93
Czech Republic 3 7
Denma k 6 49
Finland 1 7
F ance 38 225
Ge many 38 203
Uni ed Kingdom 66 535
I eland 1 2
I aly 11 270
Ko ea (Rep.) 1 12
Li huania 1 1
Ne he lands 9 28
Poland 5 18
Romania 1 3
Russian Fede a ion 3 4
Spain 7 62
Sweden 6 19
Swi ze land 4 11
Tu key 4 13
Uni ed S a es 28 155
To al 241 1,726
714
M.Muelle , B.Swoboda
The inal da ase included 1,726 mode decisions ac oss 144 coun ies om
2010–2020; hese decisions we e made by 241 ECFs o igina ing in 22 home coun-
ies (see Table1). We used mul ile el modeling o accoun o ou hie a chically
s uc u ed da a and employed a maximum likelihood es ima o wi h obus s anda d
e o s and es s a is ics.
3.2 Measu emen
We used di e en sou ces o ou measu es, all o which we e ma ched one yea
p io o mode implemen a ion, as discussed, because FDI is planned p io o i s
implemen a ion (e.g., Swoboda e al., 2007). This app oach esul ed in a ime lag o
he assessmen o in e nal and ex e nal ac o s (e.g., Li e al. 2021).
Dependen Va iable. We measu ed wholly owned FDI (1) s. non (wholly
owned) FDI (0) as a dicho omous a iable cap u ing in e na ional ECFs’ ope a-
ion mode decisions in each hos coun y o e ime (e.g., He e al., 2013; S allkamp
e al., 2023). The measu emen o mode choices as bina y decisions has commonly
been used in IB (e.g., Kos o a e al., 2020, showing o en consis en e ec s o me ic
measu emen ). We used a successi e da a selec ion p ocedu e in ol ing se e al da a
sou ces and eco ded he ini ial yea o an ECF’s coun y websi e alongside he local
domain, language, and cu ency (as no exac day/mon h da a we e a ailable in 80%
o he cases). Wholly owned FDI includes ypical o ms o long- e m in es men in
a coun y, such as wa ehouses and logis ics o se ice/ma ke ing subsidia ies ha
a e wholly owned by an ECF (e.g., Cahen & Bo ini, 2020). We ca e ully assessed
each case o wholly owned FDI o ensu e ha i was p edominan ly ela ed o he
ECF’s online ac i i ies by e i ying i s loca ion and unc ions as an ope a ion mode
o online selling documen ed by he i m (e.g., logis ics, ul illmen , online cus-
ome se ice). In con as , modes wi hou wholly owned FDI (e.g., coun y-speci ic
domain only, local pa ne ships, pa ial owne ship) in logis ics o se ices o he
online business we e classi ied as non (wholly owned) FDI.
Independen Va iables. In e na ional online expe ience cap u ed i ms’ o e all
knowledge abou o eign ma ke accumula ed h ough i ual p esence o e ime
(e.g.,Dow & La imo, 2009; Wan e al., 2023). We measu ed he numbe o yea s
o which an ECF was in e na ionally ac i e online wi h ega d o each mode deci-
sion (i.e., one-yea lagged), which is he mos equen ly used p oxy o in e na ional
expe ience o e ime (e.g.,Tang & Gude gan, 2018; Wan e al., 2023). We e e ed
o he en e ed coun y da a p o ided by Digi al Comme ce 360 (2021) and collec ed
he ini ial ime o en y and subsequen ope a ion mode decisions h ough he a i-
ous abo emen ioned sou ces. We es ed he al e na i e measu e o he numbe o
coun ies (which has been used less equen ly) in ou s abili y checks (Ipsmille
e al., 2022), while o he measu es, such as he di e si y o expe ience a ios, we e
una ailable (e.g., Cla ke e al., 2013).
Fi ms’ hos coun y online expe ience was cap u ed as a ime-based indica o
e lec ing knowledge abou a pa icula o eign ma ke h ough i ual p esence o e
ime (e.g.,Cla ke e al., 2013; Wan e al., 2023). This expe ience was measu ed by
he numbe o yea s ha an in e na ional ECF was ac i e online in a speci ic coun y
715
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
one yea be o e a mode was implemen ed based on ou collec ion o each ope a ion
mode decision h ough a ious sou ces. Al e na i e measu es, such as he numbe o
ope a ions o loca ions in a coun y, we e no a ailable o mos ECFs (e.g., Tang &
Gude gan, 2018).
ECFs’ FDI online expe ience is conside ed a ime-based measu e o i ms’
knowledge abou implemen ing FDI in o eign ma ke s o e ime (e.g., Cui e al.,
2013). We measu ed his a iable as he numbe o yea s since an ECF i s imple-
men ed FDI in a coun y in addi ion o a coun y-speci ic domain (e.g., Schwens
e al., 2018). We e e ed o ou (lagged) da a on subsequen FDI o e ime p o ided
by he abo e sou ces. Al e na i e measu es, such as he numbe o FDI ins ances in
a coun y o a ea o e ime, we e no a ailable (e.g.,Ba kema & D ogendijk, 2007;
Hong & Lee, 2015).
We measu ed logis ics pe o mance in a coun y ia he six-dimensional logis ics
pe o mance index (A is e al., 2018, p. 8; Schu & Mo sche , 2017). This indica-
o e lec s he quali y o ade in as uc u e, ease o logis ics se ices and cus oms,
acking abili y, and imely shipmen s wi hin a coun y (Wo ld Bank, 2022a). The
da a (one-yea lagged) co e ele en yea s in wo-yea pe iods. We used he app oxi-
ma e mean alues o wo yea s o measu e his a iable in he in e im yea s (e.g.,
2010 and 2012 o 2011; Blomk is & D ogendijk, 2016).
We measu ed in e ne use popula ion wi hin a coun y as he loga i hm o he
o al popula ion using he in e ne on he basis o da a ob ained om Wo ld Bank
(2022c). This measu e was based on a coun y’s in e ne pene a ion and he o al
popula ion o e he obse ed pe iod o ele en yea s wi h a one-yea ime lag. In e -
ne use is mos o en cap u ed by in e ne pene a ion, which we es ed in ou s abil-
i y checks (e.g., Ba sakis e al., 2023a). Howe e , his ac o uses only he p opo ion
o in e ne use s ( a he han he ac ual numbe o use s) o indica e ECFs’ po en-
ial cus ome s. Fo example, while a e y small coun y cha ac e ized by high-le el
in e ne pene a ion may be ini ially appealing o FDI, he numbe o po en ial cus-
ome s in his coun y may be e y low, and FDI may no be wo hwhile (e.g., Muel-
le & Swoboda, 2025).
We measu ed di e ences in coun ies’ ROL ega ding he ele an dimension o
he WGIs o e ele en yea s wi h a one-yea ime lag (Wo ld Bank, 2022b). This
dimension ep esen s he ex en o which agen s ha e con idence in and abide by he
ules o socie y (i.e., he quali y o con ac en o cemen , p ope y igh s, police, and
cou s as well as he likelihood o c ime and iolence) and is pa icula ly ele an
o ECFs (Oxley & Yeung, 2001; Schu & Mo sche , 2017). Howe e , egula o y
ins i u ions a e b oade and cap u ed mos equen ly by an index o he ins i u ional
dimension o he Global Compe i i eness Repo (GCR; Kos o a e al., 2020), ol-
lowed by an index o all six WGIs. As WGIs we e shown o be ele an o ECFs’
in e na ionaliza ion speed (Swoboda & Sinning, 2022), we es ed bo h indices in
s abili y checks bu assumed hem o be less ele an o FDI decisions.
Con ols. A bo h le els, impo an a iables we e con olled and measu ed one
yea p io o FDI.
A he i m le el, we con olled o i m size as a adi ionally impo an capabil-
i y o FDI (Pezde ka & Sinko ics, 2011). Fi m size was measu ed by he loga i hm
o i ms’ online sales and ob ained om Digi al Comme ce 360. Al e na i es (e.g.,
716
M.Muelle , B.Swoboda
numbe o employees) we e no a ailable o us; ECFs hi e ewe employees (Luo
e al., 2005).
We con olled o i ms’ online age, which was measu ed as he numbe o
yea s since an ECF s a ed selling online, as no all ECFs ounded hei businesses
online (which hus likely ep esen s a digi al capabili y). This ac o may a ec
wholly owned FDI, as such ECFs may ha e mo e ime o acqui e digi al esou ces
o suppo FDI (e.g., Sinko ics e al., 2013). Da a we e collec ed, e.g., om ECFs’
websi es.
The home ma ke size o an ECF was con olled as i migh a ec i s expansion
and wholly owned FDI (Swoboda & Sinning, 2022). A small ( s. la ge) home ma -
ke size o e s limi ed domes ic g ow h oppo uni ies and may o ce in e na ional
in es men . Ma ke size was measu ed by he coun y’s loga i hmized GDP pe cap-
i a based on Wo ld Bank (2022c) da a.
We con olled whe he ECFs an an online ma ke place (i.e., a hyb id pla o m
wi h physical deli e y) in addi ion o hei co e e ail business (Meye e al., 2023).
Wholly owned FDI may be equi ed o such ECFs aiming o p o ide and con ol
ul illmen se ices in o eign ma ke s o a ac local endo s o o e local cus ome
se ice, no o he wise gua an eed (B ou he s e al., 2022). We measu ed dicho o-
mously: 1=yes (ma ke place) and 0=no (Digi al Comme ce 360, 2021).
Fi ms’ mul ichannel s a egy was con olled. Fi ms ope a ing only online ha e a
g ea e p opo ion o in e ne -based ac i i ies han do i ms ope a ing mul ichannel
(e.g., o line channels, Schu e al., 2016). ECFs wi h addi ional o line channels a e
mo e lexible and can le e age exis ing o ganiza ional capabili ies o make online
ela ed FDI and in eg a e o line and online supply chain ac i i ies (e.g., e u ning
in-s o e pu chases o wa ehouses; Swoboda & Sinning, 2022). We measu ed his
dicho omously: 1=also ope a ing o line s o es and 0=no (Digi al Comme ce 360,
2021).
P io in e na ional and hos coun y expe ience wi h an o line s o e was included
as a con ol because i ms such as Apple may ha e gained expe ience in a coun y
p io o online- ela ed FDI (e.g., ia o line s o es), which hey could use o scale
and suppo u he FDI ega ding hei online ope a ions. This ac o was measu ed
dicho omously (1=p io in e na ional/hos coun y expe ience and 0=no such expe-
ience). We also collec ed hese da a om a ious sou ces.
We con olled o popula ion densi y a he coun y le el as one possible indica-
o o ma ke a ac i eness (as a possible ba ie o loca ion ad an age), pa icula ly
ele an o comme ce i ms, as i may a ec hei wholly owned FDI, which is o en
loca ed in densely popula ed egions wi h high demand (Chan e al., 2011). We
measu ed his a iable as people pe sq. km o land a ea acco ding o da a ob ained
om Wo ld Bank (2022c) as hese da a co e all ou hos coun ies o e ime and
a oid po en ial collinea i y issues wi h he o he coun y-le el a iables. We con-
duc ed addi ional analyses o es al e na i e measu es o a ac i eness in s abili y
checks. Fu he measu es, such as online ma ke sales o numbe s o online cus om-
e s, we e a ailable only o ce ain Eu opean coun ies and only o e he las i e
yea s.
Table 2 shows desc ip i e s a is ics and co ela ions o all a iables. Co e-
la ions a he i m le el did no aise collinea i y conce ns, and all we e assessed
717
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
Table 2 Desc ip i e s a is ics and co ela ions
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
Mean 0.140 3.210 0.210 1.170 18.320 10.120 10.530 0.150 0.670 0.050 0.540 3.463 0.846 16.167 281.998
SD 0.350 4.405 1.153 3.076 2.320 6.143 0.450 0.358 0.470 0.222 0.498 0.520 0.937 1.501 972.473
VIF
(max.)
– 1.912 1.107 1.709 1.169 1.336 1.097 1.414 2.822 1.080 2.492 1.233 1.093 1.169 1.064
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
Wholly owned
FDI (1/0) (1)
1
In e na ional
online expe i-
ence (2)
0.255*** 1
Hos coun y
online expe i-
ence (3)
0.407*** 0.270*** 1
FDI online
expe ience (4)
0.220*** 0.625*** 0.224*** 1
Fi m size (log.)
(5)
0.113*** 0.218*** 0.097*** 0.230*** 1
Online age (6) 0.023ns 0.350*** 0.058* 0.268*** 0.269*** 1
Home ma ke
size (log.) (7)
0.077** 0.141*** 0.052* 0.126*** − 0.070** 0.139*** 1
Ma ke place
(1/0) (8)
0.008ns 0.018ns − 0.018ns 0.096*** 0.130*** − 0.105*** − 0.105*** 1
Mul ichannel
(1/0) (9)
− 0.077** − 0.052* − 0.064** − 0.047† − 0.039ns 0.134*** 0.151*** − 0.502*** 1
P io hos coun-
y expe ience
(1/0) (10)
0.165*** 0.039ns 0.004ns 0.006ns − 0.060* 0.008ns 0.111*** − 0.091*** 0.147*** 1
P io in e -
na ional
expe ience
(1/0) (11)
− 0.008ns 0.076** − 0.016ns 0.008ns − 0.022ns − 0.002ns 0.173*** − 0.408*** 0.739*** 0.204*** 1
Logis ics
pe o -
mance
(12)
0.182*** 0.094*** 0.104*** 0.031ns − 0.022ns 0.029ns 0.063** 0.018ns − 0.186*** 0.099*** − 0.081** 1
718
M.Muelle , B.Swoboda
Table 2 (con inued)
In e ne
use
popula-
ion
(log.)
(13)
0.148*** 0.063** 0.061* 0.067** − 0.011ns 0.005ns − 0.024ns 0.073** − 0.178*** 0.087*** − 0.132*** 0.319*** 1
Rule o law
(14)
0.156*** 0.071** 0.100*** − 0.005ns − 0.046† 0.010ns 0.085*** − 0.028ns − 0.133*** 0.079** − 0.046† 0.857*** 0.009ns 1
Popula ion
densi y
(15)
0.079** 0.059* − 0.002ns 0.048* 0.046† 0.063** 0.004ns 0.029ns 0.010ns 0.068** 0.026ns 0.172*** 0.149*** − 0.069** 1
Ns No signi ican , FDI Fo eign di ec in es men , SD S anda d de ia ion, VIF Va iance in la ion ac o
* p < 0.05
** p < 0.01
*** p < 0.01
† p < 0.10
719
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
indi idually; howe e , we de ec ed a high co ela ion o .857 be ween logis ics pe -
o mance and ROL a he coun y le el. This may esul om logis ics pe o mance
elying o some deg ee on he ROL in a hos coun y (e.g., as public in es o s ely
on s ong en o ceable con ac s, cons uc ion pe mi s, o he imely implemen a ion
o in as uc u e p ojec s due o high ROL, e.g., Cumming e al., 2023; Janje ic &
Winkenbach, 2020). To a oid collinea i y issues, hese wo a iables we e used only
in sepa a e eg essions. We u he included heo e ically ele an con ols and con-
duc ed nume ous obus ness checks (Kalnins, 2022; Lindne e al., 2022) by e e -
ence o independen heo e ical a ionales conce ning he impo an oles played by
bo h ac o s among ECFs’ FDIs. We obse ed a iance in la ion ac o s below he
usual h eshold in all models, and he highes sco e emained below h ee (O’B ien,
2007).
3.3 Me hod
Hypo heses we e es ed using mul ile el modeling in Mplus 9 o accoun o nes ed
da a conside ing i m- and coun y-le el e ec s and a iance bo h be ween and wi hin
coun ies simul aneously (Hox e al., 2018, p. 215). Mul ile el modeling accoun s o
a iances pe aining o he dependen a iable bo h wi hin and be ween g oups; o
example, e ec s o he independen a iables may di e be ween g oups (i.e., hos
coun ies), leading o mo e accu a e es ima ion ac oss coun ies (which a e no consid-
e ed in s anda d eg ession analysis; e.g., Finch & Bolin, 2017, pp. 33–37). Mul ile el
modeling accoun s o ou hie a chically s uc u ed da a (e.g., expe ience is ied o a
speci ic i m, while ROL is linked wi h a speci ic coun y, e.g., He nández & Nie o,
2015) and includes ele an in o ma ion simul aneously in he analysis. The la e
app oach educes he isk o biased esul s; hus, his me hod has been ecommended in
IB esea ch (e.g., Su ana e al., 2024).
We compu ed s epwise andom in e cep and slope models, and he Akaike in o ma-
ion c i e ion (AIC), Bayesian in o ma ion c i e ion (BIC), and −2 log likelihood we e
used o assess model i (Hox e al., 2018, p. 123, Kau man, 1996).
Fi s , a baseline model wi h only he con ols was calcula ed. We hen added all he
coun y-le el independen a iables sepa a ely, which we e included as clus e means o
accoun o a ia ion wi hin clus e s o e ime. Fi m-le el independen a iables we e
subsequen ly es ed sepa a ely be o e all he a iables we e en e ed simul aneously in o
he model. All he independen a iables we e g and mean cen e ed and s anda dized
(inc easing in e p e abili y o in e cep s and ensu ing he compa abili y o coe icien
es ima es; Hox e al., 2018, pp. 61–63; S allkamp e al., 2023). The ollowing equa ion
desc ibes he model wi h only he con ols and i m-le el independen a iables (
ILij)
i s sepa a ely, hen join ly:
(1)
Logi
(FDI
ij
)=𝛽
0j
+𝛽
1j(
IL
ij)
+𝛽
con ols
ILC
ij
+
j
720
M.Muelle , B.Swoboda
wi h logi FDIij= loga i hmized odds o he p obabili y o ECF i choosing wholly
owned FDI in coun y j; In OnExpij = ECF’s in e na ional online expe ience;
HoCoOnExpij= hos coun y online expe ience; FDIOnExpij= FDI online expe i-
ence; β0j = i s -le el in e cep ; β1j, β2j, β3j = eg ession sco es o i m-le el inde-
penden a iables;
ILCij
= i m-le el con ols; CLCj= coun y-le el con ol; i j =
i s -le el e o e m. Equa ion (3) shows he model wi h a coun y-le el independ-
en a iable (
CLj)
only.
We hen allowed he in e cep β alues o a y ac oss coun ies as ollows:
wi h γ00 = second-le el in e cep o he dependen a iable; γ10 and γ20 = second-
le el in e cep s o he andom slopes o he i m-le el p edic o s; u0j = coun y-le el
esidual a iance; u1j, u2j, and u3j = andom slopes o he i m-le el a iables a he
coun y le el. Subs i u ion led o he ollowing:
whe e LPj = LP o hos coun y j; INTj and ROLj = coun y in e ne use popula ion
and ROL;
γ
03, γ04, and γ05 = ixed e ec s coe icien s o he coun y-le el p edic-
o s; and
𝜀
= esidual a iances and e o s.
3.4 Resul s
The esul s o he hypo hesis es s a e p esen ed in Table3. We s anda dized all he
a iables p io o es ima ion o assess he ela i e e ec sizes and magni udes as
well as he epo ed odds a ios (Exp(B), Kau man, 1996). Owing o he s ong co -
ela ion obse ed in his con ex , logis ics pe o mance and ROL we e assessed in
sepa a e models.
(2)
Logi (FDI
ij
)=𝛽
0j
+𝛽
1j
(In OnExp
ij
)+𝛽
2j
(HoCoOnExp
ij
)
+𝛽3
j
(FDIOnExp
ij
)+ 𝛽
con ols
ILC
ij
+𝛽
con ols
CLC +
ij
(3)
Logi
(FDI
ij
)=𝛽
0j
+𝛾
03(
CL
j)
+𝛾
ILC
ILC
ij
+𝛾
CLC
CLC
j
+
𝜀
(4)
𝛽0j=𝛾00 +u0j
(5)
𝛽1j=𝛾10 +u1j
(6)
𝛽2j=𝛾20 +u2j
(7)
𝛽3j=𝛾30 +u3j
(8)
Logi
(FDIij)=𝛾00 +𝛾10(In OnExpij)+𝛾20(HoCoOnExpij)+𝛾30(FDIOnExpij
)
+𝛾03(LPj)+𝛾04(INTj)+𝛾05(ROLj)+u0j+u1j(In OnExpij)
+u2j(HoCoOnExpij)+u3j(FDIOnExpij)
+𝛾
ILC
ILC
ij
+𝛾
CLC
CLC
j
+𝜀
721
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
Fo ECFs wi h mo e in e na ional online expe ience, he esul s indica ed an
inc easing likelihood o wholly owned FDI (b = 0.420, p < 0.001, Exp(B) = 1.522),
suppo ing Hypo hesis 1. A one-s anda d-de ia ion inc ease in his expe ience
inc eased he odds o FDI by a ac o o 1.522 in Models 9 and 10. The mo e expe i-
ence ECFs ob ained online in e na ionally, he mo e likely hey we e o choose FDI.
ECFs’ hos coun y online expe ience inc eased hei likelihood o choosing
wholly owned FDI (bModel 9 = 3.165, p < 0.001, Exp(B) = 23.689; bModel 10 = 3.305,
p < 0.001, Exp(B) = 27.250). A one-s anda d-de ia ion inc ease in hos coun y
expe ience inc eased he odds o FDI by a ac o o 23.689 in Model 9 and 27.250
in Model 10. The mo e expe ience ECFs ha e gained in speci ic o eign online ma -
ke s, he mo e likely hey a e o choose FDI o e ime. Hypo hesis 2 is suppo ed.
ECFs’ FDI online expe ience inc eased he likelihood o hese i ms choosing
wholly owned FDI when assessed indi idually (bModel 4 = 0.439, p < 0.001); how-
e e , no signi ican e ec was obse ed when in e na ional and hos coun y online
expe ience we e included (bModel 9 = 0.125, p > 0.10; bModel 10 = 0.114, p > 0.10).
The e o e, Hypo hesis 3 was ejec ed.
Logis ics pe o mance inc eased ECFs’ likelihood o choosing wholly owned
FDI (b = 0.526, p < 0.001, Exp(B) = 1.692), suppo ing Hypo hesis 4. A one-s and-
a d-de ia ion inc ease in hos coun y logis ics pe o mance inc eased he odds o
ECFs choosing wholly owned FDI by a ac o o 1.692. The be e a hos coun y’s
logis ical in as uc u e is, he mo e likely ECFs a e o choose FDI, such as h ough
local wa ehouses o subsidia ies.
The in e ne use popula ion inc eases he likelihood o ECFs choosing wholly
owned FDI (bModel 9 = 0.553, p < 0.001, Exp(B) = 1.738; bModel 10 = 0.379, p <
0.001, Exp(B) = 1.461). A one-s anda d-de ia ion inc ease in he in e ne use pop-
ula ion inc eases he odds o ECFs choosing FDI by a ac o o 1.738 when ROL is
accoun ed o and by 1.461 when logis ics pe o mance is accoun ed o . The la ge
a coun y’s in e ne use popula ion is, he mo e likely ECFs a e o use FDI o
expansion. These esul s suppo Hypo hesis 5.
ROL inc eases he likelihood o ECFs’ wholly owned FDI (b = 0.515, p < 0.001,
Exp(B) = 1.674). A one-s anda d-de ia ion inc ease in he hos coun y’s ROL
inc eased he odds o ECFs choosing he FDI mode by a ac o o 1.674. The mo e
anspa en and e ec i e a coun y’s egula ions a e, he mo e likely ECFs a e o op
o FDI. Thus, Hypo hesis 6 is suppo ed.
As con ols, i m size, ma ginal ECFs’ pla o m business model, and p io hos
coun y expe ience signi ican ly and consis en ly a ec ed he likelihood o ECFs
choosing FDI.
3.5 S abili y Checks
To ensu e s abili y, al e na i e models we e es ed. Fi s , a andom spli -hal es was
conduc ed. The esul s o he i m- and coun y-le el a iables emained s able (see
Web Appendix A).
Second, we eplaced se e al independen a iables wi h al e na i es (see Web
Appendix B).
722
M.Muelle , B.Swoboda
Table 3 Resul s
In e cep Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10
bpbpbpbpbpbpbpbpbp Exp(B) bp Exp(B)
− 2.109***
(0.146)
− 2.290***
(0.219)
− 1.760***
(0.164)
− 2.323***
(0.207)
− 1.688***
(0.178)
− 2.692***
(0.156)
− 2.344***
(0.136)
−
2.473***
(0.153)
− 2.742***
(0.212)
− 2.665***
(0.216)
Fi m Le el
In e na ional
Online Expe ience
→ Wholly owned
FDI (1/0)
0.675***
(0.085)
0.381**
(0.111)
0.420***
(0.114)
1.522 0.420***
(0.113)
1.522
Hos Coun y
Online Expe ience
→ Wholly owned
FDI (1/0)
4.274***
(0.770)
4.256***
(0.808)
3.165***
(0.627)
23.689 3.305***
(0.634)
27.250
FDI Online Expe i-
ence
→ Wholly owned
FDI (1/0)
0.439***
(0.081)
0.120ns
(0.088)
0.125ns
(0.086)
1.133 0.114ns
(0.086)
1.121
Coun y Le el
Logis ics Pe -
o mance
→ Wholly owned
FDI (1/0)
0.633***
(0.098)
0.526***
(0.103)
1.692
In e ne Use
Popula ion
(log.)
→ Wholly owned
FDI (1/0)
0.505***
(0.101)
0.553***
(0.102)
1.674 0.379***
(0.117)
1.461
Rule o Law → Wholly owned
FDI (1/0)
0.466***
(0.091)
0.515***
(0.094)
1.738
Con ols (Fi m Le el)
Fi m Size
(log.)
→ Wholly
owned FDI
(1/0)
0.641***
(0.139)
0.524***
(0.129)
0.405**
(0.134)
0.466***
(0.126)
0.318**
(0.113)
0.627***
(0.133)
0.651***
(0.138)
0.630***
(0.135)
0.347**
(0.115)
1.415 0.336**
(0.110)
1.399
Ma ke place
(1/0)
→ Wholly
owned FDI
(1/0)
− 0.187*
(0.076)
− 0.172*
(0.069)
− 0.144†
(0.079)
− 0.233**
(0.081)
− 0.165*
(0.077)
− 0.131†
(0.074)
− 0.199*
(0.080)
− 0.132†
(0.073)
− 0.143†
(0.079)
0.867 − 0.151†
(0.079)
0.860
Mul ichannel
(1/0)
→ Wholly
owned FDI
(1/0)
− 0.519***
(0.101)
− 0.290*
(0.118)
− 0.409**
(0.118)
− 0.464***
(0.109)
− 0.280*
(0.126)
− 0.404***
(0.098)
− 0.474***
(0.100)
− 0.439***
(0.097)
− 0.181ns
(0.121)
0.834 − 0.181ns
(0.119)
0.834
Home Ma ke
Size (log.)
→ Wholly
owned FDI
(1/0)
0.236*
(0.100)
0.159†
(0.095)
0.180ns
(0.112)
0.166†
(0.091)
0.123ns
(0.106)
0.202*
(0.098)
0.241***
(0.099)
0.206*
(0.098)
0.092ns
(0.101)
1.096 0.095ns
(0.103)
1.100
729
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
p essu e. Mo eo e , socie ies ace dynamic changes in IT ha equi e egula ion o
enable he popula ion and indus ies o keep pace wi h hese shi s.
5 Limi a ions andFu u e Resea ch
This s udy has limi a ions ha sugges di ec ions o u u e esea ch.
We explo e longi udinal da a conce ning leading ECFs, bu an ex ended da abase
will suppo u he conclusions, e.g., ega ding ECFs wo ldwide, business- o-busi-
ness ECFs, o egional di e ences in wholly owned FDI (e.g., Yang e al., 2025).
Examining smalle ECFs o he g owing pla o m business wi hin ECFs can p o-
ide new insigh s in o FDI (ou con ols o i m size and he ma ke place a ec FDI
bo h posi i ely and nega i ely). The con ol o p e ious expe ience in a hos coun y
(beyond online) is also posi i e, sugges ing u he esea ch on going s. bo n digi-
als (Ba sakis e al., 2023a).
Wi h espec o ou measu emen s, we conside FDI bu canno iden i y acqui-
si ions o coope a i e modes (bo h and esp. ECFs’ a ious pa ne ships; e.g.,
B ou he s e al., 2022). Measu ing ECFs’ deg ee o con ol, mode swi ches, o
lea ning om ailu e o e ime is challenging bu could e eal s epwise decisions
a ec ed by an eceden s (e.g., Muelle & Swoboda, 2025). As no ed, da a on mo e
ECF-speci ic an eceden s ac oss na ions could become a ailable o u u e esea ch
(e.g., ECFs’ online ma ke ing capabili ies, such as social media usage o web a -
ic in coun ies). Online ma ke po en ial, local e-comme ce compe i ion, o speci ic
ROL and media ba ie s ha a ec ECFs’ in e na ionaliza ion can be s udied by e -
e ence o u u e da a conce ning e-ma ke s (e.g.,Ba sakis e al., 2023a; Pezde ka &
Sinko ics, 2011).
In ou amewo k, we cap u e he local en i onmen wi h in as uc u al ac o s
and he ROL, while no ma i e o cul u al ins i u ions may be s udied (e.g., Lee
e al., 2023). We o en e e o oppo uni ies o cus ome needs, which a e c i i-
cal o ECFs, ye wha d i es he adap a ion o online o e s o se ices is unclea
bu impo an (e.g., asso men , paymen op ions; Ipsmille e al., 2022). I is also
unclea how adap a ion/s anda diza ion decisions a ec ECFs’ local o gene al pe -
o mance o g ow h. The same applies o compe i ion o digi al capabili ies (e.g.,
Fleu y e al., 2024).
6 Conclusions
Despi e many oppo uni ies o digi al in e na ionaliza ion highligh ed in li e a u e,
no s udy has in es iga ed whe he and how digi al and especially ECFs choose
be ween i ual s. physical p esence, i.e., websi es wi h local domains and in es -
men s in own wa ehouses o subsidia ies ab oad. Such mode choices can ha e sig-
ni ican implica ions. De eloping heo y abou he unde lying mechanisms o his
decision, we imp o e ou limi ed knowledge o digi al i ms’ expansion.
In pa icula , we con ibu e o heo y by cla i ying and explaining how ope a-
ion mode decisions ela e o he exploi a ion o ECFs’ online knowledge and hos
730
M.Muelle , B.Swoboda
coun y in as uc u e, i.e., how business model-speci ic ac o s ac as de e minan s
o his choice. We ind suppo o mos o ou hypo heses; howe e , he ac o s we
analyzed explain FDI o di e en deg ees. The ele ance o elemen s o ECFs’ i -
ual and physical business models o his majo decision is hus, o he i s ime,
heo e ically and empi ically cla i ied and con as ed wi h i-businesses’ o manu ac-
u e s’ digi al in e na ionaliza ion.
No es
1. We conduc ed a sys ema ic li e a u e e iew in >50 jou nals ollowing Ha zing’s
jou nal quali y lis on he basis o Gau and Kuma (2018). We used he key-
wo ds ECFs, digi al i ms, i-businesses, FDI, mode o en y s a egy, i ual o
physical p esence, in e na ional o hos coun y expe ience, logis ics and in e ne
in as uc u e, ROL o ba ie s ( o s udies since 2010 and c oss-ci a ions). As
indica ed in he in oduc ion, empi ical s udies on “o line i ms” (manu ac u ing,
se ice/IT) ha e analyzed many an eceden s o mode decisions, albei mos ly
a he i m le el. Fo example, Dow and La imo (2009) highligh ed he ole o
in e na ional expe ience, He e al. (2013) ha o hos coun y expe ience, and Cui
e al. (2013) ha o FDI expe ience in he bina y decision “FDI s. o he modes.”
The insigh s a e no always consis en (e.g., non-signi ican Dow & La imo, 2009,
nega i e He nández & Nie o, 2015, posi i e Tang & Gude gan, 2018). O he
an eceden s a e shown by me a-analyses (e.g., Wan e al., 2023 wi h ins i u ional
dis ance, R&D, ad e ising in ensi y). Signi ican ly ewe s udies ha e ocused on
mode choices among se ice/IT i ms. Mos unde line he posi i e ole o in e -
na ional expe ience on FDI (Czinko a e al., 2009; Ekeledo & Si akuma , 2004a;
E ans e al., 2008), while Pla-Ba be e al. (2014) show a posi i e ole o hos
coun y and a nega i e o FDI expe ience. O he an eceden s a e also add essed
(e.g., dis ances by Pico -Coupey e al., 2014; s a egic mo i es by Sánchez Pei-
nado & Pla Ba be , 2006). Fewe s udies add ess coun y-le el an eceden s and
o he bes o ou knowledge, mos ly speci ic o manu ac u ing o se ice i ms.
Some s udies ha e highligh ed he ele ance o egula o y ac o s, including ROL,
o manu ac u e s’ FDI (e.g.,Bailey, 2018; S oian & Filippaios, 2008). O he
esea ch, o example, a ecen me a-analysis o 224 s udies (Cue o‐Cazu a
e al., 2023), has add essed poli ical isks, ma ke a ac i eness, unce ain y o
legal ac o s. In an online con ex , only wo s udies conside ope a ion mode
decisions. Among manu ac u ing i ms wi h a i ual p esence, Ipsmille e al.
(2022) e ealed a posi i e e ec o en ep eneu ial o ien a ion on he choice o
wo ypes o i ual p esence. Among online se ice/IT i ms, S allkamp e al.
(2023) e ealed posi i e e ec s o dis ance on FDI choice. To ou knowledge, no
empi ical s udies ha e conside ed coun y-le el an eceden s o examined an e-
ceden s ha pe ain speci ically o ECFs’ business model and mode decisions.
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps:// doi.
o g/ 10. 1007/ s11575- 025- 00575-7.
731
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
Funding Open Access unding enabled and o ganized by P ojek DEAL. 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.
Decla a ions
Con lic o in e es None.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License,
which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e
Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his
a icle a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line
o he ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended
use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mis-
sion di ec ly om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/
licenses/by/4.0/.
Re e ences
Abou You (2021). Abou you expands wa ehouse ne wo k wi h loca ion in Slo akia Re ie ed
20.06.2023, om h ps:// co po a e. abou you. de/ en/ p esse/ abou - you- e wei e - lage - ne zw e k- um-
s and o - in- de - slowa kei.
Aga wal, J., & Wu, T. (2015). Fac o s in luencing g ow h po en ial o e-comme ce in eme ging econo-
mies: An ins i u ion-based N-OLI amewo k and esea ch p oposi ions. Thunde bi d In e na ional
Business Re iew, 57(3), 197–215.
Ahi, A. A., Sinko ics, N., & Sinko ics, R. R. (2023). E-comme ce policy and he global economy: A pa h
o mo e inclusi e de elopmen ? Managemen In e na ional Re iew, 63(1), 27–56.
Alibaba (2024). Global loca ion. Re ie ed 20.12.2024, om h ps:// www. aliba bag o up. com/ en- US/
global- loca ion.
Amazon (1998). Amazon.com en e s Eu opean book ma ke . Re ie ed 20.06.2023, om h ps:// p ess.
abou amazon. com/ 1998/ 10/ amazon- com- en e s- eu op ean- book- ma ke .
Amazon (2024). Amazonglobal expo coun ies and egions. Re ie ed 20.12.2024, om h ps:// www.
amazon. com/ gp/ help/ cus o me / displ ay. h ml? nodeId= GCBBS ZMUXA 6U2P8R.
A in, M. B., P adhan, R. P., & Nai , M. (2021). Unco e ing in e links among ICT connec i i y and
pene a ion, ade openness, o eign di ec in es men , and economic g ow h: The case o he G-20
coun ies. Telema ics and In o ma ics, 60, 101567.
A is J-F., Ojala L., Wiede e C., Shephe d B., Raj A., Dai abaye a K., & Kiiski T. (2018). Connec ing
o compe e 2018: T ade logis ics in he global economy - The logis ics pe o mance index and i s
indica o s, The Wo ld Bank G oup.
Bailey, N. (2018). Explo ing he ela ionship be ween ins i u ional ac o s and FDI a ac i eness: A
me a-analy ic e iew. In e na ional Business Re iew, 27(1), 139–148.
Ba kema, H. G., & D ogendijk, R. (2007). In e na ionalising in small, inc emen al o la ge s eps? Jou -
nal o In e na ional Business S udies, 38(7), 1132–1148.
Ba sakis, G., Kona a, P., & Theoha akis, V. (2023a). Digi al sales channels and he ela ionship be ween
p oduc and in e na ional di e si ica ion: E idence om going digi al e ail MNEs. Global S a -
egy Jou nal, 13(4), 830–856.
Ba sakis, G., Theoha akis, V., Li, C., & Kona a, P. (2023b). In e na ionaliza ion and digi aliza ion: Thei
di e ing ole on g oce and non-g oce e aile pe o mance. Jou nal o Re ailing, 99(3), 400–419.
Benmamoun, M., Singh, N., Lehne , K., & Lee, S. B. (2019). In e na ionaliza ion o e-comme ce co po-
a ions (ECCs): Ad anced s eme ging ma ke s. Mul ina ional Business Re iew, 27(4), 317–338.
Beugelsdijk, S., Kos o a, T., Kuns , V. E., Spada o a, E., & Van Essen, M. (2018). Cul u al dis ance and
i m in e na ionaliza ion: A me a-analy ical e iew and heo e ical implica ions. Jou nal o Man-
agemen , 44(1), 89–130.
732
M.Muelle , B.Swoboda
Blomk is , K., & D ogendijk, R. (2016). Chinese ou wa d o eign di ec in es men s in Eu ope. Eu o-
pean Jou nal o In e na ional Managemen , 10(3), 343–358.
Blyde, J., & Molina, D. (2015). Logis ic in as uc u e and he in e na ional loca ion o agmen ed p o-
duc ion. Jou nal o In e na ional Economics, 95(2), 319–332.
B ou he s, K. D., Chen, L., Li, S., & Shahee , N. (2022). Cha ing new cou ses o en e o eign ma ke s:
Concep ualiza ion, heo e ical amewo k, and esea ch di ec ions on non- adi ional en y modes.
Jou nal o In e na ional Business S udies, 53(9), 2088–2115.
B ou he s, K. D., Geisse , K. D., & Ro hlau , F. (2016). Explaining he in e na ionaliza ion o ibusiness
i ms. Jou nal o In e na ional Business S udies, 47(5), 513–534.
Cahen, F., & Bo ini, F. M. (2020). In e na ional digi al compe ence. Jou nal o In e na ional Manage-
men , 26(1), 100691.
Cen e o Re ail Resea ch (2023). Value o online e ail sales in he Uni ed Kingdom (UK) om 2012 o
2022. Re ie ed 20.01.2025, om h ps:// www. s a i s a. com/ s a i s ics/ 315506/ online- e ail- sales- in-
he- uni ed- kingd om/.
Chan, P., Finnegan, C., & S e nquis , B. (2011). Coun y and i m le el ac o s in in e na ional e ail
expansion. Eu opean Jou nal o Ma ke ing, 45(6), 1005–1022.
Cla ke, J. E., Tamaschke, R., & Liesch, P. W. (2013). In e na ional expe ience in in e na ional business
esea ch: A concep ualiza ion and explo a ion o key hemes. In e na ional Jou nal o Manage-
men Re iews, 15(3), 265–279.
Col on, D. A., Ro h, M. S., & Bea den, W. O. (2010). D i e s o in e na ional e- ail pe o mance: The
complexi ies o o ien a ions and esou ces. Jou nal o In e na ional Ma ke ing, 18(1), 1–22.
Con ac o , F. J., Dangol, R., Nu uzzaman, N., & Raghuna h, S. (2020). How do coun y egula ions
and business en i onmen impac o eign di ec in es men (FDI) in lows? In e na ional Business
Re iew, 29(2), 101640.
Con ac o , F. J., Nu uzzaman, N., Dangol, R., & Raghuna h, S. (2021). How FDI in lows o eme ging
ma ke s a e in luenced by coun y egula o y ac o s: An explo a o y s udy. Jou nal o In e na-
ional Managemen , 27(1), 100834.
Co iello, N., Kano, L., & Liesch, P. W. (2017). Adap ing he Uppsala model o a mode n wo ld: Mac o-
con ex and mic o ounda ions. Jou nal o In e na ional Business S udies, 48(9), 1151–1164.
Cue o-Cazu a, A., Du an, P., A egle, J. L., & an Essen, M. (2023). Hos coun y poli ics and in e na-
ionaliza ion: A me a-analy ic e iew. Jou nal o Managemen S udies, 60(1), 204–241.
Cui, L., Li, Y., & Li, Z. (2013). Expe ien ial d i e s o o eign di ec in es men by la e-come Asian
i ms: The Chinese e idence. Jou nal o Business Resea ch, 66(12), 2451–2459.
Cumming, D., Johan, S., Khan, Z., & Meye , M. (2023). E-comme ce policy and In e na ional business.
Managemen In e na ional Re iew, 63(1), 3–25.
Czinko a, M. R., G ossman, D. A., Ja algi, R. R. G., & Nugen , N. (2009). Fo eign ma ke en y mode o
se ice i ms: The case o US MBA p og ams. Jou nal o Wo ld Business, 44(3), 274–286.
Demi bag, M., Ta oglu, E., & Glais e , K. W. (2008). Fac o s a ec ing pe cep ions o he choice be ween
acquisi ion and g een ield en y: The case o Wes e n FDI in an eme ging ma ke . Managemen
In e na ional Re iew, 48(1), 5–38.
Digi al Comme ce 360 (2021). Digi al comme ce 360 esea ch & da a. Re ie ed 20.06.2023, om
h ps:// www. digi alcom me ce 360. com/ abou - ou - esea ch/.
Dow, D., & La imo, J. (2009). Challenging he concep ualiza ion and measu emen o dis ance and in e -
na ional expe ience in en y mode choice esea ch. Jou nal o In e na ional Ma ke ing, 17(2),
74–98.
Du, J., Lu, Y., & Tao, Z. (2008). FDI loca ion choice: Agglome a ion s ins i u ions. In e na ional Jou -
nal o Finance & Economics, 13(1), 92–107.
Dunning, J. H., & Wymbs, C. (2001). The challenge o elec onic ma ke s o in e na ional business he-
o y. In e na ional Jou nal o he Economics o Business, 8(2), 273–301.
ECDB (2024). Why did amazon ail in China: local compe i ion de ea s he e ail gian . Re ie ed
10.11.2024, om h ps:// ecomm e cedb. com/ insig h s/ why- amazon- ailed- in- china- ma ke - dynam
ics- s ong- compe i ion/ 4670.
Edua dsen, J., Ma ino a, S., Leonidou, L. C., & Ch is odoulides, P. (2023). O ganiza ional In luences
and pe o mance impac o c oss-bo de E-comme ce ba ie s: The mode a ing ole o home
coun y digi al in as uc u e and o eign ma ke in e ne pene a ion. Managemen In e na ional
Re iew, 63(3), 433–467.
Ekeledo, I., & Si akuma , K. (2004a). In e na ional ma ke en y mode s a egies o manu ac u ing i ms
and se ice i ms: A esou ce-based pe spec i e. In e na ional Ma ke ing Re iew, 21(1), 68–101.
733
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
Ekeledo, I., & Si akuma , K. (2004b). The impac o e-comme ce on en y-mode s a egies o se ice
i ms: A concep ual amewo k and esea ch p oposi ions. Jou nal o In e na ional Ma ke ing,
12(4), 46–70.
E ans, J., Ma ondo, F. T., & B idson, K. (2008). Psychic dis ance: An eceden s, e ail s a egy implica-
ions, and pe o mance ou comes. Jou nal o In e na ional Ma ke ing, 16(2), 32–63.
Fila o che , I., S ange, R., Piesse, J., & Lien, Y.-C. (2007). FDI by i ms om newly indus ialised econ-
omies in eme ging ma ke s: Co po a e go e nance, en y mode and loca ion. Jou nal o In e na-
ional Business S udies, 38(1), 556–572.
Financial Times (2024). Shein, emu and he pa cel wa s. Re ie ed 10.11.2024, om h ps:// www. .
com/ con e n / 67319 cc8- 1 0e- 4155- 8 5e- 664 e 8603d 2d.
Finch, H., & Bolin, J. (2017). Mul ile el modeling using Mplus. CRC P ess.
Fleu y, A., Fleu y, M. T. L., Oli ei a, L., & Leao, P. (2024). Going digi al EMNEs: The ole o digi al
ma u i y capabili y. In e na ional Business Re iew, 33(4), 102271.
Gab ielsson, M., Raa ikainen, M., & Julkunen, S. (2022). Accele a ed in e na ionaliza ion among inexpe-
ienced digi al en ep eneu s: Towa d a holis ic en ep eneu ial decision-making model. Manage-
men In e na ional Re iew, 62(2), 137–168.
Gong, C., He, X., & Lengle , J. (2024). In e na ionalisa ion h ough digi al pla o ms: A sys ema ic
e iew and u u e esea ch agenda. In e na ional Ma ke ing Re iew, 41(5), 938–980.
Halaszo ich, T. F., & Kin a, A. (2020). The impac o dis ance, na ional anspo a ion sys ems and logis-
ics pe o mance on FDI and in e na ional ade pa e ns: Resul s om Asian global alue chains.
T anspo Policy, 98, 35–47.
HDE (2024). Online-moni o 2024. Re ie ed 20.01.2025, om h ps:// www. s a i s a. com/ s a i s ics/
454371/ e- comme ce- e en ue- ge ma ny/.
He, X., B ou he s, K. D., & Fila o che , I. (2013). Resou ce-based and ins i u ional pe spec i es on
expo channel selec ion and expo pe o mance. Jou nal o Managemen , 39(1), 27–47.
He, Y., Wu, R., & Choi, Y.-J. (2021). In e na ional logis ics and c oss-bo de E-comme ce ade: Who
ma e s whom? Sus ainabili y, 13(4), 1745–1765.
Henna , J.-F. (2022). How much is new in B ou he s e al.’s new o eign en y modes, and do hey chal-
lenge he ansac ion cos heo y o en y mode choice? Jou nal o In e na ional Business S udies,
53(9), 2116–2132.
He nández, V., & Nie o, M. J. (2015). The e ec o he magni ude and di ec ion o ins i u ional dis ance
on he choice o in e na ional en y modes. Jou nal o Wo ld Business, 50(1), 122–132.
Hong, S. J., & Lee, S.-H. (2015). Reducing cul u al unce ain y h ough expe ience gained in he
domes ic ma ke . Jou nal o Wo ld Business, 50(3), 428–438.
Hox, J. J., Moe beek, M., & Van de Schoo , R. (2018). Mul ile el analysis: Techniques and applica-
ions. Rou ledge.
Hua, L., & Wu, Y. (2024). S a egic analysis o e ical in eg a ion in c oss-bo de e-comme ce
logis ics se ice supply chains. T anspo a ion Resea ch Pa E: Logis ics and T anspo a ion
Re iew, 188, 103626.
Ipsmille , E., Diko a, D., & B ou he s, K. D. (2022). Digi al in e na ionaliza ion o adi ional i ms:
Vi ual p esence and en ep eneu ial o ien a ion. Jou nal o In e na ional Managemen , 28(4),
100940.
ITA (2024). 2024 eComme ce size & sales o ecas . Re ie ed 20.12.2024, om h ps:// www. ade.
go / ecomm e ce- sales- size- o ec as .
Janje ic, M., & Winkenbach, M. (2020). Cha ac e izing u ban las -mile dis ibu ion s a egies in
ma u e and eme ging e-comme ce ma ke s. T anspo a ion Resea ch Pa A: Policy and P ac-
ice, 133, 164–196.
Kalnins, A. (2022). When does mul icollinea i y bias coe icien s and cause ype 1 e o s? a econcili-
a ion o Lindne , Puck, and Ve beke (2020) wi h Kalnins (2018). Jou nal o In e na ional Busi-
ness S udies, 53(7), 1536–1548.
Kau man, R. L. (1996). Compa ing e ec s in dicho omous logis ic eg ession: A a ie y o s anda d-
ized coe icien s. Social Science Qua e ly, 77(1), 90–109.
Klie , H., Schwens, C., Zapkau, F. B., & Diko a, D. (2017). Which esou ces ma e how and whe e?
a me a-analysis on i ms’ o eign es ablishmen mode choice. Jou nal o Managemen S udies,
54(3), 304–339.
Kos o a, T., Beugelsdijk, S., Sco , W. R., Kuns , V. E., Chua, C. H., & an Essen, M. (2020). The
cons uc o ins i u ional dis ance h ough he lens o di e en ins i u ional pe spec i es:
734
M.Muelle , B.Swoboda
Re iew, analysis, and ecommenda ions. Jou nal o In e na ional Business S udies, 51(4),
467–497.
Ko ha, S., Rindo a, V. P., & Ro hae mel, F. T. (2001). Asse s and ac ions: Fi m-speci ic ac o s in
he in e na ionaliza ion o US in e ne i ms. Jou nal o In e na ional Business S udies, 32(4),
769–791.
Kwon, Y. C., & Konopa, L. J. (1993). Impac o hos coun y ma ke cha ac e is ics on he choiceo
o eign ma ke en y mode. In e na ional Ma ke ing Re iew, 10(2), 60–76.
La i , Z., La i , S., Ximei, L., Pa han, Z. H., Salam, S., & Jianqiu, Z. (2018). The dynamics o ICT,
o eign di ec in es men , globaliza ion and economic g ow h: Panel es ima ion obus o he -
e ogenei y and c oss-sec ional dependence. Telema ics and In o ma ics, 35(2), 318–328.
Lee, H. S., Pa naik, C., & Gau , A. S. (2023). In e na ionaliza ion o I-business i ms: The ole o
dis ance on loca ion choice. Jou nal o Business Resea ch, 164, 114010.
Li, Y., Zhang, B., Fan, D., & Li, Z. (2021). Digi al media, con ol o co up ion, and eme ging mul ina-
ional en e p ise’s FDI en ymode choice. Jou nal o Business Resea ch, 130, 247–259.
Lin, F. (2015). Es ima ing he e ec o he In e ne on in e na ional ade. The Jou nal o In e na-
ional T ade & Economic De elopmen , 24(3), 409–428.
Lindne , T., Puck, J., & Ve beke, A. (2022). Beyond add essing mul icollinea i y: Robus quan i a i e
analysis and machine lea ning in in e na ional business esea ch. Jou nal o In e na ional Busi-
ness S udies, 53(7), 1307–1314.
Luo, Y., Zhao, J. H., & Du, J. (2005). The in e na ionaliza ion speed o e-comme ce companies: An
empi ical analysis. In e na ional Ma ke ing Re iew, 22(6), 693–709.
Mello, G. (2019). Amazon.com s a s di ec sales o me chandise in B azil a e delays. Re ie ed
20.06.2023, om h ps:// www. eu e s. com/ a ic le/ us- amazon- com- b azil- idUSK CN1PG 0AG.
Meye , K. E., Li, J., B ou he s, K. D., & Jean, R.-J.B. (2023). In e na ional business in he digi al age:
Global s a egies in a wo ld o na ional ins i u ions. Jou nal o In e na ional Business S udies,
54(4), 577–598.
Mi hani, M. A. (2023). Scaling digi al and non-digi al business models in o eign ma ke s: The case
o inancial ad ice indus y in he Uni ed S a es. Jou nal o Wo ld Business, 58(4), 101457.
Mo sche , D., Sch amm-Klein, H., & Swoboda, B. (2010). Decades o esea ch on ma ke en y
modes: Wha do we eally know abou ex e nal an eceden s o en y mode choice? Jou nal o
In e na ional Managemen , 16(1), 60–77.
Muelle , M., & Swoboda, B. (2025). E ec s o In e na ional E-comme ce i ms’ dep h e sus b ead h
o con ol on o eign sales pe o mance and he ole o ma ke size and g ow h as con ex ac-
o s. Managemen In e na ional Re iew. In P ess.
Nielsen, B. B., Asmussen, C. G., & Wea he all, C. D. (2017). The loca ion choice o o eign di ec in es -
men s: Empi ical e idence and me hodological challenges. Jou nal o Wo ld Business, 52(1),
62–82.
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(5), 673–690.
Oxley, J. E., & Yeung, B. (2001). E-comme ce eadiness: Ins i u ional en i onmen and in e na ional
compe i i eness. Jou nal o In e na ional Business S udies, 32(4), 705–723.
Pan, Y., & Tse, D. K. (2000). The hie a chical model o ma ke en y modes. Jou nal o In e na ional
Business S udies, 31(4), 535–554.
Pash, C. (2017). Amazon is coming o Aus alia wi h ’low p ices, as selec ion, and as deli e y.
Re ie ed 20.06.2023, om h ps:// www. smh. com. au/ busin ess/ compa nies/ amazon- is- coming- o-
aus alia- wi h- low- p ices- as - selec ion- and- as - deli e y- 20170 420- g o6lb. h ml.
Paul, J., & Gup a, P. (2014). P ocess and in ensi y o in e na ionaliza ion o IT i ms–e idence om
India. In e na ional Business Re iew, 23(3), 594–603.
Pezde ka, N., & Sinko ics, R. R. (2011). A concep ualiza ion o e- isk pe cep ions and implica ions o
small i m ac i e online in e na ionaliza ion. In e na ional Business Re iew, 20(4), 409–422.
Pico -Coupey, K., Bu , S. L., & Clique , G. (2014). Re aile s׳ expansion mode choice in o eign ma -
ke s: An eceden s o expansion mode choice in he ligh o in e na ionaliza ion heo ies. Jou nal
o Re ailing and Consume Se ices, 21(6), 976–991.
Pla-Ba be , J., Villa , C., & León-Da de , F. (2014). Augmen ing e sus exploi ing en y modes in so
se ices: Reconside ing he ole o expe ien ial knowledge. In e na ional Ma ke ing Re iew, 31(6),
621–636.
Raku en (2024). Ou businesses. Re ie ed 20.12.2024, om h ps:// global. aku en. com/ co p/ abou /
compa ny/.
735
De e minan s o Wholly Owned Fo eign Di ec In es men s in…
Reu e s (2023). Amazon, o he e aile s e amp ’ ee’ shipping as cos s soa . Re ie ed 20.06.2023, om
h ps:// www. eu e s. com/ busin ess/ e ail- consu me / amazon- o he - e ai le s- e amp- ee- shipp
ing- cos s- soa - 2023- 03- 24/.
Rind leisch, A., Mal e , A. J., Ganesan, S., & Moo man, C. (2008). C oss-sec ional e sus longi udi-
nal su ey esea ch: Concep s, indings, and guidelines. Jou nal o Ma ke ing Resea ch, 45(3),
261–279.
Ro hae mel, F. T., Ko ha, S., & S eensma, H. K. (2006). In e na ional ma ke en y by US in e ne i ms:
An empi ical analysis o coun y isk, na ional cul u e, and ma ke size. Jou nal o Managemen ,
32(1), 56–82.
Sánchez Peinado, E., & Pla Ba be , J. (2006). Mode o en y in se ice i ms: S a egic a iables and
cha ac e is ics o se ices in luencing he in e na ionaliza ion p ocess. In e na ional Ma ke ing
Resea ch, 17, 159–192.
Scayle GmbH (2024). In e na ional ecomme ce: how o e ec i ely g ow in eas e n Eu ope Re ie ed
15.05.2024, om h ps:// www. scayle. com/ blog/ in e na io nal- ecomm e ce- eas e n- eu ope/.
Schu, M., & Mo sche , D. (2017). Fo eign ma ke selec ion o online e aile s—A pa h-dependen pe -
spec i e on in luence ac o s. In e na ional Business Re iew, 26(4), 710–723.
Schu, M., Mo sche , D., & Swoboda, B. (2016). In e na ionaliza ion speed o online e aile s: A
esou ce-based pe spec i e on he in luence ac o s. Managemen In e na ional Re iew, 56(5),
733–757.
Schwens, C., Zapkau, F. B., B ou he s, K. D., & Hollende , L. (2018). Limi s o in e na ional en y mode
lea ning in SMEs. Jou nal o In e na ional Business S udies, 49(7), 809–831.
Sinko ics, N., Sinko ics, R. R., & Jean, R.-J. (2013). The in e ne as an al e na i e pa h o in e na ionali-
za ion? In e na ional Ma ke ing Re iew, 30(2), 130–155.
S allkamp, M., Chen, L., & Li, S. (2023). Boo s on he g ound: Fo eign di ec in es men by bo n digi al
i ms. Global S a egy Jou nal, 13(4), 805–829.
S oian, C., & Filippaios, F. (2008). Dunning’s eclec ic pa adigm: A holis ic, ye con ex speci ic ame-
wo k o analysing he de e minan s o ou wa d FDI: E idence om in e na ional G eek in es -
men s. In e na ional Business Re iew, 17(3), 349–367.
Su ana, A., Cha an, M., Kuma , V., & Chi ico, F. (2024). The in e na ionaliza ion p ocess: A con ex ual
analysis o Indian ibusiness i ms. In e na ional Business Re iew, 33(4), 102255.
Swoboda, B., Schwa z, S., & Hälsig, F. (2007). Towa ds a concep ual model o coun y ma ke selec ion:
Selec ion p ocesses o e aile s and C&C wholesale s. In e na ional Re iew o Re ail, Dis ibu ion
and Consume Resea ch, 17(3), 253–282.
Swoboda, B., & Sinning, C. (2022). E ec s o In e na ionaliza ion hy hm and speed on E-Comme ce
Fi ms’ g ow h and he ole o ins i u ional dis ances. Managemen In e na ional Re iew, 62(2),
169–201.
Tang, R. W., & Gude gan, S. P. (2018). A Me a-analysis o he in e na ional expe ience–owne ship s a -
egy ela ionship: A dynamic capabili ies iew. Managemen In e na ional Re iew, 58(4), 541–570.
Tols oy, D., Hånell, S. M., & Özbek, N. (2023). E ec ual ma ke c ea ion in he c oss-bo de e-comme ce
o small-and medium-sized en e p ises. In e na ional Small Business Jou nal, 41(1), 35–54.
Tols oy, D., No dman, E. R., Hånell, S. M., & Özbek, N. (2021). The de elopmen o in e na ional e-com-
me ce in e ail SMEs: An e ec ua ion pe spec i e. Jou nal o Wo ld Business, 56(3), 101165.
U.S. Census Bu eau (2010). Qua e ly e ail E-comme ce sales. Re ie ed 05.11.2024, om h ps://
www2. census. go / e ail/ elea ses/ his o ical/ ecomm/ 10q4. pd .
U.S. Census Bu eau (2022). Annual e ail ade su ey. Re ie ed 05.11.2024, om h ps:// www. census.
go / da a/ ables/ 2022/ econ/ a s/ suppl emen al- ecomm e ce. h ml.
Vahlne, J.-E., & Johanson, J. (2017). F om in e na ionaliza ion o e olu ion: The uppsala model a 40
yea s. Jou nal o In e na ional Business S udies, 48(9), 1087–1102.
Ve beke, A., Coeu de oy, R., & Ma , T. (2018). The u u e o in e na ional business esea ch on co po-
a e globaliza ion ha ne e was…. Jou nal o In e na ional Business S udies, 49(9), 1101–1112.
Visse , R. (2019). The e ec o he in e ne on he ma gins o ade. In o ma ion Economics and Policy,
46, 41–54.
Wan, C., Sousa, C. M., Lengle , J., & Tan, Q. (2023). En y mode choice: A me a-analysis o an eceden s
and ou comes. Managemen In e na ional Re iew, 63(2), 193–246.
Wen up, R. (2016). The online–o line balance: In e na ionaliza ion o Swedish online se ice p o id-
e s. Jou nal o In e na ional En ep eneu ship, 14(4), 562–594.
Wo ld Bank (2022a). Logis ics pe o mance index. Re ie ed 20.06.2023, om h ps:// lpi. wo ld bank. o g/
in e na io nal.
736
M.Muelle , B.Swoboda
Au ho s and A ilia ions
Ma iusMuelle 1· Be nha dSwoboda1
* Be nha d Swoboda
b.swoboda@uni- ie .de
1 Chai o Ma ke ing andRe ailing, T ie Uni e si y, T ie , Ge many
Wo ld Bank (2022b). Wo ldwide go e nance indica o s. Re ie ed 20.06.2023, om h ps:// da ab ank.
wo ld bank. o g/ sou ce/ wo ld wide- go e nance- indic a o s.
Wo ld Bank (2022c). Wo ld de elopmen indica o s. Re ie ed 20.06.2023, om h ps:// da ab ank. wo ld
bank. o g/ epo s. aspx? sou ce= 2& se ies= LP. LPI. OVRL. XQ& coun y.
Yang, B., Bai, W., Chen, Y., & Rong, K. (2025). In e na ionaliza ion o digi al i ms: A sys ema ic e iew
and esea ch agenda. Jou nal o Business Resea ch, 189, 115124.
Yu, S., Beugelsdijk, S., & de Haan, J. (2015). T ade, us and he ule o law. Eu opean Jou nal o Poli i-
cal Economy, 37, 102–115.
Zalando (2015). Zalando o s a ope a ions in i s I alian sa elli e ul illmen cen e . Re ie ed 20.02.2024,
om h ps:// co po a e. zalan do. com/ en/ compa ny/ zalan do- s a - ope a ions- i s- i ali an- sa el li e- ul i
llmen - cen e .
Zalando (2023). Annual epo 2022. Re ie ed 20.06.2023, om h ps:// co po a e. zalan do. com/ en/ in es
o - ela ions/ annual- epo - 2022.
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps
and ins i u ional a ilia ions.
