Vous êtes sur la page 1sur 14

International Business Review 19 (2010) 575588

Contents lists available at ScienceDirect

International Business Review


journal homepage: www.elsevier.com/locate/ibusrev

External uncertainty and entry mode choice: Cultural distance, political


risk and language diversity
Cristina Lopez-Duarte 1,*, Marta M. Vidal-Suarez 1
Universidad de Oviedo, Dpto. de Administracion de Empresas., Escuela Jovellanos Gijon, Avda. Luis Moya Blanco, n8 261, 33203 Gijon, Asturias, Spain

A R T I C L E I N F O A B S T R A C T

Article history: This paper aims at analyzing the effect of external uncertainty on the entry mode choice
Received 27 July 2009 when investing abroad. We consider the effect of uncertainty coming from the formal host
Received in revised form 24 March 2010 countrys environment (political risk) and from the informal one (cultural distance). The
Accepted 25 March 2010
potential existence of an interaction effect between both of them is also analyzed. In
particular, we analyze how language diversity between the home and host countries may
Keywords:
condition the inuence of this interaction effect on the entry mode choice. In order to
Cultural distance
empirically test our predictions, a database of foreign direct investments made by Spanish
Entry mode
Foreign direct investment rms is used.
Joint venture 2010 Elsevier Ltd. All rights reserved.
Language
Language barriers
Language diversity
Political risk
Wholly owned subsidiary

1. Introduction

The analysis of the entry mode choice related to foreign direct investments (FDIs) constitutes a classic topic of study in
International Business. This choice implies deciding on the degree of commitment that the investing rm wants to assume in
the host market; in this sense, the investing company should decide if it will invest alone maintaining 100% of the equity of
the rm located in the host market that is, investing through a wholly owned subsidiary (WOS) or if it will share the
equity with (at least) another rm investing through a joint venture (JV).
The entry mode choice has been broadly analyzed in the literature from different theoretical perspectives, being the
Transaction Cost Theory (TCT) among the most commonly used frameworks. From TCT perspective the external uncertainty
affecting the internationalization process arises as one of the main factors conditioning the entry mode choice. In such a
framework, the cultural distance (CD) between the home and the host countries of the FDI (informal environment) and the
host countrys political risk (PR) (formal environment) become the two main sources of such external uncertainty. Both
variables have been widely analyzed in the literature on entry mode choice, although almost always in an isolated way,
ignoring, therefore, the existence of potential interaction effects between both of them. Said in other words, the moderating/
intensifying effect of one of these variables on the role played by the other one has been ignored (Brouthers & Brouthers,


Financial support from Spains Ministerio de Ciencia y Tecnologa (Project: SEC 2009-07786) is gratefully acknowledged. An earlier version of this paper
has been published by Fundacion de las Cajas de Ahorros (FUNCAS) working paper series #476.
* Corresponding author. Tel.: +34985182179; fax: +34985182161.
E-mail addresses: clduarte@uniovi.es (C. Lopez-Duarte), mmvidal@uniovi.es (M.M. Vidal-Suarez).
1
Both authors contributed equally to this work.

0969-5931/$ see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ibusrev.2010.03.007
576 C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588

2001). One of the purposes of this paper is to contribute empirical evidence on the potential existence of such an interaction
effect.
In addition, external uncertainty may be conditioned by a third factor which has traditionally been bundled within the
cultural distance concept, thus, ignoring its individual inuence on the choice of mode of entry: the language diversity (LD)
between the home and host countries. Linguistic distance between the home and host countries is an important component
of psychic distance which is likely to inuence transaction costs (Demirbag, Tatoglu, & Glaister, 2007), and which is not
always captured by cultural distance measures (West & Graham, 2004). The literature analyzing the role of language on
internationalization patterns2 points out to this factor as one of the main sources of conict in international business
administration. In spite of these facts, language has been an almost overlooked factor in entry mode literature; and the very
scarce existing literature is far from conclusive about the role played by LD on the entry mode choice. On the one side, some
authors argue that language diversity increases the risk perceived by foreign investors and, thus, their tendency to invest
through JVs rather than through wholly owned subsidiaries see, for instance, Demirbag et al. (2007). On the contrary, other
researchers have found that the higher the linguistic barrier, the greater the control degree of parent companies over their
subsidiaries (Harzing & Feely, 2008).3 We seek to analyze how language diversity conditions the role played by the above
mentioned interaction effect between cultural distance and political risk on the choice of entry mode.
For empirical testing of our hypotheses we used a database collecting 334 FDIs carried out between 1989 and 2003 by
listed Spanish companies whose shares are traded on the Madrid Stock Exchange. These FDIs are located in 33 different
countries, endowing the sample with a high degree of diversity both in cultural distance regarding Spain, and in host
countries political risk. Such a diversity, together with the fact that the home country is Spain, constitutes an additional
advantage of this paper, since most empirical evidence on entry mode choice tends to show a US bias (being the US the home
or host country of investments), as already pointed out by Brouthers and Brouthers (2001).
The paper has been organized in the following way: in Section 2, we present a literature review on entry mode choice
based on TCT. This review is focused on the analysis of the relationship between external uncertainty and entry mode choice.
Next, we revise the traditionally established predictions related to external uncertainty by considering the interaction effect
between the cultural distance and the political risk; as well as the language diversity between the home and host countries.
Our predictions are tested with a sample of FDIs carried out by Spanish rms whose features are presented in Section 3;
Section 4 shows empirical evidence and a discussion of our results. Finally, we summarize the main conclusions and
managerial implications.

2. Literature review

Basing on the framework provided by TCT, the external uncertainty associated to the FDI constitutes a main factor
conditioning the entry mode choice. Such an external uncertainty is related to the uncertainty perceived by the investing
company in the formal and informal institutional environment of the host country (Delios & Henisz, 2003b; Henisz & Delios,
2002; Slangen & van Tulder, 2009). In the literature about entry mode, the uncertainty associated to the informal
environment of the host country has been traditionally conceptualized and measured through the cultural distance, while
that related to the formal aspect of the environment has been linked with the country risk; in particular, the political risk
(Zhao, Luo, & Suh, 2004).
Following Hofstede (1980, 2001), culture is dened as collective mental programmes shared by a group of people; these
programmes being different from one group to another. The culture is, thus, what distinguishes one group from another.
National culture refers to such programmes when the identied group of people share the national environment.4 The
cultural distance between two nations reects the existent differences in certain values, norms and behaviour rules between
them (Shenkar, 2001). These differences increase the liability of foreignness or the difculties that the investing rm should
overcome when it seeks to develop its activities in a new country.
The country risk refers to the volatility of the political, economic, and social factors of the target country; while the political
risk is dened as the likelihood of an unfavourable change in the governmental regime of the country and/or in the policies
issued by such a regime (Henisz, 2000). Although such a difference between country and political risk exists, it becomes
difcult to isolate the three types of factors that congure the country risk, since political, social and economic causes of risk
tend to be highly correlated. The higher the host market volatility, the more difcult for the foreign investor to obtain,
interpret and organize the information to successfully carry out a FDI (Delios & Henisz, 2003b).
Although TCT points to the external uncertainty as a decisive factor conditioning the entry mode choice, it is not clear how
such uncertainty inuences the investing rms decision. On the contrary, TCT can accommodate opposite predictions
related to this issue (Brouthers & Brouthers, 2001; Harzing, 2003; Shenkar, 2001).5 On the one side, the investing rm may
prefer to invest through a joint venture in order to gain access to local knowledge and contacts. Since these are specic assets

2
See Demirbag et al. (2007) for a review.
3
The paper by Arora and Fosfuri (2000), relative to the choice between WOSs and licensing agreements in the chemical industry, empirically proves the
relevance of language when choosing entry mode.
4
In the strictest sense, the culture concept is easier to be applied to societies than to nations. Historically, however, many groups of people sharing
national contexts have developed collective programmes, which allow a national culture approach.
5
These papers highlight such double hypothesis based on TCT, but all of them focus on the analysis of external uncertainty derived from cultural distance.
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588 577

difcult to be obtained through market contracting, the foreign rm may decide to share the investment project with a local
partner which provides the access to such specic resources. Regarding the CD, and in terms of Gatignon and Anderson
(1988), the local partner in a JV is the bridge that allows the foreign investor to save the cultural gap between both nations.
The foreign company may rely on the local partner some control and coordination tasks taking advantage of its familiarity
with the host countrys culture as, for instance, the local employees management or the development of distribution
networks. Regarding the political risk, by means of investing through a JV, the foreign investor limits its commitment of
resources on the host country and shares risk with (at least) a second partner. Additionally, joint ventures are more exible as
entry modes than WOSs, since they can be more easily dissolved: its partial commitment in the FDI facilitates the foreign
investor disinvestment when unfavourable changes take place on the environment. In summary, when the formal and
informal external environment is highly uncertain, foreign investors will prefer equity joint ventures over wholly owned
subsidiaries.
On the other side, greater external uncertainty leads to higher transaction costs when cooperating with a local partner, so
foreign investors may prefer to invest through a WOS in order to avoid such costs. Following Erramilli and Rao (1993) and
Hennart (1988), costs derived from negotiating a JV contract increase as cultural distance does: differences related to values,
norms, and behaviour rules make it more difcult the search of potential partners, as well as the negotiation with them. Later
enforcement of contracts also becomes trickier. Investing through a WOS makes it unnecessary to cooperate with a partner
whose decision and behaviour rules are not well-known and/or understood by the foreign investor. External uncertainty
coming from the formal host country environment that is, political risk may lead to the same decision: volatility in the
host countrys formal environment may make it impossible for the foreign investor to anticipate all contingencies (Agarwal,
1994). Additionally, a high degree of volatility in the host country may hinder the foreign investors ability to enforce
cooperative agreements (Brouthers & Brouthers, 2001). In synthesis, an increase in the host country risk can alter the subtle
balance of contributions, roles and benets of local and foreign partners (Meschi & Riccio, 2008).
In spite of the existence of two confronting theoretical approaches, empirical evidence related to the effect of country risk
on entry mode choice is quite conclusive: the meta-analysis carried out by Zhao et al. (2004) shows the scarce probability of
investing through a WOS in contexts of high country risk.6 On the contrary, empirical evidence related to the effect of cultural
distance on entry mode choice shows clearly contradictory results see exhaustive reviews of the literature by Brouthers
and Brouthers (2001), Harzing (2003) and Shenkar (2001) giving place to what is known in literature as the CD paradox
(Brouthers & Brouthers, 2001; p. 177) or the myopia regarding the CD (Harzing, 2003, p. 75).The existence of confronting
empirical evidence persists when reviewing the most recent literature.7
Different factors may underlie this CD paradox, among them, the fact that most papers aimed at analyzing the effect of
external uncertainty on entry mode choice has traditionally analyzed uncertainty coming from the formal environment
(political risk) and/or from the informal one (cultural distance) in an isolated way.8 Therefore, the potential moderating/
intensifying effect of one of these variables on the role played by the other one has been ignored only the papers by Agarwal
(1994) and Brouthers and Brouthers (2001) have explored the existence of an interaction effect between both variables.
Although we think it is crucial to analyze the role of such an interaction effect, the joint analysis of both variables does not
solve by itself the problem derived from the existence of two opposite proposals. On the one side, it can be expected that when
both the formal an informal sources of external uncertainty increase, rms prefer investing through entry modes which
guarantee a high level of control over their internationalization project. On the other, rms facing such a context of high
external uncertainty may show a clear preference for exibility, limited resource commitment, shared risk and participation
of a local partner who is familiar with the local cultural and political environment.
A third variable exists, however, which may help to solve this paradox: the language diversity between the home and host
countries of the FDI. The interaction between both sources of external uncertainty must be analyzed under the light provided
by this variable, which has been traditionally forgotten in the literature on the entry mode choice. As we will see in the
following section, the local partner can play a role reducing external uncertainty only when there is no language diversity
between the home and host countries of the FDI. When language diversity exists, the local partner does not contribute to
reducing external uncertainty; just on the contrary, he enhances problems derived from such uncertainty.

6
See, for instance, Agarwal and Ramaswami (1992), Arora and Fosfuri (2000), Akhter and Lusch (1988), Aulakh and Kotabe (1997), Delios and Beamish
(1999), Delios and Henisz (2000), Erramilli and Rao (1993), Henisz (2000), Hill et al. (1990), Kim & Hwang (1992), and Shrader et al. (2000). A few studies
exist, nevertheless, which show a non-decisive inuence of host country risk on entry mode choiceNakos and Brouthers (2002), Pinho (2007)and/or
point out to the existence of a moderating effect of third variables in the relationship between country risk and entry modeBrouthers and Brouthers
(2001), Slangen and van Tulder (2009), Tsang (2005).
7
The studies by Chen and Hu (2002), Fisher and Ranasinghe (2001), Kim and Gray (2009) and Tsang (2005) show the foreign investors preference for
WOSs when investing in high cultural distant host countries. Those by Arora and Fosfuri (2000), Chang and Rosenzweig (2001), Pak and Park (2004), Quer
et al. (2007), Tatoglu et al. (2003), Tsang (2005) and Yiu and Makino (2002) point out towards a preference for cooperative ventures in high cultural distant
contexts. Finally, those by Demirbag et al. (2009), Luo (2001), Rajan and Pangarkar (2000) and Ruz-Moreno, Mas-Ruz, and Nicolau-Gonzalez (2007)
conclude that the CD between both countries does not signicantly affect the entry mode choice, while others show different results depending on the type
of cooperative ventures (Lopez-Duarte & Garca-Canal, 2002, 2004).
8
The papers by Arora and Fosfuri (2000), Delios and Henisz (2003a), Demirbag et al. (2009), Quer et al. (2007) and Yiu and Makino (2002) have analyzed
the role of both cultural distance and political risk on the choice of entry mode, although not considering the potential interaction effect between them.
Empirical results are quite different, probably due to the fact that these studies have used quite heterogeneous samples regarding periods of time, industries
and home and host countries.
578 C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588

2.1. Language: the forgotten factor

Due to the broad acceptance of cultural distance as a main factor conditioning the entry mode choice, other factors related
to the diversity among countries have been systematically obviated in the literature. Following Harzing (2003), it becomes
necessary to enrich the cultural distance concept with other distance concepts and to recover the psychic distance (PD)
concept, which was rstly introduced in entry mode literature by the Uppsala School (Johanson & Vahlne, 1977; Johanson &
Wiedersheim-Paul, 1975). This concept includes traditional cultural dimensions, as well as differences in new dimensions,
among others, those related to the languages spoken in the home and host countries. As Demirbag et al. (2007) point out,
linguistic diversity between the home and host countries is an important component of psychic distance which is likely to
inuence transaction costs.
Language barriers between the home and host countries are part of the liability of foreignness (Luo & Shenkar, 2006);
therefore, they may signicantly inuence internationalization decisions and entry mode choice. Language has been,
however, a neglected factor in international business literature, being known as the forgotten factor (Harzing & Feely,
2008; Marschan-Piekkari, Welch, & Welch, 1997). In addition, as West and Graham (2004) point out, LD is not always
captured by cultural distance measures.
Different reasons may underlie for such a lack of attention: rstly, language has been bundled into the broader term
culture, so that researchers have implicitly analyzed its inuence on rms decisions when analyzing the role of cultural
distance between countries and, thereby, it has been thought unnecessary to focus on its independent role (Luo & Shenkar,
2006; Welch, Welch, & Marschan-Piekkari, 2005). Pre-eminence of American researchers, together with the dominance of
the English language in international business, may be a second factor underlying the systematic omission of this variable in
the studies on the entry mode choice (Fredriksson, Barner-Rasmussen, & Piekkari, 2006; Harzing & Feely, 2008). Finally, it
becomes particularly difcult to disentangle language effects from broader cultural inuences, as far as language inuences
cultural and managerial values in international contexts (West & Graham, 2004): it is not only that the language a person
learns as a child may inuence his/her values and way of thinking, but also that language diversity may inuence or
condition the perceived cultural distance between two countries (Harzing, 2005; Harzing & Maznevski, 2002). Such an
inuence would be developed through cultural accommodation or ethnic reinforcement processes carried out by
individuals who develop professional activities in countries where the spoken language is not their mother language.
Through a cultural accommodation process, individuals working in a second (not mother) language acquire some of the
cultural attitudes and values associated with that language, as far as they are inuenced (consciously or subconsciously) by
the culture of that language. On the contrary, through the ethnic reinforcement process these individuals show a stronger
endorsement of their natural cultural values; that is, the use of a second language makes their ethnicity more salient.
As mentioned above, when the FDIs context is characterized by high external uncertainty, the local partner may play two
different roles: he may contribute to reduce such uncertainty due to his familiarity with the local cultural and political
environment, or he may strengthen problems derived from both working in a unfamiliar environment and cooperating with
a partner whose values and behaviour rules are not correctly understood.
The particular role played by the local partner critically depends on his relationship with the foreign investor. Such a
relationship is, in turn, language dependent. In fact, inter-partner communication is considered an essential factor in order to
achieve a successful alliance development and to prevent JVs failure see Tallman and Shenkar (1994) and Luo (2001) for an
exhaustive review building trust and collaboration between partners requires personal and structural attachments which
depend on effective communication. This communication is, in turn, hardly conditioned by existing language barriers
between partners although the language barrier is wider than the language diversity dimension, the latter is a key factor
(Harzing & Feely, 2008); we will, thus, focus on this dimension.
Both formal and informal communication are most likely to occur in instances where the people involved share a
common language. Language diversity between both partners may not only difcult successful communication between
them, but also emerge as a critical source of conict. As a starting point, the functional language of the venture must be
selected; that is, the language formally designed for verbal and written use by both partners in order to be able to work
together (Luo & Shenkar, 2006). Naturally, the design of a functional language depends on the bargaining power of each party
and entails a negotiation process. This functional language becomes a control mechanism (Luo & Shenkar, 2006; Root, 1994):
if a venture uses party As native language as the functional language, party A will be in a stronger position than party B in
controlling information and inuencing decisions. Such a situation empowers party A to control management and
administrative processes. As Harzing and Feely (2008) point out, joint venture partners who accommodate to the other
partners language feel that they have relinquished some of the control over the relationship: the partner whose mother
tongue is the ventures functional language might start to dominate the relationship power distortion. In summary,
selection of the functional language may introduce a superiorsubordinate relationship in the venture which does not
necessarily reect the actual commitment, responsibility or decision power assigned to each partner according to contract
and/or equity share (Luo & Shenkar, 2006).
As a second point, language barriers between JV partners belonging to different language groups disturb international
ows of information as Kogut and Zander (1992) point out, for cross-border information and knowledge transfer
(especially tacit knowledge) to be effective, communication integrity is vital. Even if those partners are relatively competent
in the language of the other party, loss of theoretical skill is always present (Harzing & Feely, 2008). Misunderstandings are
therefore easily caused, due to, among others, ltration messages are only partially transmitted and distortion processes
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588 579

intended meaning is altered during the transmission of the message (Marschan-Piekkari et al., 1997). In summary,
language barriers between JV partners may not only prevent or disturb the ow of information between them, but also derive
in a loss of credibility and trust between them (Harzing & Feely, 2008; Luo, 2001). The bigger the language barrier between
the JVs partners, the higher the costs and difculties related to their cooperation.
Yet, two additional conicts may emerge when both partners do not share the same native language, both of them related
to the workforces management: on the one side, language can emerge as a source of power for some individuals and, on the
other, language diversity can derive in factions and the creation of groups within the venture.
In an international business context, language skills may deliver power and opportunities to some individuals who,
otherwise, would not enjoy such a position see, among others, Andersen and Rasmussen (2004), Harzing and Feely (2008),
Luo and Shenkar (2006), Marschan-Piekkari et al. (1997), Marschan-Piekkari, Welch, and Welch (1999a, 1999b), Neal (1998),
Welch et al. (2005). Language facility can give these individuals the power to act as informal gatekeepers, allowing them to
control the nature and ow of communication. In such a context, they may effectively inuence management processes,
disturbing the intended hierarchy and modifying the authority channels; in summary, introducing a new source of conict in
the JVs management.
In addition, language diversity between both partners is a strong candidate for the denition of group boundaries within
the venture: as Harzing and Feely (2008) point out, in international contexts, language emerges as a dominant factor dening
the group boundaries and composition. When partners do not share the same mother tongue, language-based clusters are
likely to emerge see, for instance, Harzing and Feely (2008), Fredriksson et al., 2006, Marschan-Piekkari et al. (1997, 1999a,
1999b), Neal (1998) , and tensions between different clusters are likely to arise: interpersonal relationships and
information transfers are easily achieved within each cluster, but not among different clusters. Such a context creates a
strong sense of exclusion (we and they), blocking the effective collaboration process of both partners and avoiding the
emergence of trust and shared values.
To resume, when different partners in an international JV do not share a common language, transaction costs related to
cooperation are seriously enhanced (Luo & Shenkar, 2006), and it becomes particularly difcult for the local partner to play
his role as a bridge between the foreign investor and the local context.
Basing on these ideas, we pose the following hypothesis:
When external uncertainty derived from the interaction between cultural distance and political risk is high, and there
is not language diversity between the FDIs home and host countries, investing rms will prefer to invest through JVs
rather than through WOSs.

3. Empirical analysis

3.1. Database, dependent variable and methodology

In order to carry out our analysis, we created a database collecting the FDIs made through joint ventures and wholly
owned subsidiaries carried out between 1989 and 2003 by listed Spanish companies whose shares are traded on the Madrid
Stock Exchange it has to be addressed that Spanish outward FDI ows were not totally liberalized until 1988; as a
consequence, almost no FDI processes took place prior to the beginning of the 1990s.9
FDIs made by Spanish rms have been identied by following this process: after preparing a list including all Spanish
rms listed in the Madrid Stock Exchange for each year within our study, an exhaustive and systematic research into PRENSA
BARATZ database and Expansion and Cinco Das archives was carried out for each identied company.10 As a double check, we
also completed and veried the gathered information by using the reports presented by each rm to the Stock Exchange
National Commission, which is responsible for supervising the Madrid Stock Exchange, as well as other corporate reports.
This research allowed us to identify 509 new FDIs; that is, FDIs that are the rst investment relative to a particular
internationalization project (thus, accumulations or further investments made by the same rm and related to the same
project have not been included in our sample). FDIs located in host countries which pose legal constrains to full ownership by
foreign investors and FDIs related to privatization processes were eliminated from this database, since in both cases the
foreign investor is not free to choose the entry mode. Additionally, FDIs located in host countries lacking cultural distance
measures and/or political risk rates (according to the measures referred to in the following section) were also eliminated. The
nal sample has been composed of 334 FDIs located in 34 different host countries and made up by 63 different rms some
of these rms have carried out multiple FDI projects. In Table 1a a breakdown of host countries can be found. The second part
of the 1990s accounts for a very high volume of the FDIS registered in the database. Latin America is the host region that
receives the highest volume of Spanish FDIs almost 40% of FDIs in the database , followed by the European Union (37%)
and other OECD countries (15%).11 Regarding industry breakdown, it has to be said that service industries (nance, regulated
services or other services) account for more than 70% of collected FDIs.

9
See Guillen (2005) for a detailed analysis of the accelerated international process experienced by Spanish rms since the beginning of the 1990s.
10
Expansion and Cinco Das are the leading newspapers on economy in Spain.
11
An exhaustive literature review relative to Spanish FDI industry and geographical distribution can be found in Duran (2006).
580 C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588

Table 1a
Breakdown of host countries.

Host country FDIs Investing rms Host country FDIs Investing rms Host country FDIs Investing rms

Argentina 47 14 Greece 1 1 Philipinnes 6 5


Australia 1 1 Guatemala 1 1 Poland 5 4
Austria 7 6 Hungary 3 3 Portugal 39 15
Brazil 21 13 Ireland 2 2 Russia 1 1
Canada 5 2 Israel 1 1 Singapore 1 1
Colombia 17 9 Italy 16 10 South Africa 2 2
Denmark 3 2 Japan 2 2 Switzerland 1 1
Ecuador 2 2 Korea 1 1 Thailand 2 2
Finland 1 1 Mexico 35 17 Ukraine 1 1
France 17 12 Morocco 7 7 U. Kingdom 15 13
Germany 20 16 Netherlands 7 4 USA 35 18
Venezuela 9 6

Graph 1. Time distribution of FDIs collected in each subsample.

In order to classify these FDIs in JVs and WOSs, we have followed the approach by Brouthers and Hennart (2007) and
Slangen and van Tulder (2009) followed by most of the empirical literature on entry mode choice the distinctive
characteristic of a JV is the method chosen to remunerate input providers; that is, each partner is paid ex post from the prots
of the venture. Once again following previous research, a rms ownership share over 95% of equity is dened as full
ownership and a share in the 1095% as shared ownership ownership shares under 10% has not been included in the
database, as they can be understood as portfolio rather than direct investments.
In order to test the inuence of the language diversity on the role played by the interaction effect between CD and PR, this
original sample was split into two different subsamples according to the language spoken in the host country. The rst
subsample collects 132 FDIs located in host countries where Spanish is an ofcial language (although not necessarily the only
ofcial language), while the second subsample collects 202 FDIs located in host countries where Spanish is not an ofcial
language. 100% of host countries in the former subsample are Latin American Countries. Graph 1 shows the time distribution
of FDIs collected in each subsample, while Table 1b shows a breakdown of entry modes depending upon the language spoken
in the host country.
Several binomial regression models have been estimated. The dependent variable, mode of entry, is a dichotomous one
which values 1 when the FDI has been carried out through a WOS and 0 otherwise. In these models the likelihood that the
investment is made through a wholly owned subsidiary is explained by the independent variables dened below. In these
estimations, the coefcients obtained for every independent variable evaluate the effect of the increments of these variables
on the likelihood of the dependent variable equalling 1.

3.2. Independent and control variables

Firstly, to analyze the effect of aggregated cultural distance, we have used the Kogut & Singh Index based on Hofstedes
cultural dimensions (HOF IND): power distance, uncertainty avoidance, individualism and masculinity.12 These are the measure
and tool, respectively, most frequently used in the literature in order to measure the extent to which different cultures are
similar or different.13 Such a tool is liable to some criticism, among them, the fact that identical values of CD can be reached
for pairs of countries whose individual cultural dimensions are different, as well as the assumption of equivalence (Shenkar,
2001). Regarding the latter, Hofstede (2001) points out that cultural gaps in different dimensions are not equally important

12
The fth dimension identied by Hofstede and Bond (1988) the long term orientation has not been used in this paper as it is available for only 23
countries.
13
See both, literature reviews by Harzing (2003) and Shenkar (2001), as well as the review of most recent literature carried out in section 2 of this paper.
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588 581

Table 1b
Breakdown of entry modes in each subsample.

Non-Spanish-speaking subsample Spanish-speaking subsample

JVs 62.87% 86.36%


WOSs 37.13% 13.64%

in conditioning rms decisions and entry mode choice nevertheless, very few studies have analyzed the effect of
distance relative to each cultural dimension on entry mode choice, among them those by Barkema and Vermeulen (1998)
and Brouthers and Brouthers (2001). In order to analyze the effect of individual cultural dimension on entry mode
decisions, we have taken the absolute value of the difference between the home and host countries values for each
cultural attribute. These variables are labelled as PD (power distance), UA (uncertainty avoidance), IND (individualism) and
MAS (masculinity).
The Euromoney Risk Ratings over the period of study have been used to measure the political risk affecting the host
country a variable labelled as PR. Euromoney ratings are comprehensive country risk ratings which assemble accounts for
nancial, economic and political risk. These ratings use a scale from 0 (the highest possible country risk) to 100 (the lowest
possible country risk). In order to make it easier to analyze the interaction effects between CD and PR, we have transformed
this variable into 100 Euromoney Risk Rating. Thereby, the higher the value of the PR variables, the higher the host
countrys political risk.
Different interaction effects between each one of the CD variables and the PR variable were estimated as the product of
the two variables (CD  PR). With the aim of increasing interpretability of these interaction effects and avoiding
multicolinearity problems, CD and PR variables have been mean-centered. Each variable has been centered by subtracting
the mean score of the variable from each data-point. The mean value of each new centered variable is, thus, 0.
Several control variables have been included in the analysis in order to control the effect that certain factors related to
investing rms, industries and host countries have on the entry mode choice.
Firstly, different variables related to the investing rms characteristics have been included; in particular, their size and
experience. The investing rms size has been measured through their market capitalization calculated on the 31st of
December of the year immediately before the FDI was made. A logarithmic transformation of this variable was done (SIZE).
Regarding the investing rms experience, two variables have been introduced: its international or multinational experience
and its experience related to a particular host country. In both cases it has been measured as the number of months between
the rst FDI rst FDI in absolute terms or rst FDI is the host country, respectively, carried out by the investing company
and the FDI collected in the database INT EX and HC EXP, respectively. We have used the number of months, instead of the
number of years (most frequently used in the literature) in order to more precisely measure the Spanish rms experience;
bearing in mind the very high time concentration degree of Spanish outward FDI ows.
As our sample shows a high bias towards service rms, we have included in the model a dichotomous variable (MANUF)
which values one when the investing rm competes in a manufacturing industry and 0 when it competes in a service
industry.
The ECDEV 1 and ECDEV 2 variables seek to approach differences in economic development between Spain and the host
countries collected in the database. When doing so we have used data from the World Development Report by The World
Bank. In this report, the World Bank clusters different economies in four different groups attending each countrys degree of
economic development high, middle-high, middle-low, and low economic development. As only two FDIs collected in our
database were in the fourth category, we gathered together the third and fourth clusters. Taking into account that Spain
remains in the rst cluster (high economic development) throughout the 15 years in the study, the ECDEV 1 is a dichotomous
variable which values 1 when the host country is clustered in the third or fourth categories (lower economic development)
and 0 otherwise, while the ECDEV 2 is a dichotomous variable which values 1 when the host country is clustered in the
second category (middle-high economic development). However, in the Spanish-speaking subsample both variables were
gathered together, as there were too few FDIs in the third category and no FDIs in the fourth. Finally, by including the host
countrys GDP growth rate (GDP GR), we attempt to account for the growth of host countries markets. Values for this
variable have been obtained from the World Banks database.
Tables 2 and 3 show both correlations matrixes of the variables used in our empirical tests.

4. Results

Tables 4 and 5 report results from different logistic regression models. Table 4 reports results obtained using the Spanish-
speaking subsample, while Table 5 reports results relative to the non-Spanish-speaking subsample. Logistic regressions have
been estimated using thereto different specications: (1) control variables only, (2) main effects related to aggregate cultural
distance and political risk added and (3) interaction effect between CD (composed index) and PR added. Models (4)(7)
include the interaction effects between individual cultural dimensions and political risk.
The x2 of estimated models is statistically signicant in all cases, while different observations are satisfactorily classied
at percentages which range from 62 to 70% non-Spanish-speaking subsample and from 81 to 89% Spanish-speaking
582
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588
Table 2
Correlation matrix Spanish-speaking subsample.

WOSs PR HOF IND PD UA IND MAS SIZE INT EXP HC EXP MANUF ECDEV 1 GDP GR

WOSs 1 0.144 0.055 0.091 0.128 0.037 0.02 0.266(**) 0.169 0.123 0.340(**) 0.02 0.003
PR 1 0.028 0.089 0.387(**) 0.071 0.263(**) 0.228(**) 0.219(*) 0.099 0.042 0.286(**) 0.306(**)
HOF IND 1 0.785(**) 0.495(**) 0.904(**) 0.842(**) 0.071 0.191(*) 0.218(*) 0.044 0.011 0.05
PD 1 0.412(**) 0.524(**) 0.705(**) 0.057 0.119 0.099 0.046 0.221(*) 0.126
UA 1 0.646(**) 0.026 0.133 0.122 0.001 0.163 0.687(**) 0.062
IND 1 0.619(**) 0.098 0.203(*) 0.208(*) 0.082 0.402(**) 0.029
MAS 1 0 0.117 0.218(*) 0.016 0.404(**) 0.094
SIZE 1 0.768(**) 0.380(**) 0.559(**) 0.11 0.186(*)
INT EXP 1 0.467(**) 0.318(**) 0.065 0.027
HC EXP 1 0.186(*) 0.035 0.003
MANUF 1 0.145 0.133
ECDEV 1 1 0.153
GDP GR 1
Mean 0.14 0 0 0 0 0 0 9.61 88.95 19.88 0.14 0.31 3.22
Std Dev 0.344 6.882 0.506 7.316 4.297 12.772 7.817 0.891 48.677 38.249 0.352 0.465 3.414

*p < 0.05.
**p < 0.01.
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588
Table 3
Correlation Matrix non-Spanish-speaking subsample.

WOSs PR HOF IND PD UA IND MAS SIZE INT EXP HC EXP MANUF ECDEV 1 ECDEV 2 GDP GR

WOSs 1 0.055 0.184(**) 0.201(**) 0.122 0.087 0.182(**) 0.067 0.059 0.067 0.143(*) 0.072 0.051 0.038
PR 1 0.263(**) 0.026 0.194(**) 0.423(**) 0.132 0.166(*) 0.122 0.115 0.219(**) 0.802(**) 0.338(**) 0.154(*)
HOF IND 1 0.621(**) 0.814(**) 0.461(**) 0.593(**) 0.05 0.036 0.121 0.016 0.082 0.056 0.025
PD 1 0.414(**) 0.173(*) 0.377(**) 0.002 0.051 0.183(**) 0.068 0.181(*) 0.103 0.041
UA 1 0.594(**) 0.242(**) 0.055 0.089 0.059 0.004 0.04 .243(**) 0.008
IND 1 0.012 0.02 0.115 0.097 0.093 0.332(**) 0.158(*) 0.119
MAS 1 0.157(*) 0.05 0.09 0.016 0.133 0.249(**) 0.125
SIZE 1 0.527(**) 0.189(**) 0.509(**) 0.183(**) 0.004 0.192(**)
INT EXP 1 0.385(**) 0.115 0.059 0.021 0.05
HC EXP 1 0.106 0.101 0.091 0.05
MANUF 1 0.205(**) 0.104 0.043
ECDEV 1 1 0.065 0.179(*)
ECDEV 2 1 0.077
GDP GR 1
Mean 0.37 0 0 0 0 0 0 9.04 59.79 12.05 0.38 0.04 0.09 2.82
Std Dev 0.484 13.822 0.763 10.152 16.046 10.609 9.962 0.858 55.211 26.873 0.486 0.196 0.293 2.155

*p < 0.05.
**p < 0.01.

583
584 C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588

Table 4
Results of estimated binomial regression models Spanish-speaking subsample.

Model (1) Model (2) Model (3) Model (4) Model (5) Model (6) Model (7)

CONST 2.495 1.795 0.685 1.474 0.172 0.064 1.023


PR 0.079 (0.044)* 0.107 (0.050)** 0.092 (0.049)* 0.155 (0.067)** 0.130 (0.054)** 0.087 (0.046)*
HOF IND 0.288 (0.531) 0.917 (0.681)
HOF IND  PR 0.167 (0.089)*
PD 0.044 (0.044)
PD  PR 0.010 (0.007)
UA 0.390 (0.136)***
UA  PR 0.064 (0.018)***
IND 0.036 (0.028)
IND  PR 0.009 (0.004)**
MAS 0.029 (0.044)
MAS  PR 0.003 (0.005)
SIZE 0.529 (0.529) 0.518 (0.553) 0.276 (0.585) 0.190 (0.595) 0.405 (0.603) 0.354 (0.582) 0.443 (0.574)
INT EXP 0.003 (0.008) 0.007 (0.009) 0.007 (0.009) 0.006 (0.009) 0.016 (0.010) 0.009 (0.009) 0.007 (0.009)
HC EXP 0.008 (0.013) 0.010 (0.013) 0.012 (0.013) 0.012 (0.013) 0.006 (0.013) 0.011 (0.013) 0.010 (0.013)**
MANUF 1.427 (0.782)* 1.684 (0.817)** 1.836 (0.856)** 1.926 (0.877)** 2.327 (1.015)** 1.835 (0.856)** 1.724 (0.824)
ECDEV 1 0.146 (0.599) 0.358 (0.632) 0.503 (0.665) 0.131 (0.713) 1.563 (0.936)* 0.987 (0.723) 0.265 (0.739)
GDP GR 0.068 (0.089) 0.120 (0.096) 0.151 (0.095) 0.136 (0.095) 0.256 (0.122)** 0.190 (0.100)* 0.125 (0.096)
x2 **
14.19 17.67 **
21.61***
20.59** ***
38.84 23.72 ***
17.93**
% Satisf. 84.1 84.8 83.3 81.8 89.4 84.1 83.3
classied
*
p < 0.1.
**
p < 0.05.
***
p < 0.01.

Table 5
Results of estimated binomial regression models non-Spanish-speaking subsample.

Model (1) Model (2) Model (3) Model (4) Model (5) Model (6) Model (7)

CONST 1.504 2.141 0.682 0.877 1.139 1.164 0.526


PR 0.051 (0.028)** 0.049 (0.029)* 0.051 (0.027)* 0.033 (0.028) 0.004 (0.026) 0.043 (0.030)
HOF IND 0.723 (0.231)*** 0.710 (0.235)**
HOF IND  PR 0.006 (0.017)
PD 0.069 (0.019)***
DP  PR 0.001 (0.001)
UA 0.021 (0.011)*
UA  PR 0.000 (0.001)
IND 0.032 (0.017)*
IND  PR 0.000 (0.001)
MAS 0.065 (0.021)***
MAS  PR 0.001 (0.002)
Size 0.072 (0.254) 0.031 (0.263) 0.013 (0.269) 0.030 (0.267) 0.037 (0.264) 0.048 (0.260) 0.156 (0.277)
INT EXP 0.004 (0.004) 0.003 (0.004) 0.003 (0.004) 0.005 (0.004) 0.003 (0.004) 0.005 (0.004) 0.003 (0.004)
HC EXP 0.008 (0.006) 0.011 (0.006)* 0.011 (0.006)* 0.015 (0.006)** 0.009 (0.006) 0.010 (0.006)* 0.011 (0.006)*
MANUF 0.848 (0.375)** 0.812 (0.385)** 0.826 (0.388)** 0.735 (0.387)* 0.828 (0.382)** 0.822 (0.381)** 0.713 (0.387)*
ECDEV 1 0.922 (0.865) 2.169 (1.190)* 2.132 (1.196)* 2.093 (1.216)* 1.305 (1.157) 1.286 (1.131) 2.761 (1.603)*
ECDEV2 0.936 (0.585) 2.827 (1.262)** 2.759 (1.279)** 3.022 (1.306)** 2.280 (1.289)* 1.365 (1.167) 2.549 (1.258)**
GDP GR 0.085 (0.076) 0.103 (0.080) 0.096 (0.082) 0.091 (0.079) 0.096 (0.082) 0.083 (0.078) 0.146 (0.085)*
x2 11.31 22.14*** 22.24** 27.51*** 15.44 16.04* 24.99***
% Satisf. 62.4 69.8 69.3 70.3 65.3 67.3 68.8
classied
*
p < 0.1.
**
p < 0.05.
***
p < 0.01.

subsample. In order to test the robustness of the estimated models, they were replicated using a logarithmic transformation
of both experience variables. Results maintain robust.
Our results conrm the existence of a moderating/intensifying effect of political risk on the role played by CD on the entry
mode choice, although such an effect is only statistically signicant when there is no language diversity between the home
and host countries of the FDI14 Graph 2 shows a graphic representation of interaction effects in this subsample. Results
relative to model (2) in Tables 4 and 5 show that external uncertainty coming from the formal environment positively
inuences the likelihood of choosing WOSs as entry mode the PR variable shows a positive and statistically signicant
coefcient in both tables. The same seems to occur when measuring the external uncertainty coming from the informal

14
Results relative to the interaction effects between CD and PR measures keep robust when both subsamples are combined in just one, and language
diversity is measured through a dichotomous variable.
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588 585

Graph 2. Interaction effects Spanish-speaking subsample. Plots show the pattern of the interaction effect, but not accurate values of the dependent
variable, as far as all control variables included in the regression models have not been standardized.

environment (cultural distance), at least in the non-Spanish-speaking subsample: although the coefcient of the CD variable
shows a positive sign in both tables, it is statistically signicant only in Table 5.
However, results become different when the interaction effect between CD (HOF IND) and PR is taken into account: as can
be seen in model (3) Table 4, the interaction effect shows a negative and statistically signicant sign; on the contrary, this
interaction effect does not show statistical signicance in Table 5. Results relative to model (3) in both tables provide support
for our hypothesis: when political risk is high and CD increases, foreign rms prefer to invest through JVs rather than through WOS;
however, such a preference for JVs takes place only when language barriers between partners (derived, in turn, from language
diversity) do not exist. When the FDIs environment is characterized by both sources of external uncertainty, Spanish foreign
investors show a clear preference for entry modes which provide higher exibility, lower resource commitment and allow
sharing risk with a partner who enjoys familiarity with the host countrys formal and informal environment. However, this is
the case only when both partners share a common language which facilitates successful formal and informal communication
between them and avoids the costs and conicts derived from language diversity. Nevertheless, it has to be pointed out that
results in Table 5 do not provide conclusive empirical evidence relative to a preference for WOSs when the FDIs environment
is characterized by both sources of external uncertainty and language barriers exist.
On the whole, results maintain robust when including in the model interaction effects between individual cultural
dimensions and political risk; although some issues must be pointed out. As can be seen in models (4)(7) in Table 4, when
using Hofstedes individual cultural dimensions, all coefcients relative to the interaction effects are negative, and those
related to the Uncertainty Avoidance and Individualism dimensions are statistically signicant. On the contrary, none of these
586 C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588

interaction effects is statistically signicant when using the non-Spanish-speaking subsample (models (4)(7) in Table 5).
These results provide some support to the idea that such a preference for JVs in contexts featured by high CD and PR only
occurs when there is no language diversity between partners. However, interactions effects relative to Power Distance and
Masculinity do not show statistical signicance in Table 4. It seems, therefore, that cultural differences in different
dimensions are not equally important in conditioning Spanish rms entry mode choice and do not equally interact with
other sources of external uncertainty. Particularly surprising is the result relative to the Power Distance dimension: following
Hofstede (1980, 2001) this cultural dimension plays (together with Uncertainty Avoidance) the most relevant role on rms
decision making. As far as the interaction effect relative to this relevant dimension is not statistically signicant, results
cannot be said to be conclusive.
Regarding control variables, the MANUF variable shows a positive and statistically signicant sign in almost all estimated
models relative to both subsamples (Tables 4 and 5). This is, therefore, a robust result which points towards the fact that
Spanish rms competing in manufacturing industries prefer WOSs (that is, high control entry modes) rather than JVs when
investing abroad such a result has been already found in previous research relative to Spanish rms (Quer, Claver, & Molina,
2006). On the contrary, the rms size and its experience particularly its international experience relative to the
internationalization process itself do not show statistical signicance, as does the GDP growth variable (both Tables 4 and
5). It seems that, in the particular case of Spanish rms, and once the interaction effects between political risk and cultural distance
variables have been taken into account, these variables do not play a relevant role on the choice of entry mode, irrespective of
the existence of language diversity between the home and host countries. These results are in line with some recent research
on this topic: Brouthers (2002), Delios and Beamish (1999), and Lu (2002) do not nd empirical evidence to prove the
inuence of the investing rms international experience on entry mode choice. In the same way, the host markets growth or
the rms size do not always play a relevant role on such an election, as is shown in Brouthers (2002) and Meyer (2001),
respectively. On the contrary, the rms experience relative to each particular host country seems to play a statistically
signicant role, increasing the likehood of investing through WOSs, but only in the non-Spanish-speaking subsample
(Table 5); that is, this experience becomes particularly relevant when language barriers between both countries exist. The
host countrys economic development seems to play also a role in this subsample: results of ECDEV variables in Table 5 point
towards some preference for WOSs when investing in less developed countries, although they do not maintain robust across
different models and cannot be understood as conclusive.

5. Conclusions and managerial implications

This paper aims at analyzing the effect of external uncertainty on entry mode choice; that is, on the choice between
wholly owned subsidiaries and joint ventures. Using as a basis the insights of received TCT, we have differentiated external
uncertainty coming from the host countrys informal and formal environments cultural distance and political risk,
respectively. An exhaustive review of existing literature shows quite conclusive results regarding the role played by PR, but
conicting results regarding that of the CD. Such contradictory evidence has given place to the CD paradox (Brouthers &
Brouthers, 2001) or the myopia regarding the CD (Harzing, 2003). This paper aims at contributing to explain this paradox
by analyzing the potential interaction effect between both variables. In addition, the paper seeks to unbundle language from
the broader concept of cultural distance and to explore the role it plays on the relationship between external uncertainty and
entry mode choice.
In order to carry out our empirical test, we have used a database collecting FDIs made by means of wholly owned
subsidiaries and joint ventures carried out between 1989 and 2003 by listed Spanish companies whose shares were traded
on the Madrid Stock Exchange. This database was then split into two different subsamples depending on the language
spoken in the host country. Several logistic regression models have been estimated.
We believe our study to provide some useful insights into the research of the effect of external uncertainty on the choice
of entry mode. Firstly, our results hint at the existence of an interaction effect between the two uncertainty dimensions most
usually analyzed in the literature on entry mode: cultural distance and political risk. Such interaction effect must be
analyzed, however, under the light provided by a third variable which has been traditionally forgotten in this literature:
language diversity between the home and host countries of the FDI. Our results point out to a preference for JVs over WOSs
when the FDIs environment is characterized by both high PR and CD. Nevertheless, such a preference for JVs takes place only
when language barriers between partners, derived, in turn, from language diversity, do not exist. In such a context, Spanish
foreign investors show a clear preference for entry modes which provide higher exibility and lower resource commitment
and allow sharing risk with a second partner. It seems, therefore, that when language barriers between the investing rm and
the local partner exist, impeding effective communication between both partners, sharing the venture with a local partner is
not an effective way to reducing external uncertainty.
To sum up, our framework suggests that managers interested in expanding through foreign direct investments should
consider language diversity as a main factor inuencing the entry mode choice. Such diversity clearly conditions the
potential role played by the local partner in decreasing the external uncertainty coming from both the informal and formal
environment of the host country. When language diversity does not exist, managers should look for a local partner to share
the investment and reduce external uncertainty. The foreign investor may gain access to local knowledge and contacts and
take advantage of the local partners familiarity with the host countrys formal and informal environment. On the contrary,
when language barriers exist, cooperating with a local partner aggravates the problems and costs derived from investing in a
C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588 587

high uncertain environment. Effective communication is an essential factor in order to achieve real cooperation between
partners which derives in a balanced distribution of contributions, benets and roles.
However, these results cannot be considered to be conclusive. It is to be noticed that the particular features of our sample
may be inuencing the results obtained: the over-representation of service rms and FDIs located in OECD and Latin
American countries may be underlying such results. Firstly, the industry may signicantly condition the internationalization
patterns of rms. Secondly, up to 90% of the FDIs collected in the database are located in OECD/European Union countries
countries which show a very low degree of political risk and Latin America countries which show a very low degree of
cultural distance from Spain. These characteristics highlight certain issues that further research should address. Firstly,
future research should shed light on the applicability of our results to contexts other than FDIs carried out by Spanish rms in
the 19892003 period. Empirical studies dealing with a new period of time might analyze in depth the inuence of language
diversity and external uncertainty on the choice of mode of entry. The main forces driving Spanish rms internationalization
in the period of study addressed here have been Spains entrance into the European Union, as well as the deregulation of
different industries and privatization of state-owned companies, mainly in Latin America. As a consequence, Spanish FDI has
highly concentrated in these geographical areas, as well as in banking and regulated industries Additionally, during this
period of time, large Spanish rms have expanded abroad and become large multinationals (Guillen, 2005); while at the
beginning of the 1980s most of them were exclusively focused on the Spanish local market. The environment and challenges
faced by Spanish rms are now very different. It is therefore to be expected that Spanish outward FDI ows change, giving
rise to FDIs located in new geographical areas and made by Spanish rms competing in new industries.
Empirical studies dealing with investing rms coming from different countries, and with different measures of cultural
distance and political risk can also enrich the analysis.

References

Agarwal, S. (1994). Socio-cultural distance and the choice of joint ventures: A contingency perspective. Journal of International Marketing, 2(2), 6380.
Agarwal, S., & Ramaswami, S. N. (1992). Choice of foreign market entry mode: Impact of ownership, location and internalization factors. Journal of International
Business Studies, 23(1), 127.
Akhter, H., & Lusch, R. F. (1988). Political risks and the evolution of control of foreign subsidiaries: Equity, earnings, and marketing mix. Journal of Global Marketing,
1(3), 109128.
Andersen, H., & Rasmussen, E. S. (2004). The role of language skills in corporate communication, Corporate Communications. An International Journal, 9(3), 231242.
Arora, A., & Fosfuri, A. (2000). Wholly owned subsidiary versus technology licensing in the worldwide chemical industry. Journal of International Business Studies,
31(4), 555572.
Aulakh, P. S., & Kotabe, M. (1997). Antecedents and performance implications of channel integration in foreign markets. Journal of International Business Studies,
28(1), 145175.
Barkema, H. G., & Vermeulen, F. (1998). International expansion through start-up or acquisition: A learning perspective. Academy of Management Journal, 47(1), 7
26.
Brouthers, K. (2002). Institutional, cultural and transaction cost inuences on entry mode choice and performance. Journal of International Business Studies, 33(2),
203221.
Brouthers, K., & Brouthers, L. E. (2001). Explaining the national distance culture paradox. Journal of International Business Studies, 32(1), 177189.
Brouthers, K., & Hennart, J. F. (2007). Boundaries of the rm: Insights from international entry mode research. Journal of Management, 33(3), 395425.
Chang, S. J., & Rosenzweig, P. M. (2001). The choice of entry mode in sequential foreign direct investment. Strategic Management Journal, 22(8), 747776.
Chen, H., & Hu, M. Y. (2002). An analysis of determinants of entry mode and its impact on performance. International Business Review, 11(2), 193210.
Delios, A., & Beamish, P. W. (1999). Ownership strategy of Japanese rms: Transactional, institutional, and experience inuences. Strategic Management Journal,
20(10), 915933.
Delios, A., & Henisz, W. J. (2000). Japanese rms investment strategies in emerging economies. Academy of Management Journal, 43(3), 305323.
Delios, A., & Henisz, W. J. (2003a). Policy uncertainty and the sequence of entry by Japanese rms, 19801998. Journal of International Business Studies, 34(3), 227
241.
Delios, A., & Henisz, W. J. (2003b). Political hazards, experience and sequential entry strategies: The international expansion of Japanese rms 19801998.
Strategic Management Journal, 24(11), 11531164.
Demirbag, M., Tatoglu, E., & Glaister, K. (2007). Institutional and transaction cost inuences on MNEs ownership strategies of their afliates: Evidence form an
emerging market. Journal of World Business, 42(4), 418434.
Demirbag, M., Tatoglu, E., & Glaister, K. (2009). Equity based entry modes of emerging country multinationals: Lessons from Turkey. Journal of World Business,
44(4), 445462.
Duran, J. J. (2006). El auge de la empresa multinacional espanola. Boletn Economico de ICE, 2881, 1333.
Erramilli, M. K., & Rao, C. P. (1993). Service rms international entry mode choice: A modied transaction cost analysis approach. Journal of Marketing, 57, 1938.
Fisher, T., & Ranasinghe, M. (2001). Culture and foreign companies choice of entry mode: The case of the Singapore building and construction industry.
Construction Management and Economics, 19(4), 343353.
Fredriksson, R., Barner-Rasmussen, W., & Piekkari, R. (2006). The multinational corporation as a multilingual organization. Corporate Communication: An
International Journal, 11(4), 406423.
Gatignon, H., & Anderson, E. (1988). The multinational corporation degree of control over subsidiaries: An empirical test of a transaction cost explanation. Journal
of Law, Economics and Organization, 4(2), 305336.
Guillen, M. F. (2005). The rise of the spanish multinational rm: European business in the global economy. Cambridge: Cambridge University Press.
Harzing, A. W. K. (2003). The role of culture in entry mode studies: From negligence to myopia? Advances in International Management, 15, 75127.
Harzing, A. W. K. (2005). The use of English questionnaires in cross-national research: Does cultural accommodation obscure national differences? International
Journal of Cross-Cultural Management, 5(2), 213224.
Harzing, A. W. K., & Feely, A. J. (2008). The language barrier and its implications for HQ-subsidiary relationships. Cross Cultural Management: An International
Journal, 15(1), 4960.
Harzing, A. W. K., & Maznevski, M. (2002). The interaction between language and culture: A test of the cultural accommodation hypothesis in seven countries.
Language and Intercultural Communication, 2(2), 120139.
Henisz, W. J. (2000). The institutional environment for multinational investment. Journal of Law Economics and Organization, 16(2), 334364.
Henisz, W. J., & Delios, A. (2002). Learning about the institutional environment. In P. Ingram & B. Silverman (Eds.), The new institutionalism in strategic management.
Advanced in international management (pp. 339372). New York: JAI Press.
Hennart, J. F. (1988). A transaction cost theory of equity joint ventures. Strategic Management Journal, 9(4), 361374.
Hill, C., Hwang, P., & Kim, W. (1990). An eclectic theory of the choice of international entry mode. Strategic Management Journal, 11(2), 117128.
588 C. Lopez-Duarte, M.M. Vidal-Suarez / International Business Review 19 (2010) 575588

Hofstede, G. (1980). Cultures consequences: International differences in work related values. Beverly Hills, CA: Sage.
Hofstede, G. (2001). Cultures consequences: International differences in work-related values (2 ed.). Beverly Hills: Sage.
Hofstede, G., & Bond, M. H. (1988). The Confucius connection: From cultural roots to economic growth. Organizational Dynamics, 16(4), 421.
Johanson, J., & Vahlne, J. E. (1977). The internationalization process of the rmsA model of knowledge development and increasing foreign market
commitments. Journal of International Business Studies, 8(1), 2332.
Johanson, J., & Wiedersheim-Paul, F. (1975). The internationalization of the rmFour Swedish cases. Journal of Management Studies, October: 305322.
Kim, W. C., & Hwang, P. (1992). Global strategy and multinationals entry mode choice. Journal of International Business Studies, 23(1), 2953.
Kim, Y., & Gray, S. J. (2009). An assessment of alternative empirical measures of cultural distance: Evidence from the Republic of Korea. Asia Pacic Journal of
Management, 26(1), 5574.
Kogut, B., & Singh, H. (1988). Entering the United States by joint venture: Competitive rivalry and industry structure. In F. J. Contractor & P. Lorange (Eds.),
Cooperative strategies in international business. Lexington: Lexington Books.
Kogut, B., & Zander, U. (1992). Knowledge of the rm, combinative capabilities, and the replication of technology. Organization Science, 3(3), 383397.
Lopez-Duarte, C., & Garca-Canal, E. (2002). Adverse selection and the choice between joint-ventures and acquisitions: Evidence from Spanish rms. Journal of
Institutional and Theoretical Economics (JITE), 158(2), 304324.
Lopez-Duarte, C., & Garca-Canal, E. (2004). The choice between joint ventures and acquisitions in foreign direct investment: The role of partial acquisitions and
accrued experience. Thunderbird International Business Review, 46(1), 3968.
Lu, J. W. (2002). Intra and inter organizational imitative behaviour: Institutional inuences on Japanese rms entry mode. Journal of International Business Studies,
33(1), 1937.
Luo, Y. (2001). Determinants of entry in an emerging economy: A multilevel approach. Journal of Management Studies, 38(3), 443472.
Luo, Y., & Shenkar, O. (2006). The multinational corporation as a multilingual community: Language and organization in a global context. Journal of International
Business Review, 37(3), 321339.
Marschan-Piekkari, R., Welch, L., & Welch, D. (1997). Language the forgotten factor in multinational management. European Management Journal, 15, 591598.
Marschan-Piekkari, R., Welch, L., & Welch, D. (1999a). In the shadow: The impact of language on structure, power and communication in the multinational.
International Business Review, 8(4), 421440.
Marschan-Piekkari, R., Welch, L., & Welch, D. (1999b). Adopting a common corporate language: IHRM implications. The International Journal of Human Resource
Management, 10(3), 377390.
Meschi, P. X., & Riccio, E. L. (2008). Country risk, national cultural differences between partners and survival of international joint ventures in Brazil. International
Business Review, 17(3), 250266.
Meyer, K. E. (2001). Institutions, transaction costs and entry mode choice in eastern Europe. Journal of International Business Studies, 32(2), 257267.
Nakos, G., & Brouthers, K. D. (2002). Entry mode choice of SMEs in Central and Eastern Europe. Entrepreneurship Theory & Practice, Fall: 4763.
Neal, M. (1998). The Culture Factor: Cross National Management and the Foreign Venture. Basingstoke: McMillan Press.
Pak, Y. S., & Park, Y. (2004). Global ownership strategy of Japanese multinationals: A test of internalization theory. Management International Review, 44(1), 321.
Pinho, J. C. (2007). The impact of ownership. Location specic advantages and managerial characteristics on SME foreign entry mode choices. International
Marketing Review, 24(6), 715734.
Quer, D., Claver, E., & Andreu, R. (2007). Foreign market entry mode in the hotel industry: The impact of country and rm specic factors. International Business
Review, 16(3), 362376.
Quer, D., Claver, E., & Molina, J. (2006). Riesgo, distancia cultural y estrategia de entrada: nueva evidencia emprica en el proceso de internacionalizacion de la
empresa espanola. Boletn Economico del ICE, 2875, 1321.
Rajan, K. S., & Pangarkar, N. (2000). Mode of entry choice: An empirical study of Singaporean multinationals. Asia Pacic Journal of Management, 17(1), 4966.
Root, F. R. (1994). Entry strategies in international markets. Washington: Lexington Books.
Ruz-Moreno, F., Mas-Ruz, F. J., & Nicolau-Gonzalez, J. L. (2007). Two stage process of FDI: Ownership structure and diversication mode. Journal of Business
Research, 60(7), 795805.
Shenkar, O. (2001). Cultural distance revisited: Towards a more rigorous conceptualization and measurement of cultural differences. Journal of International
Business Studies, 32(3), 519535.
Shrader, R. C., Oviatt, B. M., & McDougall, P. P. (2000). How new ventures exploit trade-offs among international risk factors: Lessons for the accelerated
internationalization of the 21st century. Academy of Management Journal, 43(6), 12271247.
Slangen, A. H. L., & van Tulder, R. J. (2009). Cultural distance, political risk, or governance quality? Towards a more accurate conceptualization and measurement of
external uncertainty in foreign entry mode research. International Business Review, 18(3), 276291.
Tallman, S. B., & Shenkar, O. (1994). A managerial decision model of international cooperative venture formation. Journal of International Business Studies, 25(1),
91114.
Tatoglu, E., Glaister, K., & Erdal, F. (2003). Determinants of foreign ownership in Turkish manufacturing. Eastern European Economics, 41(2), 541.
Tsang, E. W. K. (2005). Inuences of foreign ownership level and entry mode choice in Vietnam. International Business Review, 14(4), 441463.
Welch, D. E., Welch, L. S., & Marschan-Piekkari, R. (2005). Speaking in tongues: Language and international management. International Studies of Management &
Organization, 35(1), 1027.
West, J., & Graham, J. L. (2004). A linguistic based measure of cultural distance and its relationship to managerial values. Management International Review, 44(3),
239260.
Yiu, D., & Makino, S. (2002). The choice between joint ventures and wholly owned subsidiary: An institutional perspective. Organization Science, 13(6), 667683.
Zhao, H., Luo, Y., & Suh, T. (2004). Transaction costs determinants and ownership based entry mode choice a meta-analytical review. Journal of International
Business Studies, 35(6), 524544.