Journal of International Business Studies

https://doi.org/10.1057/s41267-023-00657-4

RESEARCH NOTE

Bilateral political tension and the signaling role of patenting in a host

country
Nan Zhou1 · Jiatao Li2 · Jue Wang3

Received: 1 September 2021 / Revised: 23 August 2023 / Accepted: 31 August 2023

© The Author(s) 2023

Abstract
The current increasing volatility in international politics makes it more important to understand how multinational enter-
prises respond to political tension between host and home countries. This paper explains the impact of macro-level bilateral
political tension on micro-level strategy of multinationals in the host country. We developed the idea that patenting may be
used to signal a firm’s commitment and contribution to the host country’s economy and development. Data on 437 large
multinationals and interviews with senior managers of 20 foreign subsidiaries in China show that patenting local innovation
does indeed help an investing firm signal its usefulness to the host country government. It can thus serve as a response to
bilateral political tension. The relationship between political tension and local patenting also depends on the relative trade
dependence of the home and host countries and on the investing firm’s technology level and its stake in China. The greater
the dependence of an MNE and its home country government on the host country, the more likely patenting of local innova-
tions would increase in times of bilateral political tension.

Keywords Political tension · Patenting · Signaling · Multinational firms · China

Introduction manifests as “disagreement over policy issues, hostility

between leaders, and negative public sentiment” (Davis
Firms do not operate in a frictionless world. The growing & Meunier, 2011: 628). Political risk influences all MNEs
importance of political considerations and their unpredict- operating in a host country, but political tension mainly
ability increasingly challenge the managers of multination- applies to MNEs from a home country in a tense political
als (MNEs) (Grodal & O'Mahony, 2017). Political tension relationship with the host country. Some level of political
tension is not unusual, but in the era of de-globalization,
political tensions are more common than before. That poses
Accepted by Alain Verbeke, Consulting Editor, 31 August 2023.
This article has been with the authors for four revisions. new challenges for MNEs.
Despite their importance, scholarly understanding of
* Jue Wang how political tension influences MNEs’ strategies remains
[email protected] limited (Aguilera, Henisz, Oxley, & Shaver, 2019). Most
Nan Zhou scholarly work has focused either on normal relationships or
[email protected] on extreme cases such as war (e.g., Li & Vashchilko, 2010).
Jiatao Li However, political tension could even determine the survival
[email protected] of an MNE subsidiary, so it is therefore important to under-
1
Advanced Institute of Business, School of Economics stand its influence and how MNEs can respond.
and Management, Tongji University, 1500 Siping Road, This study investigated one potential response: patenting.
Shanghai 200092, People’s Republic of China It tested the idea that when bilateral political tension arises
2
Management Department, School of Business between the home and a host country, an MNE might pat-
and Management, Hong Kong University of Science ent more local innovations in the host country as a signal of
and Technology, Clear Water Bay, Kowloon, Hong Kong its usefulness to local stakeholders, seeking to enhance its
3
School of Business Administration, Southwestern University legitimacy. The resource-dependence perspective was then
of Finance and Economics, 555 Liu Tai Da Dao, Wenjiang applied to explore the moderating roles of several factors
District, Chengdu 610030, People’s Republic of China

13
Vol.:(0123456789)
 Journal of International Business Studies

which might influence that effort’s probability. Specifically, study thus examined the possibility that MNEs might use
the two economies’ trade inter-dependence, the firm’s inno- patenting to respond to bilateral political tension.
vation capacity, and its stake in the host country were tested China’s legal system nominally protects trademarks, pat-
as potentially important moderators. The intuition was that ents, and copyrights. China has joined all the major inter-
the greater the dependence of an MNE and its home-country national intellectual property rights-related conventions.
government on the host country, the greater should be the Although it is gradually improving, China’s legal system in
need for the MNE to use patenting of local innovations as a fact remains relatively weak. In practice, patents are some-
response to bilateral political tension. times not really enforceable (Cheng, 1998). As a result, pat-
These ideas were tested using data describing the sub- enting in China can instead lead to local illegal imitation
sidiaries of 437 large MNEs in China during the 1992–2020 if a patented product appears profitable (You & Katayama,
period. In addition, 48 senior managers from 20 foreign 2005). Weak intellectual property rights (IPR) protection
subsidiaries in China were interviewed about the topic. The implies that MNEs patenting in China likely have other pur-
findings help to clarify the impact of political relations on poses such as signaling in mind.
MNEs and their responses. They show that patenting of local
innovations can serve as an effective response, and they also
offer a new understanding of the value of patenting. Hypothesis development
It is important, however, to keep in mind that China as a
host country is in many ways an exceptional case. China is Political tension and patenting
a large market, and it plays an important role in global sup-
ply chains. Exiting from it would not be an option for many When an MNE invests in a foreign country, it steps into
MNEs, making patenting a particularly useful response an implicit contract with the host-country government
in this case. The study’s findings may apply in other large under which the latter is expected to protect the investor
economies such as the U.S. and E.U., but perhaps less well and refrain from expropriating its property (Asiedu, Jin, &
in smaller economies where an MNE can exit more easily. Nandwa, 2009). The host-country government should pro-
vide institutional support for the MNE’s operations (Kobrin,
1987). When political tension arises between the host and
home countries, the host-country government’s interests may
Research background begin to diverge from those of an MNE to the extent that
the MNE is seen as representing the national interests of its
There have been previous investigations of the impact of home country (Gilpin, 1975). MNEs from a home country
political relationships on firm strategy (e.g., Li, Van-Ass- involved in political tension must therefore anticipate a lack
che, Li, & Qian, 2022), but most have considered only how of institutional support from the host-country government.
MNEs might benefit from good host country–home coun- They may even receive distinctly unfavorable treatment
try relationships (Li, Meyer, Zhang, & Ding, 2018). War (Bundy, Pfarrer, Short, & Coombs, 2017).
has received some scholarly attention, but not more normal If political tension escalates, wise managers may consider
worsening of bilateral relationships. In an era of de-globali- divestment (Blake & Moschieri, 2017), but a firm will not
zation, deteriorating or sustained poor political relations take such a dramatic step lightly. MNEs in China need to
between two countries has become more likely. Given the consider the size of the Chinese market and China’s role in
potentially significant impact of political tension on MNEs, the global value chain (Hatani, 2009). They need other ways
how they might best respond should be examined. That was to respond to political tensions.
the aim of this study. A host-country government may hesitate to treat an
One primary motive for patenting is of course to shield a MNE unfavorably if doing so would impede local innova-
firm’s competitive advantages from imitation. Besides pro- tion. Local innovation generates new knowledge which can
tection, however, firms could also file patents as a signal sometimes have great value. It can to some extent determine
(Somaya, 2012). They can signal a firm’s ability to inno- an economy’s competitiveness (Dodgson, 2009). Astute pol-
vate and that it has potentially useful or valuable technology iticians will therefore seek to enhance local innovation for
(Levitas & McFadyen, 2009). That could signal the firm’s the sake of the economy (Zhou, Velamuri, & Dauth, 2017).
quality to stakeholders such as investors (Somaya, 2012) When a host-country government contemplates which
and alliance partners (Larson, 1992), but this study focused MNEs to punish in the context of political tension, there
on another type of signaling. The intuition was that patent- is information asymmetry between the government and the
ing could potentially enhance an MNE’s legitimacy in the firms regarding the firms’ contributions to local innovation.
eyes of a host-country government by signaling the firm’s MNEs can signal such capability to the government by pat-
willingness and ability to contribute to local innovation. This enting their local innovations. Local innovation is costly

13
Journal of International Business Studies

because it requires a firm to invest in local R&D and train in a host country is strengthened when its home country
local researchers (Beise, 2004), which takes much time and depends heavily on the host country.
effort. That is what makes it an effective signal.
Sending signals through patenting local innovations calls Resource dependence theory also predicts that when the
for support from an MNE’s headquarters. When political host country relies on an MNE for key resources, the firm
tensions arise, the headquarters may shift the assignment holds the whip hand. When an MNE invests in a host coun-
of IP ownership from the parent to its subsidiary so that try, often one of the most important resources it can bring is
the foreign subsidiary can have more autonomy and tools to better technology. Knowledge spillover through demonstra-
respond to the challenges brought by the political tension. tion effects and employee turnover can then help the host
In other words, it might give the foreign subsidiary a supply country substantially (Zhang, Li, Li, & Zhou, 2010). The
of available patents to serve as a signal. better the technology an MNE can offer, the more oppor-
To summarize, a host-country government might hesi- tunities the host country has to acquire the technology to
tate to subject an MNE to unfavorable treatment if its local enhance industry competitiveness, and the more dependent
subsidiary has signaled its ability to innovate locally. That the host-country government will be (Lall, 1990). That tends
is how patenting local innovations could serve as a useful to give high-technology MNEs greater safety when political
response to bilateral political tensions. tension arises. It is less influenced by political tension and
less motivated to use patenting as a signal.
Hypothesis 1 (H1) An MNE’s patent applications pertain-
ing to local innovations in a host country increase with the Hypothesis 3 (H3) The relationship between bilateral politi-
level of bilateral political tension between the host and its cal tension and an MNE’s patenting of local innovations is
home country. weaker when the MNE’s level of technology is higher.

Resource dependence theory suggests some factors which An MNE’s dependence on the host country is also influ-
might influence the relationship between patenting and enced by how much investment in the host country is at stake.
political tension (Pfeffer & Salancik, 1978). When the host A heavily committed MNE is in a less favorable position.
country depends on the home country and/or its MNEs for MNEs rely on the host country for resources such as labor and
resources, it is in a weak position. The MNEs then have less raw materials. The larger their stake in the host country, the
need to signal their usefulness. One such factor is general more they depend on the host country for resources. To make
economic power. As Hirschman (1980: 16) has observed, things worse, large size attracts greater public attention and
“Power to interrupt commercial or financial relations with government intervention because of the firm’s great economic
any country, considered as an attribute of national sover- significance and high visibility (Moon & Lado, 2000). MNEs
eignty, is the root cause of the influence or power position invested heavily or with numerous subsidiary operations in a
which a country acquires in other countries.” country are thus more influenced by political tensions between
However, it is necessary to consider both the home coun- the host and home countries and their motivation to use pat-
try’s dependence on the host country and the host country’s enting to minimize the impact of political tension is stronger.
dependence on the home country because they may not
be symmetric. When the home country has a dependence Hypothesis 4 (H4) The relationship between bilateral politi-
advantage over the host country, it can threaten to reduce cal tension and an MNE’s patenting of local innovation is
or terminate economic exchanges. Such a threat is more strengthened when the MNE’s subsidiaries in the host coun-
convincing the greater the host country’s dependence. In try are numerous.
that case, home-country MNEs will have less need to signal
through patenting. The same logic applies in reverse when
the host country holds the better cards. Home-country MNEs
are then more influenced by bilateral political tensions and Methods
signaling might then be very helpful.
Data
Hypothesis 2a (H2a) The relationship between bilateral
political tension and an MNE’s patenting of local innova- These hypothese were tested using data describing the pat-
tion in a host country is weakened when the host country enting by large MNEs in China. China is an appropriate set-
depends heavily on the firm’s home country. ting because most large MNEs actively engage in patenting
in China. The necessary data are available from 1992, so the
Hypothesis 2b (H2b) The relationship between bilateral sample period was from 1992 to 2020. The Fortune Global 500
political tension and an MNE’s patenting of local innovation listings for 1995 (the first), 2000, 2010, and 2020 were used to

13
 Journal of International Business Studies

identify the MNEs active in China with the largest sales vol- dependence (Ross, 2019). A home country’s trade depend-
umes. A total of 853 non-Chinese MNEs appeared in the lists. ence on China (Dh) was represented by the home country’s
Information about the firms’ patent applications came imports from China and its exports to China as a proportion
from the incoPat patent database. It has compiled more than of the home country’s GDP. China’s trade dependence on
100 million pieces of patent information from 117 official a home country (Dc) was computed similarly. The China
patent offices and business vendors. For each MNE, the import and export data came from the United Nations’
incoPat database was searched to identify any patent applica- Comtrade database. The GDP data were from the World
tions it filed in China. Either the parent MNE or a subsidiary Development Indicators compiled by the World Bank. Fol-
could be the applicant. The search produced 1,181,929 pat- lowing the lead of Gulati and Sytch (2007), a spline specifi-
ent applications filed by 474 MNEs between 1992 and 2020. cation was used to measure the direction of asymmetry and
Data describing the MNEs were extracted from the Osiris differentiate between the dependence advantage of a home
database, which covers nearly 80,000 publicly listed com- country and that of China (Johnston, 1997). That allowed
panies worldwide. It offered financial information for 444 exploring possible variations in the effect of a dependence
of the 853 MNEs. Because certain country-level data were advantage that a single variable approach would not be able
missing, this yielded a total of 437 MNEs and 10,380 firm- to uncover. Home country’s dependence advantage was
year observations from 1992 to 2020 as the final sample. quantified by subtracting the dependence of a home country
on China from the dependence of China on that home coun-
try (Dc − Dh) if Dc > Dh, and was taken as zero otherwise.
Measures China’s dependence advantage was computed similarly.
Patenting should of course depend to some extent on
The dependent variable in the analyses was number of local R&D intensity. That was quantified using R&D spending as
patent applications, which was the number of invention pat- a proportion of a firm’s operating revenue in the previous
ent applications filed by a local subsidiary of an MNE that year (Kong & Su, 2021). Then, host-country subsidiary was
were the first applications for a given technology globally. another candidate predictor. It was the number of subsidiar-
The inventor’s location was used to identify local inventions. ies that an MNE had in China. The more subsidiaries the
The main predictor variable was political tension in a greater an MNE’s stake in China.
country’s relationship with China. The data came from a Various other potentially important predictors were also
foreign relations dataset developed by the Institute of Inter- included in the analyses. Return on assets (ROA), current
national Relations at Tsinghua University (Yan, 2010; Yan ratio, firm age, firm size, and time in China described the
& Zhou, 2004). It traces the bilateral relationships between parent firms. An MNE with a better ROA or current ratio
China and major countries from 1950 to 2020 on a monthly may have had more resources to commit to applying for pat-
basis. It has been used to measure bilateral relations between ents. Firm age was the number of years since an MNE’s
China and other countries by scholars in areas such as eco- inception. Longer experience should allow an MNE to accu-
nomics (Du, Ju, Ramirez, & Yao, 2017) and international mulate more resources, making it more likely to file patent
relations (Davis, Fuchs, & Johnson, 2019). The details of applications. Firm size was considered potentially important
how tension was quantified are explained in Appendix 1. because larger firms have more resources to support local
The computation yielded the continuous variable politi- patenting. The natural logarithm of an MNE’s total assets
cal tension. Its value was the absolute value of the change was used to quantify firm size. Time in China was the num-
in bilateral relations between an MNE’s home country and ber of years since an MNE first entered China. Longer opera-
China if the bilateral relations score decreased by at least tions in China might increase the probability that an MNE
one point during a year. Its value was zero otherwise. A would invest in local R&D leading to patent applications.
one-point change represents a substantial deterioration in The level of IPR protection in China might also be impor-
bilateral relations. For example, the China–US score in tant. The IPR protection score in “Report on China’s Intel-
December 2015 was 1.6, but − 0.5 in December 2016. The lectual Property Development” released by China’s National
score dropped by 2.1 points in 1 year when Donald J. Trump Intellectual Property Administration was used to quantify
became the U.S. president, thus the value of political tension it. A higher IPR protection score might encourage MNEs to
between the U.S. and China in 2016 was 2.1. Version 5.0 of innovate in China and file for patent protection.
the Militarized Interstate Dispute database was also searched Some home-country factors were also considered as
for relevant events. Table A1.2 in Appendix 1 summarizes potential predictors. GDP growth was the annual percentage
the political tensions identified using those data sources and rate of GDP growth at market prices in constant local cur-
involving home countries with MNEs in the sample. rency. Slow GDP growth at home should encourage MNEs
Four moderators were included in the modeling. Eco- to place more emphasis on foreign markets. Openness was
nomic dependence is usually measured in terms of trade net FDI inflows (new investment inflows less disinvestment)

13
Journal of International Business Studies

as a percentage of the economy’s GDP. The data came from

− 0.14
1.00
the World Bank’s World Development Indicators.

14
At the industry level, number of competitors was the
logarithm of the number of firms in China in the same four-

0.19
0.08
1.00
13
digit industry classification (using China’s industry coding).
For firms operating in multiple industries, the industry in

− 0.37
− 0.31
0.00
1.00
which the most revenue was generated was used. The data

12
were gathered from industry statistics and statistical year-

− 0.22
− 0.01
− 0.19
0.29
1.00
books. The stiffer the local competition, the more likely that

11
an MNE would file local patent applications to protect its
competitiveness.

− 0.29
− 0.21
0.01
0.39
0.26
1.00
Industry and year dummies were also included in the

10
analyses. The industry dummies were based on the two-digit

− 0.22
− 0.15
0.04
0.11
0.18
0.17
1.00
industry classification reported in the Osiris dataset.

− 0.01
0.00
− 0.02
0.02
− 0.05

0.01
0.06
1.00
Modeling

N = 10,158. Pearson correlations, two-tailed. Correlations with an absolute value ≥ 0.021 are significant at the 0.01 level of confidence.
8
The dependent variable, the number of local invention patent

0.08
0.05
− 0.06
0.17
− 0.03
− 0.03
− 0.04
0.02
1.00
applications, was a non-negative integer, but approximately

7
50% of that variable’s observations were zero. The data were

− 0.05
0.02
0.00
0.00
0.06
− 0.01

0.05
− 0.04

0.09
1.00
over-dispersed, so zero-inflated negative binomial models
were therefore evaluated. The zinb command in the Stata 6
software was used in the analyses. To account for the lon-

0.01
− 0.03
0.01
0.03
0.12
0.08
0.07
0.10
0.08
1.00
0.03
gitudinally clustered nature of the data, they were clustered
5

by MNE when running the regressions, using Stata’s cluster

− 0.29
− 0.03
0.10
0.40
0.33
0.13
− 0.08

0.08
− 0.04

0.01
0.04
option to generate robust standard errors. 1.00
4

− 0.18
0.37
− 0.38
0.17
− 0.30
− 0.27
− 0.23
− 0.06

0.01
− 0.37

0.10
0.04
1.00

Results
3

− 0.05
− 0.07
− 0.04
0.27
0.20
0.08
− 0.03

0.01
− 0.01

0.02
− 0.08
0.00
0.03
1.00

Table 1 shows the means, standard deviations, and correla-

2

tions describing the variables. All significant correlations

0.06
− 0.02
0.01
0.09
0.14
0.11
0.06
0.00
− 0.07

0.01
0.15
0.08
0.10
1.00
0.03

were below 0.40, which suggests that there was no serious

multicollinearity problem. The largest variance inflation fac-
1

tor among the regression models was 3.15, which is below

10.84
0.58
2.54
123.29
8.01
1.31
44.68
1.82
8.03
13.63
0.02
4.48
17.57
0.71
0.04
S.D.

the recommended cutoff of 5 (Ryan, 1997).

Table 2 tests for any relationship between political tension 10.04
0.43
1.85
144.82
4.57
16.96
55.07
1.45
6.52
2.26
4.63
1.94
0.21
0.03
0.01

and patenting of local innovations. The dependent variable

Mean
Table 1  Descriptive statistics and correlation matrix

is number of local patent applications. Model 1 is the base-

line model, which includes only control variables. Political
Home country’s dependence advantage

tension was added in Model 2. Its coefficient is positive and

Number of local patent applications

significant (p = 0.002), which shows that MNEs in China are

China’s dependence advantage

more likely to patent local innovations when political ten-

sion between their home country and China is more serious.
Host-country subsidiary

Number of competitors

The coefficient remained positive and significant in models

2–6. Therefore, H1 is supported. Using the coefficients in
Political tension

Model 2, the expected number of local patent applications

R&D intensity

IPR protection
Time in China
Current ratio

GDP growth

by an MNE would increase by 33.63% if political tension

Openness
Firm size
Firm age
Variable

between its home country and China increased from zero to

ROA

one, holding all the other variables in the model constant.

Models 3–7 employ alternative independent variables.
Models 3 and 4 test different ways of identifying the political
15
14
13
12
11
10
9
8
7
5
6
1
2
3
4

13
 Journal of International Business Studies

Table 2  Coefficients of zero-inflated negative binomial regressions predicting patent applications (main effect)
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
IV = change in 0.5 as the 1.5 as the IV = change in IV = percentage IV = value of the
bilateral rela- threshold threshold bilateral rela- change in bilat- bilateral relations
tions score in defining a tense defining a tense tions score eral relations score if the score
tension period period period score is < 0

Political tension 0.29 0.33 0.33 0.12 0.08 0.09

(0.09) [0.002] (0.11) [0.003] (0.10) [0.001] (0.06) [0.034] (0.02) [0.000] (0.05) [0.088]
Control variables
Home country’s − 3.78 − 2.81 − 3.80 − 3.54 − 4.22 − 4.72 − 2.85
dependence (5.45) [0.488] (5.46) [0.607] (5.38) [0.481] (5.38) [0.510] (5.39) [0.434] (5.40) [0.382] (5.46) [0.601]
advantage
China’s depend- 9.80 10.38 10.46 10.29 9.90 9.82 9.89
ence advan- (5.06) [0.053] (4.94) [0.036] (4.94) [0.034] (4.95) [0.038] (4.97) [0.046] (5.19) [0.059] (4.98) [0.047]
tage
R&D intensity 0.03 0.03 0.03 0.03 0.03 0.04 0.03
(0.02) [0.050] (0.02) [0.041] (0.02) [0.040] (0.02) [0.044] (0.02) [0.049] (0.02) [0.046] (0.02) [0.050]
Host-country 0.06 0.06 0.06 0.06 0.06 0.06 0.06
subsidiary (0.04) [0.123] (0.04) [0.109] (0.04) [0.099] (0.04) [0.101] (0.04) [0.104] (0.04) [0.114] (0.04) [0.106]
ROA 0.02 0.01 0.01 0.01 0.02 0.02 0.02
(0.01) [0.008] (0.01) [0.009] (0.01) [0.010] (0.01) [0.009] (0.01) [0.007] (0.01) [0.008] (0.01) [0.008]
Current ratio 0.58 0.58 0.58 0.58 0.58 0.56 0.61
(0.23) [0.012] (0.23) [0.011] (0.23) [0.011] (0.23) [0.011] (0.23) [0.013] (0.24) [0.020] (0.23) [0.007]
Firm age 0.00 0.00 0.00 0.00 0.00 0.00 0.00
(0.00) [0.520] (0.00) [0.454] (0.00) [0.450] (0.00) [0.454] (0.00) [0.472] (0.00) [0.524] (0.00) [0.518]
Firm size 0.89 0.83 0.82 0.82 0.83 0.86 0.83
(0.22) [0.000] (0.21) [0.000] (0.21) [0.000] (0.21) [0.000] (0.21) [0.000] (0.23) [0.000] (0.22) [0.000]
Time in China 0.07 0.07 0.07 0.07 0.07 0.07 0.07
(0.02) [0.000] (0.02) [0.000] (0.02) [0.000] (0.02) [0.000] (0.02) [0.000] (0.02) [0.000] (0.02) [0.000]
IPR protection 0.05 0.04 0.04 0.04 0.04 0.05 0.04
(0.02) [0.006] (0.02) [0.036] (0.02) [0.046] (0.02) [0.040] (0.02) [0.038] (0.02) [0.011] (0.02) [0.035]
GDP growth − 0.07 − 0.06 − 0.07 − 0.06 − 0.06 − 0.05 − 0.05
(0.05) [0.206] (0.05) [0.210] (0.05) [0.262] (0.05) [0.226] (0.05) [0.252] (0.05) [0.329] (0.05) [0.326]
Openness 0.02 0.02 0.02 0.02 0.02 0.02 0.02
(0.01) [0.055] (0.01) [0.070] (0.01) [0.079] (0.01) [0.076] (0.01) [0.063] (0.01) [0.100] (0.01) [0.059]
Number of 0.10 0.08 0.09 0.09 0.10 0.11 0.08
competitors (0.08) [0.196] (0.07) [0.325] (0.08) [0.271] (0.08) [0.250] (0.08) [0.198] (0.08) [0.157] (0.08) [0.293]
Year and indus- Included Included Included Included Included Included Included
try dummies
Wald’s chi- 8084.06 8265.62 5984.11 8201.62 8370.65 8094.99 7644.15
squared
Log pseudo- − 4658.57 − 4650.66 − 4650.86 − 4651.54 − 4655.49 − 4653.11 − 4656.02
likelihood

N = 10,158. Two-tailed tests. Standard errors are reported in parentheses. p values are reported in brackets.

tension period. Instead of using 1 as the threshold, 0.5 and the value. That too was reverse-coded. Political tension
1.5 were tested. Political tension remained a positive and remained a positive and significant (p = 0.000) predictor.
significant predictor in both models (p = 0.003, p = 0.001). Finally, instead of using the score change to quantify politi-
Model 5 tested the political relations score of the current cal tension, political tension was considered to exist when
year minus the score in the previous year as a measure of the bilateral foreign relationship score between two countries
political tension. It was reverse-coded by multiplying it by was below zero. In that case the score was used, otherwise
– 1, so that a positive value means an increase in politi- it was 0. The results are summarized in Model 7. Political
cal tension. Model 5 shows that political tension was still tension measured this way was not a significant predictor,
a positive and significant (p = 0.034) predictor. In model 6, showing that signaling through local innovation may not act
the political relations score was a percentage, rather than as an effective response to political tension that lasts for a

13
Journal of International Business Studies

long time. MNEs may only intend to signal in the first year Model 2 includes a term representing an interaction
when bilateral relations deteriorate. between political tension and China’s dependence advan-
Table 3 summarizes the results testing for moderating tage. Its coefficient is positive and significant (p = 0.042).
effects. Model 1 includes a term representing an interac- This means that the impact of political tension on the num-
tion between political tension and home country’s depend- ber of local patent applications is stronger when China’s
ence advantage. Its coefficient is negative and significant dependence advantage over the home country is greater.
(p = 0.000). That means that the impact of political tension The interaction term remained positive and significant in
on the number of local patent applications is weaker when the full model (model 5). Therefore, H2b is supported.
the home country is less dependent on China. The inter- Model 3 includes a term interacting political tension
action term remained negative and significant in the full with R&D intensity. Its coefficient is negative and signifi-
model (Model 5). Therefore, H2a is supported. cant (p = 0.034). This means that the relationship between

Table 3  Coefficients of zero-inflated negative binomial regressions predicting patent applications (moderating effects)
Model 1 Model 2 Model 3 Model 4 Model 5

Political tension 0.51 0.21 0.46 0.53 0.23

(0.10) [0.000] (0.09) [0.017] (0.14) [0.001] (0.16) [0.001] (0.11) [0.044]
Political relations × home country’s dependence advantage − 4.49 − 4.25
(1.13) [0.000] (1.14) [0.000]
Political relations × China’s dependence advantage 6.90 8.49
(3.39) [0.042] (3.10) [0.006]
Political relations × R&D intensity − 0.05 − 0.01
(0.02) [0.034] (0.00) [0.048]
Political relations × host-country subsidiary 0.17 0.14
(0.09) [0.046] (0.04) [0.000]
Control variables
Home country’s dependence advantage − 2.97 − 2.98 − 3.06 − 2.87 − 3.92
(5.64) [0.599] (5.50) [0.587] (5.53) [0.580] (5.45) [0.599] (5.74) [0.494]
China’s dependence advantage 10.63 10.59 10.18 10.47 8.51
(4.90) [0.030] (4.97) [0.033] (4.91) [0.038] (4.94) [0.034] (4.96) [0.086]
R&D intensity 0.03 0.03 0.03 0.03 0.03
(0.01) [0.019] (0.02) [0.042] (0.02) [0.033] (0.02) [0.041] (0.01) [0.020]
Host-country subsidiary 0.06 0.06 0.06 0.06 0.06
(0.04) [0.111] (0.04) [0.107] (0.04) [0.109] (0.04) [0.109] (0.04) [0.113]
ROA 0.01 0.01 0.01 0.01 0.01
(0.01) [0.049] (0.01) [0.010] (0.01) [0.009] (0.01) [0.009] (0.01) [0.056]
Current ratio 0.58 0.58 0.59 0.54 0.58
(0.22) [0.009] (0.23) [0.011] (0.22) [0.009] (0.22) [0.015] (0.22) [0.008]
Firm age 0.00 0.00 0.00 0.00 0.00
(0.00) [0.466] (0.00) [0.459] (0.00) [0.489] (0.00) [0.450] (0.00) [0.466]
Firm size 0.85 0.83 0.84 0.83 0.86
(0.21) [0.000] (0.21) [0.000] (0.21) [0.000] (0.21) [0.000] (0.21) [0.000]
Time in China 0.07 0.07 0.07 0.07 0.07
(0.02) [0.000] (0.02) [0.000] (0.02) [0.000] (0.02) [0.000] (0.02) [0.000]
IPR protection 0.04 0.04 0.04 0.04 0.04
(0.02) [0.024] (0.02) [0.041] (0.02) [0.036] (0.02) [0.039] (0.02) [0.032]
GDP growth − 0.08 − 0.06 − 0.06 − 0.06 − 0.06
(0.05) [0.128] (0.05) [0.217] (0.05) [0.221] (0.05) [0.210] (0.05) [0.206]
Openness 0.02 0.02 0.02 0.02 0.02
(0.01) [0.061] (0.01) [0.073] (0.01) [0.073] (0.01) [0.069] (0.01) [0.074]
Number of competitors 0.08 0.07 0.08 0.08 0.07
(0.08) [0.322] (0.08) [0.396] (0.08) [0.322] (0.08) [0.335] (0.08) [0.389]
Year and industry dummies Included Included Included Included Included
Wald’s chi-squared 7637.32 7358.04 7518.07 7560.69 6591.52
Log pseudo-likelihood − 4643.25 − 4649.58 − 4647.65 − 4649.83 − 4638.06

N = 10,158. Two-tailed tests. Standard errors are in the parentheses. p values are in the brackets.

13
 Journal of International Business Studies

political tension and the number of applications is weaker and significant in the full model (model 5). Therefore, H4 is
when an MNE is doing more R&D. The interaction term also supported. All previous results hold in the full model.
remains negative and significant in the full model (model The magnitudes of the marginal effects of the moderat-
5). Therefore, H3 is supported. ing variables were assessed using the approach proposed by
Model 4 includes a term interacting political tension with Busenbark, Graffin, Campbell and Lee (2022). The marginal
host-country subsidiary. Its coefficient is positive and sig- effect of a predictor on the dependent variable was plotted
nificant (p = 0.046). So the relationship between political over different levels of the moderating variable. Figures 1,
tension and patenting is stronger when an MNE has more 2, 3, and 4 are based on the coefficients of models 1 to 4
of a stake in China. The interaction term remains positive of Table 3. The x-axis is the level of the moderator, and

Fig. 1  Marginal effect of politi- 6

cal tension on the number of
local patent applications at dif-
ferent levels of home country’s
dependence advantage
4
2
0
-2

0 .02 .04 .06 .08 .1

Home Country’s Dependence Advantage
95% CI Marginal effect

Fig. 2  Marginal effect of

15

political tension on the number

of local patent applications
at different levels of China’s
dependence advantage
10
5
0

0 .02 .04 .06 .08 .1

China’s Dependence Advantage
95% CI Marginal effect

13
Journal of International Business Studies

Fig. 3  Marginal effect of politi-

5
cal tension on the number of
local patent applications at dif-
ferent levels of R&D intensity

4
3
2
1
0

0 2 4 6 8 10
R&D intensity

95% CI Marginal effect

Fig. 4  Marginal effect of

30

political tension on the number

of local patent applications at
different levels of host-country
subsidiaries
25
20
15
10
5

0 5 10 15 20
Host country subsidiaries

95% CI Marginal effect

the y-axis is the marginal effect with its 95% confidence the value of China’s dependence advantage increases, the
interval. In all four figures, the marginal effects at lower marginal effects remain positive and continues to increase.
levels of the moderator are significantly different from those The patterns shown in Fig. 3 are consistent with H2b. Fig-
at higher levels, supporting significant moderating effects. ures 3 and 4 follow similar patterns to Figs. 2 and 3 and are
The marginal effects are also significantly different from 0, consistent with H3 and H4.
supporting the main effect. In Fig. 1, as the value of home Table A2.1 in Appendix 2 presents the results of some
country’s dependence advantage increases, the marginal robustness checks using an alternative dependent variable,
effects gradually change from positive to negative. The pat- alternative model formulations, and lagged independent
terns shown in Fig. 2 are consistent with H2a. In Fig. 2, as variables.

13
 Journal of International Business Studies

Post hoc analysis bilateral relations. Patenting local innovations enhances

their legitimacy. This study thus not only enriches under-
These findings were further tested in interviews with 48 sen- standing of MNEs’ responses to changes in the politi-
ior managers of 20 MNE subsidiaries in China. Appendix 3 cal environment, but also extends understanding of the
describes the details of the interview process and the sample. motives underlying patenting.
Its country and industry distribution corresponded with the As previously mentioned, this study’s focus on China
sample of the secondary data analyses. Table A3.2 presents may have compromised the generalizability of the findings.
some quotes from the interviews. Although MNEs located anywhere might have to deal with
The interviews confirmed that foreign subsidiaries patent bilateral political tensions, political considerations can be par-
in China for reasons other than protection (see Table A3.3). ticularly important in China because China’s bilateral rela-
Twelve of the interviewees stated that their firms filed patent tionships with many nations have been deteriorating in recent
applications for other reasons. Many mentioned that patent- years. Also, exit from China may be a rather extreme response
ing could serve as a signal to enhance the firm’s reputa- due to the large size of the Chinese market for most goods.
tion. As one of the managers said, “The government was Patenting rather than divesting may be a response particularly
also pleased to witness such a large patent output…We then relevant to China. Future studies should examine other host
could get a lot of help in bidding and applying for certain countries to test the generalizability of this study’s findings.
policy certifications.” Note too that the sample here was restricted to very large,
Table A3.4 summarizes some quotes from the inter- listed firms. Their responses may to some extent be unrep-
views dealing with foreign subsidiaries filing for patents as resentative of those of smaller firms. Smaller MNEs may be
a response to bilateral political tension. As one of the man- more flexible in their strategies and more inclined to exit.
agers said, “In fact, patent application transmitted a signal Future research might usefully exploit data on a broader set
to the government. You will then get more policy [support] of MNEs to examine the generalizability of these results.
and financing, which are quite important to your operation Signaling through local innovation has been shown to
in the Chinese market. Especially since in recent years serve as a response to deteriorating bilateral relations, but its
our Japan headquarters is not so confident about doing benefits may weaken over time. Future studies could explore
business in China. So, after you initially send a signal to MNEs’ responses when political tension persists over an
the government, the government may in turn give certain extended period.
support for your operations there. This is critical for you In any such follow-up research, the measure of bilateral
to deal with risks and shocks.” political tension might be improved by using a greater vari-
ety of data sources. The Tsinghua University dataset has its
advantages, but it only considers relations between China
Discussion and certain important countries. Some potentially important
bilateral tensions may thus have been overlooked. Future
The recent de-globalization trend has further highlighted studies should seek to correct that.
the importance of political considerations for MNEs.
This study tested the idea that bilateral political tension
between the host and home countries influences MNEs’ Conclusions
patenting. The findings extend scholarly understanding
of the impact of the political environment on MNE strat- When bilateral political tension arises between an MNE’s
egy. The data show that patenting of local innovation can home and host countries, patenting local innovations in the
serve as a signal of an MNE’s usefulness to a host-country host country can serve as a signal of the firm’s usefulness to
government (at least in China). The findings clarify how local stakeholders to enhance its legitimacy. The usefulness
political tension influences MNEs’ operations, and also of patenting depends on the relative dependence of the home
incorporate signaling theory into the study of political country and the firm on the host country, just as resource
influences on firm strategy. dependence theory predicts. We found support for our argu-
Persistent political tension could become a new nor- ments in a sample of foreign subsidiaries operating in China.
mal which MNEs must adapt to and deal with. Continuing The findings of this study thus enrich scholarly understand-
to operate in the face of bilateral political tension calls ing of the impact of political factors on MNEs, and of the
for paying close attention to the bilateral political situa- signaling role of patenting as a non-market strategy, both of
tion. While previous studies have identified other ways to which deserve further investigation.
respond to political risk or other crises, such as CSR (Zhou
Supplementary Information The online version contains supplemen-
& Wang, 2020), the findings here expand the toolbox for tary material available at https://d​ oi.o​ rg/1​ 0.1​ 057/s​ 41267-0​ 23-0​ 0657-4.
MNE managers that decide to stay despite deteriorating

13
Journal of International Business Studies

Acknowledgements This research was supported in part by National embeddedness on a manufacturer’s performance in procurement
Natural Science Foundation of China (Project Numbers: 72122016, relationships. Administrative Science Quarterly, 52(1), 32–69.
71902091, 71673227, 71832012); and by the Research Grants Council Hatani, F. (2009). The logic of spillover interception: The impact of
of Hong Kong (HKUST# 16506622) and HKUST Institute for Emerg- global supply chains in China. Journal of World Business, 44(2),
ing Market Studies (IEMS) Research Grant (IEMS21BM05). 158–166.
Hirschman, A. (1980). National power and the structure of foreign
Open Access This article is licensed under a Creative Commons Attri- trade. Berkeley, CA: University of California Press.
bution 4.0 International License, which permits use, sharing, adapta- Johnston, J. (1997). Econometric methods (4th ed.). New York:
tion, distribution and reproduction in any medium or format, as long McGraw-Hill.
as you give appropriate credit to the original author(s) and the source, Kobrin, S. J. (1987). Testing the bargaining hypothesis in the manufac-
provide a link to the Creative Commons licence, and indicate if changes turing sector in developing countries. International Organization,
were made. The images or other third party material in this article are 41(4), 609–638.
included in the article’s Creative Commons licence, unless indicated Kong, L., & Su, H. (2021). On the market reaction to capitalization of
otherwise in a credit line to the material. If material is not included in R&D expenditures: Evidence from Chinext. Emerging Markets
the article’s Creative Commons licence and your intended use is not Finance & Trade, 57(5), 1300–1311.
permitted by statutory regulation or exceeds the permitted use, you will Lall, S. (1990). Building international competitiveness in developing
need to obtain permission directly from the copyright holder. To view a countries. Paris: OECD.
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Larson, A. (1992). Network dyads in entrepreneurial settings: A study
of the governance of exchange relationships. Administrative Sci-
ence Quarterly, 37(1), 76–104.
Levitas, E., & McFadyen, M. A. (2009). Managing liquidity in research-
intensive firms: Signaling and cash flow effects of patents and alli-
ance activities. Strategic Management Journal, 30(6), 659–678.
References Li, J., Meyer, K. E., Zhang, H., & Ding, Y. (2018). Diplomatic and
corporate networks: Bridges to foreign locations. Journal of Inter-
Aguilera, R., Henisz, W., Oxley, J. E., & Shaver, J. M. (2019). Special national Business Studies, 49(6), 659–683.
issue introduction: International strategy in an era of global flux. Li, J., Van-Assche, A., Li, L., & Qian, G. (2022). Foreign direct invest-
Strategy Science, 4(2), 61–69. ment along the belt and road: A political economy perspective.
Asiedu, E., Jin, Y., & Nandwa, B. (2009). Does foreign aid mitigate the Journal of International Business Studies, 53(5), 902–919.
adverse effect of expropriation risk on foreign direct investment? Li, Q., & Vashchilko, T. (2010). Dyadic military conflict, security alli-
Journal of International Economics, 78(2), 268–275. ances, and bilateral FDI flows. Journal of International Business
Beise, M. (2004). Lead markets: Country-specific drivers of the global Studies, 41(5), 765–782.
diffusion of innovations. Research Policy, 33(6), 997–1018. Moon, C. W., & Lado, A. A. (2000). MNC-host government bargaining
Blake, D. J., & Moschieri, C. (2017). Policy risk, strategic decisions power relationship: A critique and extension within the resource-
and contagion effects: Firm-specific considerations. Strategic based view. Journal of Management, 26(1), 85–117.
Management Journal, 38(3), 732–750. Pfeffer, J., & Salancik, G. R. (1978). The external control of organizations:
Bundy, J., Pfarrer, M. D., Short, C. E., & Coombs, W. T. (2017). Crises A resource dependence perspective. New York: Harper & Row.
and crisis management: Integration, interpretation, and research Ross, R. S. (2019). On the fungibility of economic power: China’s
development. Journal of Management, 43(6), 1661–1692. economic rise and the East Asian security order. Planning Theory,
Busenbark, J. R., Graffin, S. D., Campbell, R. J., & Lee, E. Y. (2022). A 25(1), 302–327.
marginal effects approach to interpreting main effects and modera- Ryan, T. P. (1997). Modern regression methods. New York: Wiley.
tion. Organizational Research Methods, 25(1), 147–169. Somaya, D. (2012). Patent strategy and management: An integrative
Cheng, J. (1998). China’s copyright system: Rising to the spirit of review and research agenda. Journal of Management, 38(4),
trips requires an internal focus and WTO membership. Fordham 1084–1114.
International Law Journal, 21(5), 1965–1979. Yan, X. T. (2010). Sino-foreign relations review 1950–2005: Quan-
Davis, C. L., Fuchs, A., & Johnson, K. (2019). State control and the titative measurement of the relationship between China and big
effects of foreign relations on bilateral trade. Journal of Conflict powers. Beijing: Higher Education Press.
Resolution, 63(2), 405–438. Yan, X. T., & Zhou, F. Y. (2004). A quantitative measurement of inter-
Davis, C. L., & Meunier, S. (2011). Business as usual? Economic national bi-lateral political relations. Social Sciences in China,
responses to political tensions. American Journal of Political 25(6), 90–103.
Science, 55(3), 628–646. You, K., & Katayama, S. (2005). Intellectual property rights protection
Dodgson, M. (2009). Asia’s national innovation systems: Institutional and imitation: An empirical examination of Japanese F.D.I. in
adaptability and rigidity in the face of global innovation chal- China. Pacific Economic Review, 10(4), 591–604.
lenges. Asia Pacific Journal of Management, 26(3), 589–609. Zhang, Y., Li, H., Li, Y., & Zhou, L. (2010). FDI spillovers in an
Du, Y., Ju, J., Ramirez, C. D., & Yao, X. (2017). Bilateral trade and emerging market: The role of foreign firms’ country origin diver-
shocks in political relations: Evidence from China and some of sity and domestic firms’ absorptive capacity. Strategic Manage-
its major trading partners, 1990–2013. Journal of International ment Journal, 31(9), 969–989.
Economics, 108(6), 211–225. Zhou, N., & Wang, H. (2020). Foreign subsidiary CSR as a buffer
Gilpin, R. (1975). U.S. Power and the multinational corporation: The again parent firm reputation risk. Journal of International Busi-
political economy of foreign direct investment. New York: Basic. ness Studies, 51(8), 1256–1282.
Grodal, S., & O’Mahony, S. (2017). How does a grand challenge Zhou, W., Velamuri, V. K., & Dauth, T. (2017). Changing innovation
become displaced? Explaining the duality of field mobilization. roles of foreign subsidiaries from the manufacturing industry in
Academy of Management Journal, 60(5), 1801–1827. China. International Journal of Innovation Management, 21(1),
Gulati, R., & Sytch, M. (2007). Dependence asymmetry and joint 39–47.
dependence in interorganizational relationships: Effects of

13
 Journal of International Business Studies

Publisher's Note Springer Nature remains neutral with regard to Fellow of the AIB and an editor of the Journal of International Business
jurisdictional claims in published maps and institutional affiliations. Studies. His current research interests are in the areas of organiza-
tional learning, strategic alliances, corporate governance, innovation,
and entrepreneurship, with a focus on issues related to global firms and
Nan Zhou is Professor at School of Economics and Management, those from emerging economies.
Tongji University in China. She received a Ph.D. degree from Wharton
School, University of Pennsylvania. Her research addresses questions Jue Wang is Professor and Dean of School of Business Administration,
that intersect the fields of corporate strategy and international business, Southwestern University of Finance and Economics. He received a PhD
focusing primarily on understanding how strategic decisions of MNEs degree from Birkbeck College, University of London, UK. His research
are influenced by firm resources and institutional environments in the interests include organizational learning, knowledge spillover, talent
context of emerging markets. management, cross-border mergers and acquisitions, and corporate
social responsibility of emerging economy multinational enterprises.
Jiatao Li is Chair Professor of Management and Lee Quo Wei Profes-
sor of Business, and Director of the Center for Business Strategy and
Innovation, Hong Kong University of Science and Technology. He is a

13