Description
According to transaction cost theory, the risk of opportunism is a major obstacle to cooperative relationships,
such that multinational corporations need to exert control over their international joint
ventures (IJVs). However, inconsistent findings indicate that the controleperformance linkage is complex
and poorly understood. This study develops a mediating framework and suggests that perceived value
gap and information asymmetry have a direct effect on IJV performance. Perceived value gap and information
asymmetry can also mediate the relationship between control mechanisms (e.g., centralization,
formalization, and socialization) and IJV performance. A sample of 152 Taiwanese IJVs in China
showed that the direct effects of formalization and socialization on IJV performance satisfaction were
significant. The empirical results showed that information asymmetry and perceived value gap mediated
the formalizationeperformance and socializationeperformance relationships. However, neither the
direct effect of centralization nor the indirect effect through information asymmetry mediation on IJV
performance was significant.
Reexamining the relationship between control mechanisms and international
joint venture performance: The mediating roles of perceived value gap and
information asymmetry
Ming-Chang Huang
a, *
, Hsin-Hua Hsiung
b
, Ting-Chun Lu
c
a
Department of Business Administration, Providence University, Shalu District, Taichung, Taiwan, ROC
b
Department of Business Administration, National Dong Hwa University, Shou-Feng, Hualien, Taiwan, ROC
c
Department of Business Administration, National Chengchi University, Wenshan District, Taipei, Taiwan, ROC
a r t i c l e i n f o
Article history:
Received 24 August 2012
Accepted 2 September 2014
Available online 11 March 2015
Keywords:
Control mechanism
Information asymmetry
International joint ventures
International joint ventures performance
Perceived value gap
a b s t r a c t
According to transaction cost theory, the risk of opportunism is a major obstacle to cooperative re-
lationships, such that multinational corporations need to exert control over their international joint
ventures (IJVs). However, inconsistent ?ndings indicate that the controleperformance linkage is complex
and poorly understood. This study develops a mediating framework and suggests that perceived value
gap and information asymmetry have a direct effect on IJV performance. Perceived value gap and in-
formation asymmetry can also mediate the relationship between control mechanisms (e.g., centraliza-
tion, formalization, and socialization) and IJV performance. A sample of 152 Taiwanese IJVs in China
showed that the direct effects of formalization and socialization on IJV performance satisfaction were
signi?cant. The empirical results showed that information asymmetry and perceived value gap mediated
the formalizationeperformance and socializationeperformance relationships. However, neither the
direct effect of centralization nor the indirect effect through information asymmetry mediation on IJV
performance was signi?cant.
© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier
Taiwan LLC. All rights reserved.
1. Introduction
The formation of an international joint venture (IJV) is an
important market entry strategy that has been adopted by many
multinational corporations (MNCs) in the course of their global
expansions (Gong, Shenkar, Luo, & Nyaw, 2007; Makino, Chan,
Isobe, & Beamish, 2007). To realize the potential for joint value
creation, partners must exchange information, share knowledge,
and make investments. Due to the possibility of opportunism,
however, investing valuable assets and sharing proprietary
knowledge with the other partner(s) may expose each party to a
certain degree of risk (Parkhe, 1993; Zeng & Chen, 2003). Luo
(2007a) de?ned opportunism in the context of IJVs as acts or pat-
terns of behavior performed by speci?c parties from speci?c
countries to obtain unilateral gains to the substantial detriment of
other parties from other countries. When entering into a voluntary
interorganizational arrangement such as an IJV, assets are located
overseas, a focal parent becomes vulnerable to the actions of the
other partner, and the risk of expropriation or opportunism is
greater than it would be in a domestic context (Ertug, Cuypers,
Noorderhaven, & Bensaou, 2013; Pangarkar & Klein, 2004). There
is therefore a need for control mechanisms, which can be classi?ed
as centralization, formalization, and socialization according to their
objective of control over actions, results, and personnel/culture (Li,
Tang, Okano, & Gao, 2013). There is also a need to reduce trans-
action costs and enhance parent satisfaction by decreasing the
likelihood of opportunism(Deitz, Tokman, Richey, &Morgan, 2010;
Fryxell, Dooley, & Vryza, 2002; Luo, Shenkar, & Nyaw, 2001;
Mesquita & Brush, 2008; Pangarkar & Klein, 2004). The purpose
of control is to protect an MNC's own interests while simulta-
neously trying to overcome a partner's opportunistic behavior
(Calantone & Zhao, 2000; Ren, Gray, & Kim, 2009), which can be
considered as an antecedent of IJV performance (Christoffersen,
2013; Reus & Rottig, 2009; Selekler-G€ oks¸ en & Uysal-Tez€ olmez,
2007). IJV control is very important to explaining IJV performance
and to IJV success (Liu, Vredenburg, & Steel, 2014).
* Corresponding author. Department of Business Administration, Providence
University, No. 200, Section 7, Taiwan Boulevard, Shalu District 43301, Taiwan, ROC.
E-mail address: [email protected] (M.-C. Huang).
Peer review under responsibility of College of Management, National Cheng Kung
University.
HOSTED BY
Contents lists available at ScienceDirect
Asia Paci?c Management Review
j ournal homepage: www. el sevi er. com/ l ocat e/ apmrvhttp://dx.doi.org/10.1016/j.apmrv.2014.12.004
1029-3132/© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
Asia Paci?c Management Review 20 (2015) 32e43
There are many studies ofdand numerous efforts to explor-
edthe relationships between control mechanisms and IJV out-
comes that enhance our understanding and knowledge of
international alliances, the causes and consequences of the MNC-
subsidiary relationship, and the organizational design of interna-
tional business. However, the theoretical arguments and empirical
?ndings remain incongruent. For example, Newburry, Zeira, and
Yeheskel (2003) showed that IJVs located in China with authority
are more effective in the formulation and implementation of stra-
tegic plans. Zhang and Li (2001) indicated that autonomy and
success are related to IJV performance. However, Lu and Lee (2005)
reported that the impact of centralization on IJV performance
satisfaction is insigni?cant. The meta-analysis of Reus and Rottig
(2009) indicated that hierarchical control may theoretically have
positive or negative effects on performance. In mainland China,
Calantone and Zhao (2000) indicated that the con-
troleperformance relationship is insigni?cant for Japanese IJVs in
China; however, for Korea and the USA IJVs, it is positive. Reus and
Rottig (2009) report that hierarchical control has an insigni?cant
direct effect on performance in the Chinese sample but a signi?cant
effect in the non-Chinese sample. Some studies indicate that formal
control mechanisms can enhance IJV performance (Gong et al.,
2007; Perrone, Zaheer, & McEvily, 2003), while other studies
report inconsistent results (Fryxell et al., 2002; Inkpen & Currall,
2004; Nohria & Ghoshal, 1994). Luo and Park (2004) con?rm the
positive relationship between socialization and IJV performance.
However, the empirical results of Fryxell et al. (2002) show only a
marginally signi?cant positive relationship between socialization
and IJV performance. These studies provide important insights
and syntheses, but they do not provide conclusive evidence of
controleperformance relationship (Liu et al., 2014). These incon-
sistent ?ndings indicate that the controleperformance linkage is
complex and poorly understood. The relationship between control
mechanisms and IJV performance remains unclear (Pangarkar &
Klein, 2004). We still have only a limited understanding of the
process by which speci?c control mechanisms function.
Through a careful review of the literature, this study identi?es
three research gaps. First, opportunism is a critical assumption of
human behavior in transaction cost theory, and the potential for
opportunism plays an important role in IJV success (Geringer &
Hebert, 1989). Opportunism creates a need for ex ante and ex post
safeguarding, including formal and social governance structure
(Nakos & Brouthers, 2008; Subramani & Venkatrman, 2003).
However, potentially opportunistic partners do not necessarily
behave opportunistically. This observation raises the question of
why opportunistic behavior occurs. We contend that the perceived
value gap and information asymmetry are two critical conditions
that may induce opportunistic behaviors (Williamson, 1985).
Seetoo and Hsiung (2004) refer to the perceived value gap as the
extent to which common goals or shared values diverge among
partners, and they identify information asymmetry as the level of
information distribution asymmetry throughout the alliance.
Control mechanisms must overcome these two antecedents to
prevent opportunistic behavior. Little systematic empirical research
has been conducted to test the effects of control mechanisms on the
perceived value gaps and information asymmetry. Second, many
studies assume that if a value gap and information asymmetry exist
among partners, control mechanisms will have direct positive ef-
fects on IJV performance. The indirect effects of control mecha-
nisms on an IJV's outcome through a reduction in the value gap and
information asymmetry are treated as given and ignored. There-
fore, few studies have precisely measured the extent of value gap
incongruence and information asymmetry or clari?ed the way in
which the controleperformance relationship of an IJVis affected by
the value gap and information asymmetry among partners. Given
the importance of control mechanisms in IJVs, how they function
should be addressed systematically. However, there is little
research on how control mechanisms have been executed in
practice (Li et al., 2013). Finally, the de?nitions of control mecha-
nisms differ such that the empirical results are dif?cult to compare.
Ren et al. (2009) suggested that making distinctions among control
mechanisms and understanding the impact of each mechanism on
IJV performance to enrich our understanding could be necessary for
future studies.
To answer the question of how MNCs can effectively coordinate
and control IJVs, we suggest that the underlying reasons for
opportunism should be taken into account. Based on transaction
cost theory and the information process perspective, this study
proposes a conceptual model to examine whether parent com-
panies' control mechanisms could enhance the performance of an
IJV by reducing the perceived value gap and information asym-
metry among partners. In our view, information asymmetry and
the perceived value gap have a signi?cant mediating effect on the
relationship between control mechanisms and IJV performance.
The IJV approach is particularly prevalent among MNCs in the
Chinese market (Chan, Luk, &Wang, 2005; Li et al., 2013). A sample
of 152 IJVs in China involving a Taiwanese parent company
con?rmed the mediating effects of a perceived value gap and in-
formation asymmetry on the relationships between control
mechanisms and IJV performance. This paper proposes a theoret-
ical model and contributes to the literature on the con-
troleperformance relationship of IJVs in the following ways: (1)
testing the effects of control mechanisms on value gap and infor-
mation asymmetry reduction, which allowed us to address the is-
sues of opportunistic behavior and IJV management; (2) clarifying
the direct and indirect relationships between control mechanisms
and IJV performance through this theoretical framework to
enhance our understanding of the controleperformance link.
The remainder of this paper is organized as follows. In the next
section, we review the relevant literature, discuss the main study
concepts, and explain the theoretical basis for the hypothesized
relationships. In the methodology section, we describe the sample,
the measures used, and the statistical methods. In the results sec-
tion, we present the statistical relationships that were observed
among the control mechanisms, perceived value gap, information
asymmetry, and IJV performance. Finally, we present the conclu-
sions, explain the theoretical implications, and discuss the practical
implications of our ?ndings.
2. Theoretical background and hypotheses
2.1. Facilitating conditions of opportunistic behavior
Transaction cost theory de?nes opportunism as seeking self-
interest with guile (Williamson, 1985), with decision-makers
showing a tendency to exploit their positions to advance their
own self-interest (Das & Rahman, 2010; Rind?eish & Heide, 1997;
Tsang, 2006). This phenomenon provides a foundation for IJV
control theory (Deitz et al., 2010). The potential for opportunism
arises when one party is vulnerable to the actions of another, whose
behavior is not under its control (Parkhe, 1993). The formation of a
joint venture naturally gives rise to many opportunities for
opportunism (Morris & Cadogan, 2001). While opportunism is the
cause and antecedent of a partner's opportunistic behavior
(Rind?eish & Heide, 1997), opportunism per se is not a suf?cient
condition for opportunistic behavior. Certain critical conditions or
facilitators should exist for the actual occurrence of this behavior.
Opportunistic behavior is most likely to appear in two situations.
The ?rst such situation occurs when the partners lack common
objectives or shared values. The larger a partner perceives a value
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 33
gap in the difference among partners regarding common goals and
shared values (Seetoo & Hsiung, 2004), the higher the perceived
value gap and the likelihood of opportunism will be. The second
situation occurs when there is information asymmetry among
partners. Williamson (1985) noted that opportunism results from
the incomplete disclosure of information. Information asymmetry
creates an opportunity for a party to engage in opportunistic
behavior, which may not be sensed by the other parties. Therefore,
the possibility of opportunistic behavior rises in accordance with
any increase in the perceived value gap and information
asymmetry.
2.1.1. Information asymmetry
Bounded rationality is the assumption that decision makers face
constraints to their information-processing and communication
abilities, which limit their rationality (Rind?eish & Heide, 1997).
The exchange of information may reinforce the connection be-
tween the parent and the subsidiary and improve cooperation, but
incomplete or misleading information may cause opportunistic
behavior to occur more frequently (Williamson, 1985). Deliberately
withholding information in a joint venture relationship is an un-
healthy behavior that usually causes misunderstandings, incorrect
expectations, and dysfunctional con?ict among partners that in-
crease coordination costs and reduce the decision-making quality
(Bennet & Cadogan, 2001; Kwon, 2008; Morris & Cadogan, 2001).
Information is generally not symmetrically distributed among
all of the parties to a transaction (Mishra, Heide, & Cort, 1998).
Information asymmetries exist in all exchange relationships and
are accentuated in and IJV as each parent brings very different
perspectives, priorities, and quality of resources to decision making
(Yan &Child, 2004). In a transactional or cooperative relationship, if
the parties are not equally familiar with the details of a situation,
the ?owof information is likely to be blocked. In strategic alliances,
the lack of suf?cient information may make it dif?cult for one party
to predict and respond to the other party's self-interest seeking
action (Glaister, Husan, & Buckley, 2003a). Hence, transaction cost
theory treats information asymmetry as a primary cause of
opportunistic behavior (Williamson, 1985). Information asymmetry
makes it dif?cult for one party to detect the other party's oppor-
tunism, which results in an increase in monitoring costs (Wathne &
Heide, 2000). The increasing need for monitoring and coordination
results in higher transaction costs and inhibits joint venture suc-
cess. Therefore, we propose that information asymmetry will have
detrimental effects on the performance of an IJV.
H
1
: Ceteris paribus, information asymmetry is negatively related to
IJV performance.
2.1.2. Perceived value gap
The operation of IJVs is strongly affected by the partners' values,
practices, and management systems (Danis & Parkhe, 2002).
Because each partner may have unique interests and goals for the
IJV, differences in values are very common. For instance, the Chi-
nese partners in US-Sino joint ventures are often interested in
acquiring management know-how and advanced technology, but
the foreign partners are often more interested in obtaining access
to the Chinese domestic market and cheap labor (Chan et al., 2005).
Ouchi (1980) noted that common values and beliefs can provide the
type of harmony of interests that eliminates the possibility of
opportunistic behavior. A set of shared values and non-legal sanc-
tions in the form of relational norms encourages partners to
commit to their exchange relationship, restricts the partner ?rms
from acting opportunistically (Huang, Cheng, & Tseng, 2014), and
facilitates partner cooperation (Zeng & Chen, 2003). To avoid
opportunistic behavior, the IJV management process requires
participating partners to establish a clearly shared strategic vision
(Glaister et al., 2003a). The absence of shared values signi?es the
existence of a value gap among partners and may breed opportu-
nistic motives (Seetoo & Hsiung, 2004).
In IJVs, the behavior of the collaborating partners usually affects
collaborative performance and satisfaction. In the presence of
opportunistic tendencies and competitive threats, it is dif?cult for
partners to develop collaboration (Park & Ungson, 2001). To avoid
opportunistic behaviors and alliance failures, it is quite important
for the partners to reduce goal divergence and to build shared
values (Ouchi, 1980). Shared values represent the collective goals
and aspirations of organizational members. The establishment of
shared values in the formof relational norms represents a value gap
that is less perceived among IJV partners. If partners have shared
values, each partner is less likely to sacri?ce the other party's in-
terests in pursuit of its own separate goals or ambitions.
When organization members share common perceptions about
the way they should interact with one another, they can avoid po-
tential misunderstandings and gain more opportunities for the free
exchange of ideas or resources (Tsai &Ghoshal, 1998). In contrast, if
partners have major differences in terms of their strategic goals and
management styles, their negotiation costs and transaction costs
may increase (Fryxell et al., 2002; Seetoo & Hsiung, 2004), and the
IJV performance may be impaired (Pangarkar and Klein, 2004). The
establishment of shared values in the form of relational norms,
which lessen perceived value gaps, encourages partners to commit
to their exchange relationship and prevents partner ?rms from
acting opportunistically (Das & Teng, 2001; Liu, Luo, & Liu, 2009;
Tangpong, Hung, & Ro, 2010). Therefore, according to transaction
cost theory, we expect there to be a negative relationship between
the perceived value gap and IJV performance.
H
2
: Ceteris paribus, the perceived value gap is negatively related to
IJV performance.
2.2. The mediating role of the perceived value gap and information
asymmetry
How to handle the problem of opportunism is a crucial concern
for organizations with a hybrid governance structure (Luo, 2005).
To a greater or lesser extent, the parent companies of an IJV must
implement certain control mechanisms. These are either organi-
zational or structural arrangements or processes by which a party
can in?uence the behavior and output of another party through the
use of power, authority, and a wide range of bureaucratic, cultural,
and informal mechanisms (Calantone & Zhao, 2000; Fryxell et al.,
2002; Geringer & Hebert, 1989; Pangarkar & Klein, 2004) as safe-
guards to reduce transaction costs and protect their interests
against opportunism (Mesquita & Brush, 2008). Control involves
the monitoring process and consists of formal control (bureaucratic
control) and informal control (culture control) (Baliga & Jaeger,
1984). Drawing on the relevant research, we selected three types
of control mechanisms (namely, centralization, formalization, and
socialization) as our independent variables (Fryxell et al., 2002;
Geringer & Hebert, 1989; Li et al., 2013; Nobel & Birkinshaw,
1998; Sundaram & Black, 1992). Adequate control mechanisms
may have several important functions, such as increasing the
transparency of partners' behavior (Parkhe, 1993), preventing
cheating and moral hazards (Rind?eish & Heide, 1997), achieving
expected objectives, and ensuring that the IJV can faithfully pursue
strategies that conform to the interests of the parent companies
(Pangarkar & Klein, 2004). We predicted that these types of con-
trols could enhance IJV performance by reducing the perceived
value gap and information asymmetry.
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 34
2.2.1. Centralization and information asymmetry
The allocation of decision rights is a means of safeguarding
(Schepker, Oh, Martynov, & Poppo, 2014). Centralization refers to
the amount of decision-making power each parent company ex-
ercises with regard to the venture's daily operations (Yan & Gray,
2001). Centralization means that middle and lower-level man-
agers have relatively limited authority or freedomto make strategic
and operational decisions independently of the parent companies
(Baliga &Jaeger, 1984; Glaister, Husan, &Buckley, 2003b; Newburry
et al., 2003; Nobel & Birkinshaw, 1998; Reus & Rottig, 2009).
Through centralization, an MNC can determine the manner in
which IJV activities are performed so that the IJV is in line with its
own objectives (Reus & Rottig, 2009; Yan & Gray, 2001) and posi-
tively related to parent-subsidiary integration (Newburry et al.,
2003).
Control rights may also include monitoring provisions to pro-
vide behavioral control (Schepker et al., 2014). Centralization is
therefore a crucial dimension in the structural design of MNCs,
which can affect the capacity and capability of information pro-
cessing (Tushman & Nadler, 1978). For an MNC to make strategic
decisions and maintain adequate control, it is important to gather
suf?cient information about the IJV and its partners from sub-
sidiaries. However, some scholars have indicated that centraliza-
tion might reduce the autonomy or freedom of an IJV subsidiary,
thus adversely affecting the performance (Newburry et al., 2003).
However, we believe that through effective communication and
ef?cient information systems, the potential side effects of central-
ization can be overcome.
To increase the awareness of strategic development and
enhance the understanding of foreign business operations, an MNC
needs to build an effective information reporting system that can
capture key information from its IJV subsidiaries and partners (Yan
& Child, 2004). The complete information-processing system of
MNCs involves the gathering, interpretation, synthesis, and storage
of information within the headquarters and in the subsidiaries
(Egelhoff, 1982; Tushman & Nadler, 1978). Structural features are
the sources of organizational information-processing capability in
terms of information acquisition, analysis, and interpretation
(Carson, Madhok, Varman, & John, 2003). MNCs can communicate
their expectations to IJV managers and partners through the
planning and budgeting process as well as through the particularly
important business plan agreed upon through the board (Glaister
et al., 2003b). IJV managers provide necessary information
through the board, and they act in the way expected of them
(Glaister et al., 2003b). Centralization is likely to lead to a clear
understanding between the elements of the IJV system and
decrease the information gap among partners.
Information processing can be conducted more easily in a
centralized organization than in one that is decentralized.
Furthermore, a regular, frequent, and compulsory ?ow of infor-
mation between the parent company and the subsidiary helps to
reduce information asymmetry that results in minimizing the
possibility of hiding and cheating among MNCs and its partners.
Thus, higher degrees of parent control are associated with
greater performance (Quer, Claver, & Rienda, 2007). Ceteris par-
ibus, MNCs that are centralized may possess more information
about the policies, operations, and decision-making processes of
their IJVs, and they can grasp their partners' intentions and be-
haviors more effectively (Chang & Taylor, 1999) and thus increase
performance satisfaction. Hence, we propose the following
hypothesis.
H
3
: Through the mediation of information asymmetry among IJV
partners, centralization has a positive association with IJV
performance.
2.2.2. Formalization, perceived value gap and information
asymmetry
Baliga and Jaeger (1984) treat formalization as a type of
bureaucratic control that consists of using limits and an explicit set
of rules and regulations to delineate the desired performance in
terms of output and/or behavior. Formalization emphasizes these
matters through contracts, formal written procedures, policies, and
individual performance appraisals (Wang & Fulop, 2007). Formal
control mechanisms provide a justi?cation for a parent company to
trace its subsidiaries' moves or actions. Formalization provides an
opportunity and a channel for MNCs to communicate with their
partners and share critical information. Formalization focuses on
the information of a ?rm's operation process and uses objectives to
measure the accuracy and completeness of business activities (Liu
et al., 2014). Establishing a formal reporting system facilitates the
?ow of information feedback between parent companies and their
IJV subsidiaries, and it also helps a parent company to monitor and
coordinate the activities of its IJV partners. Thus, frequent
communication through formal mechanisms enhances the trans-
parency of partner actions (Chalos & O'Connor, 2004), which may
result in lower information asymmetry.
Control rights can include monitoring provisions to provide
behavioral control (Schepker et al., 2014). The concept of formal-
ization focuses on monitoring and reporting the operational pro-
cess of business activities to increase information transparency and
decrease the misunderstanding among MNCs, IJVs, and other
partners. Therefore, formalization provides a communication tool
to enhance information-processing capabilities for the trans-
mission of information from one party to another (Cam en,
Gottfridsson, & Rundh, 2011). Formalization facilitates key infor-
mation capturing and affects an MNC's ability to monitor, coordi-
nate, and integrate activities within the context of an IJV (Yan &
Child, 2004). Information asymmetry among partners can be
decreased through formalization and monitoring (Seetoo &Hsiung,
2004).
When deliberate control and information-gathering procedures
are in place, the partners' tendency toward opportunism can be
identi?ed more easily, such that the likelihood of opportunistic
behavior is reduced (Fryxell et al., 2002). Hence, effective formal
control mechanisms can enhance IJV strategy implementation and
strategic competitiveness (Yan & Child, 2004). Because formaliza-
tion increases the information-processing capability of MNCs, we
suggest that the exercise of formal control mechanisms can reduce
information asymmetry among partners and enhance the effec-
tiveness of management control. Accordingly, we propose the
following hypothesis:
H
4
: Through the mediation of information asymmetry among IJV
partners, formalization has a positive association with IJV
performance.
The formalization mechanism involves optimizing decision
making by relying on formal records and procedures (Sundaram &
Black, 1992). Regarding the role and consequences of formalization
in IJV management, there are two opposing perspectives in the
existing literature.
On one hand, some studies tend to support the view that
formalization negatively in?uences the perceived value gap. First,
formalization involves the development of clear and well-
documented policies, procedures, and operating rules (Martinez
& Jarillo, 1997), which help to clarify the exchange partner roles
and expectations and provides clarity regarding the terms, rem-
edies and con?ict resolution procedures (Tangpong et al., 2010).
Therefore, formalization also promotes communications among
partners, which reveals an openness and willingness to rely on the
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 35
other partners (Costa e Silva, Bradley, &, Sousa, 2012). Formaliza-
tion enhances institutionalized practices and routines for dealing
with a partner organization, creating a stable context within which
interorganizational and interpersonal trust can be developed
(Zaheer, McEvily, & Perrone, 1998). This type of trust is bene?cial to
the development of common goals and shared values. Furthermore,
in the process of documentation or institutionalization, IJV partners
must communicate and clarify their speci?c needs and objectives.
The communication and clari?cation process can push IJV partners
to achieve consensus on strategic goals and to specify the behavior
that is required in the form of rules or an explicit program(Nobel &
Birkinshaw, 1998). Trust enhances the belief that a partner will
perform actions that will result in positive outcomes and will not
perform unexpected actions that will result in negative outcomes
(Kwon, 2008). Once trust, shared values, and a relational norm are
established, which means a lower perceived value gap, partners
may be reluctant to behave opportunistically (Liu et al., 2009).
Formalization as a control mechanism would carry a cost to build,
which would its increase monitoring cost. Through the develop-
ment of shared value, relational norms, and trust, which operate as
self-enforcement safeguards and a less costly governance alterna-
tive (Poppo & Zenger, 2002), partners can cooperate to offset the
weaknesses of formalization (Liu et al., 2009). Therefore, we pro-
pose that formalization can decrease the value gap among IJV
partners, which lowers the likelihood of opportunism. IJV perfor-
mance can be enhanced by formalization through a reduction of the
value gap.
H
5a
: Through the mediation of the perceived value gap among IJV
partners, formalization has a positive association with IJV
performance.
On the other hand, some studies do not support the notion that
formalization helps to reduce the value gap. Formalization causes
decision-making to be routinized through rules and procedures
(Nobel & Birkinshaw, 1998). Although ex ante and ex post safe-
guarding can mitigate opportunism, several researchers (Inkpen &
Currall, 2004; Nohria & Ghoshal, 1994) argued that formalization
does not necessarily contribute to the development of shared
values. Inkpen and Currall (2004) noted the fact that trust develops
more slowly in the context of extensive formal control. Luo and
Park (2004) argued that formalized control is often exercised
because of a lack of trust. Furthermore, the troublesome processes
of formalization and institutionalization may reduce participants'
intentions to engage in repeated renegotiation to accommodate the
changing expectations of their partners (Fryxell et al., 2002).
Relying on formal control may also cause partners to neglect the
importance of creating a compatible, harmonious, and considerate
cultural environment. The likelihood of opportunism is unlikely to
be decreased when an MNC exerts formal control over its IJV. Thus,
we provide a competing hypothesis as follows.
H
5b
: Through the mediation of the perceived value gap among IJV
partners, formalization has a negative association with IJV
performance.
2.2.3. Socialization, the perceived value gap, and information
asymmetry
Social control mechanisms usually have positive effects on in-
formation sharing and two-way communication among partners
(Ertug et al., 2013). For example, transferring and rotating managers
between the headquarters and subsidiaries can create verbal in-
formation channels (Baliga & Jaeger, 1984; Edstr€ om & Galbraith,
1977). Frequent meetings and planned visits between IJV
managers and headquarters personnel can enhance information
sharing and collaboration (Chalos & O'Connor, 2004). Furthermore,
expatriates and boundary spanners with good socialization training
can develop shared norms and a collective culture. These practices
and interactions help IJV partners communicate more openly and
honestly, thus ensuring that they will be willing to share more
sensitive information.
Ahuja and Galvin (2003) observed that applying social control
mechanisms can improve performance and members' ability to
communicate. Good communication results from consistently and
clearly informing the partner about information that is relevant to
cooperation (Glaister et al., 2003a). Frequent communication can
increase the transparency of partners' actions (Chalos & O'Connor,
2004) and improve each partner's ability to understand and predict
the behavior of the opposite party (Das & Teng, 1998). In summary,
by creating an open and cordial atmosphere, social control mech-
anisms help to improve trust and lower both the cost of informa-
tion sharing and the level of information asymmetry among
partners (Luo, 2007b) as well as mitigate potential concerns of
opportunism and reduce transaction costs to improve IJV perfor-
mance satisfaction. Accordingly, we propose the following
hypothesis.
H
6
: Through the mediation of information asymmetry among IJV
partners, socialization has a positive association with IJV
performance.
Socialization is associated with the development of shared
norms and a collective culture. Through socialization, JV managers
learn to share common attitudes and knowledge of the organiza-
tion (Chalos & O'Connor, 2004). In the context of IJVs, socialization
in a subsidiary can be achieved in a variety of ways, such as through
instruction in the parent company's rules and procedures (Gupta &
Govindarajan, 1991), structured personnel interactions (Fryxell
et al., 2002), and the rotation of personnel in key managerial po-
sitions between the parent company and the subsidiaries (Chalos &
O'Connor, 2004; Edstr€ om & Galbraith, 1977; Selekler-G€ oks¸ en &
Uysal-Tez€ olmez, 2007). Strong supportive personal relationships
between senior managers lead to partners' understanding and
commitment to the IJV (Glaister et al., 2003a). The appointment of
key personnel by MNCs may act as boundary spanners for trans-
ferring MNCs' culture and policies to align their objectives with IJVs
and other partners (Selekler-G€ oks¸ en & Uysal-Tez€ olmez, 2007).
Underlying the use of expatriate staf?ng as a culture control
mechanism is evidence that this leads to a better transmission of
common values and goals between foreign and local partners
(Chalos & O'Connor, 2004). Through its Hamburger University,
McDonald's delivers its business skill and promotes its corporate
philosophy (Wathne & Heide, 2000). Socialization in a partnership
can also be achieved through reciprocal business skill training,
mutual technical development, frequent managerial contacts, and
exercises involving relational behavior (Dong, Tse, & Cavusgil,
2008; Yilmaz, Sezen, & Ozdemir, 2005).
The aimof socialization is to encourage the alignment of partner
values and interests (Chalos & O'Connor, 2004), which reduces the
likelihood of opportunism and enhances IJV performance. The
existing literature provides cues and evidence to explain the posi-
tive link between socialization and shared values among partners.
First, a number of studies have found that socialization can help
partners create convergent goals and avoid opportunism (Dong
et al., 2008). Through socialization, personal familiarity, problem
solving, norms, identity, and cohesion can be created (Ahuja &
Galvin, 2003; S¸ engün & Wasti, 2009), and members can become
more committed to their organization and share views that
strongly in?uence their behaviors (Das & Teng, 2001). Second, the
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 36
personal and social relationships among the managers of cooper-
ative partners are crucial to resolving disputes and reducing goal
discrepancy (Glaister et al., 2003a; Wathne & Heide, 2000). Third,
cultural values and mindsets relating to speci?c issues can be
disseminated among partners through the repeated interactions of
their personnel over time and can thus become systematized and
shared (Baliga & Jaeger, 1984; Hult, Ketchen, & Slater, 2004).
Accordingly, we propose the following hypothesis.
H
7
: Through the mediation of the perceived value gap among IJV
partners, socialization has a positive association with IJV
performance.
3. Method
3.1. Data collection
The sampling frame of this study came from the 2005 Approved
Foreign Investment list published by the Investment Commission,
Ministry of Economic Affairs, Taiwan. To control cultural or
contextual factors, we selected only IJVs that were set up in China.
For each of these IJVs, one of the parent companies was a Taiwanese
MNC. Some previous studies suggest that if the ownership of any
parent is below 20% or over 80%, a partnership cannot be treated as
an IJV (Choi & Beamish, 2004). Based on this criterion, we elimi-
nated certain IJVs from the sample and were left with 692 IJVs. The
data collection comprised two stages. In the ?rst stage, each IJV was
contacted via fax and telephone. Owing to problems such as
incorrect telephone numbers and addresses or the IJV business
having closed, only 652 IJVs were contacted. Of these IJVs, 357
companies promised to participate in this study, representing a
participation rate of 54.8%. In the second stage, 357 questionnaires
were sent to these companies. Informants were asked to select an
important IJV as reference to answer our questionnaire. A total of
152 complete responses were received after several follow-ups,
representing a response rate of 42.6%.
Among the 152 participating IJVs, 12 are involved in chemistry,
12 in food, 11 in machinery, seven in textile and clothing, 31 in other
manufacturing, 59 are the electrical and electronics industry, six in
medical science and biotechnology, and six are involved in the
service industry. To test for nonresponse bias, a t test was adopted
to check the difference between early respondents and late re-
spondents, as suggested by Armstrong and Overton (1977). We
followed the recommendation of Weiss and Heide (1993): early
responses were de?ned as the ?rst 75% and the last 25% were
considered as late responses. In this test, we collected data fromthe
Taiwan Stock Market Observation Post System and compared the size
of investment (which is measured as the cumulative foreign direct
investment) and ownership structure (which is measured as equity
owned by the Taiwanese MNCs) between these two groups of re-
spondents. The results of t tests showed insigni?cant differences
between these two groups of ?rms with respect to either cumu-
lative foreign direct investment (F ¼1.292, p >0.05) or the owner-
ship structure of IJVs (F ¼0.902, p >0.05), mitigating the concern
for nonresponse bias (Armstrong & Overton, 1977).
We relied on the managers of the Taiwanese parent companies
to respond to the surveys. In accordance with the common
practice in IJV research, this study relied on data provided by one
partner. However, collecting data from a single source may result
in the problem of common method variance. Therefore, we con-
ducted a Harman's one-factor test as suggested by Podsakoff and
Organ (1986) to evaluate how serious this problem might be for
our study. The results of this test revealed that a six-factor
structure was the best solution, and the measurement items of
our six variables did not load heavily on one factor. It thus
appears that the problem of common method variance was
insigni?cant.
3.2. Measures
The research variables of this study include centralization,
formalization, socialization, perceived value gap, information
asymmetry, and IJV performance. The measures of these variables
were adopted from existing scales that had been used in previous
studies, or they were developed fromconcepts that are discussed in
the existing literature. All survey items were measured on a seven-
point scale, ranging from1 (strongly disagree) to 7 (strongly agree).
3.2.1. Dependent variable
IJV performance can be assessed using objective or subjective
measures. Because some studies have demonstrated that subjective
and objective measures of IJV performance are highly correlated
(Ari~ no, 2003; Geringer & Hebert, 1989; Mohr, 2006), several re-
searchers suggest that a subjective measure of joint venture per-
formance is appropriate (Gong et al., 2007; Pangarkar & Klein,
2004). This study adopted perceptual measures to assess the
satisfaction of the parent company in the following three areas: (1)
strategic goal achievement; (2) cooperative relationship with IJV
partners; and (3) willingness to continue this cooperative rela-
tionship. These measures have been proven to be reliable when
evaluating IJV performance in previous studies (Choi & Beamish,
2004; Luo & Park, 2004; Yan & Gray, 2001). The a reliability in
the present study was 0.90.
3.2.2. Mediators
Information asymmetry. Information asymmetry describes a
situation in which one party possesses more information than the
other party regarding the object of an exchange (Mishra et al.,
1998). In relationships characterized by information asymmetry,
it is possible for a party to act opportunistically without being
detected (Wathne & Heide, 2000). To assess information asym-
metry, we measured the level of the parent company's under-
standing of its IJV partner's actions. Respondents were asked to
evaluate how well they knew their partners in terms of actions,
strategic change, and the degree of mutual understanding. The
higher the level of the respondents' knowledge about their part-
ners, the lower the level of information asymmetry would be. The a
reliability for this scale was 0.87.
Perceived value gap. Seetoo and Hsiung (2004) de?ned the
perceived value gap as the perceived differences among partners in
goals, strategic directions, and organizational values. Following this
perspective, this study asked respondents to evaluate how they
differed from their partners in strategic vision, goals, and expec-
tations of howgoals can be achieved (Newburry et al., 2003). The a
reliability for this scale was 0.74.
3.2.3. Independent variables
Centralization. To measure the degree of centralization, this
study ?rst assessed the degree of autonomy the IJV had in decision-
making. A low degree of decision-making autonomy for IJV sub-
sidiaries would imply a high degree of centralization by the parent
companies (Geringer & Hebert, 1989; Glaister et al., 2003a;
Newburry et al., 2003). This study asked respondents to indicate
the degree of decision-making autonomy held by the IJV in relation
to ?ve operational (functional) activities, including research and
development strategy, ?nance strategy, marketing strategy,
manufacturing strategy, and human resource management strategy
(Choi & Beamish, 2004; Luo et al., 2001; Yu, Wong, & Chiao, 2006).
The a reliability for this scale was 0.94.
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 37
Formalization. Formalization refers to the extent to which the
parent company adopts written regulations, objectives, processes,
rules, and duties to de?ne the expected behavior and the standards
of output or quality (Poppo & Zenger, 2002; Wang & Fulop, 2007).
Respondents in this study were asked to evaluate the degree of
formalization in quality control procedures, cost control pro-
cedures, and operation performance reporting (Baliga & Jaeger,
1984; Jaussaud & Schaaper, 2006). The a reliability for this scale
was 0.80.
Socialization. In the present study, socialization refers to the
extent to which the parent company creates shared norms and
values and controls the behavior and decision making of its IJV
through social interactions (Chalos & O'Connor, 2004; Nohria &
Ghoshal, 1994). To measure socialization, three items were used:
(1) the frequency of the local partner's employees visiting the MNC
headquarters; (2) the frequency of group communication among
partners (Chalos & O'Connor, 2004; Glaister et al., 2003a); and (3)
the frequency of the local partner's employees being invited to join
training activities hosted by the MNC headquarters (Glaister et al.,
2003a; Jaussaud & Schaaper, 2006). The a reliability of this scale
was 0.80.
3.2.4. Control variables
Several control variables were included in this study. While we
provided no formal hypotheses for these factors, we controlled for
their impacts in our model. These control variables were industry,
IJV age, size, and ownership. Two dummy variables were used to
code the industry. The industry effect of the high technology sector,
which included the electrical and electronics industry, medical
science and the biotechnology sector, was coded 1; for other sec-
tors, it was coded 0. The industry effect of the service sector was
coded 1; for other sectors, it was coded 0. IJV age was computed by
taking the difference (in years) between the year 2011, when the
data were collected, and the year when the IJV's establishment was
initially reported to the Taiwanese government. IJV size was
measured by capital and by looking at the records of IJVs' average
annual sales over the past 3 years. Ownership was measured as the
percentage of IJV equity controlled by an MNC.
3.3. Reliability and validity
Although our scales were adopted or developed from previous
studies on related issues, we interviewed four CEOs of IJVs and
asked for their feedback to con?rm the validity of the measures.
The Cronbach a of the six scales all exceeded 0.70, indicating good
reliabilities. To evaluate the construct validity of the measures, we
conducted con?rmatory factor analysis using structural equation
modeling with the maximum likelihood method. The con?rmatory
factor analysis results demonstrated that our six-factor measure-
ment model had a satisfactory ?t (c
2
¼229.42, df ¼155, c
2
/
df ¼1.48, goodness of ?t index ¼0.87, normed ?t index ¼0.95,
comparative ?t index ¼0.98, root mean square error of approx-
imation ¼0.06). Although the goodness of ?t index value was
slightly below the recommended threshold of 0.90, the values of
both normed ?t index and comparative ?t index exceeded 0.90.
These ?ndings indicate that the measurement model has a good
model-to-data ?t. Furthermore, the ratio of chi-square to the de-
gree of freedomwas 1.48 (any value below 2 for this ratio indicates
a good ?t). Table 1 presents the values of composite reliability, the
squared multiple correlation, and the average variance extracted
(AVE). The composite reliability values of our study variables were
all above 0.70, the squared multiple correlation values were all
above 0.40, and the AVE values were all above 0.50. These results
indicate that our measures had good internal consistency and
convergent validity (Bagozzi and Yi, 1988).
To assess discriminant validity, Fornell and Larcker (1981) sug-
gested that the square root of AVE should be greater than the cor-
relation coef?cient in the corresponding columns and rows. Table 2
shows the means, standard deviations, and correlation coef?cients
of the study variables. This table reveals that the square root of each
Table 1
The values of squared multiple correlation (SMC), composite reliability (CR), and average variance extracted (AVE) for main research variables.
Variable Mean (SD) Estimate (SE) t value SMC CR AVE
Centralization
Marketing strategy of subsidiary 3.64 (1.78) 0.91 (0.06) 14.30
**
0.82 0.94 0.76
Capital investment of subsidiary 3.67 (1.53) 0.79 (0.07) 11.54
**
0.63
Manufacturing strategy of subsidiary 3.48 (1.75) 0.92 (0.06) 14.80
**
0.85
Human resource strategy of subsidiary 3.45 (1.76) 0.87 (0.07) 13.32
**
0.75
Research and development strategy of subsidiary 3.53 (1.77) 0.87 (0.07) 13.30
**
0.75
Formalization
Formal quality control procedure 5.75 (1.00) 0.69 (0.08) 8.90
**
0.58 0.81 0.59
Formal cost control procedure 5.86 (0.89) 0.81 (0.08) 10.70
**
0.65
Formal operation performance report 6.07 (0.83) 0.81 (0.08) 10.69
**
0.65
Socialization
The frequency of the local partner's employees visiting the MNC headquarters 5.10 (1.47) 0.77 (0.07) 10.55
**
0.60 0.82 0.60
The frequency of group communication among partners 5.37 (1.12) 0.79 (0.07) 10.93
**
0.63
The frequency of the local partner's employees being invited to join training
activities hosted by the MNC headquarters
4.75 (1.56) 0.77 (0.07) 10.50
**
0.59
Perceived Value Gap
Same vision among partners 5.28 (1.01) 0.72 (0.08) 9.41
**
0.51 0.75 0.51
Same strategic goals among partners 5.13 (1.13) 0.75 (0.08) 9.97
**
0.56
Parent company can achieve its strategic goals through this IJV 5.39 (0.87) 0.66 (0.08) 8.38
**
0.53
Information Asymmetry
There exist suf?cient information about partner's actions 5.33 (1.01) 0.85 (0.07) 12.69
**
0.72 0.86 0.68
There exist some channels to understand partner's strategic change 5.16 (1.10) 0.69 (0.07) 9.48
**
0.48
There are numerous effort to foster mutual understanding 5.23 (1.05) 0.92 (0.06) 14.41
**
0.85
IJV performance
Strategic goal achievement 5.18 (1.08) 0.89 (0.07) 13.75
**
0.80 0.90 0.75
Cooperative relationship with IJV's partners 5.36 (0.97) 0.91 (0.06) 14.07
**
0.82
Willingness to continue this cooperative relationship 5.51 (1.00) 0.80 (0.07) 11.62
**
0.64
n ¼152;
*
p
According to transaction cost theory, the risk of opportunism is a major obstacle to cooperative relationships,
such that multinational corporations need to exert control over their international joint
ventures (IJVs). However, inconsistent findings indicate that the controleperformance linkage is complex
and poorly understood. This study develops a mediating framework and suggests that perceived value
gap and information asymmetry have a direct effect on IJV performance. Perceived value gap and information
asymmetry can also mediate the relationship between control mechanisms (e.g., centralization,
formalization, and socialization) and IJV performance. A sample of 152 Taiwanese IJVs in China
showed that the direct effects of formalization and socialization on IJV performance satisfaction were
significant. The empirical results showed that information asymmetry and perceived value gap mediated
the formalizationeperformance and socializationeperformance relationships. However, neither the
direct effect of centralization nor the indirect effect through information asymmetry mediation on IJV
performance was significant.
Reexamining the relationship between control mechanisms and international
joint venture performance: The mediating roles of perceived value gap and
information asymmetry
Ming-Chang Huang
a, *
, Hsin-Hua Hsiung
b
, Ting-Chun Lu
c
a
Department of Business Administration, Providence University, Shalu District, Taichung, Taiwan, ROC
b
Department of Business Administration, National Dong Hwa University, Shou-Feng, Hualien, Taiwan, ROC
c
Department of Business Administration, National Chengchi University, Wenshan District, Taipei, Taiwan, ROC
a r t i c l e i n f o
Article history:
Received 24 August 2012
Accepted 2 September 2014
Available online 11 March 2015
Keywords:
Control mechanism
Information asymmetry
International joint ventures
International joint ventures performance
Perceived value gap
a b s t r a c t
According to transaction cost theory, the risk of opportunism is a major obstacle to cooperative re-
lationships, such that multinational corporations need to exert control over their international joint
ventures (IJVs). However, inconsistent ?ndings indicate that the controleperformance linkage is complex
and poorly understood. This study develops a mediating framework and suggests that perceived value
gap and information asymmetry have a direct effect on IJV performance. Perceived value gap and in-
formation asymmetry can also mediate the relationship between control mechanisms (e.g., centraliza-
tion, formalization, and socialization) and IJV performance. A sample of 152 Taiwanese IJVs in China
showed that the direct effects of formalization and socialization on IJV performance satisfaction were
signi?cant. The empirical results showed that information asymmetry and perceived value gap mediated
the formalizationeperformance and socializationeperformance relationships. However, neither the
direct effect of centralization nor the indirect effect through information asymmetry mediation on IJV
performance was signi?cant.
© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier
Taiwan LLC. All rights reserved.
1. Introduction
The formation of an international joint venture (IJV) is an
important market entry strategy that has been adopted by many
multinational corporations (MNCs) in the course of their global
expansions (Gong, Shenkar, Luo, & Nyaw, 2007; Makino, Chan,
Isobe, & Beamish, 2007). To realize the potential for joint value
creation, partners must exchange information, share knowledge,
and make investments. Due to the possibility of opportunism,
however, investing valuable assets and sharing proprietary
knowledge with the other partner(s) may expose each party to a
certain degree of risk (Parkhe, 1993; Zeng & Chen, 2003). Luo
(2007a) de?ned opportunism in the context of IJVs as acts or pat-
terns of behavior performed by speci?c parties from speci?c
countries to obtain unilateral gains to the substantial detriment of
other parties from other countries. When entering into a voluntary
interorganizational arrangement such as an IJV, assets are located
overseas, a focal parent becomes vulnerable to the actions of the
other partner, and the risk of expropriation or opportunism is
greater than it would be in a domestic context (Ertug, Cuypers,
Noorderhaven, & Bensaou, 2013; Pangarkar & Klein, 2004). There
is therefore a need for control mechanisms, which can be classi?ed
as centralization, formalization, and socialization according to their
objective of control over actions, results, and personnel/culture (Li,
Tang, Okano, & Gao, 2013). There is also a need to reduce trans-
action costs and enhance parent satisfaction by decreasing the
likelihood of opportunism(Deitz, Tokman, Richey, &Morgan, 2010;
Fryxell, Dooley, & Vryza, 2002; Luo, Shenkar, & Nyaw, 2001;
Mesquita & Brush, 2008; Pangarkar & Klein, 2004). The purpose
of control is to protect an MNC's own interests while simulta-
neously trying to overcome a partner's opportunistic behavior
(Calantone & Zhao, 2000; Ren, Gray, & Kim, 2009), which can be
considered as an antecedent of IJV performance (Christoffersen,
2013; Reus & Rottig, 2009; Selekler-G€ oks¸ en & Uysal-Tez€ olmez,
2007). IJV control is very important to explaining IJV performance
and to IJV success (Liu, Vredenburg, & Steel, 2014).
* Corresponding author. Department of Business Administration, Providence
University, No. 200, Section 7, Taiwan Boulevard, Shalu District 43301, Taiwan, ROC.
E-mail address: [email protected] (M.-C. Huang).
Peer review under responsibility of College of Management, National Cheng Kung
University.
HOSTED BY
Contents lists available at ScienceDirect
Asia Paci?c Management Review
j ournal homepage: www. el sevi er. com/ l ocat e/ apmrvhttp://dx.doi.org/10.1016/j.apmrv.2014.12.004
1029-3132/© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
Asia Paci?c Management Review 20 (2015) 32e43
There are many studies ofdand numerous efforts to explor-
edthe relationships between control mechanisms and IJV out-
comes that enhance our understanding and knowledge of
international alliances, the causes and consequences of the MNC-
subsidiary relationship, and the organizational design of interna-
tional business. However, the theoretical arguments and empirical
?ndings remain incongruent. For example, Newburry, Zeira, and
Yeheskel (2003) showed that IJVs located in China with authority
are more effective in the formulation and implementation of stra-
tegic plans. Zhang and Li (2001) indicated that autonomy and
success are related to IJV performance. However, Lu and Lee (2005)
reported that the impact of centralization on IJV performance
satisfaction is insigni?cant. The meta-analysis of Reus and Rottig
(2009) indicated that hierarchical control may theoretically have
positive or negative effects on performance. In mainland China,
Calantone and Zhao (2000) indicated that the con-
troleperformance relationship is insigni?cant for Japanese IJVs in
China; however, for Korea and the USA IJVs, it is positive. Reus and
Rottig (2009) report that hierarchical control has an insigni?cant
direct effect on performance in the Chinese sample but a signi?cant
effect in the non-Chinese sample. Some studies indicate that formal
control mechanisms can enhance IJV performance (Gong et al.,
2007; Perrone, Zaheer, & McEvily, 2003), while other studies
report inconsistent results (Fryxell et al., 2002; Inkpen & Currall,
2004; Nohria & Ghoshal, 1994). Luo and Park (2004) con?rm the
positive relationship between socialization and IJV performance.
However, the empirical results of Fryxell et al. (2002) show only a
marginally signi?cant positive relationship between socialization
and IJV performance. These studies provide important insights
and syntheses, but they do not provide conclusive evidence of
controleperformance relationship (Liu et al., 2014). These incon-
sistent ?ndings indicate that the controleperformance linkage is
complex and poorly understood. The relationship between control
mechanisms and IJV performance remains unclear (Pangarkar &
Klein, 2004). We still have only a limited understanding of the
process by which speci?c control mechanisms function.
Through a careful review of the literature, this study identi?es
three research gaps. First, opportunism is a critical assumption of
human behavior in transaction cost theory, and the potential for
opportunism plays an important role in IJV success (Geringer &
Hebert, 1989). Opportunism creates a need for ex ante and ex post
safeguarding, including formal and social governance structure
(Nakos & Brouthers, 2008; Subramani & Venkatrman, 2003).
However, potentially opportunistic partners do not necessarily
behave opportunistically. This observation raises the question of
why opportunistic behavior occurs. We contend that the perceived
value gap and information asymmetry are two critical conditions
that may induce opportunistic behaviors (Williamson, 1985).
Seetoo and Hsiung (2004) refer to the perceived value gap as the
extent to which common goals or shared values diverge among
partners, and they identify information asymmetry as the level of
information distribution asymmetry throughout the alliance.
Control mechanisms must overcome these two antecedents to
prevent opportunistic behavior. Little systematic empirical research
has been conducted to test the effects of control mechanisms on the
perceived value gaps and information asymmetry. Second, many
studies assume that if a value gap and information asymmetry exist
among partners, control mechanisms will have direct positive ef-
fects on IJV performance. The indirect effects of control mecha-
nisms on an IJV's outcome through a reduction in the value gap and
information asymmetry are treated as given and ignored. There-
fore, few studies have precisely measured the extent of value gap
incongruence and information asymmetry or clari?ed the way in
which the controleperformance relationship of an IJVis affected by
the value gap and information asymmetry among partners. Given
the importance of control mechanisms in IJVs, how they function
should be addressed systematically. However, there is little
research on how control mechanisms have been executed in
practice (Li et al., 2013). Finally, the de?nitions of control mecha-
nisms differ such that the empirical results are dif?cult to compare.
Ren et al. (2009) suggested that making distinctions among control
mechanisms and understanding the impact of each mechanism on
IJV performance to enrich our understanding could be necessary for
future studies.
To answer the question of how MNCs can effectively coordinate
and control IJVs, we suggest that the underlying reasons for
opportunism should be taken into account. Based on transaction
cost theory and the information process perspective, this study
proposes a conceptual model to examine whether parent com-
panies' control mechanisms could enhance the performance of an
IJV by reducing the perceived value gap and information asym-
metry among partners. In our view, information asymmetry and
the perceived value gap have a signi?cant mediating effect on the
relationship between control mechanisms and IJV performance.
The IJV approach is particularly prevalent among MNCs in the
Chinese market (Chan, Luk, &Wang, 2005; Li et al., 2013). A sample
of 152 IJVs in China involving a Taiwanese parent company
con?rmed the mediating effects of a perceived value gap and in-
formation asymmetry on the relationships between control
mechanisms and IJV performance. This paper proposes a theoret-
ical model and contributes to the literature on the con-
troleperformance relationship of IJVs in the following ways: (1)
testing the effects of control mechanisms on value gap and infor-
mation asymmetry reduction, which allowed us to address the is-
sues of opportunistic behavior and IJV management; (2) clarifying
the direct and indirect relationships between control mechanisms
and IJV performance through this theoretical framework to
enhance our understanding of the controleperformance link.
The remainder of this paper is organized as follows. In the next
section, we review the relevant literature, discuss the main study
concepts, and explain the theoretical basis for the hypothesized
relationships. In the methodology section, we describe the sample,
the measures used, and the statistical methods. In the results sec-
tion, we present the statistical relationships that were observed
among the control mechanisms, perceived value gap, information
asymmetry, and IJV performance. Finally, we present the conclu-
sions, explain the theoretical implications, and discuss the practical
implications of our ?ndings.
2. Theoretical background and hypotheses
2.1. Facilitating conditions of opportunistic behavior
Transaction cost theory de?nes opportunism as seeking self-
interest with guile (Williamson, 1985), with decision-makers
showing a tendency to exploit their positions to advance their
own self-interest (Das & Rahman, 2010; Rind?eish & Heide, 1997;
Tsang, 2006). This phenomenon provides a foundation for IJV
control theory (Deitz et al., 2010). The potential for opportunism
arises when one party is vulnerable to the actions of another, whose
behavior is not under its control (Parkhe, 1993). The formation of a
joint venture naturally gives rise to many opportunities for
opportunism (Morris & Cadogan, 2001). While opportunism is the
cause and antecedent of a partner's opportunistic behavior
(Rind?eish & Heide, 1997), opportunism per se is not a suf?cient
condition for opportunistic behavior. Certain critical conditions or
facilitators should exist for the actual occurrence of this behavior.
Opportunistic behavior is most likely to appear in two situations.
The ?rst such situation occurs when the partners lack common
objectives or shared values. The larger a partner perceives a value
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 33
gap in the difference among partners regarding common goals and
shared values (Seetoo & Hsiung, 2004), the higher the perceived
value gap and the likelihood of opportunism will be. The second
situation occurs when there is information asymmetry among
partners. Williamson (1985) noted that opportunism results from
the incomplete disclosure of information. Information asymmetry
creates an opportunity for a party to engage in opportunistic
behavior, which may not be sensed by the other parties. Therefore,
the possibility of opportunistic behavior rises in accordance with
any increase in the perceived value gap and information
asymmetry.
2.1.1. Information asymmetry
Bounded rationality is the assumption that decision makers face
constraints to their information-processing and communication
abilities, which limit their rationality (Rind?eish & Heide, 1997).
The exchange of information may reinforce the connection be-
tween the parent and the subsidiary and improve cooperation, but
incomplete or misleading information may cause opportunistic
behavior to occur more frequently (Williamson, 1985). Deliberately
withholding information in a joint venture relationship is an un-
healthy behavior that usually causes misunderstandings, incorrect
expectations, and dysfunctional con?ict among partners that in-
crease coordination costs and reduce the decision-making quality
(Bennet & Cadogan, 2001; Kwon, 2008; Morris & Cadogan, 2001).
Information is generally not symmetrically distributed among
all of the parties to a transaction (Mishra, Heide, & Cort, 1998).
Information asymmetries exist in all exchange relationships and
are accentuated in and IJV as each parent brings very different
perspectives, priorities, and quality of resources to decision making
(Yan &Child, 2004). In a transactional or cooperative relationship, if
the parties are not equally familiar with the details of a situation,
the ?owof information is likely to be blocked. In strategic alliances,
the lack of suf?cient information may make it dif?cult for one party
to predict and respond to the other party's self-interest seeking
action (Glaister, Husan, & Buckley, 2003a). Hence, transaction cost
theory treats information asymmetry as a primary cause of
opportunistic behavior (Williamson, 1985). Information asymmetry
makes it dif?cult for one party to detect the other party's oppor-
tunism, which results in an increase in monitoring costs (Wathne &
Heide, 2000). The increasing need for monitoring and coordination
results in higher transaction costs and inhibits joint venture suc-
cess. Therefore, we propose that information asymmetry will have
detrimental effects on the performance of an IJV.
H
1
: Ceteris paribus, information asymmetry is negatively related to
IJV performance.
2.1.2. Perceived value gap
The operation of IJVs is strongly affected by the partners' values,
practices, and management systems (Danis & Parkhe, 2002).
Because each partner may have unique interests and goals for the
IJV, differences in values are very common. For instance, the Chi-
nese partners in US-Sino joint ventures are often interested in
acquiring management know-how and advanced technology, but
the foreign partners are often more interested in obtaining access
to the Chinese domestic market and cheap labor (Chan et al., 2005).
Ouchi (1980) noted that common values and beliefs can provide the
type of harmony of interests that eliminates the possibility of
opportunistic behavior. A set of shared values and non-legal sanc-
tions in the form of relational norms encourages partners to
commit to their exchange relationship, restricts the partner ?rms
from acting opportunistically (Huang, Cheng, & Tseng, 2014), and
facilitates partner cooperation (Zeng & Chen, 2003). To avoid
opportunistic behavior, the IJV management process requires
participating partners to establish a clearly shared strategic vision
(Glaister et al., 2003a). The absence of shared values signi?es the
existence of a value gap among partners and may breed opportu-
nistic motives (Seetoo & Hsiung, 2004).
In IJVs, the behavior of the collaborating partners usually affects
collaborative performance and satisfaction. In the presence of
opportunistic tendencies and competitive threats, it is dif?cult for
partners to develop collaboration (Park & Ungson, 2001). To avoid
opportunistic behaviors and alliance failures, it is quite important
for the partners to reduce goal divergence and to build shared
values (Ouchi, 1980). Shared values represent the collective goals
and aspirations of organizational members. The establishment of
shared values in the formof relational norms represents a value gap
that is less perceived among IJV partners. If partners have shared
values, each partner is less likely to sacri?ce the other party's in-
terests in pursuit of its own separate goals or ambitions.
When organization members share common perceptions about
the way they should interact with one another, they can avoid po-
tential misunderstandings and gain more opportunities for the free
exchange of ideas or resources (Tsai &Ghoshal, 1998). In contrast, if
partners have major differences in terms of their strategic goals and
management styles, their negotiation costs and transaction costs
may increase (Fryxell et al., 2002; Seetoo & Hsiung, 2004), and the
IJV performance may be impaired (Pangarkar and Klein, 2004). The
establishment of shared values in the form of relational norms,
which lessen perceived value gaps, encourages partners to commit
to their exchange relationship and prevents partner ?rms from
acting opportunistically (Das & Teng, 2001; Liu, Luo, & Liu, 2009;
Tangpong, Hung, & Ro, 2010). Therefore, according to transaction
cost theory, we expect there to be a negative relationship between
the perceived value gap and IJV performance.
H
2
: Ceteris paribus, the perceived value gap is negatively related to
IJV performance.
2.2. The mediating role of the perceived value gap and information
asymmetry
How to handle the problem of opportunism is a crucial concern
for organizations with a hybrid governance structure (Luo, 2005).
To a greater or lesser extent, the parent companies of an IJV must
implement certain control mechanisms. These are either organi-
zational or structural arrangements or processes by which a party
can in?uence the behavior and output of another party through the
use of power, authority, and a wide range of bureaucratic, cultural,
and informal mechanisms (Calantone & Zhao, 2000; Fryxell et al.,
2002; Geringer & Hebert, 1989; Pangarkar & Klein, 2004) as safe-
guards to reduce transaction costs and protect their interests
against opportunism (Mesquita & Brush, 2008). Control involves
the monitoring process and consists of formal control (bureaucratic
control) and informal control (culture control) (Baliga & Jaeger,
1984). Drawing on the relevant research, we selected three types
of control mechanisms (namely, centralization, formalization, and
socialization) as our independent variables (Fryxell et al., 2002;
Geringer & Hebert, 1989; Li et al., 2013; Nobel & Birkinshaw,
1998; Sundaram & Black, 1992). Adequate control mechanisms
may have several important functions, such as increasing the
transparency of partners' behavior (Parkhe, 1993), preventing
cheating and moral hazards (Rind?eish & Heide, 1997), achieving
expected objectives, and ensuring that the IJV can faithfully pursue
strategies that conform to the interests of the parent companies
(Pangarkar & Klein, 2004). We predicted that these types of con-
trols could enhance IJV performance by reducing the perceived
value gap and information asymmetry.
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 34
2.2.1. Centralization and information asymmetry
The allocation of decision rights is a means of safeguarding
(Schepker, Oh, Martynov, & Poppo, 2014). Centralization refers to
the amount of decision-making power each parent company ex-
ercises with regard to the venture's daily operations (Yan & Gray,
2001). Centralization means that middle and lower-level man-
agers have relatively limited authority or freedomto make strategic
and operational decisions independently of the parent companies
(Baliga &Jaeger, 1984; Glaister, Husan, &Buckley, 2003b; Newburry
et al., 2003; Nobel & Birkinshaw, 1998; Reus & Rottig, 2009).
Through centralization, an MNC can determine the manner in
which IJV activities are performed so that the IJV is in line with its
own objectives (Reus & Rottig, 2009; Yan & Gray, 2001) and posi-
tively related to parent-subsidiary integration (Newburry et al.,
2003).
Control rights may also include monitoring provisions to pro-
vide behavioral control (Schepker et al., 2014). Centralization is
therefore a crucial dimension in the structural design of MNCs,
which can affect the capacity and capability of information pro-
cessing (Tushman & Nadler, 1978). For an MNC to make strategic
decisions and maintain adequate control, it is important to gather
suf?cient information about the IJV and its partners from sub-
sidiaries. However, some scholars have indicated that centraliza-
tion might reduce the autonomy or freedom of an IJV subsidiary,
thus adversely affecting the performance (Newburry et al., 2003).
However, we believe that through effective communication and
ef?cient information systems, the potential side effects of central-
ization can be overcome.
To increase the awareness of strategic development and
enhance the understanding of foreign business operations, an MNC
needs to build an effective information reporting system that can
capture key information from its IJV subsidiaries and partners (Yan
& Child, 2004). The complete information-processing system of
MNCs involves the gathering, interpretation, synthesis, and storage
of information within the headquarters and in the subsidiaries
(Egelhoff, 1982; Tushman & Nadler, 1978). Structural features are
the sources of organizational information-processing capability in
terms of information acquisition, analysis, and interpretation
(Carson, Madhok, Varman, & John, 2003). MNCs can communicate
their expectations to IJV managers and partners through the
planning and budgeting process as well as through the particularly
important business plan agreed upon through the board (Glaister
et al., 2003b). IJV managers provide necessary information
through the board, and they act in the way expected of them
(Glaister et al., 2003b). Centralization is likely to lead to a clear
understanding between the elements of the IJV system and
decrease the information gap among partners.
Information processing can be conducted more easily in a
centralized organization than in one that is decentralized.
Furthermore, a regular, frequent, and compulsory ?ow of infor-
mation between the parent company and the subsidiary helps to
reduce information asymmetry that results in minimizing the
possibility of hiding and cheating among MNCs and its partners.
Thus, higher degrees of parent control are associated with
greater performance (Quer, Claver, & Rienda, 2007). Ceteris par-
ibus, MNCs that are centralized may possess more information
about the policies, operations, and decision-making processes of
their IJVs, and they can grasp their partners' intentions and be-
haviors more effectively (Chang & Taylor, 1999) and thus increase
performance satisfaction. Hence, we propose the following
hypothesis.
H
3
: Through the mediation of information asymmetry among IJV
partners, centralization has a positive association with IJV
performance.
2.2.2. Formalization, perceived value gap and information
asymmetry
Baliga and Jaeger (1984) treat formalization as a type of
bureaucratic control that consists of using limits and an explicit set
of rules and regulations to delineate the desired performance in
terms of output and/or behavior. Formalization emphasizes these
matters through contracts, formal written procedures, policies, and
individual performance appraisals (Wang & Fulop, 2007). Formal
control mechanisms provide a justi?cation for a parent company to
trace its subsidiaries' moves or actions. Formalization provides an
opportunity and a channel for MNCs to communicate with their
partners and share critical information. Formalization focuses on
the information of a ?rm's operation process and uses objectives to
measure the accuracy and completeness of business activities (Liu
et al., 2014). Establishing a formal reporting system facilitates the
?ow of information feedback between parent companies and their
IJV subsidiaries, and it also helps a parent company to monitor and
coordinate the activities of its IJV partners. Thus, frequent
communication through formal mechanisms enhances the trans-
parency of partner actions (Chalos & O'Connor, 2004), which may
result in lower information asymmetry.
Control rights can include monitoring provisions to provide
behavioral control (Schepker et al., 2014). The concept of formal-
ization focuses on monitoring and reporting the operational pro-
cess of business activities to increase information transparency and
decrease the misunderstanding among MNCs, IJVs, and other
partners. Therefore, formalization provides a communication tool
to enhance information-processing capabilities for the trans-
mission of information from one party to another (Cam en,
Gottfridsson, & Rundh, 2011). Formalization facilitates key infor-
mation capturing and affects an MNC's ability to monitor, coordi-
nate, and integrate activities within the context of an IJV (Yan &
Child, 2004). Information asymmetry among partners can be
decreased through formalization and monitoring (Seetoo &Hsiung,
2004).
When deliberate control and information-gathering procedures
are in place, the partners' tendency toward opportunism can be
identi?ed more easily, such that the likelihood of opportunistic
behavior is reduced (Fryxell et al., 2002). Hence, effective formal
control mechanisms can enhance IJV strategy implementation and
strategic competitiveness (Yan & Child, 2004). Because formaliza-
tion increases the information-processing capability of MNCs, we
suggest that the exercise of formal control mechanisms can reduce
information asymmetry among partners and enhance the effec-
tiveness of management control. Accordingly, we propose the
following hypothesis:
H
4
: Through the mediation of information asymmetry among IJV
partners, formalization has a positive association with IJV
performance.
The formalization mechanism involves optimizing decision
making by relying on formal records and procedures (Sundaram &
Black, 1992). Regarding the role and consequences of formalization
in IJV management, there are two opposing perspectives in the
existing literature.
On one hand, some studies tend to support the view that
formalization negatively in?uences the perceived value gap. First,
formalization involves the development of clear and well-
documented policies, procedures, and operating rules (Martinez
& Jarillo, 1997), which help to clarify the exchange partner roles
and expectations and provides clarity regarding the terms, rem-
edies and con?ict resolution procedures (Tangpong et al., 2010).
Therefore, formalization also promotes communications among
partners, which reveals an openness and willingness to rely on the
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 35
other partners (Costa e Silva, Bradley, &, Sousa, 2012). Formaliza-
tion enhances institutionalized practices and routines for dealing
with a partner organization, creating a stable context within which
interorganizational and interpersonal trust can be developed
(Zaheer, McEvily, & Perrone, 1998). This type of trust is bene?cial to
the development of common goals and shared values. Furthermore,
in the process of documentation or institutionalization, IJV partners
must communicate and clarify their speci?c needs and objectives.
The communication and clari?cation process can push IJV partners
to achieve consensus on strategic goals and to specify the behavior
that is required in the form of rules or an explicit program(Nobel &
Birkinshaw, 1998). Trust enhances the belief that a partner will
perform actions that will result in positive outcomes and will not
perform unexpected actions that will result in negative outcomes
(Kwon, 2008). Once trust, shared values, and a relational norm are
established, which means a lower perceived value gap, partners
may be reluctant to behave opportunistically (Liu et al., 2009).
Formalization as a control mechanism would carry a cost to build,
which would its increase monitoring cost. Through the develop-
ment of shared value, relational norms, and trust, which operate as
self-enforcement safeguards and a less costly governance alterna-
tive (Poppo & Zenger, 2002), partners can cooperate to offset the
weaknesses of formalization (Liu et al., 2009). Therefore, we pro-
pose that formalization can decrease the value gap among IJV
partners, which lowers the likelihood of opportunism. IJV perfor-
mance can be enhanced by formalization through a reduction of the
value gap.
H
5a
: Through the mediation of the perceived value gap among IJV
partners, formalization has a positive association with IJV
performance.
On the other hand, some studies do not support the notion that
formalization helps to reduce the value gap. Formalization causes
decision-making to be routinized through rules and procedures
(Nobel & Birkinshaw, 1998). Although ex ante and ex post safe-
guarding can mitigate opportunism, several researchers (Inkpen &
Currall, 2004; Nohria & Ghoshal, 1994) argued that formalization
does not necessarily contribute to the development of shared
values. Inkpen and Currall (2004) noted the fact that trust develops
more slowly in the context of extensive formal control. Luo and
Park (2004) argued that formalized control is often exercised
because of a lack of trust. Furthermore, the troublesome processes
of formalization and institutionalization may reduce participants'
intentions to engage in repeated renegotiation to accommodate the
changing expectations of their partners (Fryxell et al., 2002).
Relying on formal control may also cause partners to neglect the
importance of creating a compatible, harmonious, and considerate
cultural environment. The likelihood of opportunism is unlikely to
be decreased when an MNC exerts formal control over its IJV. Thus,
we provide a competing hypothesis as follows.
H
5b
: Through the mediation of the perceived value gap among IJV
partners, formalization has a negative association with IJV
performance.
2.2.3. Socialization, the perceived value gap, and information
asymmetry
Social control mechanisms usually have positive effects on in-
formation sharing and two-way communication among partners
(Ertug et al., 2013). For example, transferring and rotating managers
between the headquarters and subsidiaries can create verbal in-
formation channels (Baliga & Jaeger, 1984; Edstr€ om & Galbraith,
1977). Frequent meetings and planned visits between IJV
managers and headquarters personnel can enhance information
sharing and collaboration (Chalos & O'Connor, 2004). Furthermore,
expatriates and boundary spanners with good socialization training
can develop shared norms and a collective culture. These practices
and interactions help IJV partners communicate more openly and
honestly, thus ensuring that they will be willing to share more
sensitive information.
Ahuja and Galvin (2003) observed that applying social control
mechanisms can improve performance and members' ability to
communicate. Good communication results from consistently and
clearly informing the partner about information that is relevant to
cooperation (Glaister et al., 2003a). Frequent communication can
increase the transparency of partners' actions (Chalos & O'Connor,
2004) and improve each partner's ability to understand and predict
the behavior of the opposite party (Das & Teng, 1998). In summary,
by creating an open and cordial atmosphere, social control mech-
anisms help to improve trust and lower both the cost of informa-
tion sharing and the level of information asymmetry among
partners (Luo, 2007b) as well as mitigate potential concerns of
opportunism and reduce transaction costs to improve IJV perfor-
mance satisfaction. Accordingly, we propose the following
hypothesis.
H
6
: Through the mediation of information asymmetry among IJV
partners, socialization has a positive association with IJV
performance.
Socialization is associated with the development of shared
norms and a collective culture. Through socialization, JV managers
learn to share common attitudes and knowledge of the organiza-
tion (Chalos & O'Connor, 2004). In the context of IJVs, socialization
in a subsidiary can be achieved in a variety of ways, such as through
instruction in the parent company's rules and procedures (Gupta &
Govindarajan, 1991), structured personnel interactions (Fryxell
et al., 2002), and the rotation of personnel in key managerial po-
sitions between the parent company and the subsidiaries (Chalos &
O'Connor, 2004; Edstr€ om & Galbraith, 1977; Selekler-G€ oks¸ en &
Uysal-Tez€ olmez, 2007). Strong supportive personal relationships
between senior managers lead to partners' understanding and
commitment to the IJV (Glaister et al., 2003a). The appointment of
key personnel by MNCs may act as boundary spanners for trans-
ferring MNCs' culture and policies to align their objectives with IJVs
and other partners (Selekler-G€ oks¸ en & Uysal-Tez€ olmez, 2007).
Underlying the use of expatriate staf?ng as a culture control
mechanism is evidence that this leads to a better transmission of
common values and goals between foreign and local partners
(Chalos & O'Connor, 2004). Through its Hamburger University,
McDonald's delivers its business skill and promotes its corporate
philosophy (Wathne & Heide, 2000). Socialization in a partnership
can also be achieved through reciprocal business skill training,
mutual technical development, frequent managerial contacts, and
exercises involving relational behavior (Dong, Tse, & Cavusgil,
2008; Yilmaz, Sezen, & Ozdemir, 2005).
The aimof socialization is to encourage the alignment of partner
values and interests (Chalos & O'Connor, 2004), which reduces the
likelihood of opportunism and enhances IJV performance. The
existing literature provides cues and evidence to explain the posi-
tive link between socialization and shared values among partners.
First, a number of studies have found that socialization can help
partners create convergent goals and avoid opportunism (Dong
et al., 2008). Through socialization, personal familiarity, problem
solving, norms, identity, and cohesion can be created (Ahuja &
Galvin, 2003; S¸ engün & Wasti, 2009), and members can become
more committed to their organization and share views that
strongly in?uence their behaviors (Das & Teng, 2001). Second, the
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 36
personal and social relationships among the managers of cooper-
ative partners are crucial to resolving disputes and reducing goal
discrepancy (Glaister et al., 2003a; Wathne & Heide, 2000). Third,
cultural values and mindsets relating to speci?c issues can be
disseminated among partners through the repeated interactions of
their personnel over time and can thus become systematized and
shared (Baliga & Jaeger, 1984; Hult, Ketchen, & Slater, 2004).
Accordingly, we propose the following hypothesis.
H
7
: Through the mediation of the perceived value gap among IJV
partners, socialization has a positive association with IJV
performance.
3. Method
3.1. Data collection
The sampling frame of this study came from the 2005 Approved
Foreign Investment list published by the Investment Commission,
Ministry of Economic Affairs, Taiwan. To control cultural or
contextual factors, we selected only IJVs that were set up in China.
For each of these IJVs, one of the parent companies was a Taiwanese
MNC. Some previous studies suggest that if the ownership of any
parent is below 20% or over 80%, a partnership cannot be treated as
an IJV (Choi & Beamish, 2004). Based on this criterion, we elimi-
nated certain IJVs from the sample and were left with 692 IJVs. The
data collection comprised two stages. In the ?rst stage, each IJV was
contacted via fax and telephone. Owing to problems such as
incorrect telephone numbers and addresses or the IJV business
having closed, only 652 IJVs were contacted. Of these IJVs, 357
companies promised to participate in this study, representing a
participation rate of 54.8%. In the second stage, 357 questionnaires
were sent to these companies. Informants were asked to select an
important IJV as reference to answer our questionnaire. A total of
152 complete responses were received after several follow-ups,
representing a response rate of 42.6%.
Among the 152 participating IJVs, 12 are involved in chemistry,
12 in food, 11 in machinery, seven in textile and clothing, 31 in other
manufacturing, 59 are the electrical and electronics industry, six in
medical science and biotechnology, and six are involved in the
service industry. To test for nonresponse bias, a t test was adopted
to check the difference between early respondents and late re-
spondents, as suggested by Armstrong and Overton (1977). We
followed the recommendation of Weiss and Heide (1993): early
responses were de?ned as the ?rst 75% and the last 25% were
considered as late responses. In this test, we collected data fromthe
Taiwan Stock Market Observation Post System and compared the size
of investment (which is measured as the cumulative foreign direct
investment) and ownership structure (which is measured as equity
owned by the Taiwanese MNCs) between these two groups of re-
spondents. The results of t tests showed insigni?cant differences
between these two groups of ?rms with respect to either cumu-
lative foreign direct investment (F ¼1.292, p >0.05) or the owner-
ship structure of IJVs (F ¼0.902, p >0.05), mitigating the concern
for nonresponse bias (Armstrong & Overton, 1977).
We relied on the managers of the Taiwanese parent companies
to respond to the surveys. In accordance with the common
practice in IJV research, this study relied on data provided by one
partner. However, collecting data from a single source may result
in the problem of common method variance. Therefore, we con-
ducted a Harman's one-factor test as suggested by Podsakoff and
Organ (1986) to evaluate how serious this problem might be for
our study. The results of this test revealed that a six-factor
structure was the best solution, and the measurement items of
our six variables did not load heavily on one factor. It thus
appears that the problem of common method variance was
insigni?cant.
3.2. Measures
The research variables of this study include centralization,
formalization, socialization, perceived value gap, information
asymmetry, and IJV performance. The measures of these variables
were adopted from existing scales that had been used in previous
studies, or they were developed fromconcepts that are discussed in
the existing literature. All survey items were measured on a seven-
point scale, ranging from1 (strongly disagree) to 7 (strongly agree).
3.2.1. Dependent variable
IJV performance can be assessed using objective or subjective
measures. Because some studies have demonstrated that subjective
and objective measures of IJV performance are highly correlated
(Ari~ no, 2003; Geringer & Hebert, 1989; Mohr, 2006), several re-
searchers suggest that a subjective measure of joint venture per-
formance is appropriate (Gong et al., 2007; Pangarkar & Klein,
2004). This study adopted perceptual measures to assess the
satisfaction of the parent company in the following three areas: (1)
strategic goal achievement; (2) cooperative relationship with IJV
partners; and (3) willingness to continue this cooperative rela-
tionship. These measures have been proven to be reliable when
evaluating IJV performance in previous studies (Choi & Beamish,
2004; Luo & Park, 2004; Yan & Gray, 2001). The a reliability in
the present study was 0.90.
3.2.2. Mediators
Information asymmetry. Information asymmetry describes a
situation in which one party possesses more information than the
other party regarding the object of an exchange (Mishra et al.,
1998). In relationships characterized by information asymmetry,
it is possible for a party to act opportunistically without being
detected (Wathne & Heide, 2000). To assess information asym-
metry, we measured the level of the parent company's under-
standing of its IJV partner's actions. Respondents were asked to
evaluate how well they knew their partners in terms of actions,
strategic change, and the degree of mutual understanding. The
higher the level of the respondents' knowledge about their part-
ners, the lower the level of information asymmetry would be. The a
reliability for this scale was 0.87.
Perceived value gap. Seetoo and Hsiung (2004) de?ned the
perceived value gap as the perceived differences among partners in
goals, strategic directions, and organizational values. Following this
perspective, this study asked respondents to evaluate how they
differed from their partners in strategic vision, goals, and expec-
tations of howgoals can be achieved (Newburry et al., 2003). The a
reliability for this scale was 0.74.
3.2.3. Independent variables
Centralization. To measure the degree of centralization, this
study ?rst assessed the degree of autonomy the IJV had in decision-
making. A low degree of decision-making autonomy for IJV sub-
sidiaries would imply a high degree of centralization by the parent
companies (Geringer & Hebert, 1989; Glaister et al., 2003a;
Newburry et al., 2003). This study asked respondents to indicate
the degree of decision-making autonomy held by the IJV in relation
to ?ve operational (functional) activities, including research and
development strategy, ?nance strategy, marketing strategy,
manufacturing strategy, and human resource management strategy
(Choi & Beamish, 2004; Luo et al., 2001; Yu, Wong, & Chiao, 2006).
The a reliability for this scale was 0.94.
M.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 32e43 37
Formalization. Formalization refers to the extent to which the
parent company adopts written regulations, objectives, processes,
rules, and duties to de?ne the expected behavior and the standards
of output or quality (Poppo & Zenger, 2002; Wang & Fulop, 2007).
Respondents in this study were asked to evaluate the degree of
formalization in quality control procedures, cost control pro-
cedures, and operation performance reporting (Baliga & Jaeger,
1984; Jaussaud & Schaaper, 2006). The a reliability for this scale
was 0.80.
Socialization. In the present study, socialization refers to the
extent to which the parent company creates shared norms and
values and controls the behavior and decision making of its IJV
through social interactions (Chalos & O'Connor, 2004; Nohria &
Ghoshal, 1994). To measure socialization, three items were used:
(1) the frequency of the local partner's employees visiting the MNC
headquarters; (2) the frequency of group communication among
partners (Chalos & O'Connor, 2004; Glaister et al., 2003a); and (3)
the frequency of the local partner's employees being invited to join
training activities hosted by the MNC headquarters (Glaister et al.,
2003a; Jaussaud & Schaaper, 2006). The a reliability of this scale
was 0.80.
3.2.4. Control variables
Several control variables were included in this study. While we
provided no formal hypotheses for these factors, we controlled for
their impacts in our model. These control variables were industry,
IJV age, size, and ownership. Two dummy variables were used to
code the industry. The industry effect of the high technology sector,
which included the electrical and electronics industry, medical
science and the biotechnology sector, was coded 1; for other sec-
tors, it was coded 0. The industry effect of the service sector was
coded 1; for other sectors, it was coded 0. IJV age was computed by
taking the difference (in years) between the year 2011, when the
data were collected, and the year when the IJV's establishment was
initially reported to the Taiwanese government. IJV size was
measured by capital and by looking at the records of IJVs' average
annual sales over the past 3 years. Ownership was measured as the
percentage of IJV equity controlled by an MNC.
3.3. Reliability and validity
Although our scales were adopted or developed from previous
studies on related issues, we interviewed four CEOs of IJVs and
asked for their feedback to con?rm the validity of the measures.
The Cronbach a of the six scales all exceeded 0.70, indicating good
reliabilities. To evaluate the construct validity of the measures, we
conducted con?rmatory factor analysis using structural equation
modeling with the maximum likelihood method. The con?rmatory
factor analysis results demonstrated that our six-factor measure-
ment model had a satisfactory ?t (c
2
¼229.42, df ¼155, c
2
/
df ¼1.48, goodness of ?t index ¼0.87, normed ?t index ¼0.95,
comparative ?t index ¼0.98, root mean square error of approx-
imation ¼0.06). Although the goodness of ?t index value was
slightly below the recommended threshold of 0.90, the values of
both normed ?t index and comparative ?t index exceeded 0.90.
These ?ndings indicate that the measurement model has a good
model-to-data ?t. Furthermore, the ratio of chi-square to the de-
gree of freedomwas 1.48 (any value below 2 for this ratio indicates
a good ?t). Table 1 presents the values of composite reliability, the
squared multiple correlation, and the average variance extracted
(AVE). The composite reliability values of our study variables were
all above 0.70, the squared multiple correlation values were all
above 0.40, and the AVE values were all above 0.50. These results
indicate that our measures had good internal consistency and
convergent validity (Bagozzi and Yi, 1988).
To assess discriminant validity, Fornell and Larcker (1981) sug-
gested that the square root of AVE should be greater than the cor-
relation coef?cient in the corresponding columns and rows. Table 2
shows the means, standard deviations, and correlation coef?cients
of the study variables. This table reveals that the square root of each
Table 1
The values of squared multiple correlation (SMC), composite reliability (CR), and average variance extracted (AVE) for main research variables.
Variable Mean (SD) Estimate (SE) t value SMC CR AVE
Centralization
Marketing strategy of subsidiary 3.64 (1.78) 0.91 (0.06) 14.30
**
0.82 0.94 0.76
Capital investment of subsidiary 3.67 (1.53) 0.79 (0.07) 11.54
**
0.63
Manufacturing strategy of subsidiary 3.48 (1.75) 0.92 (0.06) 14.80
**
0.85
Human resource strategy of subsidiary 3.45 (1.76) 0.87 (0.07) 13.32
**
0.75
Research and development strategy of subsidiary 3.53 (1.77) 0.87 (0.07) 13.30
**
0.75
Formalization
Formal quality control procedure 5.75 (1.00) 0.69 (0.08) 8.90
**
0.58 0.81 0.59
Formal cost control procedure 5.86 (0.89) 0.81 (0.08) 10.70
**
0.65
Formal operation performance report 6.07 (0.83) 0.81 (0.08) 10.69
**
0.65
Socialization
The frequency of the local partner's employees visiting the MNC headquarters 5.10 (1.47) 0.77 (0.07) 10.55
**
0.60 0.82 0.60
The frequency of group communication among partners 5.37 (1.12) 0.79 (0.07) 10.93
**
0.63
The frequency of the local partner's employees being invited to join training
activities hosted by the MNC headquarters
4.75 (1.56) 0.77 (0.07) 10.50
**
0.59
Perceived Value Gap
Same vision among partners 5.28 (1.01) 0.72 (0.08) 9.41
**
0.51 0.75 0.51
Same strategic goals among partners 5.13 (1.13) 0.75 (0.08) 9.97
**
0.56
Parent company can achieve its strategic goals through this IJV 5.39 (0.87) 0.66 (0.08) 8.38
**
0.53
Information Asymmetry
There exist suf?cient information about partner's actions 5.33 (1.01) 0.85 (0.07) 12.69
**
0.72 0.86 0.68
There exist some channels to understand partner's strategic change 5.16 (1.10) 0.69 (0.07) 9.48
**
0.48
There are numerous effort to foster mutual understanding 5.23 (1.05) 0.92 (0.06) 14.41
**
0.85
IJV performance
Strategic goal achievement 5.18 (1.08) 0.89 (0.07) 13.75
**
0.80 0.90 0.75
Cooperative relationship with IJV's partners 5.36 (0.97) 0.91 (0.06) 14.07
**
0.82
Willingness to continue this cooperative relationship 5.51 (1.00) 0.80 (0.07) 11.62
**
0.64
n ¼152;
*
p