Content uploaded by Quan Li
Author content
All content in this area was uploaded by Quan Li on May 27, 2021
Content may be subject to copyright.
Leviathan as foreign investor: Geopolitics
and sovereign wealth funds
Di Wang
1
, Robert J. Weiner
2
,
Quan Li
3
and
Srividya Jandhyala
4
1
Department of Government, The University of
Texas at Austin, Austin, TX 78712-1704, USA;
2
International Business Department, George
Washington University, Washington, DC 20052,
USA;
3
Department of Political Science, Texas A&M
University, College Station, TX 77843-4348, USA;
4
ESSEC Business School, 5 Nepal Park,
Singapore 139408, Singapore
Correspondence:
D Wang, Department of Government, The
University of Texas at Austin, Austin,
TX 78712-1704, USA
e-mail: diwang@austin.utexas.edu
Abstract
Sovereign wealth funds (SWFs) are important but understudied state investors.
We investigate whether geopolitics influences SWFs foreign acquisitions, asking
if and how their FDI patterns differ from those of private firms. Theoretical
expectations are mixed. On the one hand, limited managerial control of target
firms suggests that SWFs may be unable to pursue political goals, and thus they
are no more sensitive to geopolitics than private firms. On the other hand, state
ownership of SWFs can generate national security externalities and thereby
makes SWFs more sensitive to geopolitics. Utilizing novel big-data measures of
cooperative and adversarial relations based on media reporting and three
different tests, we examine over 5800 cross-border acquisitions by SWFs and
private firms. We find that home–host conflict hinders SWFs more than private
firms whereas cooperation helps SWFs more than private firms. Hence, despite
SWFs’ lack of managerial control of target firms, state ownership moderates
geopolitical influences on their internationalization and makes them more
sensitive than private firms to interstate relations. Our findings suggest that
government concern over FDI by state entities goes beyond their operational
activities.
Journal of International Business Studies (2021).
https://doi.org/10.1057/s41267-021-00415-4
Keywords: liability of foreignness; political relationships; government; MNE–host
country relations; logistic regression; state-owned enterprises
INTRODUCTION
The 21st century has witnessed a dramatic rise in state capitalism.
State entities, which used to operate primarily domestically, have
become some of the world’s largest international actors, and have
attracted wide attention in academic and policy circles (Bremmer,
2009; Musachio, Lazzarini, & Aguilera, 2015; Kurlantzick, 2016). A
vibrant stream of research, surveyed in Cuervo-Cazurra et al. (2014)
and Musacchio and Lazzarini (2018), studies whether and how
state-owned multinational firms differ from private-sector firms in
FDI patterns (Duanmu, 2014; Knutsen et al., 2011). This literature
focuses primarily on comparing the foreign expansion of private
firms with state-owned operating enterprises (termed SOEs here).
Surprisingly, another class of state-led investors – sovereign
wealth funds (SWFs) – has received limited attention in the IB
literature.
1
SWFs are state entities that invest their home countries’
budgetary surpluses abroad, through both mergers & acquisitions
Supplementary Information The on-
line version contains supplementary material
available at https://doi.org/10.1057/s41267-
021-00415-4.
Received: 7 June 2018
Revised: 4 December 2020
Accepted: 18 January 2021
Journal of International Business Studies (2021)
ª2021 Academy of International Business All rights reserved 0047-2506/21
www.jibs.net
(M&A) and purchases of securities. SWF capital
often derives from oil or other commodities (e.g.,
Abu Dhabi, Kuwait, Saudi Arabia), or from trade
surpluses (e.g., China, Singapore, Hong Kong).
There are now over 90 SWFs worldwide (SWF
Institute, 2020). They are some of the world’s
largest international investors, with about $8 tril-
lion in assets, estimated to exceed those of hedge
funds and private equity combined (Megginson &
Fotak, 2015). SWFs account for roughly the same
cross-border acquisition value as SOEs (Gilson &
Milhaupt, 2009; Guedhami, 2012).
This article seeks to shed light on SWFs as
understudied state investors. Expanding the scope
of the small SWF literature in IB that focuses
primarily on organizational issues (Aguilera et al.,
2016; Vasudeva, 2013; Vasudeva et al., 2018),
2
we
investigate whether geopolitics influences SWFs’
foreign acquisitions, asking if and how their FDI
patterns differ from those of private firms.
Existing literature suggests two competing views.
One view holds that since SWFs are state-owned
like SOEs, they are perceived to invest based on
interests of their home country rather than market
factors; thus, SWFs are more sensitive than private
firms to geopolitical factors. SWFs are more likely
than private firms to invest in countries that share
cooperative ties with their home country and less
likely to do so when the relations are adversarial. In
other words, state ownership moderates the effect
of geopolitics on firm international expansion.
An alternative view posits that although SWFs are
state-owned like SOEs, they differ in some crucial
dimensions. For example, according to Cuervo-
Cazurra & Li (forthcoming), ‘‘The indirect govern-
ment ownership of companies [through SWFs or
state banks] is usually driven by a desire to obtain a
return on investment…’’ SWFs are typically passive
investors, rather than operating entities, taking
minority positions and rarely involved in the
management of target firms (Kotter and Lel, 2011;
Gilson and Milhaupt, 2009; Megginson and Fotak,
2015). Hence, they do not expand abroad based on
their experience operating at home, and do not
build firm-level operational capabilities to be
exploited in foreign markets. Without exercising
operational control, SWFs may be less willing or
able to pursue non-commercial goals on behalf of
home governments. Thus, geopolitics should not
influence the host’s security concerns, foreign
investors’ political risk, or create differential access
to resources between SWFs and private investors.
Changes in home-host relations ought to affect the
international expansion of SWFs and private firms
similarly.
The two competing views suggest that whether
SWFs are more sensitive to geopolitics than private
firms is an empirical issue. To address it, we build a
dataset that covers over 5800 cross-border acquisi-
tions by SWFs and private-sector firms from ten
major SWF home countries in 88 target countries
from 1982 to 2012. M&A transactions provide a
good arena for examining geopolitics because host
governments tend to scrutinize such deals more
closely than greenfield investment.
We design three tests to answer three related
questions: (1) Taking the location of an acquisition
as given, do interstate relations affect the likelihood
that a deal is undertaken by an SWF instead of a
private firm? (2) (How) do interstate relations
influence the location choices of SWFs and private
investors? (3) Do interstate relations influence the
annual share of SWF deals in cross-border acquisi-
tions? The premise of our approach is that consis-
tent findings across tests provide convincing
evidence in adjudicating competing views. Hence,
one important contribution of our research is to
shed light on the investment behavior of a signif-
icant but understudied class of state investors.
Our research also contributes to the IB literature
by showcasing a novel database for measuring
home-host relations. We utilize a global database
of media reports of events between countries to
capture and separate conflict and cooperation in
home-host relations. The ‘‘big data’’ design relies on
text-mining algorithms and political events data
compiled by GDELT (Global Database of Events,
Language, and Tone), a large, freely-available,
comprehensive database of news media in over
100 languages from around the world.
Our statistical findings are broadly consistent in
two important ways: across all three tests noted
above and among three comparisons – SWFs vs.
private financial firms, SWFs vs. private non-finan-
cial firms, and SWFs vs. all private firms. Conflict
events between home and host countries hinder
SWFs more than private firms, whereas cooperation
events help SWFs more than private firms. Thus,
SWFs are more sensitive to geopolitical changes
than private firms in international expansion.
Despite SWFs’ lack of managerial control of target
firms, state ownership matters for international
investment. Our research demonstrates that gov-
ernment concern over FDI by state entities goes
beyond their operational activities.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
GEOPOLITICS AND SWFS
Definitions of geopolitics vary; we follow Caldaraand
Iacoviello’s (2018) as interactions among nations
associated with alliances, cooperation, wars, and
interstate tensions. Geopolitical relations between
countries can be cooperative or adversarial, which are
qualitatively different; their effects on foreign invest-
ment need not be similar in magnitude.
As a large literature in international relations
argues, geopolitics influences international eco-
nomic transactions – trade and investment are
higher among allies and lower between countries in
adversarial relationships (Pollins, 1989a,1989b;
Mansfield & Bronson, 1997; Morrow, Siverson &
Tabares, 1998; Li & Sacko, 2002; Kastner, 2007;
Biglaiser & DeRouen, 2007; Li, 2008; Li & Vash-
chilko, 2010).
The importance of geopolitics is in part due to the
national-security externalities that trade and invest-
ment generate, i.e., costs or benefits that fall on third
parties who are not part of a transaction (Kastner,
2007). In the context of foreign acquisitions, for
instance, managers of acquirer and target firms may
not consider national-security implications, but their
business transaction can generate security external-
ities for both home and host countries.
From the home state’s perspective, positive exter-
nalities may arise if firms repatriate earnings from
foreign operations, or obtain information contribut-
ing to economic, social, and military development at
home. By contrast, negative externalities could
include offshoring of jobs, capital, technology, and
managerial capabilities (Fors & Kokko, 2001). A key
concern is that rents from foreign investment are
converted into military or economic strength by the
host country (Kastner, 2007).
The host state’s concerns are analogous. A posi-
tive externality can arise, for instance, when an
acquisition spurs capital or technology transfer to
the host country (Smeets, 2008; Wei & Liu, 2006),
which may lead to military, economic, or other
benefits. On the other hand, host governments
often worry about negative ramifications that arise
from foreign control of domestic firms (Shi et al.,
2016). For instance, the host government’s ability
to leverage domestic commerce in the national
interest is reduced, or acquisition by foreign
investors can adversely affect the intellectual prop-
erty, efficiency, productivity, and competitiveness
of domestic companies.
Because of these potential externalities, both
home and host governments treatment of FDI
depends on geopolitics. In particular, home-gov-
ernment actions toward the host country – coop-
erative or adversarial – influence whether and how
a home-country investor expands in the host
country. If relations are cooperative, FDI is sup-
ported by the home government and welcomed in
the host country. For instance, home governments
provide diplomatic assistance (Gertz, 2018), subsi-
dize political risk insurance (e.g., OPIC in the
United States), provide privileged access to capital
(e.g. state-owned banks in China), and share earlier
and better information to develop business oppor-
tunities in the host country (Li et al., 2018).
Similarly, host governments facilitate investment
from friendly countries through targeted actions of
investment promotion agencies (Anderson, Suther-
land, & Severe, 2015), and by providing assistance
in identifying the right government offices, and
compliance with regulations.
In contrast, if home-country actions toward the
host country are adversarial, both home and host
governments may restrict investments. Home gov-
ernments can impose export controls or sanctions,
and host governments may raise regulatory barriers
(Bertrand et al., 2016; Shi et al., 2016; Heinemann,
2012). Taken together, cooperative relations
between countries should facilitate FDI, while
adversarial relations have the opposite effect.
Key to our analysis is whether these effects of
cooperation and conflict differ between SWFs and
private investors. On this issue, the literature
suggests two opposing views. The view that despite
state ownership, SWFs are no more sensitive to
geopolitics than private firms is based on SWFs’
limited managerial control. Consistent with SWFs
not playing an operating role, Kotter and Lel (2011)
find that long-term value changes are insignificant
in SWF-acquired firms. When SWFs invest in target
firms, they usually (but not always) acquire minor-
ity stakes (Gilson & Milhaupt, 2009); for example,
Megginson and Fotak (2015) report a median stake
of 20%. Even when they do acquire majority
positions, SWFs are seldom represented on targets’
boards of directors (Bortolotti et al., 2015: Table 2)
and are not typically involved in the management
of target firms.
3
Limited managerial control of target firms sug-
gests that SWFs may be unwilling or unable to
pursue political or other non-commercial goals.
Event-study evidence appears to be consistent with
this view; investments by SWFs are less political
than those by other state actors (such as ministries,
local and regional governments), and
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
announcements of SWF and private acquisitions
have similar effects on target value (Holland, 2019).
This implies that national security externalities
related to investment by SWFs and private investors
are similar. Consequently, geopolitical relations –
whether friendly or adversarial – do not influence
SWF and private investors differently.
Nonetheless, national security externalities could
differ between SWFs and private investors because
state ownership could influence SWF investments.
SWF managers may report to government officials,
both on their boards and as external monitors.
Indeed, Bernstein, Lerner, and Schoar (2013) find
that when politicians play a management role, SWFs
invest differently and obtain worse financial results
relative to SWFs without politician involvement.
Other scholars find that political motivations help
explain SWF’s investment patterns (Dyck and Morse,
2011; Grigoryan, 2016). Kaminski (2017) argues that
Chinese SWFs target European energy companies
and install representatives on company boards to
increase their political influence in Europe.
The discussion of cooperative and adversarial
relations above suggests that when home-host rela-
tions are cooperative, a foreign SWF’s geopolitical
goals are often aligned with those of the host
government such that SWF acquisitions result in
greater economic, political, and even military bene-
fits than acquisitions by private firms. SWFs, like
SOEs, are therefore better positioned than private
investors to leverage their home government’s diplo-
matic ties (Li et al., 2018), as well as more favorable
conditions in the host country. In contrast, acquisi-
tions by SWFs from countries with adversarial rela-
tions may cause acute national security concern in
the host country. Because home governments can
direct their SWFs more readily than their private
firms, host governments may be more sensitive to
investments by the former than those by the latter.
Thus, investment by state entities may be viewed as a
red flag (Shi et al., 2016), and SWFs may face more
scrutiny, greater restrictions, and lower legitimacy
than FDI by private firms from the same home
country (Li, Xia, and Lin, 2017).
RESEARCH DESIGN
Data on Cross-Border Acquisitions by SWFs
and Private-Sector Firms
To analyze how geopolitics influences foreign
investments of SWFs vis-a
`-vis private investors, we
compare their foreign acquisition transactions.
Foreign acquisitions provide a propitious area for
studying the impact of geopolitics, for two reasons.
First, as shown in Table 1, countries with the largest,
most active SWFs tend to be autocracies with some-
times-strained relations with the largest host coun-
tries (mostly OECD members). These autocracies
have limited political competition and oversight by
NGOs and the media, leaving leaders freer to utilize
state economic power to pursue political objectives.
Second, foreign acquisitions are more politically
sensitive than greenfield FDI, and host governments
often apply a variety of formal and informal tools to
regulate, discourage, or prevent them (Dinc & Erel,
2013; Heinemann, 2012).
We focus on cross-border acquisitions by SWFs
and private investors from ten home countries
from which most SWF FDI originates. We collect
and combine transaction-level data on cross-border
acquisitions from SDC Platinum, Sovereign Wealth
Fund Institute’s list of SWFs, Bureau van Dijk’s
Orbis, Zephyr, and Capital IQ (see Online Appendix
A for data collection procedures). Our initial dataset
contains 6811 acquisition deals by SWFs and
private firms. We restrict the sample by including
only acquisitions in which acquirers purchase
equity shares at or above the 10% ownership
threshold, per the UNCTAD definition of FDI.
Due to missing data, the final sample covers 5855
transactions by 3541 acquirers from ten SWF home
countries in 88 target countries from 1982 to 2012.
Table 1presents summary information in three
panels. Panel A lists the numbers of private and
SWF deals for the ten acquirer countries.
4
Among
the 5855 acquisitions, 401 deals are by 17 SWFs and
the rest by private firms. Panel B lists the numbers
of SWF and private deals for major target countries.
Comparing SWFs with private acquirers raises the
question of what should constitute the benchmark
group: private financial or non-financial firms.
Panel C presents data on target choices by three
types of acquirers: SWF, private financial, and
private non-financial. The majority of acquisitions
by SWFs (about 64%) target non-financial firms in
industries such as mining, oil and gas, transporta-
tion and shipping, and wholesale trade-durable
goods, compared to 60% by private financial
acquirers and 89% by private non-financial firms.
Since we only consider acquisitions where acquirers
have significant equity shares, we have no strong
theoretical reasons to believe that the difference in
sensitivity to geopolitics between SWFs and private
financial acquirers should differ from that between
SWFs and private non-financial acquirers, leaving
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
Table 1 Summary of Acquisition Transactions
Panel A: Deal count by acquirer country and type
Home Country Private deals SWF deals Total
Singapore 2588 221 2809
United Arab Emirates 372 82 454
Qatar 54 37 91
Kazakhstan 34 21 55
China 495 16 511
Malaysia 1310 10 1320
Oman 37 5 42
Kuwait 132 4 136
Libya 5 3 8
South Korea 427 2 429
Total 5454 401 5855
Panel B: Deal count by target country and type
Host country Private deals SWF deals Total Host country Private deals SWF deals Total
United Kingdom 221 48 269 Russia 33 12 45
United States 470 46 516 Thailand 265 12 277
China 1080 45 1125 Brazil 26 11 37
Australia 493 23 516 Singapore 348 11 359
Indonesia 456 23 479 France 56 10 66
India 249 22 271 Germany 99 10 109
Canada 120 15 135 Japan 136 10 146
Switzerland 37 13 50 South Korea 53 10 63
Malaysia 423 12 435 Others with \10 SWF deals 889 68 957
Panel C: Deal Count by acquirer and target type
Target firms Total
Financial Non-financial
SWF 144 257 401
Private Acquirers Financial 819 1230 2049
Non-financial 384 3021 3405
Total 1347 4508 5855
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
this benchmark choice as a largely empirical issue.
Therefore, we compare SWFs with (a) private finan-
cial acquirers, (b) private non-financial acquirers,
and (c) all private acquirers.
Three Empirical Tests
We test the impact of geopolitics on SWF and
private acquisitions using three related designs. The
first test draws on acquirer-characteristics models in
the M&A literature; the second draws on invest-
ment location choice models; the third draws on
the country-dyad model in Arikan and Shenkar
(2013).
5
The three tests address related questions;
consistent findings give us more confidence in our
analysis. Due to space constraints, control-variable
definitions, data sources, and summary statistics are
in the Online Appendix. For each test, the key
independent variables are cooperation and conflict
events, which are discussed below.
Test 1: Acquirer-characteristics model
The first test is at deal level, investigating why
acquirer characteristics differ, conditional on FDI
location. This approach is widely used in finance and
accounting (Avendan
˜o & Santiso, 2011; Bird et al.,
2017; Johan et al., 2013; Karolyi & Liao, 2017). It
assumes that acquirers – here SWFs and private
investors – compete for targets based on their
attractiveness and that winners are observed but
competing offers are not. Following this approach,
we model the likelihood of a transaction being
undertaken by an SWF instead of a private investor
as a function of home–host relations and other
covariates. This test focuses on the question: taking
the investment location as given, do cooperation
and conflict affect the likelihood that a target is
acquired by an SWF instead of a private firm?
The estimation sample has 5855 observations,
each a realized acquisition. The dependent variable
– acquirer characteristic – is coded one if the
acquirer in a deal is SWF and zero if it is a private
firm. If state ownership matters, cooperation
should increase the probability that the acquirer is
an SWF, and conflict should reduce that probabil-
ity. But if SWFs are similar to private investors,
cooperation and conflict should not affect that
probability. We estimate logit models, with robust
standard errors clustered over country dyads.
Test 2: Location-choice model
The second test is at potential-deal level, drawing
on the location-choice modeling for studying
determinants of FDI in the IB literature. In this
approach, a manager considers numerous potential
target locations and chooses one as the destination.
Our data structure, however, is more complex than
in the typical location-choice model. For each
potential deal, a potential target location is consid-
ered by both SWFs and private investors, and could
be chosen by an SWF, a private firm, or neither. We
test the impact of home-host relations on these
outcomes across potential deals, employing data
from Test 1 and constructing a new sample, in
which each observation is defined by ‘‘deal, poten-
tial-target-country’’. For any single deal, not all 88
target countries are included in the risk set of
potential investment destinations; a country is
included as a plausible deal destination only if the
target industry in the potential-target country has
at least one transaction during the sample period.
Hence, the sample includes 88 target countries,
though not for every deal.
Using this sample, Test 2 addresses the questions:
(how) do home-host relations influence the loca-
tion choices of SWFs and private investors? With
this sample design, we consider two aspects:
First, do cooperation and conflict influence loca-
tion choices of SWFs and private investors? Second,
do SWFs and private firms differ in sensitivity to
geopolitics? To answer these questions, we con-
struct a dependent variable that takes on three
nominal values: 0 if there is no deal for an
observation, 1 for each deal by a private acquirer,
and 2 for each deal by an SWF.
Since the dependent variable has three nominal
values, we estimate multinomial-logit (MNL) mod-
els with robust standard errors. MNL allows us to
test whether a covariate such as cooperation or
conflict affects the likelihood of the dependent
variable taking on a value of 1 (private investment)
relative to 0 (no investment), the likelihood of it
taking on a value of 2 (SWF investment) relative to
0, and the likelihood of it taking on a value of 2
relative to 1. We expect cooperation to increase and
conflict to decrease: (1) the likelihood of a private
firm investing in a potential target country relative
to no investment, (2) the likelihood of SWF invest-
ing in a potential target country relative to no
investment, and (3) the likelihood of SWF investing
in a potential target country relative to a private
firm (the focus of our analysis).
We also estimate the models in Test 2 using
conditional logit (CL), since CL is often employed
in location-choice modeling. It is worth noting that
MNL explains location choice using investor char-
acteristics as explanatory variables, whereas CL
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
conditions these out, and explains location choice
by host-country characteristics (Greene, 2003; Long
and Freese, 2006). Here outcomes are driven by
home-host relations, which are neither purely
home nor purely host characteristics. Hence, nei-
ther MNL nor CL is perfect. We employ MNL
because it compares acquirors directly, whereas CL
relies on interaction terms that are difficult to
interpret.
1
Consistency across MNL and CL should
provide more confidence in Test 2 results.
Test 3: Country-dyad model of annual share of SWF
acquisition deals
The third test follows the country-dyad model in
Arikan and Shenkar (2013). The dependent variable
is the annual share of SWF deals over total acqui-
sition deals between two countries. As Karolyi and
Liao (2017), we aggregate data to home–host-
country-pair level, which reduces noise inherent
in transaction-level data. Each observation is
defined by ‘‘acquirer-country, target-country year.’’
Test 3 addresses the question: do cooperation and
conflict impact the annual share of SWF deals? If
SWFs are more sensitive to geopolitics, cooperation
should raise the share of SWF acquisitions, while
conflict decreases it. Since the dependent variable is
continuous but bounded between 0 and 1 and
mostly = 0, we estimate Tobit models (see Wool-
dridge, 2010 for details), with robust standard
errors clustered over dyads; OLS produces similar
results.
Measuring Interstate Conflict and Cooperation
Using Events Data
Many studies in IB and finance measure interstate
relations using an indicator of similarity or dissim-
ilarity of UN General Assembly voting records
between home and host countries, whose limita-
tions are discussed later. Instead, we exploit data
from text mining ‘‘events’’ – interactions between
home and host countries – as recorded in news
media. Prior research shows that content analysis of
press reports can provide time-varying and cross-
nationally-comparable measures that are better
suited to theoretical constructs (Zelner et al.,
2009).
6
Relative to the UN voting measure of
foreign-policy distance, events data has an advan-
tage of particular relevance to IB research – it allows
home–host relations to be categorized as coopera-
tive, neutral, or adversarial. As noted in the theo-
retical section, host- and home-government
responses to FDI differ between friendly and
unfriendly partners. Moreover, distinguishing
effects of cooperative and adversarial relations on
FDI relaxes the assumption that cooperation and
conflict effects are of equal magnitude.
A second advantage is that media-based events
data are more likely to be managerially relevant.
Media reports can directly affect managerial deci-
sion-making, including M&A (Liu & McConnell,
2013). Media reports often are a part of MNE
managers’ scanning process, thus influencing deci-
sion-making on internationalization. Because they
cover a wide variety of behavior, as shown in
Online Appendix Table B1, events data capture
more types of interactions that are relevant to
managers than just military disputes, sanctions,
troop deployment, and alliances studied in previ-
ous IB research.
Our cooperation and conflict measures are drawn
from GDELT, a novel comprehensive database of
global events (Leetaru & Schrodt, 2013; see https://
www.gdeltproject.org). It draws on the world’s
news media in print, broadcast, and web formats in
over 100 languages; it has been used in political
science research (e.g., Davis et al., 2019), but has
not appeared in the IB literature. An important
characteristic of the database is that news reports
are coded using automated machine-based algo-
rithms. The ‘‘big data’’ approach helps to identify
and interpret far more news than previous genera-
tions of events data used in studies of FDI flows
(Desbordes, 2010; Nigh, 1985).
Each news event includes the actors involved
(their location, ethnic, religious, and group attri-
butes) and the actions performed, and is tagged
with attributes that help to assess the effect and
tone involved. Moreover, each event is mapped
onto Goldstein’s (1992) conflict-cooperation scale
(Online Appendix Table B1), which reflects the
nature and intensity of the event and is widely used
in international relations research. Negative scores
represent conflict events, positive scores represent
cooperation events, and zero indicates neutral
events. The scale ranges from - 10 (most conflict-
ual) to + 10 (most cooperative). The larger the
absolute value of the Goldstein score, the more
important or significant the event.
1
Our CL estimation employs a dichotomous DV (1 if invest-
ment, 0 if no investment) and adds to the IVs an SWF dummy
variable (1 if acquiror firm is an SWF, 0 otherwise), and
interaction terms between SWF and 1) cooperation and 2)
conflict. Results are reported under sensitivity analysis. However,
in non-linear models a moderator’s effects and p-value do not
depend solely on its estimated coefficient and standard error
(Hoetker, 2007; Wiersema and Bowen, 2009).
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
We take the GDELT data on events from the
home government toward the host government,
select events that involve significant cooperation
and conflict actions above meaningful Goldstein
score thresholds,
7
and create dyadic measures of
cooperation based on the following formula:
cooperationeventsijt ¼log Pfcoopwcoop þ1
;where
subscripts i, j, and trepresent home, host, and
year, respectively, f
coop
represents the frequency of
each type of significant cooperation events, and
w
coop
represents the Goldstein score for that type.
We add one to the formula so that dyads with only
neutral or insignificant events take on zero rather
than missing values. Following Davis et al. (2019),
we log-transform the variable to address skewness
and reflect diminishing effects of multiple events.
Higher values indicate more-cooperative relations.
Conflict is measured as conflicteventsijt ¼
log Pfconf wconf
þ1
, where f
conf
and w
conf
repre-
sent the frequency of each type of significant
conflict events and its corresponding Goldstein
score, respectively, and the absolute value is
applied to remove the negative sign of w
conf
in the
Goldstein scale for conflict events. Higher values
indicate more-adversarial relations. To illustrate the
data and measures, Figure 1plots significant coop-
eration and conflict events between China and the
United States from 1982 to 2012.
Notwithstanding its relevance and advantages, a
weakness of GDELT data is worth noting. In
maximizing its collection of news reports on
events, GDELT often fails to eliminate duplicates
(Ward et al., 2013). While this poses problems for
analysts of events themselves, duplication resulting
from multiple reports of an event can indicate
event salience to managers, and thus may be a
meaningful signal when they process news and
conduct risk assessment.
FINDINGS
Tables 2,3, and 4present estimation results from
three tests, respectively: the acquirer-characteristics
model of acquisition deals, the investment loca-
tion-choice model of potential deals, and the
country-dyad model of annual share of SWF acqui-
sition deals. For each test, we estimate three
models: Models 1–3 compare SWFs with all private
firms, private financial acquirers, and private non-
financial acquirers, respectively.
Test 1: Acquirer-Characteristics Model
The results in Table 2provide broad support for the
proposition that SWFs are more sensitive to geopol-
itics than private firms. The effect of cooperation is
positive and statistically significant in all but Model
Figure 1 Events between China and USA, 1982–2012.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
2(b(cooperation) = 0.1798, p= 0.0756; b(con-
flict) = -0.1430, p= 0.0577 in Model 1). The effect
of conflict events is negative and statistically
significant in all three models.
How big are the effect sizes of cooperation and
conflict? In a non-linear model like logit, the effect
of one variable on the probability of success
depends on the values of other covariates. The best
way to demonstrate effect size is to compute the
change in probability of SWF acquisition between
two substantive scenarios. Using the coefficient
estimates in Model 1, we first compute the proba-
bility of SWF acquisition at a baseline scenario –
non-financial target, non-energy target, no British
common law, no common language, and average
values for continuous variables (institutional dis-
tance, regime distance, geographical distance, tar-
get fuel export, home GDP, host GDP, cooperation
events, and conflict events). We then compute the
probability of SWF acquisition by raising the value
of conflict or cooperation to one standard deviation
above mean. Finally, we compute the percent
change in the probability of SWF acquisition from
mean to one standard deviation above mean to
illustrate the importance of the substantive effect.
At the baseline scenario and setting conflict and
cooperation events at their mean levels (6.06, 5.28),
the probability of SWF acquisition is 0.0399. With a
one-standard-deviation increase in cooperation
events (6.06 + 1.92 = 7.98) while holding other
variables constant, the probability rises to 0.0562.
Hence, a one-standard-deviation increase in coop-
eration events above its mean is associated with a
40.85% increase in the probability of SWF
acquisition.
In terms of conflict, with a one-standard-devia-
tion increase in conflict events (5.28 + 2.16 = 7.44)
while holding other variables constant, the proba-
bility of SWF acquisition declines to 0.03. Hence, a
one-standard-deviation increase in conflict events
above its mean is associated with a 24.62% decrease
in the probability of SWF acquisition.
Table 2 Effects of conflict and cooperation in acquirer-characteristics models
Variables (1) (2) (3)
SWF versus all private SWF versus private financial SWF versus private non-financial
Dependent variable = likelihood of acquisition by SWF
Cooperation events 0.1798 0.1344 0.2246
(0.0756) (0.1715) (0.0271)
Conflict events -0.1430 -0.1430 -0.1729
(0.0577) (0.0480) (0.0452)
Target financial 0.8709 0.0586 1.7476
(0.0000) (0.7541) (0.0000)
Target energy 0.7768 1.1024 0.6187
(0.0074) (0.0001) (0.0591)
Institutional distance 0.3143 0.3130 0.3251
(0.0137) (0.0198) (0.0086)
Geographical distance 0.4323 0.4385 0.3717
(0.0073) (0.0258) (0.0110)
Regime distance 0.1697 0.1566 0.1826
(0.0000) (0.0000) (0.0000)
Target fuel export 0.0111 0.0076 0.0157
(0.0623) (0.2447) (0.0235)
British common law -0.0337 -0.1396 0.1193
(0.9012) (0.6011) (0.7159)
GDP home -0.5313 -0.4105 -0.6306
(0.0000) (0.0007) (0.0000)
GDP target 0.0661 0.0911 0.0543
(0.4529) (0.3446) (0.5758)
Common language -0.2941 -0.1421 -0.4735
(0.3220) (0.6254) (0.1704)
Constant 3.2315 0.9654 6.7012
(0.3752) (0.7946) (0.0894)
Observations 5855 2450 3806
Logit coefficient estimates, with robust pvalue in parentheses.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
Results for control variables are largely as
expected, although some are not statistically sig-
nificant. For example, relative to private firms,
SWFs are more likely to acquire targets (1) in the
financial and energy sectors, and (2) in host
countries that have high fuel exports, a higher
institutional and geographical distance from the
home country, and different political regime types.
Test 2: Investment Location Choice Model
Table 3reports model results for the same three
benchmarks as in Table 2. The only difference is
that because the dependent variable has three
nominal values, we report three sets of estimates
for each model: private investment (1) vs. no
investment (0), SWF investment (2) vs. no invest-
ment (0), and SWF (2) vs. private investment (1).
Across all three models, the effects of cooperation
are consistently positive and statistically
significant, as expected. Both private investors
and SWFs are more likely to invest in host countries
with which their home countries experience sig-
nificant cooperation events. Moreover, this positive
impact is significantly larger for SWFs than private
firms (b= 0.1519, p= 0.0171 in Model 1).
The effects of conflict, however, while consistent
across all three models, are not always as expected;
specifically, the effect of conflict on the likelihood
of investment by private firms is not negative, but
positive and significant across all three models.
8
The effect of conflict on the likelihood of invest-
ment by SWFs is negative as expected, but statisti-
cally insignificant. Nonetheless, the effect of
conflict is significantly stronger for SWFs than
private firms across all three models
(b=-0.1514, p= 0.0004 in Model 1).
How important substantively are the estimated
effects of cooperation and conflict? We follow the
Table 3 Effects of conflict and cooperation in location-choice models
(1) SWF versus all private (2) SWF versus private financial (3) SWF versus private non-
financial
1 versus 0 2 versus 0 2 versus 1 1 versus 0 2 versus 0 2 versus 1 1 versus 0 2 versus 0 2 versus 1
Dependent variable = 0 if no deal, 1 if private deal, 2 if SWF deal
Cooperation events 0.1251 0.2770 0.1519 0.1266 0.2562 0.1296 0.1147 0.2760 0.1613
(0.0000) (0.0000) (0.0171) (0.0000) (0.0000) (0.0490) (0.0000) (0.0000) (0.0129)
Conflict events 0.1368 -0.0146 -0.1514 0.1579 -0.0178 -0.1756 0.1351 -0.0126 -0.1476
(0.0000) (0.7220) (0.0004) (0.0000) (0.6685) (0.0002) (0.0000) (0.7602) (0.0009)
Target financial -0.2109 0.5317 0.7427 -0.1844 -0.2324 -0.0480 -0.3671 1.1398 1.5068
(0.0000) (0.0000) (0.0000) (0.0000) (0.0182) (0.6749) (0.0000) (0.0000) (0.0000)
Target energy -0.0734 0.8004 0.8738 -0.2314 0.7722 1.0036 -0.0743 0.7354 0.8097
(0.0307) (0.0000) (0.0000) (0.0025) (0.0000) (0.0000) (0.0754) (0.0000) (0.0000)
Institutional distance -0.0540 0.1614 0.2154 -0.0587 0.1692 0.2279 -0.0607 0.1662 0.2269
(0.0001) (0.0223) (0.0033) (0.0072) (0.0145) (0.0025) (0.0008) (0.0155) (0.0018)
Geographical
distance
-0.5041 -0.3868 0.1173 -0.4926 -0.3748 0.1178 -0.5073 -0.3895 0.1178
(0.0000) (0.0000) (0.0931) (0.0000) (0.0000) (0.1014) (0.0000) (0.0000) (0.1020)
Regime distance -0.0165 0.1571 0.1736 -0.0175 0.1312 0.1487 -0.0229 0.1565 0.1794
(0.0000) (0.0000) (0.0000) (0.0007) (0.0000) (0.0000) (0.0000) (0.0000) (0.0000)
Target fuel export -0.0068 0.0006 0.0074 -0.0047 0.0000 0.0047 -0.0101 0.0007 0.0107
(0.0000) (0.7946) (0.0040) (0.0000) (0.9962) (0.0768) (0.0000) (0.7797) (0.0000)
British common law -0.0580 -0.0873 -0.0293 -0.0839 -0.0624 0.0215 -0.0291 -0.0775 -0.0484
(0.1011) (0.4789) (0.8181) (0.1313) (0.6107) (0.8712) (0.5196) (0.5282) (0.7082)
GDP home -0.1286 -0.6104 -0.4819 -0.1638 -0.4897 -0.3259 -0.0872 -0.6162 -0.5289
(0.0000) (0.0000) (0.0000) (0.0000) (0.0000) (0.0003) (0.0000) (0.0000) (0.0000)
GDP target 0.2666 0.3265 0.0599 0.2316 0.3422 0.1107 0.2882 0.3342 0.0460
(0.0000) (0.0000) (0.2249) (0.0000) (0.0000) (0.0326) (0.0000) (0.0000) (0.3558)
Common language 0.9052 0.7373 -0.1678 0.7678 0.7474 -0.0204 0.9728 0.7390 -0.2337
(0.0000) (0.0000) (0.2074) (0.0000) (0.0000) (0.8858) (0.0000) (0.0000) (0.0846)
Constant -4.2259 0.7201 4.9460 -2.5948 -1.4437 1.1511 -5.7802 1.0053 6.7855
(0.0000) (0.7627) (0.0462) (0.0020) (0.5544) (0.6724) (0.0000) (0.6541) (0.0044)
Observations 165,657 165,657 165,657 69,773 69,773 69,773 107,281 107,281 107,281
Multinomial logit coefficient estimates, with robust pvalues in parentheses.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
set-up in Test 1 to compute the substantive effects
using Model 1 estimates and report the results
below. At the baseline scenario and setting conflict
and cooperation events at their mean levels (3.46,
4.44), the probability of private investment is
0.0161 and the probability of SWF investment is
0.0008.
For cooperation events, with a one-standard-devi-
ation increase (4.44 + 2.33 = 6.77) while holding
other variables constant, the probability of private
investment rises to 0.0214 and the probability of
SWF investment rises to 0.0015. Hence, a one-
standard-deviation increase in cooperation events
above its mean is associated with a 32.92% increase
in the probability of private investment and an
87.50% rise in the probability of SWF investment.
The increase in the probability of SWF investment is
much larger than that of private investment.
For conflict events, with a one-standard-devia-
tion increase (3.46 + 2.51 = 5.97) while holding
other variables constant, the probability of private
investment rises to 0.0226 and the probability of
SWF investment declines to 0.0007. Hence, a one-
standard-deviation increase in conflict events
above its mean is associated with a 40.37% increase
in the probability of private investment and a
2.63% decline in the probability of SWF invest-
ment, corresponding to a large difference in the
changes in probability between private firms and
SWFs.
Test 3: Country-Dyad Model of Annual Share
of SWF Acquisition Deals
Table 4reports the results for three models; across
all three, cooperation events increase the annual
share of SWF deals in total acquisitions between
two countries, while conflict events reduce it. These
effects are in the expected directions (b(coopera-
tion) = 0.2094, p= 0.0016; b(conflict) = -0.0722,
p= 0.1022 in Model 1).
Since conflict and cooperation variables are log
transformed but not the dependent variable, the
effect size should be coefficient * log(1.01) or
roughly coefficient/100 for a 1% change in an
independent variable. Based on the estimate of
cooperation (0.2094) in Model 1, a 1% increase in
cooperation events is associated with a 0.0021-unit
increase in the share of SWF deals, and a 10%
increase is associated with a 0.02-unit increase.
Considering the annual share of SWF deals has a
Table 4 Effects of conflict and cooperation in models of annual share of SWF deals
Variables (1) (2) (3)
SWF versus private SWF versus private financial SWF versus private non-financial
Dependent variable = annual share of SWF deals over total deals in dyad
Cooperation events 0.2094 0.2043 0.2923
(0.0016) (0.0201) (0.0014)
Conflict events -0.0722 -0.1293 -0.1294
(0.1022) (0.0565) (0.0393)
Institutional distance 0.1815 0.2360 0.1990
(0.0108) (0.0257) (0.0326)
Geographical distance 0.0503 0.1738 0.0316
(0.6004) (0.2927) (0.7941)
Regime distance 0.0899 0.1252 0.1215
(0.0000) (0.0000) (0.0000)
Target fuel export 0.0038 0.0058 0.0104
(0.2217) (0.2366) (0.0392)
British common law -0.1653 -0.2710 -0.2670
(0.2620) (0.2386) (0.1969)
GDP home -0.3460 -0.2882 -0.5295
(0.0001) (0.0110) (0.0001)
GDP target 0.1139 0.1440 0.1304
(0.0600) (0.1037) (0.1390)
Common language 0.2337 0.1816 0.2282
(0.1665) (0.4605) (0.3442)
Constant 2.5076 -0.7073 6.3587
(0.2927) (0.8290) (0.0776)
Observations 1435 897 1132
Tobit coefficient estimates, with robust pvalue in parentheses.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
mean of 0.086, the effect size of a 10% increase in
cooperation events is nearly a quarter of the mean
level.
Based on the estimate of conflict (-0.0722) in
Model 1, a 1% increase in conflict events is
associated with a 0.00072-unit decrease in the
annual share of SWF deals, and a 10% increase is
associated with a 0.0069-unit decrease. Considering
the annual share of SWF deals has a mean of 0.086,
the effect size of a 10% increase in conflict events is
about 8% of the mean level.
Given the large number of estimation results and
substantive scenarios, we summarize the substan-
tive effects across the three groups of tests for reader
convenience; Table 5collects and presents the
empirical estimates discussed above, showing that
SWFs are consistently more sensitive to geopolitical
changes than private investors.
SENSITIVITY ANALYSES
Multicollinearity
One possible issue with the results in Tables 2-4is
that conflict and cooperation are highly correlated
(0.85, 0.78, 0.79 for the three test samples, respec-
tively; see Appendix Table B4 for correlation matri-
ces). As widely discussed in the statistics literature
and recently noted in JIBS (Lindner et al., 2020),
multicollinearity does not bias coefficient estimates
or standard errors, and ttest estimates remain valid.
However, it does increase standard errors, making it
harder to find significant results (e.g., insignificant
effect of cooperation in Model 2 of Test 1);
estimates could become sensitive to small changes
in data and model specification. As widely noted,
multicollinearity becomes a serious concern when
the diagnostic statistic–variance inflation factor
(VIF) – exceeds the threshold value 10. Computed
VIF statistics for conflict and cooperation in Model
1 are 4.71 and 3.91 in Test 1, 3.07 and 3.01 in Test
2, and 3.32 and 2.99 in Test 3, respectively.
Aggregate measure of conflict-cooperation:
diplomatic risk
As discussed earlier, cooperation and conflict influ-
ence investors via separate mechanisms. Thus their
effects should be tested separately. Indeed, across the
three tests, effect sizes differ between conflict and
cooperation. However, cooperation and conflict
variables are highly correlated, making standard
errors larger than usual and our tests more conser-
vative. An alternative approach, as in Desbordes’s
(2010) study of FDI flows, constructs a net measure of
conflict and cooperation, assuming that the mar-
ginal effect of geopolitical influence is constant
along the scale from conflict to cooperation. Des-
bordes (2010) terms his measure diplomatic risk:
Diplomatic riskijt ¼
Pfcoopwcoop þ0þPfconf wconf
Pfcoop þPfconf þfneut
;
where the subscript neut represents neutral events,
and w
neutral
equals 0 (in the Goldstein scale), hence
Table 5 Summary of effect magnitudes
Test 1 Test 2 Test 3
Probability of acquisition by
SWFs
Probability of investment outcome Annual
share of SWF
deals over
total deals
Baseline One sd
increase
Percentage
change
SWF investment Private investment 10%
increase
Baseline One sd
increase
Percentage
change
Baseline One sd
increase
Percentage
change
Cooperation
events
0.0399 0.0562 +40.85% 0.0008 0.0015 +87.5% 0.0161 0.0214 +32.92% +0.02
Conflict
events
0.0399 0.030 -24.62% 0.0226 0.0007 -2.63% 0.0007 0.0226 +40.37% -0.0069
Notes 1. Baseline scenario-non-financial target, non-energy target, no British common law, no common
language, and average values for continuous variables (institutional distance, regime distance,
geographical distance, target fuel export, home GDP, host GDP, cooperation events, and conflict events)
2. Substantive effects computed based on Model 1 in Tables 2,3, and 4, respectively
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
the zero in the numerator. Higher values represent
greater diplomatic risk, so the variable should have
a negative effect in all three tests to be consistent
with our main results. The results are reported in
Online Appendix Table B5. With the exception of
its effect on the likelihood of private investment in
Test 2, the estimated effects are negative, and
almost all statistically significant.
UN voting dissimilarity
As noted above, a widely used measure of interstate
relations is similarity or dissimilarity of UN General
Assembly voting records between home and host
countries (see Bailey et al., 2017 for details and
construction, Bertrand et al., 2016; Li et al., 2018;
Knill et al., 2012 for applications). While this mea-
sure is available for many dyad-years, conceptual
and methodological issues limit its relevance in IB.
The measure does not and is not designed to capture
political, diplomatic, or other interactions, nor does
it capture affinity between countries. Rather, it
indicates how aligned the foreign policies of two
countries are, based on UN resolutions on global
issues. Indeed, most votes are about third parties
rather than relations between countries (Becker
et al., 2015), and many dyads with significant
tensions actually have high UN voting similarity;
e.g., Iran–Iraq, India–Pakistan, Peru–Ecuador, and
Eritrea–Ethiopia (Voeten, 2013). Finally, bloc vot-
ing, symbolic voting, and vote-buying raise ques-
tions about the measure’s relevance for managers.
In a study related to ours, Knill et al. (2012) find that
SWFs tend to invest in countries with more UNGA
voting dissimilarity with their home countries, con-
cluding thatSWFs invest in countries with which they
have weaker political relations. This result is incon-
sistent with the consensus in the IR and IB literatures
that better relations fosters trade and FDI (see meta-
analysis by Moons & Van Bergeijk, 2017).
9
Our approach differs from Knill et al. (2012)in
two ways. We employ media events to measure
geopolitics, and benchmark SWFs against private
investors, while they examined only SWFs. As a
sensitivity analysis, we re-estimate all models using
UNGA voting; results are in Online Appendix
Table B6. None of the estimated effects reach
conventional statistical significance levels, except
for its effect on private investment in Test 2.
Additional robustness tests
We conduct and report additional tests in Online
Appendix due to space constraints. First, to avoid
potential bias from measuring conflict and
cooperation using the same data source, we replace
cooperation with a military-alliance dummy from
the Correlates of War Project and re-estimate all
models. Appendix Table B7 shows that alliance has
the expected positive and significant effect in
almost all models across the three tests. The effect
of conflict events is much weaker and less consis-
tent. Since the alliance dummy does not capture all
cooperation events, the effect of conflict is con-
founded by omitted cooperation events.
Second, over half of our SWF transactions are
Singapore-based (see Table 1). For sensitivity anal-
ysis, we exclude Singapore transactions and re-
estimate all models. In Appendix Table B8, the
effect of cooperation remains robust across three
tests; the effect of conflict remains robust in Test 2,
and has the expected sign in Tests 1 and 3 but
becomes largely insignificant at the 10% level.
Third, the UNCTAD FDI threshold of 10% own-
ership excludes some block acquisitions; many
countries use 5% as the cut-off to characterize
blockholders, whose stock ownership must be
disclosed. Holdings of at least 5% are also associ-
ated with the ability of activist shareholders to
influence management, and announcements are
associated with stock price movements (Edmans &
Holderness, 2017). For sensitivity analysis, we re-
estimate all models using the 5% cut-off. The
number of deals rises to 6843, of which 491 are
by SWFs. In Appendix Table B9, the effects of
cooperation and conflict events are consistent with
and often more significant than Tables 2,3, and 4.
Finally, as noted above, we re-estimate the mod-
els in Test 2 using CL. In Appendix Table B10, the
results are broadly consistent with MNL models,
except that coefficients on the interaction term
between SWF and cooperation turn insignificant.
Not surprisingly, the results are weaker than in
MNL because of more serious multicollinearity.
DISCUSSION AND CONCLUSION
SWFs are major actors in the global economy; as
part of the rise of state capitalism, both their
numbers and assets have increased dramatically
(World Economic Forum, 2017). China Investment
Corporation has more than $940 billion in total
assets while SWFs of Abu Dhabi, Kuwait, and Hong
Kong each have more than $500 billion in assets
(SWF Institute, 2020).
In spite of their rapid rise, SWFs are less well
understood than other state actors such as SOEs
and have raised great concern in host countries.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
The countries with most active SWFs tend to be
autocracies with sometimes-strained relations with
the largest host countries (mostly OECD members,
see Panel B Table 1), as well as limited political
competition and oversight by NGOs and the media,
leaving leaders freer to utilize state economic power
to pursue political objectives.
Concern over acquisitions from these countries is
expressed in part through claims that SWFs should
pursue ‘‘commercial objectives,’’ and led to creation
of a supranational institution for voluntary self-
regulation, the International Working Group of
Sovereign Wealth Funds (IWGSWF), now known as
International Forum of Sovereign Wealth Funds
https://www.ifswf.org/. Indeed, SWF investment
charters typically state that ‘‘the fund seeks to
maximize financial returns’’ or ‘‘the fund invests in
assets on a commercial basis’’ (IWGSWF, 2008).
This concern may be surprising, given the passive
nature of SWF investments. However, large share-
holders not involved in management may never-
theless influence firm governance (Aguilera et al.,
2016; Appel et al., 2016), including information
disclosure. SWF investment can also affect target
operations and competitiveness, through reducing
its financial constraints.
10
Concern may reflect
host-country discomfort with state capitalism itself.
As Shi et al. (2016: 14) observe, state ownership
may be interpreted as a representation of the
sovereign state, and globalization of the entity
may ‘‘challenge the national sovereignty of target
states and affect international relations between
home states and target states.’’ Hence, by shedding
light on SWF investment, our research should be of
interest to scholars, policymakers, and managers.
Our analysis also contributes to three broader
streams of IB research. First, our study contributes
to the literature on home-host relations. Media
events offer a time-varying measure that allows for
longitudinal analysis. They also reflect more closely
how managers assess the international landscape.
Thus, they can complement widely-used measures
such as colonial history, cultural distance, institu-
tional distance, trust and animosity (Arikan and
Shenkar, 2013), formal alliances (Kandogan &
Hiller, 2018; Li and Vashchilko, 2010) and interna-
tional investment agreements (Jandhyala & Wei-
ner, 2014).
While our focus is construction of measures of
international relations, media-based event data also
provide IB scholars opportunities to create cross-
country comparative measures, e.g. stakeholder
networks (Odziemkowska and Henisz, 2020),
political risk. Their availability at high frequency
also allows exploration of crisis dynamics (Berkman
et al., 2011). Given the automated coding algo-
rithms, these data allow us to capture events on a
scale far greater than archival methods.
Second, our research contributes to the small
literature on FDI by financial investors (Dai &
Nahata, 2016; Guler & Guille
´n, 2010; Wright et al.,
2005). Like venture-capital firms, SWFs take large
stakes in their targets. Our location-choice model
results show that FDI by private financial investors
is sensitive to geopolitics, though less so than FDI
by SWFs.
Third, our study contributes to the literature on
home–host relations. Media events offer a time-
varying measure that allows for longitudinal anal-
ysis. They also reflect more closely how managers
assess the international landscape. Thus, they can
complement widely used measures such as colonial
history, cultural distance, institutional distance,
trust and animosity (Arikan & Shenkar, 2013),
formal alliances (Kandogan & Hiller, 2018;Li&
Vashchilko, 2010) and international investment
agreements (Jandhyala & Weiner, 2014).
For scholars of state ownership, examining SWFs
offers empirical advantages over SOEs in compar-
isons with private firms. SOEs differ from private
firms along several dimensions beyond ownership,
hindering identification of determinants of differ-
ences in FDI patterns. Potential variation in inter-
national expansion outcomes between SOEs and
private firms may stem from differences in opera-
tional capabilities, and in their experiences at
home. In contrast, SWFs are not operating entities,
and invest abroad.
Moreover, SOEs receive a variety of benefits from
state ownership, which may not be geopolitical in
nature; e.g., preferential access to capital through
soft budget constraints, explicit or implicit finan-
cial guarantees, freedom from paying dividends,
and subsidized loans from state banks (Luo et al.,
2010; Wang et al., 2012).
11
SWFs as financial
investors, however, are a source rather than the
user of funds. Hence, they do not need soft budget
constraints, government bailouts, or other forms of
state financial assistance.
Finally, many SOEs are hybrid organizations –
both publicly listed and partially owned by private
shareholders. This mixed ownership affects their
objectives, external monitoring, and internal gov-
ernance (Bruton et al., 2015; Musachio et al., 2015;
D’Souza & Nash, 2017; Musachio & Lazzarini,
2018). In contrast, SWFs are wholly state-owned
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
entities. As a result, SWF FDI more likely reflects
state interests rather than conflicts among owners
with different objectives, which are more severe for
acquisitions (Wehrheim et al., 2020).
We acknowledge limitations of several types.
Employing events data can be tricky because the
accuracy and unbiasedness of text-parsing algo-
rithms in capturing interstate interactions across
multiple languages and media is difficult to assess.
SWFs as research sites have limitations as well.
First, while their financial-institution nature facil-
itates identification, it also raises issues of external
validity. While SWFs are ubiquitous, those engag-
ing in FDI are primarily from emerging markets.**
Cuervo-Cazurra (2018b) suggests that emerging-
market SOEs may be particularly unwelcome
abroad, potentially limiting generalizability. Sec-
ond, SWFs vary significantly in their transparency
and corporate governance. Our analysis does not
explore this variation even though these differ-
ences may induce varying investment strategies.
Moreover, the fact that SWFs have no reporting
requirements prevents researchers from seeing their
investment portfolios; only changes are observable.
Finally, we are unable to address the possibility that
countries might use SOE investment to signal
interest in improving relations (Duanmu, 2014).
Future research may explore these issues.
NOTES
1
SWFs are excluded from Rygh’s (2019) compre-
hensive survey, and scarcely mentioned in a recent
collection of research on state-owned multination-
als (Cuervo-Cazurra, 2018a).
2
Exceptions include Johan et al. (2013) on private
vs. public status of SWF targets, and Calluzzo et al.
(2017) on SWF investments in US firms.
3
For instance, Temasek, one of the largest SWFs,
states ‘‘Our portfolio companies are guided and
managed by their respective boards and
management; we do not direct their business
decisions or operations.’’ https://www.temasek.
com.sg/en/who-we-are/our-purpose.html.
4
Some finance studies include SWF portfolio
investment (e.g., Kotter & Lel, 2011), generating
larger samples. For example, Norway’s SWF does
not undertake FDI and hence is not in our data.
5
We thank an anonymous reviewer for suggesting
location-choice and country-dyad-level modeling
approaches.
6
Media text mining has also been used to develop
univariate country-level measures of economic
policy uncertainty (Baker et al., 2016) and global
geopolitical risk (Caldara & Iacavello, 2018).
7
Cooperation events with Goldstein scores above
5.2 (‘‘promise material support’’) and conflict
events with scores below -2.2 (‘‘cancel or post-
pone planned events’’) are selected, as minor events
are unlikely to sway firm managers or policymakers.
Our findings are robust to alternative thresholds.
8
One possible explanation is that frequent con-
flictual interactions may correlate with both fre-
quent cooperative actions, and investment
opportunities. For example, despite strained rela-
tions, Chinese private firms may find the United
States more attractive than countries with which
China exhibits fewer interactions – positive or
negative.
9
This counterintuitive result may be an artifact of
SWF home countries being mostly autocracies and
main host countries being industrialized democra-
cies. Autocracies such as Singapore and the UAE
tend to vote differently from industrialized democ-
racies at the UNGA, despite strong diplomatic ties.
10
Kotter and Lel (2011) find that SWFs target
financially constrained firms. In some cases, SWF
capital has been critical in alleviating target finan-
cial distress; e.g., the Qatar SWF’s investment in
Barclay’s Bank in 2008.
11
Finance research (surveyed in Megginson,
2017) finds that state ownership lowers firms’ cost
of capital, and hence their investment decisions.
REFERENCES
Aguilera, R. V., Capape
´, J., & Santiso, J. 2016. Sovereign wealth
funds: A strategic governance view. Academy of Management
Perspectives, 30(1): 5–23.
Anderson, J., Sutherland, D., & Severe, S. 2015. An event study
of home and host country patent generation in Chinese MNEs
undertaking strategic asset acquisitions in developed markets.
International Business Review, 24(5): 758–771.
Appel, I. R., Gormley, T. A., & Keim, D. B. 2016. Passive
investors, not passive owners. Journal of Financial Economics,
121(1): 111–141.
Arikan, I., & Shenkar, O. 2013. National animosity and cross-
border alliances. Academy of Management Journal, 56(6):
1516–1544.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
Avendan
˜o, R., & Santiso, J. 2011. Are sovereign wealth funds
politically biased? A comparison with other institutional
investors. International Finance Review, 12: 313–353.
Bailey, M. A., Strezhnev, A., & Voeten, E. 2017. Estimating
dynamic state preferences from United Nations voting data.
Journal of Conflict Resolution, 61(2): 430–456.
Baker, S. R., Bloom, N., & Davis, S. J. 2016. Measuring economic
policy uncertainty. Quarterly Journal of Economics, 131(4):
1593–1636.
Becker, R. N., Hillman, A. L., Potrafke, N., & Schwemmer, A. H.
2015. The preoccupation of the United Nations with Israel:
Evidence and theory. The Review of International Organizations,
10(4): 413–437.
Berkman, H., Jacobsen, B., & Lee, J. B. (2011). Time-varying rare
disaster risk and stock returns. Journal of Financial Economics,
101, 313–332.
Bernstein, S., Lerner, J., & Schoar, A. (2013). The Investment
Strategies of Sovereign Wealth Funds. Journal of Economic
Perspectives,27(2), 219–238.
Bertrand, O., Betschinger, M. A., & Settles, A. 2016. The
relevance of political affinity for the initial acquisition premium
in cross-border acquisitions. Strategic Management Journal,
37(10): 2071–2091.
Biglaiser, G., & DeRouen, K., Jr. 2007. Following the flag: Troop
deployment and US foreign direct investment. International
Studies Quarterly, 51(4): 835–854.
Bird, A., Edwards, A., & Shevlin, T. 2017. Does US foreign
earnings lockout advantage foreign acquirers? Journal of
Accounting and Economics, 64(1): 150–166.
Bortolotti, B., Fotak, V., & Megginson, W. L. 2015. The
sovereign wealth fund discount: Evidence from public equity
investments. Review of Financial Studies, 28(11): 2993–3035.
Bremmer, I. 2009. State capitalism comes of age-the end of the
free market. Foreign Affairs, 88(3): 40–55.
Bruton, G. D., Peng, M. W., Ahlstrom, D., Stan, C., & Xu, K.
2015. State-owned enterprises around the world as hybrid
organizations. Academy of Management Perspectives, 29(1):
92–114.
Caldara, D., & Iacoviello, M. 2018. Measuring geopolitical risk.
FRB International Finance Discussion Paper 1222. https://
www2.bc.edu/matteo-iacoviello/gpr_files/GPR_PAPER.pdf
Calluzzo, P., Dong, G. N., & Godsell, D. 2017. Sovereign wealth
fund investments and the US political process. Journal of
International Business Studies, 48(2): 222–243.
Cuervo-Cazurra, A. (Ed). 2018a. State-owned multinationals:
Governments in global business. Springer.
Cuervo-Cazurra, A. 2018b. Thanks but no thanks: State-owned
multinationals from emerging markets and host-country
policies. Journal of International Business Policy, 1(3–4):
128–156.
Cuervo-Cazurra, A., & Li, C. forthcoming. State ownership and
internationalization: The advantage and disadvantage of
stateness. Journal of World Business.
Cuervo-Cazurra, A., Inkpen, A., Musacchio, A., & Ramaswamy,
K. 2014. Governments as owners: State-owned multinational
companies. Journal of International Business Studies, 45:
919–942.
D’Souza, J., & Nash, R. 2017. Private benefits of public control:
Evidence of political and economic benefits of state owner-
ship. Journal of Corporate Finance, 46: 232–247.
Dai, N., & Nahata, R. 2016. Cultural differences and cross-
border venture capital syndication. Journal of International
Business Studies, 47(2): 140–169.
Davis, C. L., Fuchs, A., & Johnson, K. 2019. State control and the
effects of foreign relations on bilateral trade. Journal of Conflict
Resolution, 63(2): 405–438.
Desbordes, R. 2010. Global and diplomatic political risks and
foreign direct investment. Economics & Politics, 22(1): 92–125.
Dinc, S. I., & Erel, I. 2013. Economic nationalism in mergers and
acquisitions. Journal of Finance, 68(6): 2471–2514.
Duanmu, J. L. 2014. State-owned MNCs and host country
expropriation risk: The role of home state soft power and
economic gunboat diplomacy. Journal of International Business
Studies, 45(8): 1044–1060.
Dyck, I.J., and Morse, A., 2011. Sovereign wealth fund portfo-
lios. Chicago Booth Research Paper, (11-15).
Edmans, A. and Holderness, C.G., 2017. Blockholders: A survey
of theory and evidence. In The handbook of the economics of
corporate governance, Vol. 1: 541–636. North-Holland.
Fors, G., & Kokko, A. 2001. Home country effects of FDI: Foreign
production and structural change in home-country opera-
tions. In M. Blomstrom, & L. S. Goldberg (Eds), Topics in
empirical international economics: A Festschrift in honor of Robert
E. Lipsey. University of Chicago Press.
Gertz, G. 2018. Commercial diplomacy and political risk.
International Studies Quarterly, 62(1): 94–107.
Gilson, R. J., & Milhaupt, C. J. 2009. Sovereign wealth funds and
corporate governance: A minimalist response to the new
mercantilism. Revue d’e
´conomie financie
`re, 9(1): 345–362.
Goldstein, J. S. 1992. A conflict-cooperation scale for WEIS
events data. Journal of Conflict Resolution, 36(2): 369–385.
Greene, W. 2003. Econometric Analysis. New Jersey: Prentice
Hall.
Grigoryan, A. (2016). The ruling bargain: sovereign wealth
funds in elite-dominated societies. Economics of Governance,
17(2), 165–184.
Guedhami, O. 2012. Characteristics of government acquisitions
over time: International evidence and crisis effect. Privatization
Barometer Report: 30–43.
Guler, I., & Guille
´n, M. F. 2010. Institutions and the interna-
tionalization of US venture capital firms. Journal of International
Business Studies, 41(2): 185–205.
Heinemann, A. 2012. Government control of cross-border
M&A: Legitimate regulation or protectionism? Journal of
International Economic Law, 15(3): 843–870.
Hoetker, G. 2007. The use of logit and probit models in strategic
management research: Critical issues. Strategic Management
Journal, 28(4): 331–343.
Holland, K. 2019. Government investment in publicly traded
firms. Journal of Corporate Finance, 56: 319–342.
Jandhyala, S., & Weiner, R. J. 2014. Institutions sans frontie
`res:
International agreements and foreign investment. Journal of
International Business Studies, 45(6): 649–669.
Johan, S. A., Knill, A., & Mauck, N. 2013. Determinants of
sovereign wealth fund investment in private equity vs public
equity. Journal of International Business Studies, 44(2):
155–172.
Kaminski, T. (2017). Sovereign Wealth Fund Investments in
Europe as an Instruement of Chinese Energy Policy. Energy
Policy,101, 733–339.
Kandogan, Y., & Hiller, J. 2018. Alliances in international
governmental organizations, regional trade agreement for-
mation, and multinational enterprise regionalization strategy.
Journal of International Business Studies, 49(6): 729–742.
Karolyi, G. A., & Liao, R. C. 2017. State capitalism’s global reach:
Evidence from foreign acquisitions by state-owned companies.
Journal of Corporate Finance, 42, 367–391.
Kastner, S. L. 2007. When do conflicting political relations affect
international trade? Journal of Conflict Resolution, 51(4):
664–688.
Knill, A., Lee, B. S., & Mauck, N. 2012. Bilateral political relations
and sovereign wealth fund investment. Journal of Corporate
Finance, 18(1): 108–123.
Knutsen, C. H., Rygh, A., & Hveem, H. 2011. Does state
ownership matter? Institutions’ effect on foreign direct invest-
ment revisited. Business and Politics, 13(1): 1–31.
Kotter, J., & Lel, U. 2011. Friends or foes? Target selection
decisions of sovereign wealth funds and their consequences.
Journal of Financial Economics, 101(2): 360–381.
Kurlantzick, J. 2016. State capitalism: How the return of statism is
transforming the world. Oxford University Press.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
Leetaru, K. and Schrodt, P.A., 2013, April. Gdelt: Global data on
events, location, and tone, 1979–2012. In ISA annual conven-
tion, Vol. 2, No. 4: 1–49. Citeseer.
Li, J., Meyer, K. E., Zhang, H., & Ding, Y. 2018. Diplomatic and
corporate networks: Bridges to foreign locations. Journal of
International Business Studies, 49: 659–683.
Li, J., Xia, J., & Lin, Z. 2017. Cross-border acquisitions by state-
owned firms: How do legitimacy concerns affect the comple-
tion and duration of their acquisitions? Strategic Management
Journal, 38(9): 1915–1934.
Li, Q. 2008. Foreign direct investment and interstate military
conflict. Journal of International Affairs, 62: 53–66.
Li, Q., & Sacko, D. H. 2002. The (Ir) relevance of militarized
interstate disputes for international trade. International Studies
Quarterly, 46(1): 11–43.
Li, Q., & Vashchilko, T. 2010. Dyadic military conflict, security
alliances, and bilateral FDI flows. Journal of International
Business Studies, 41(5): 765–782.
Lindner, T., Puck, J., & Verbeke, A. 2020. Misconceptions about
multicollinearity in international business research: Identifica-
tion, consequences, and remedies. Journal of International
Business Studies, 51(3): 283–298.
Liu, B., & McConnell, J. 2013. The role of the media in corporate
governance: Do the media influence managers’ capital allo-
cation decisions? Journal of Financial Economics, 110(1): 1–17.
Long, J. S., & Freese, J. 2006. Regression models for categorical
dependent variables using Stata. Stata Press.
Luo, Y., Xue, Q., & Han, B. 2010. How emerging market
governments promote outward FDI: Experience from China.
Journal of World Business, 45(1): 68–79.
Mansfield, E. D., & Bronson, R. 1997. Alliances, preferential
trading arrangements, and international trade. American
Political Science Review, 91(1): 94–107.
Megginson, W. L., & Fotak, V. 2015. Rise of the fiduciary state: A
survey of sovereign wealth fund research. Journal of Economic
Surveys, 29(4): 733–778.
Megginson, W. L. 2017. State capitalism and state ownership of
business in the 21st century. Available at SSRN 3094412.
Moons, S. J. V., & van Bergeijk, P. A. 2017. Does economic
diplomacy work? A meta-analysis of its impact on trade and
investment. World Economy, 40(2): 336–368.
Morrow, J. D., Siverson, R. M., & Tabares, T. E. 1998. The
political determinants of international trade: the major pow-
ers, 1907–1990. American Political Science Review, 92(3):
649–661.
Musacchio, A., & Lazzarini, S. G. 2018. State-owned enterprises
as multinationals: Theory and research directions: 255–276.
Palgrave Macmillan.
Musacchio, A., Lazzarini, S. G., & Aguilera, R. V. 2015. New
varieties of state capitalism: Strategic and governance impli-
cations. Academy of Management Perspectives, 29(1):
115–131.
Nigh, D. 1985. The effect of political events on United States
direct foreign investment: A pooled time-series cross-sectional
analysis. Journal of International Business Studies, 16(1): 1–17.
Pollins, B. M. 1989a. Conflict, cooperation, and commerce: The
effect of international political interactions on bilateral trade
flows. American Journal of Political Science, 33: 737–761.
Pollins, B. M. 1989b. Does trade still follow the flag? American
Political Science Review, 83(2): 465–480.
Rygh, A. 2019. Bureaucrats in international business: A review of
five decades of research on state-owned MNEs. In A. Chidlow,
et al. (Eds), The changing strategies of international business:
49–69. London: Palgrave Macmillan.
Shi, W., Hoskisson, R. E., & Zhang, Y. A. 2016. A geopolitical
perspective into the opposition to globalizing state-owned
enterprises in target states. Global Strategy Journal, 6(1):
13–30.
Smeets, R. 2008. Collecting the pieces of the FDI knowledge
spillovers puzzle. The World Bank Research Observer, 23(2):
107–138.
SWF Institute. Top 91 largest Sovereign Wealth Fund rankings by
total assets. Sovereign Wealth Fund Institute. https://www.
swfinstitute.org/fund-rankings/sovereign-wealth-fund
Train, K. E. 2009. Discrete choice methods with simulation (2nd
ed.). Cambridge University Press.
Vasudeva, G. 2013. Weaving together the normative and
regulative roles of government: How the Norwegian sovereign
wealth fund’s responsible conduct is shaping firms’ cross-
border investments. Organization Science, 24(6): 1662–1682.
Vasudeva, G., Nachum, L., & Say, G. D. 2018. Overcoming
information asymmetry in internationalization: The signaling
effect of a sovereign wealth fund as an Institutional interme-
diary. Academy of Management Journal, 61(4): 1583–1611.
Voeten, E. 2013. Data and analyses of voting in the UN General
Assembly. In B. Reinalda (Ed), Routledge handbook of interna-
tional organization: 80–92. Routledge.
Wang, C., Hong, J., Kafouros, M., & Wright, M. 2012. Exploring
the role of government involvement in outward FDI from
emerging economies. Journal of International Business Studies,
43(7): 655–676.
Ward, M. D., Beger, A., Cutler, J., Dickenson, M., Dorff, C., &
Radford, B. 2013. Comparing GDELT and ICEWS event data.
Analysis, 21(1): 267–297.
Wehrheim, D., Dalay, H. D., Fosfuri, A., & Helmers, C. 2020.
How mixed ownership affects decision making in turbulent
times: Evidence from the digital revolution in telecommuni-
cations. Journal of Corporate Finance, 64: 101626.
Wiersema, M. F., & Bowen, H. P. 2009. The use of limited
dependent variable techniques in strategy research: Issues and
methods. Strategic Management Journal, 30(6): 679–692.
World Economic Forum. 2017. What is a sovereign wealth fund?
http://www.weforum.org/agenda/2017/10/what-you-need-
to-know-about-sovereign-wealth-funds
Wei, Y., & Liu, X. 2006. Productivity spillovers from R&D,
exports and FDI in China’s manufacturing sector. Journal of
International Business Studies, 37(4): 544–557.
Wooldridge, J. M. 2010. Econometric analysis of cross section and
panel data. MIT Press.
Wright, M., Pruthi, S., & Lockett, A. 2005. International venture
capital research: From cross-country comparisons to crossing
borders. International Journal of Management Reviews, 7(3):
135–165.
Zelner, B. A., Henisz, W. J., & Holburn, G. L. 2009. Contentious
implementation and retrenchment in neoliberal policy reform:
The global electric power industry, 1989–2001. Administrative
Science Quarterly, 54(3): 379–412.
ABOUT THE AUTHORS
Di Wang is Lecturer at the Department of
Government and International Relations and Glo-
bal Studies, University of Texas at Austin. She tea-
ches courses and conducts research on the politics
of economic globalization, sovereign wealth funds,
and business–government relations. She received a
PhD in Political Science from Texas A&M
University.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
Robert J. Weiner is Professor of International
Business, Public Policy & Public Administration,
and International Affairs, George Washington
University. Using the global petroleum industry as
a research laboratory, he has examined corruption,
FDI, political risk, SOEs, speculation, and trans-
parency. His doctorate in Business Economics is
from Harvard University. He is a US citizen born in
the Azores.
Quan Li is Professor of Political Science at Texas
A&M University. His research focuses on political
risk for multinational corporations, political
violence, economic globalization, and democratic
governance. He received a PhD in Political Science
from Florida State University.
Srividya Jandhyala is an Associate Professor at
ESSEC Business School, Singapore. Her research
examines business–government relations, includ-
ing cooperation and conflicts between firms and
governments. She received a PhD in Management
from The Wharton School, University of
Pennsylvania.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional
affiliations.
Accepted by Alain Verbeke, Editor-in-Chief, 18 January 2021. This article has been with the authors for four revisions.
Leviathan as foreign investor Di Wang et al.
Journal of International Business Studies
