International Business Review 19 (2010) 4658

Contents lists available at ScienceDirect

International Business Review

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

More than just national cultural distance: Testing new distance scales on
FDI in Slovakia
Douglas Dow a,*, Sonia Ferencikova b
a
b

Centre for the Practice of International Trade, Melbourne Business School, The University of Melbourne, 200 Leicester Street, Carlton, Victoria 3053, Australia
University of Economics Bratislava, Dolnozemska Cesta 1, Bratislava 852 35, Slovakia

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 5 September 2008
Received in revised form 20 May 2009
Accepted 18 November 2009

Over the past decade, numerous calls have been made within the international business
literature for a broader conceptualization and measurement of non-geographic forms
distance amongst countries. One promising response to this call has been a set of psychic
distance stimuli scales put forward by Dow, D., & Karunaratna, A. (2006). Developing a
multidimensional instrument to measure psychic distance stimuli. Journal of International
Business Studies, 37(5), 575577. However, to date, these new scales have only been tested
in one very limited setting predicting bi-lateral trade ows. This paper extends the
generalizability of the Dow and Karunaratna scales by testing their criterion-related
validity with respect to three specic foreign direct investment (FDI) issues: predicting
market selection, entry mode choice and performance. The results indicate that the Dow
and Karunaratna scales are signicantly stronger predictors of market selection and FDI
performance than the traditional Kogut and Singh index; and that researchers should go
beyond using national cultural distance as their sole measure of distance amongst
countries. The results for predicting entry mode choice are more ambiguous; however, the
authors argue that the ambiguity may reect the inadequacies of the classic TCE-based
approach to predicting entry mode, rather than shortcomings in the measurement of the
distance construct.
2009 Elsevier Ltd. All rights reserved.

Keywords:
Cultural distance
Entry mode
FDI
Hofstede
Market selection
Performance
Psychic distance

1. Introduction
The concept of psychic distance and the closely related subsidiary construct of cultural distance have enjoyed substantial
prominence in the international business (IB) literature over several decades. After an inconspicuous start (Beckerman,
1956), psychic distance emerged in the 1970s as one of the corner stones of the Uppsala internationalization process model
(Johanson & Wiedersheim-Paul, 1975). A decade later, Kogut and Singh (1988) similarly raised the prole of cultural distance
by converting the Hofstede (1980) national culture dimensions into a formative index. Since that time, psychic and cultural
distance have played such a prominent role in a wide range of empirical IB studies that Cho and Padmanabhan (2005, p. 309)
concluded that almost . . . no international business study can be complete unless there is an explicit variable controlling for
cultural distance. Even more recently, the concept of institutional distance has gained attention within the IB literature (Xu
& Shenkar, 2002). Yet, despite this plethora of conceptualizations of distance, until recently, very little effort has been put
into developing scales for measuring them. Instead, the vast majority of researchers have chosen to simply employ the Kogut

* Corresponding author. Tel.: +61 3 9349 8149.

E-mail addresses: d.dow@mbs.edu (D. Dow), ferencik@euba.sk (S. Ferencikova).
0969-5931/$ see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ibusrev.2009.11.001

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

47

and Singh index (Harzing, 2003) as their sole indicator of distance. This is especially surprising given the consistent stream of
ambiguous and weak empirical results which have caused commentators to call for a broader conceptualization and
measurement of the distance construct (Harzing, 2003; Shenkar, 2001; Tihanyi, Grifth, & Russell, 2005; Zhao, Luo, & Suh,
2004).
One promising response to the call for better measures of the distance construct has been a set of psychic distance
stimuli scales recently published, and made publicly available, by Dow and Karunaratna (2006). However, until now, this
new set of scales has only been validated in one specic setting predicting trade ows amongst countries. Thus, the primary
objective of this paper is to test the Dow and Karunaratna scales across a broader range of managerial issues specically,
predicting FDI market selection, entry mode choice and FDI performance.
In addition to testing the criterion-related validity of the Dow and Karunaratna scales, this paper makes two more modest
contributions to the empirical FDI literature.
b One potential criticism of the Kogut and Singh index is the fact that the underlying Hofstede data was collected from one
single rm more than 40 years ago. However, an extremely extensive meta-analysis has recently provided updated
estimates of Hofstedes national cultural distances based on much broader populations (Taras & Steel, 2006). These updated
national cultural distance estimates have been incorporated in our analyses to determine if the age and bias of the original
Hofstede data is possibly contributing to the weak results.
b A second concern about the vast majority of empirical FDI research is that it is very heavily concentrated on three host
countries: the USA, Japan and China. For that reason, we have chosen to conduct our research in a relatively unique
situation the small but rapidly developing Central Eastern European country of Slovakia.
2. Literature review and hypotheses
2.1. More than just national cultural distance
Despite both its prominence in the literature and its intuitive appeal, psychic distance is a relatively complex construct,
and is often misunderstood and misapplied. Beckerman (1956, p. 38) initially coined the term to explain unexpected
patterns in intra-European trade. In effect, he was arguing that the location theory emphasis on minimising transportation
costs (Weber, 1929) did not adequately explain intra-European trade patterns. Unfortunately, despite coining the term,
Beckerman contributed relatively little to our understanding of the construct; other than to suggest it might involve
language difculties. Thus, the rst formal denition of psychic distance was left to Johanson and Wiedersheim-Paul
(1975, p. 308):
factors preventing or disturbing the ow of information between rm and market. Examples of such factors are differences
in language, culture, political systems, level of education, level of industrial development, etc.
Since that time, a variety of denitions have been debated (e.g. Brewer, 2007; Evans & Mavondo, 2002; OGrady & Lane,
1996) emphasizing issues such as perceptions versus actual differences, and the distinction between distances and
differences. However, two core aspects of the construct (1) that it is a multidimensional construct, and (2) that it is about
the ability (or inability) of parties in different markets to communicate and accurately understand each other appear have
stood the test of time. Indeed, several commentators (Harzing, 2003; Shenkar, 2001; Tihanyi et al., 2005) have recently
echoed a call to return to the multidimensional nature of the construct.
Despite broad agreement that psychic distance is a multidimensional construct including (or inuenced by) factors such
as differences in language, religion, culture, education, industrial development, and political systems (Dow & Karunaratna,
2006; Evans & Mavondo, 2002; Johanson & Wiedersheim-Paul, 1975; Shenkar, 2001), at the empirical level the vast
majority of researchers have defaulted to employing a single metric, specically Kogut and Singhs (1988) composite index
of Hofstedes (1980) four national culture dimensions. Tihanyi et al. (2005) found that 55 of the 66 samples they reviewed
employed the Hofstede index as their sole indicator of psychic distance. We are not disputing here whether the Hofstede
dimensions may be appropriate measures of some aspects of national culture, but rather arguing that the Hofstede index is
only a narrow component of psychic distance. As stated by Zhao et al. (2004): the use of Hofstedes cultural index (used by
most of the studies) as a measure of uncertainty seems ineffective to capture the diversity and subtlety of cultural
inuences. In contrast, the recently published Dow and Karunaratna scales (2006) cover a much broader range of factors
potentially disturbing the ow of information, namely differences in language, religion, industrial development,
education and political systems. However, to date, these new scales have only been tested with respect to predicting trade
ow amongst countries. Thus an underlying theme of each of our hypotheses is that while national cultural distance may be
inuencing a variety of aspects of foreign direct investment, a broader array of psychic distance stimuli, as represented by
the Dow and Karunaratna scales (2006), will have an additional impact over and above the effects of national cultural
distance.
Before proceeding on to the specic hypotheses, it is important to expand upon why this paper is focussing on the Dow
and Karunaratna (2006) approach to conceptualising and measuring distance, as opposed to say the institutional distance
approach (Xu, Pan, & Beamish, 2004), or the perceptual approach (Ellis, 2008), or even another psychic distance-based
national level difference approach (e.g. Brewer, 2007).

48

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

b While the underlying theory of institutional distance is reasonably well developed and dened (Xu & Shenkar, 2002), so far,
exceptional little effort has been expended on actually measuring institutional distance. The majority of researchers (e.g.
Gaur, Delios, & Singh, 2007; Gaur & Lu, 2007; Xu et al., 2004) have relied on opportunistic third party surveys such as the
World Competitiveness Report (e.g. 1994) as indicators of the various dimensions of institutional distance. Moreover, when
one disaggregates the concept of institutional distance, the overlap between psychic distance and institutional distance
becomes quite substantial. Indeed, Demirbag, Glaister, and Tatoglu (2007), one of the few not to rely on World
Competitiveness Report scales, use cultural distance and differences in language as their measures of institutional distance.
Similarly, differences in religion, education systems, and political systems are all cited by Johanson and Wiedersheim-Paul
(1975) as underlying dimensions of psychic distance; and yet, they can also be consider forms of institutional distance. As a
result, we argue that in the same manner as cultural distance, institutional distance is a subset of the broader concept of
psychic distance, and the efforts to develop new scales within that eld are not as advanced as with psychic distance.
b Even if one accepts the argument that psychic distance is a broader level construct, and thus encompasses both cultural
distance and institutional distance, numerous commentators (e.g. Ellis, 2008; Evans & Mavondo, 2002; Sousa & Bradley,
2006) have argued that psychic distance should be measured in terms of the decision makers perceptions, rather than in
terms of exogenous differences between countries. This approach has many merits; however, it also has some very serious
constraints. First of all, most major IB decisions are relatively infrequent and difcult to predict events. As a result, is it
virtually impossible to measure a managers a priori perceptions; and thus, most researcher default to using post hoc
perceptions of distance (e.g. Ellis, 2008). This approach creates problems in terms of identifying the direction of causality. A
manager invests in a market and then is subsequently interviewed and claims to perceive the market as close; but did the
manager invest there because the country was close, or is the country perceived to be close because of the managers post
hoc experience in that market? In addition to that, Evans and Mavondo (2002, p. 517) argue that the true explanatory
power of psychic distance can only be revealed when the individual elements are fully measured. As a result, we argue that
measuring psychic distance in terms of the underlying factors (or stimuli) is an important step forward, although we would
also strongly endorse the efforts of Sousa and Bradley (2006) and Hakanson and Ambos (2007) in attempting to test the
linkages between the stimuli and the actual perceptions.
b A third and nal issue here is why adopt the Dow and Karunaratna (2006) scales, as opposed to other exogenous psychic
distance scales such as Brewers (2007)? The sad truth of the situation is that the Dow and Karunaratna (2006) scales and the
Brewer (2007) scales represent the only multidimensional efforts to measure a broad range of psychic distance stimuli. While
the Brewer scales raise some interesting issues concerning distance-bridging factors, such as informational and sporting ties
amongst countries, there is substantial overlap between the two (e.g. differences in culture, language, and level of industrial
development), and the Dow and Karunaratna scales cover a broader range of the factors most commonly associated with
psychic distance (e.g. Evans & Mavondo, 2002; Harzing, 2003; Stottinger & Schlegelmilch, 1998). Moreover, the Brewer scales
are constructed purely from an Australian perspective and many of its unique dimensions are not available for a broad range of
countries. As a result, the Dow and Karunaratna scales appear to be the most comprehensive available.
2.2. As a predictor of FDI market selection
In light of our preceding arguments for researchers to include a broader set of factors than just national cultural distance,
we now need to turn to the various criterion variables for which we can test our assertion.
The Uppsala school (Johanson & Wiedersheim-Paul, 1975) initially popularised the concept of psychic distance by using it
to explain both export and FDI market selection. They argued that factors disrupting the ow of information between specic
markets reduce a rms awareness of business opportunities in other markets, as well as raising the risk that the rm may
either be mistaken about the opportunity, or unable to effectively capitalize on it. Since that time, psychic or cultural distance
has played a consistent role as a predictor variable for bi-lateral FDI ows (Habib & Zurawicki, 2002; Razin & Tong, 2005), the
source of inward FDI (Grosse & Goldberg, 1991; Grosse & Trevino, 1996), the destination of outward FDI (Davidson, 1983),
and the order of market entries for FDI (Benito & Gripsrud, 1992; Erramilli, 1991). Unfortunately, as highlighted earlier, ve
of these seven studies used Hofstede as their sole indicator of distance, and only a slim majority (4 of 7) of these studies found
a statistically (negative) relationship. Thus, while there is some evidence that a distance-market selection relationship exists,
national cultural distance appears to be a weak surrogate. This leads us to our rst hypothesis:
H1. The likelihood of foreign direct investment occurring between two countries will be negatively correlated with:
a. the national culture distance between the countries, and
b. other aspects of psychic distance, such as differences in language, religion, industrial development, education and political
systems.
2.3. As a predictor of entry mode
The second and more common application of the distance construct has been as an explanatory variable in predicting
foreign market entry mode, specically the use of joint ventures versus wholly owned subsidiaries. This application of
distance has its origins in two seminal articles (Gatignon & Anderson, 1988; Kogut & Singh, 1988). Gatignon and Anderson

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

49

(1988) included socio-cultural distance (i.e. psychic distance) as an indicator of internal uncertainty within the transaction
cost economics (TCE) framework. They predicted that large distances would increase internal uncertainty; which in turn
would encourage managers to seek lower control entry modes, such as joint ventures. Kogut and Singh (1988) based their
hypotheses on the Uppsala internationalisation process model but arrived at essentially the same prediction. This hypothesis
has been tested numerous times over the subsequent two decades, with a recent meta-analysis (Tihanyi et al., 2005) citing
66 prior studies. Yet, despite the plethora of attention, Tihanyi et al.s (2005, p. 524) regression results [have] failed to
provide statistical evidence of signicant relationships between cultural distance and entry mode choice. However, in a
more modest meta-analysis, Zhao et al. (2004) found a very small but statistically signicant effect. Nevertheless, both sets of
authors subsequently suggested that the traditional measurement instrument, based on Hofstede dimensions of national
culture, may be too narrow to capture the full impact of the distance construct. This, combined with our earlier discussions of
the denition and measurement of psychic distance, leads us to our second hypothesis:
H2. Preference for entry via joint venture, as opposed to entry via a wholly owned subsidiary will be positively correlated
with:
a. the national culture distance between host and home countries, and
b. other aspects of psychic distance, such as differences in language, religion, industrial development, education and political
systems.
2.4. As a predictor of FDI performance
The third major application of the distance construct concerns its impact on subsidiary performance. Possibly the most
famous article in this literature stream is The Psychic Distance Paradox (OGrady & Lane, 1996); however, earlier
researchers (e.g. Li & Guisinger, 1991) had already empirically explored the linkages between psychic distance and
subsidiary performance. The most common hypothesis follows the simple logic that psychic distance causes communication
problems, which in turn increases costs and the risks of making a mistake. Thus, one would expect a lower level of
performance in more psychically distant markets. However, several authors (Evans & Mavondo, 2002; OGrady & Lane, 1996)
have argued that low levels of psychic distance may cause overcondence, which may also adversely affect performance. In
their meta-analysis on the issue, Tihanyi et al. (2005) found a negative but not statistically signicant correlation between
subsidiary performance and psychic distance. Once again, the narrow nature of the instrument most commonly used to
measure psychic distance was raised as one possible explanation for these ambiguous results. These results, combined with
our earlier discussions of the denition and measurement of psychic distance, lead us to our third and nal hypothesis. We
have adopted here the more traditional performance hypothesis, but are cognisant that an argument for the reverse
direction could also be made.
H3. The performance of a foreign investment in the host country will be negatively correlated with:
a. the national culture distance between host and home countries, and
b. other aspects of psychic distance, such as differences in language, religion, industrial development, education and political
systems.
3. Research methodology
The following section outlines the methodology used to test our six hypotheses. In effect we are conducting three
separate analyses with a common thread comparing different indices for measuring the distance construct. For that reason,
the rst and most important issue is how we operationalize the distance measures.
3.1. Hofstede
In order to provide a consistent benchmark with past research, our rst measure of distance is the index based on the four
original Hofstede (1980) dimensions (Hof) of national cultural distance. The analyses presented within this paper concern
the four dimensions collapsed into one single index as recommended by Kogut and Singh (1988); however, subsidiary
analyses treating each dimension as an independent variable were also conducted.1
3.2. Taras and Steeles revised estimates of national cultural distance
In addition to the earlier criticisms of the traditional Hofstede scales being too narrow, a number of commentators
(Brewer, 2007) have also criticised the Hofstede scales for being too old and subject to possible sample bias (McSweeney,

The subsidiary analyses yielded essentially the same results with respect to the main hypotheses.

50

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

2002). These criticisms prompted a recent meta-analysis and the public release of revised estimates of national cultural
distance for 76 countries (Taras & Steel, 2006). Thus, our second measure of national cultural distance uses the updated
estimates made available by Taras and Steel (2006). Given the timeframe of our data set, the estimates for the 1990s are
employed (HofT90). The transformation recommended by Kogut and Singh (1988) is used in both cases to convert the four
dimensions of national culture into a single formative index.
3.3. Dow and Karunaratnas psychic distance stimuli
For our third distance instrument, which incorporates additional aspects of psychic distance (PDDK), we have created a
formative index based on ve of the major dimensions put forward by Dow and Karunaratna (2006): differences in language,
religion, industrial development, education and degree of democracy. We have chosen to collapse these into a single index as
most of the Dow and Karunaratna dimensions are already highly correlated, and simultaneous evaluation of the individual
dimensions raises substantial problems in terms of multicollinearity. The specic scores for these ve variables are publicly
available (Dow, 2007) and have been converted in to a single composite index using a similar methodology as for the national
cultural indices:
PDDK

5
X
Ii jk 2 =V k =5
k1

where Iijk is the distance between countries i and j for the kth dimension of psychic distance, and Vk is the variance of the kth
dimension of psychic distance across 120 countries.
The distance estimates between Slovakia (the host country for this study) and 87 countries are listed in Table 1 for all
three instruments.
As previously mentioned, the sample population for this study is FDI into Slovakia between 1990 and 2006. Consistent
with Webers (1929) locational theory, Slovakia, as a low wage country close to the European Union, has attracted a great
deal of foreign investment over the past two decades. Yet, despite the abundance of empirical research on FDI, the majority of
studies have focussed on China (e.g. Pan, 2002), the USA (e.g. Herrmann & Datta, 2006) and Japan (e.g. Delois & Henisz, 2000).
Thus, the rapid opening up of Eastern Europe over the past two decades provides a perfect opportunity to study FDI in a
different setting both in terms of the culture and the size of the economy. It may very well be the case that smaller developing
host markets attract a different spectrum of investing rms, thus changing the dynamics of market selection, entry mode
choice and performance.
The survey instrument was developed and then pre-tested on a selection of Slovak managers. Firms were identied from
a government administered database on FDI in Slovakia, and a total of 500 companies were approached in two waves. There
is no statistically signicant difference between the two waves. A nal useable sample of 154 ventures implies an effective
response rate of 31%. The surveys were completed either via in-person or telephone interviews by Slovakian PhD students,
uent in both English and Slovak. A key respondent, typically the senior executive or a director of the rm, was interviewed
in each case as the nature of the questions required detailed knowledge of the original investment and the parent companies.
3.4. Predicting market selection
In order to test the market selection hypotheses (H1a and H1b), the basic unit of analysis is each potential home nation.
Given the limited coverage of the Hofstede and Taras and Steele variables, only 87 countries (listed in Table 1) could be
considered as potential home countries. From the 154 ventures surveyed, only 24 nations are actually represented (i.e. 63 of
the original 87 nations do not have a single FDI venture in our sample). We have chosen to retain in our market selection
analyses the 63 nations with no entries because the lack of any entries is important information in itself. However, the
analyses have also been repeated with those 63 nations withheld. Except for a substantial reduction in sample size and
statistical power, the results are essentially the same.
The dependent variable for the market selection analyses is the natural logarithm of number of Slovakian FDI ventures
originating from each of the home countries (Entries_ln). This follows the methodology of Anand and Kogut (1997).
For control variables, a gravity model (e.g. Green & Cunningham, 1975; Grosse & Trevino, 1996; Razin & Tong, 2005) has
been adopted with the logarithm of the home nation population (POP_ln), the GDP per capita (GDP_pc) and the geographic
distance between the markets (Dist_ln) as the main control variables. The descriptive statistics and the correlation matrix for
all these variables are in Tables 2 and 3. The market selection hypotheses are tested using a multiple regression of the
aforementioned variables.
3.5. Predicting entry mode
In order to test the entry mode choice hypotheses (H2a and H2b), the basic unit of analysis is each FDI venture (i.e.
n = 154). For this analysis, logistic regression is employed with the initial mode of entry (JV) as the dependent variable.
Businesses where the foreign parents represent less than 95% of the equity are classied as joint ventures and are coded as
1. A 95% cut off for dening joint ventures is a common threshold within the entry mode literature (Arregle, Hebert, &

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

51

Table 1
Summary of the 87 potential home countries.a.
Countries

Hofb

HofT90c

PDDKd

# of entries

Countries

Argentina
Australia
Austria
Bangladesh
Belgium
Brazil
Bulgaria
Canada
Chile
China
Colombia
Costa Rica
Cote dIvoire
Croatia
Czech Republic
Denmark
Ecuador
Egypt
El Salvador
Estonia
Ethiopia
Finland
France
Germany
Ghana
Greece
Guatemala
Hong Kong
Hungary
India
Indonesia
Iran
Iraq
Ireland
Israel
Italy
Jamaica
Japan
Kenya
Korea, Rep. of
Kuwait
Lebanon
Libyan
Luxembourg

4.7
5.3
5.9
3.3
4.6
4.1
5.4
5.5
7.4
2.6
3.7
10.4
4.2
5.1
3.8
11.9
3.2
3.2
6.2
7.5
5.0
8.6
5.1
5.1
4.2
5.4
6.6
3.6
3.2
2.8
4.4
4.9
3.2
5.0
8.4
3.5
4.2
2.5
5.0
6.3
3.2
3.2
3.2
5.4

4.8
0.7
0.9

0.9
1.4
1.1
5.2
1.2
1.6
0.9
1.8
1.1
5.3
1.4
1.3
4.4
1.1
0.8
1.4
1.4
4.1
2.0
1.0
4.6
1.5
1.2
1.2
2.8
1.0
3.0
2.1
0.7
4.2
4.1
4.2
4.9
0.9
2.3
1.1
1.6
3.0
3.2
1.4
3.4
1.4
4.4
1.8

0
1
14
0
3
2
0
0
0
1
0
0
0
0
9
3
0
0
0
0
0
1
12
38
0
1
0
0
3
1
0
0
0
0
0
10
0
4
0
4
0
0
0
0

Malaysia
Malta
Mexico
Morocco
Mozambique
Netherlands
New Zealand
Nigeria
Norway
Pakistan
Panama
Peru
Philippines
Poland
Portugal
Romania
Russian Fed.
Saudi Arabia
Serbia
Sierra Leone
Singapore
Slovenia
South Africa
Spain
Suriname
Sweden
Syria
Switzerland
Taiwan
Tanzania
Thailand
Trinidad
Turkey
UAE
Ukraine
United Kingdom
USA
Uruguay
Uzbekistan
Venezuela
Vietnam
Zambia
Yemen
Total

4.1
2.6
2.9
0.6
1.0
0.8
1.3
1.2
2.1
1.4
0.7
1.1
1.4
1.9
1.7
0.3
1.3
1.0
2.1
1.4
0.3
1.8
1.0
0.8
1.7
0.3
4.2
0.4
1.8
0.7
0.6
1.7
1.0
0.3
0.3
0.3

Hofb
3.3
5.5
2.4
3.4
10.3
6.4
4.2
11.5
5.2
5.0
5.9
2.0
3.4
7.7
4.8
5.5
3.2
5.0
4.2
5.0
8.3
3.6
5.6
5.3
12.3
4.2
5.4
5.0
6.1
4.7
4.9
3.2
4.5
5.0
6.6
2.5
5.3
5.0

HofT90c

PDDKd

# of entries

0.8

2.6
1.0
1.4
4.7
4.1
1.3
1.4
4.2
1.5
4.8
1.3
1.5
1.5
0.8
1.0
1.1
1.1
4.6
2.3
5.7
2.7
1.0
1.6
1.4
1.4
1.5
4.5
1.2
2.4
3.6
3.2
1.1
3.1
3.7
0.9
1.3
2.0
0.9
3.7
1.3
4.5
2.7
5.5

0
0
0
0
0
8
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
2
0
5
0
5
0
0
0
0
0
0
0
4
21
0
0
0
0
0
0
154

0.6
1.7
1.4
1.1
2.1
2.1
2.4
1.5
0.6
0.8
0.8
3.8
1.7
0.3
1.1
2.1
1.1
1.3
2.0
4.6
1.5
0.3
1.2
1.7
1.0
0.7
1.3
4.9
0.3
0.8
0.7
1.3
1.3
2.1
0.3
1.7

a
These countries represent the set of all countries for which the Dow and Karunaratna (2006) variables and the Hofstede (1980, 2001) or Taras and Steel
(2006) variables are available. These countries form the basis of our market selection analyses.
b
Psychic distance from Slovakia as measured by Kogut and Singhs (1988) composite index of the Hofstede dimensions of national culture.
c
Psychic distance from Slovakia as measured by Kogut and Singhs (1988) composite index of the Hofstede dimensions of national culture but using Taras
and Steeles (2006) most recent estimated values for the 1990s.
d
Psychic distance from Slovakia based on a composite index of ve of Dow and Karunaratnas (2006) dimensions: differences in language, region,
industrial development, education and degree of democracy.

Beamish, 2006; Lu, 2002). The count of joint ventures versus wholly owned subsidiaries by home country is summarized in
Table 4.
A total of eight control variables commonly found in entry mode choice studies have been incorporated in these analyses,
in addition to our measures of distance. The rst two control variables are four point scale indicators of the R&D intensity
(R&D) and advertising intensity (Adv) of the foreign parent. These two are amongst the most commonly employed variables
(Zhao et al., 2004) in entry mode studies and are justied in the TCE literature as indicators of asset specicity and free-riding
potential.
A third control variable is an indicator of the size of the foreign parent (PSize_f), and is based on a factor score of the
number of worldwide employees and the worldwide revenues of the rm. These two measures are highly correlated
(r = 0.879) and when combined produce a Cronbach alpha of 0.706. Firm size is typically included in such studies (e.g. Arregle
et al., 2006; Delois & Henisz, 2000; Hennart & Larimo, 1998) to control for differences in the parent rms resourceabundance.

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

52

Table 2
Descriptive statistics for market selection variables.

POP_ln
Dist_ln
GDP_pc
Hof
HofT90
PDDK
Entries_ln

Expected sign wrt freq of selection

Min.

Max.

Mean

S.D.

+

+



n.a.

87
87
87
82
76
87
87

0.99
4.03
0.08
1.97
0.26
0.73
0.00

7.13
9.79
42.42
12.30
4.93
5.68
3.66

2.81
8.06
9.21
5.14
1.45
2.33
0.44

1.57
1.20
10.88
2.12
1.07
1.43
0.85

Table 3
Correlation matrix for market selection variables (n = 87, uos).

1
2
3
4
5
6
7

POP_ln
Dist_ln
GDP_pc
Hofa
HofT90b
PDDK
Entries_ln

1.00
0.191
0.210*
0.334**
0.018
0.347**
0.229*

1.00
0.282**
0.182
0.160
0.308**
0.439**

1.00
0.339**
0.062
0.301**
0.533**

1.00
0.201
0.327**
0.142

1.00
0.257*
0.002

1.00
0.286**

*p < 0.05, **p < 0.01 (two tailed signicance).

a
n = 82.
b
n = 76.

Two single item variables are included in the analysis to take in consideration diversied rms (Diversied) and ventures
which are in an industry unrelated to the parent rms normal line of business (Unrelated). In both of these situations, the
foreign parent may be looking to a joint venture partner for industry specic knowledge, in addition to country specic
knowledge (Chang & Rosenzweig, 2001; Hennart & Larimo, 1998).
The nal three control variables (GExp_f, RExp_f and Exp_Slo) reect different aspects of prior international experience:
specically, overall international experience (e.g. Arregle et al., 2006), experience in the local region (e.g. Brouthers &
Brouthers, 2003), and experience in that specic country (e.g. Lu, 2002). Both regional and global experiences are measured
as the factor score of the number of markets entered and the number of years of experience. The reliabilities of both scales are

Table 4
Summary of entry modes by home country.
Countries

# of JV

# of WOS

Hungary
Czech Republic
Greece
Austria
Italy
Belgium
France
Germany
Switzerland
Netherlands
United Kingdom
Australia
Denmark
Korea, Republic of
Spain
Finland
Norway
Sweden
Brazil
USA
Japan
India
Saudi Arabia
China

1
1
0
4
5
1
4
10
1
3
2
1
0
0
0
0
1
3
0
4
1
1
1
0

2
8
1
10
5
2
8
28
4
5
2
0
3
4
2
1
0
2
2
17
3
0
0
1

Total

43

111

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

53

Table 5
Descriptive statistics for entry mode and performance variables.

R&D
Adv
PSize_f
Age_subsid
Unrelated
Diversied
GExp_f
RExp_f
Exp_Slo
Hof
HofT90
PDDK
JV
Perf_f

Expected sign wrt JV

Expected sign wrt Perf_f

Min.

Max.

Mean

S.D.




n.a.
+
+



+
+
+
n.a.
n.a.

n.a.
n.a.
+
+

n.a.
+
+
+



n.a.
n.a.

154
154
154
154
154
154
154
154
154
154
154
154
154
145

1
1
1.14
0
1
1
1.01
0.58
0
2.48
0.26
0.73
0
2.58

4
4
5.32
16
5
5
4.88
7.19
1
12.30
4.64
5.32
1
2.39

2.47
2.37
0.00
7.21
2.55
3.29
0.04
0.24
0.23
5.50
1.28
1.43
0.28
0.00

1.21
1.05
0.93
4.85
1.13
1.13
1.04
1.16
0.42
2.23
0.85
0.63
0.45
1.00

Table 6
Correlation matrix for entry mode and performance analyses (n = 154).a.

1
2
3
4
5
6
7
8
9
10
11
12
13
14

RD
Adv
PSize_f
Age_subsid
Unrelated
Diversied
GExp_f
RExp_f
Exp_Slo
Hof
HofT90
PDDK
JV
Perf_f

10

11

12

13

1.00
0.23**
0.10
0.05
0.03
0.11
0.16
0.10
0.23**
0.02
0.04
0.08
0.14
0.08

1.00
0.06
0.07
0.01
0.14
0.07
0.03
0.05
0.04
0.12
0.16
0.15
0.04

1.00
0.22**
0.04
0.02
0.31**
0.21**
0.12
0.09
0.13
0.06
0.09
0.05

1.00
0.13
0.04
0.22**
0.08
0.17*
0.07
0.01
0.04
0.13
0.03

1.00
0.32**
0.12
0.15
0.02
0.10
0.03
0.11
0.16
0.00

1.00
0.21*
0.15
0.12
0.14
0.10
0.01
0.09
0.03

1.00
0.42**
0.20*
0.03
0.12
0.13
0.18*
0.16

1.00
0.28**
0.11
0.03
0.01
0.20*
0.13

1.00
0.12
0.07
0.00
0.13
0.14

1.00
0.36**
0.14
0.06
0.15

1.00
0.22**
0.13
0.05

1.00
0.05
0.13

1.00
0.02

*p < 0.05, **p < 0.01 (two tailed signicance).

a
The sample size of 154 applies to all variables except for the performance variable, Perf_f. For that variable, and all of its bi-variate correlations, the
sample size is 145.

within acceptable limits with Cronbach alphas of 0.618 and 0.867 respectively. Prior experience within Slovakia is
measured as a single dummy variable. Descriptive statistics and a correlation matrix for these variables are available in
Tables 5 and 6.
3.6. Predicting FDI performance
As with the entry mode analysis, in order to test the performance hypotheses (H3a and H3b), the basic unit of analysis is
each FDI venture (n = 154). Multiple regression analysis is used with the dependent variable being a factor score based on 11
ve-point perceptual scales measuring various aspects of subsidiary performance. These 11 scales are drawn from Geringer
and Hebert (1991) and Brouthers (2002). The individual items have loadings between 0.563 and 0.840, with an overall
Cronbach alpha of 0.910, indicating a very reliable construct.
A total of seven control variables are included in the analyses, in addition to our two measures of psychic distance. Five of
those control variables are identical to ones used in the entry mode analyses: size of the parent rm (PSize_f), whether the
venture is in an industry unrelated to the parent rms normal line of business (Unrelated), and the three measures of
international experiences (GExp_f, RExp_f and Exp_Slo). Size of the parent rm and its degree of international experience are
frequently included in such analyses (Brouthers, 2002; Goerzen & Beamish, 2003; Shaver, 1998) with the expectation that
larger and more experienced parent rms, with their greater abundance of tangible and intangible assets are able to confer
performance advantages onto their subsidiaries.
The age of the subsidiary (Age_subsid), measured in the number of years the venture has been operating, is a commonly
included to control for the liability of newness (Li & Guisinger, 1991, p. 211). In essence, it recognises that there may be a
positive relationship between the age of a business and its performance. In this particular instance, it will also control for a
potential sample bias. Poor performing ventures founded early in our sampling time frame (19902006) may have already
withdrawn from Slovakia; and thus, will be missing from our survey.

54

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

Table 7
Multiple regression predicting number of FDI entries into Slovakia potential host countries (regression coefcients, with t-statistics in parentheses).

Dependent variable: Entries_ln

Constant
POP_ln
Dist_ln
GDP_pc
Hof
HofT90
PDDK
Adj r2
F
p

Model 1a

Model 2b

1.632
0.262
0.243
0.036
0.006

2.148
0.257
0.296
0.038

(3.120)**
(5.849)***
(4.239)***
(5.490)***
(0.184)n.s.

(3.852)***
(5.096)***
(4.812)***
(5.662)***

0.140 (2.600)**

0.075 (1.126)n.s.
0.117 (2.157)*

0.544
20.30
0.000

0.573
21.16
0.000

***p < 0.001, **p < 0.01, *p < 0.05 (one tailed signicance).
a
Sample size of 82 countries.
b
Sample size of 76 countries.

The nal control variable for our performance analyses is the entry mode. While there is no a priori expectation that one
entry mode will yield a higher level of performance, it is considered prudent to include the mode of entry (JV) as a potential
predictor of performance (Brouthers, 2002; Li, 1995; Li & Guisinger, 1991; Shaver, 1998). Tables 5 and 6 provide detailed
descriptive statistics and a correlation matrix for the variables in the performance analyses.
4. Results
Before conducting the analyses, the data was screened to address issues such as inuential outliers and missing variables.
For several variables, specically the size of the parent rm (PSize_f) and two of the three measures of international
experience (GExp_f and RExp_f), natural logarithm transformations were considered. However, after conducting the
analyses both with and without transforming these variables, the results are effectively the same. As a result, we have
reported the results here with the variables untransformed for simplicity.
In cases where multiple indicators are employed, missing data has been estimated using the remaining indicators. In
instances, where data is missing for single indicator variables, the population mean has been imputed. This substitution has
been limited to small number of instances and for control variables only.
Table 7 provides a summary of our market selection analyses. Model 1 is highly signicant, as are all the control variables.
However, while our measure of other aspects of psychic distance, PDDK, is statistically signicant, supporting H1b; the
traditional measure of cultural distance, Hof, is not. For Model 2, which employs the updated estimates of national cultural
distance (HofT90), the coefcient for the national culture variable switches to the predicted direction and appears to be
stronger, but it still falls short of accepted standards of statistical signicance. Thus the rst hypothesis, H1a, is not
supported.
Table 8
Logistic regression predicting entry mode of FDI into Slovakia (logistic regression coefcients, with Wald statistics in parentheses).
Model 3
Dependent variable: JV
Constant
R&D
Adv
PSize_f
Unrelated
Diversied
GExp_f
RExp_f
Exp_Slo
Hof
HofT90
PDDK
Year
n
Chi Sq
df
Signif
Nagelkerke R2
% Correct

1.770
0.181
0.320
0.111
0.118
0.371
0.311
0.547
0.393
0.050

(2.748)
(1.104)
(2.499)t
(0.151)
(0.385)
(3.259)*
(1.357)
(3.333)*
(0.528)
(0.345)

0.171 (0.330)

154
21.352
10
0.019
0.187
74.0

***p < 0.001, **p < 0.01, *p < 0.05, tp < 0.10 (one tailed signicance).

Model 4

Model 5

193.5 (4.589)
0.146 (0.676)
0.298 (2.059)t
0.205 (0.503)
0.165 (0.694)
0.360 (2.912)*
0.482 (2.634)t
0.528 (3.227)*
0.198 (0.124)
0.036 (0.176)

196.8 (4.650)
0.157 (0.778)
0.250 (1.411)
0.165 (0.320)
0.147 (0.549)
0.416 (3.774)*
0.500 (2.715)*
0.573 (3.659)*
0.214 (0.146)

0.239 (0.616)
0.098 (4.670)*
154
26.230
11
0.006
0.226
72.7

0.380 (2.724)*
0.337 (1.181)
0.100 (4.758)*
154
28.784
11
0.002
0.246
75.3

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

55

Table 9
Multiple regressions predicting performance of FDI entries into Slovakia (regression coefcients, with t-statistics in parentheses).
Model 6
Dependent variable: Perf_f
Constant
PSize_f
Age_subsid
Unrelated
JV
GExp_f
RExp_f
Exp_Slo
Hof
HofT90
PDDK
n
Adj r2
F
p

0.097
0.094
0.004
0.005
0.122
0.196
0.009
0.314
0.071

Model 7
(0.236)
(0.971)
(0.210)
(0.062)
(0.648)
(2.007)*
(0.098)
(1.507)t
(1.871)

0.250 (1.738)*
145
0.039
1.643
0.109

Model 8

0.114 (0.365)

0.161
0.007
0.310
0.072

(1.799)*
(0.075)
(1.546)t
(1.945)

0.244 (1.743)*
145
0.056
2.704
0.023

0.210
0.109
0.000
0.011
0.092
0.186
0.038
0.265

Model 9
(0.518)
(1.121)
(0.021)
(0.144)
(0.485)
(1.878)*
(0.412)
(1.266)

0.152 (1.675)*
0.037 (0.412)
0.254 (1.265)

0.031 (0.307)
0.255 (1.721)*

0.044 (0.448)
0.252 (1.745)*

145
0.014
1.234
0.280

0.245 (0.879)

145
0.032
1.939
0.092

***p < 0.001, **p < 0.01, *p < 0.05, tp < 0.10 (one tailed signicance).

In contrast, the entry mode models (Table 8) are statistically signicant, but surprisingly weak in terms of the number of
non-signicant predictor variables and the proportion of explained variance. Even when adopting a generous standard of
0.10 signicance, only three of the eight control variables, and neither of the main variables of interest (Hof and PDDK),
achieve statistical signicance in Model 3. In subsequent investigations, a temporal trend in the use of joint ventures was
identied, and thus an extra variable (Year) has been introduced to control for that (Model 4); however, the non-signicance
of the two distance variables does not change. The introduction of the updated national cultural distance values (Model 5)
does improve the situation with that variable achieving 0.05 signicance; but our measure of other aspects of psychic
distance, PDDK, remains statistically non-signicant. Thus, hypothesis H2a is supported, but not H2b.
For the potential impact of psychic distance on host market performance (Table 9), the predictive power of the various
models is even weaker still. Indeed, for the initial performance model (Model 6), so few of the control variables are signicant
that the overall model fails to achieve statistical signicance. A more parsimonious performance model (Model 7) achieves
statistical signicance; however, it is still disappointingly weak in terms of its predictive power (an adjusted r2 of 5.6%). The
introduction of the updated national cultural distance estimates (HofT90 in Models 8 and 9) appears to have very little
benecial impact on the overall model. Yet, despite these limitations, our measure of other aspects of psychic distance, PDDK,
is a statistically signicant predictor of performance in all four regressions, supporting H3b. The same cannot be said for
national cultural distance. Regardless of whether the traditional measure of cultural distance, Hof, is employed, or the
updated measure (HofT90), hypothesis H3a is not supported.
5. Discussion and conclusions
5.1. Summary of contributions for researchers
In terms of achieving its original objective, this paper presents reasonably strong evidence that a formative index based
on ve key dimensions of psychic distance taken from Dow and Karunaratna (2006) is a signicant predictor of FDI market
selection and performance. In the case of predicting entry mode selection, the Dow and Karunaratna index is not a signicant
predictor variable, but neither is the original Hofstede-based index, nor are the majority of the TCE-based control variables.
This lack of statistical signicance for the entry mode analyses forms the basis of a separate discussion.
In contrast, the original Hofstede-based index of national cultural distance does not achieve statistical signicance as a
predictor for any of the three criterion variables reported in this study. Thus, with respect to measuring and operationalizing
the distance construct in empirical IB research, we believe these results argue strongly that researchers need to move beyond
simply inserting the Hofstede index. As Shenkar (2001) has already argued, the distance construct is much broader than that.
At the very least, we would recommend the inclusion of both the Hofstede and the Dow and Karunaratna indices, but
research should not end there. There may be other aspects of psychic distance which are not adequately reected in either of
these indices.
With regard to adopting the updated Taras and Steel (2006) estimates of national cultural distance, the empirical results
are not as compelling; though in each case we would argue the Taras and Steele index appears to be a superior indicator of
national cultural distance than the original Hofstede scales. For predicting entry mode choice, the Taras and Steele index is
clearly a superior predictor variable; however, for predicting market selection and performance, neither index of national
cultural distance is statistically signicant. Nevertheless, in each case, the Taras and Steele index is closer to achieving
statistical signicance (in the predicted direction) than the original Hofstede index. Thus we would argue that the Taras and
Steele index appears to be a superior measure, at least for predicting FDI behaviour in the 1990s and beyond.

56

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

Table 10
Comparison of effect sizes with prior studies (partial correlation of each variable with Entry Mode JV).

R&D intensity (R&D)

Advertising intensity
(Adv)
International experience
(GExp_f)
Cultural distance
(Hof)
(HofT90)
Psychic distance
(PDDK)

For the Slovakian sample reported here

For the Zhao et al. (2004) meta-analysis

0.11
0.16

0.055
0.063

0.13

0.101

+0.03

+0.029

+0.13
+0.08

Returning to the entry mode choice analyses, the surprisingly weak results referred to earlier, form the basis of an
unanticipated contribution of this article. Despite the fact that the overall entry mode models achieve statistical signicance,
we refer to these results as weak for two reasons. First of all, less than half of the predictor variables prove to be signicant to
0.05. Secondly, the percentage of correct predictions (75.3% in Model 5) is extremely weak given that 72.1% of all ventures in
our sample are joint ventures. In effect, Model 2 is only improving the proportion of correct predictions by 3.2%! If these
models were taken as a test of the application of TCE theory to entry mode choice, one would have to conclude that the theory
does not provide a strong explanation of entry mode selection.
Two potential reactions to our weak entry mode results might be that either the data collection process was somehow
awed, or that the TCE model does not apply in the Slovakian context. However, after reviewing both our results and the
previous literature on foreign entry mode selection, we argue there is a third and more plausible explanation. If one
compares the effect sizes for our Slovakian data (i.e. the partial correlations with the dependent variable) with a recent metaanalysis (Zhao et al., 2004), the results are remarkably similar (Table 10). In essence, our Slovakian data appears to exhibit
similar if not larger effect sizes than previous studies based in other regions! The one critical difference is that a large
proportion of entry mode studies reported in Zhao et al.s review relied on substantially larger sample sizes an average
sample size is 635 compared to 154 for our Slovakian data.
These results raise the question of whether the TCE-approach is an effective framework for predicting entry mode choice.
We would argue it does not appear to be. One can achieve statistical signicance for most TCE variables if a large enough
sample is collected, but the models only explain an alarmingly small amount of the variance. There appears to be a strong
need for researchers to return to an exploratory phase; and to investigate what other, as yet unmeasured factors, may be
driving entry mode choice. Based on our own experiences from collecting data in Slovakia, the underlying motive for the FDI
may be one of the missing factors. This aspect is reected in the work of Harzing (2002) and Sanchez-Peinado, Pla-Barber,
and Hebert, (2007); but we believe the range of FDI motives may be substantially more complex than even portrayed in those
papers.
In summary, this paper incorporates three overlapping contributions for IB researchers. The rst and foremost of these is
that it conrms the criterion-related validity and signicance of the Dow and Karunaratna (2006) scales with respect to two
distinct aspects of FDI: market selection and performance. The new scales ability to predict FDI entry mode choice is not
conrmed at this stage, but that may reect a broader problem in terms of the TCE-based approachs ability to predict entry
mode choice. Combined with continuing weak and non-signicant results with respect to the Kogut and Singh index, it
reinforces the calls for researchers to employ a much broader operationalization of psychic distance. Simply plugging in the
Hofstede index as a control variable is potentially missing a large portion of the potential impact of psychic distance.
Secondly, this paper provides some encouraging, though not conclusive evidence concerning the appropriateness of an
updated set of national culture estimates (Taras & Steel, 2006). Thirdly, and nally, this paper extends the generalisability of
these distance scales and the three predictor-variable models (market selection, entry mode and performance) to a new
geographic region Central Eastern Europe. To our knowledge only a small number of previous studies (Brouthers &
Brouthers, 2001, 2003; Dikova & van Witteloostuijn, 2007; Meyer, 2001) have explicitly addressed FDI in that region.
5.2. Summary of contributions for practitioners
The primary implication of these ndings for practitioners is in re-emphasizing and highlighting the extremely broad
range of factors which need to be taken into account when considering major foreign investment decisions be they
selecting the market or entry mode, or considering the likely returns. This excessive focus on the Hofstede dimensions of
culture has also tended to understate the importance of distance in these decisions.
5.3. Limitations and future research
The research presented here has a variety of limitations, the most salient of which is the geographic-focus. While our
sample may be representative of the population of foreign rms investing in Slovakia, it is not necessarily representative of

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

57

all Central Eastern European countries, and is certainly not representative of all countries. In particular, this study is
relatively unusual in its focus on a small developing host market. This particular focus is both one of the strengths of this
study, and one of its weaknesses. It is important for researchers to be testing their theories across a wide range of setting, and
the FDI literature in general has tended to neglect smaller developing nations; however, generalising the results to larger
markets needs to be done with caution. As a result, one of the most critical research agendas emerging from this paper
concerns the further validation of the Dow and Karunaratna scales in other host market settings.
The sample population for this study is also biased towards rms from large and/or psychically-close countries such as
the USA, Germany and the Czech Republic. This bias is important in terms of assessing the impact of home country
characteristics on entry mode choice and performance. While this bias is representative of the actual investment patterns
within Slovakia, it does mean once again that generalizing the results to rms from smaller and more distant nations needs
to be done with caution.
It should also be noted that the impact of psychic distance may vary depending on the nature of the investing rm (e.g.
small verses large rms) and/or its industry. This study has not controlled for such moderating factors, and such a line of
investigation represents an important avenue for further investigation.
The nal limitation and research implication emerging from this paper concerns the entry mode analysis. As discussed
earlier, the analyses presented in Table 8 represent surprisingly weak results in terms of the lack of statistical signicance for
most of the predictor variables. This highlights the fact that the current state of the art entry mode model does not actually
do a particularly good job of predicting rm behaviour; and that researchers have been compensating for this by seeking out
extremely large sample populations. The implication here is that a fundamental re-examination of the entry mode literature
is required in order to better understand and explain this phenomenon.
References
Anand, J., & Kogut, B. (1997). Technology capabilities of countries, rm rivalry and foreign direct investment. Journal of International Business Studies, 28(3), 445
465.
Arregle, J. L., Hebert, L., & Beamish, P. (2006). Mode of international entry: The advantages of multilevel methods. Management International Review, 46(5), 597
618.
Beckerman, W. (1956). Distance and the pattern of inter-European trade. The Review of Economics and Statistics, 38(1), 3140.
Benito, G. R. G., & Gripsrud, G. (1992). The expansion of foreign direct investments: Discrete rational location choices or a cultural learning process? Journal of
International Business Studies, 23(3), 461476.
Brewer, P. (2007). Operationalizing psychic distance: A revised approach. Journal of International Marketing, 15(1), 4466.
Brouthers, K. D. (2002). Institutional, cultural and transaction cost inuences on entry mode choice and performance. Journal of International Business Studies,
33(2), 203221.
Brouthers, K. D., & Brouthers, L. E. (2001). Explaining the national cultural distance paradox. Journal of International Business Studies, 32(1), 177192.
Brouthers, K. D., & Brouthers, L. E. (2003). Why service and manufacturing entry mode choices differ: The inuence of transaction cost factors. Journal of
Management Studies, 40(5), 11791204.
Chang, S.-J., & Rosenzweig, P. M. (2001). The choice of entry mode in sequential foreign direct investment. Strategic Management Journal, 22(8), 747776.
Cho, K. R., & Padmanabhan, P. (2005). Revisiting the role of cultural distance in MNCs foreign ownership mode choice: The moderating effect of experience
attributes. International Business Review, 14(3), 307324.
Davidson, W. H. (1983, December). Market similarity and market selection: Implications for international marketing strategy. Journal of Business Research, 11,
439456.
Delois, A., & Henisz, W. J. (2000). Japanese rms investment strategies in emerging economies. Academy of Management Journal, 43(3), 305323.
Demirbag, M., Glaister, K. W., & Tatoglu, E. (2007). Institutional and transaction cost inuences on MNE ownership strategies of their afliates: Evidence from an
emerging market. Journal of World Business, 42(4), 418434.
Dikova, D., & van Witteloostuijn, A. (2007). Foreign direct investment mode choice: Entry and establishment modes in transition economies. Journal of
International Business Studies, 38(6), 10131033.
Dow, D. (2007). Research web-page for Douglas Dow. Retrieved 2007, from http://www.mbs.edu/home/dow/research/.
Dow, D., & Karunaratna, A. (2006). Developing a multidimensional instrument to measure psychic distance stimuli. Journal of International Business Studies, 37(5),
575577.
Ellis, P. (2008). Does psychic distance moderate the market size-entry sequence relationship? Journal of International Business Studies, 39(3), 351369.
Erramilli, M. K. (1991). The experience factor in foreign market entry behavior of service rms. Journal of International Business Studies, 22(3), 479501.
Evans, J., & Mavondo, F. T. (2002). Psychic distance and organizational performance: An empirical examination of international retailing operations. Journal of
International Business Studies, 33(3), 515532.
Gatignon, H., & Anderson, E. (1988). The multinational corporations degree of control over foreign subsidiaries: An empirical test of a transaction cost
explanation. Journal of Law Economics and Organization, 4(2), 305336.
Gaur, A. S., Delios, A., & Singh, K. (2007). Institutional environments, stafng strategies, and subsidiary performance. Journal of Management, 33(4), 611636.
Gaur, A. S., & Lu, J. W. (2007). Ownership strategies and survival of foreign subsidiaries: Impacts of institutional distance and experience. Journal of Management,
33(1), 84110.
Geringer, J. M., & Hebert, L. (1991). Measuring performance of international joint ventures. Journal of International Business Studies, 22(2), 249263.
Goerzen, A., & Beamish, P. (2003). Geographic scope and MNE performance. Strategic Management Journal, 24(13), 12891306.
Green, R. T., & Cunningham, W. H. (1975). The determinants of US foreign direct investment: An empirical investigation. Management International Review, 15(2
3), 113120.
Grosse, R., & Goldberg, L. G. (1991). Foreign bank activities in the United States: An analysis by country of origin. Journal of Banking and Finance, 15(6), 10931112.
Grosse, R., & Trevino, L. J. (1996). Foreign direct investment in the United States: An analysis by country of origin. Journal of International Business Studies, 27(1),
139155.
Habib, M., & Zurawicki, L. (2002). Corruption and foreign direct investment. Journal of International Business Studies, 33(2), 291307.
Hakanson, L., & Ambos, B. (2007). The antecedents of psychic distance. Catania, Sicily: European International Business Academy.
Harzing, A. W. (2002). Acquisitions versus greeneld investments: International strategy and management of entry modes. Strategic Management Journal, 23(3),
211227.
Harzing, A. W. (2003). The role of culture in entry mode studies: From neglect to myopia. In Cheng, J., & Hitt, M. (Eds.), Advances in international management. Vol.
15 (pp.75127). Amsterdam/New York: Elsevier/JAI
Hennart, J. F., & Larimo, J. (1998). The impact of culture on the strategy of multinational enterprises: Does national origin affect ownership decisions? Journal of
International Business Studies, 29(3), 515538.

58

D. Dow, S. Ferencikova / International Business Review 19 (2010) 4658

Herrmann, P., & Datta, D. K. (2006). CEO experiences: Effects on the choice of FDI entry mode. Journal of Management Studies, 43(4), 755778.
Hofstede, G. (1980). Cultural consequences: International differences in work related values. Beverly Hills: Sage Publications.
Hofstede, G. (2001). Cultures consequences: Comparing values, behaviors, institutions, and organizations across nations. Thousand Oaks, CA: Sage Publications.
IMD. (1994). World competitiveness report. Lausanne, Switzerland.
Johanson, J., & Wiedersheim-Paul, F. (1975, October). The internationalization of the rm: Four Swedish cases. Journal of Management Studies, 12, 305322.
Kogut, B., & Singh, H. (1988). The effect of national culture on the choice of entry mode. Journal of International Business Studies, 19(3), 411432.
Li, J. (1995). Foreign entry and survival: Effects of strategic choices on performance in international markets. Strategic Management Journal, 16(5), 333351.
Li, J., & Guisinger, S. (1991). Comparative business failures of foreign-controlled rms in the united states. Journal of International Business Studies, 22(2), 209224.
Lu, J. W. (2002). Intra- and inter-organizational imitative behavior: Institutional inuences on Japanese rms entry mode choice. Journal of International Business
Studies, 33(1), 1937.
McSweeney, B. (2002). Hofstedes model of national cultural differences and their consequences: A triumph of faith a failure of analysis. Human Relations, 55(1),
89118.
Meyer, K. E. (2001). Institutions, transaction costs, and entry mode choice in Eastern Europe. Journal of International Business Studies, 32(2), 357367.
OGrady, S., & Lane, H. W. (1996). The psychic distance paradox. Journal of International Business Studies, 27(2), 309333.
Pan, Y. (2002). Equity ownership in international joint ventures: The impact of source country factors. Journal of International Business Studies, 33(2), 375384.
Razin, A. E. S., & Tong, H. (2005). Bilateral FDI ows: Threshold barriers and productivity shocks (No. 11639). Cambridge: National Bureau of Economic Research.
Sanchez-Peinado, E., Pla-Barber, J., & Hebert, L. (2007). Strategic variables that inuence entry mode choice in service rms. Journal of International Marketing,
15(1), 6791.
Shaver, J. M. (1998). Accounting for endogeneity when assessing strategy performance: Does entry mode choice affect FDI survival. Management Science, 44(4),
571585.
Shenkar, O. (2001). Cultural distance revisited: Towards a more rigorous conceptualization and measurement of cultural differences. Journal of International
Business Studies, 32(3), 519536.
Sousa, C. M. P., & Bradley, F. (2006). Cultural distance and psychic distance: Two peas in a pod? Journal of International Marketing, 14(1), 4970.
Stottinger, B., & Schlegelmilch, B. B. (1998). Explaining export development through psychic distance: Enlightening or elusive? International Marketing Review,
15(5), 357372.
Taras, V., & Steel, P. (2006). Improving cultural indices and rankings based on a meta-analysis of Hofstedes taxonomy. Beijing: Academy of International Business.
Tihanyi, L., Grifth, D. A., & Russell, C. J. (2005). The effect of cultural distance on entry mode choice, international diversication and MNE performance: A metaanalysis. Journal of International Business Studies, 36(3), 270283.
Weber, A. (1929). Theory of the location of industries. Chicago: The University of Chicago Press.
Xu, D., Pan, Y., & Beamish, P. (2004). The effect of regulative and normative distances on MNE ownership and expatriate strategies. Management International
Review, 44(3), 285307.
Xu, D., & Shenkar, O. (2002). Institutional distance and the multinational enterprise. Academy of Management Review, 27(4), 608618.
Zhao, H., Luo, Y., & Suh, T. (2004). Transaction cost determinants and ownership-based entry mode choice: A meta-analytic review. Journal of International Business
Studies, 35(6), 524544.