Master Thesis
The Effect of Psychic Distance on the Outflow
of FDI from Developing Markets
International Business & Management
by
Michiel Dijkstra – s2596415
m.h.dijkstra.2@student.rug.nl
University of Groningen
Faculty of Economics and Business
The Netherlands
Supervisor: Drs. J. van Polen
Co- Assessor: Dr. H.J. Drogendijk
January 2016
Page 3 of 47
The Effect of Psychic Distance on the Outflow
of FDI from Developing Markets
Michiel Dijkstra
Faculty of Economics and Business, University of Groningen, Groningen, the Netherlands
Abstract
This research examines the effect of psychic distance on the Outward Foreign Direct Investment (OFDI) in developing countries. The selected countries are Bulgaria and Serbia since Bulgaria has an EU-membership whereas Serbia is not an EU-member. These countries have been chosen because they share a rich history and multiple country characteristics. The analyses are based on OFDI flows from Bulgaria/Serbia to 56 countries over a time period from 2007-2012. The dimensions of psychic distance have been tested using a linear ordinary least squares (OLS) regression. The results do not show a significant relationship, however they are still interesting for future research. The results point out that the psychic distance concept cannot be used as a universal concept because developing countries tend to behave different, when choosing a market to invest in, than developed countries. Developing countries within the EU also tend to invest more in other EU countries than countries outside the EU, which is in line with the behaviour of developed countries within the EU.
Keywords: Outward Foreign Direct Investment; Psychic Distance; Developing
Page 4 of 47
Table of contents
1. Introduction ... 6
2. Theoretical Background ... 9
2.1 Psychic Distance and OFDI ... 9
2.2 European Union and the Balkan Countries ... 12
2.3 Hypothesis Development ... 14
3. Data Collection and Sample ... 18
3.1 Dependent Variable: Outflow FDI Bulgaria and Serbia ... 18
3.2 Independent Variable: Psychic Distance ... 19
3.3 Control Variables ... 21 4 Results ... 23 4.1 Descriptive Statistics ... 23 4.2 Robustness Checks ... 25 4.3 Correlations ... 26 4.4 Regression Analysis ... 28
5 Discussion & Conclusion ... 31
5.1 Limitations ... 32
5.2 Future Research ... 33
References ... 35
Appendix I: Measurement for Psychic Distance Stimuli ... 43
Appendix II: Sample Countries ... 44
Page 5 of 47
List of Acronyms
CPI Corruption Perception Indicator
EU European Union
FDI Foreign Direct Investment
GDP Gross Domestic Product
HDI Human Development Index
MNEs Multinational Enterprises
OFDI Outward Foreign Direct Investment
OLS Ordinary Least Squares
POLCON Political Constraint Index Dataset
UNCTAD United Nations Conference on Trade and Development
Page 6 of 47
1. Introduction
Markets are increasingly opening up to trade, technology and foreign direct
investment (FDI) flows offering corporations worldwide a broader range of choices on how to serve international markets (Dunning, 1999). A boom in FDI flows to
developing countries, points out that multinational enterprises (MNEs) have considered these developing countries to be profitable investment locations
(Nunnenkamp, 2002). Organizations often invest into unknown regions, in order to operate on an international level (Hofstede, 1980).
Cultures vary from country to country which causes different cultural
backgrounds (Hofstede, 2001). Differences between countries, in terms of cultural differences, affect international expansion because they increase the complexity of business transactions (Johanson & Vahlne, 1977; Ghemawat, 2001; Barkema et al., 1996; Nordström & Vahlne, 1992). It is therefore important for MNEs to understand how these cultural backgrounds might influence managerial decisions (House et al, 2004). Large distances in culture between companies will increase the risk of
misinterpretation, which in turn also increases the costs of understanding information flows (Boyacigiller, 1990). Boyacigiller (1990) indicated that these increases in
transaction costs will influence a manager’s awareness of attractiveness of doing business with a group of individuals. Therefore Kogut & Singh (1988) and Davidson (1980) state that these cultural differences between countries are predicted to influence managerial decisions such as country selection for FDI. Performance
studies show that greater distance in culture lowers the performance of MNEs in host countries, as foreign investments are affected by cultural differences (Barkema, Bell & Pennings, 1996; Benito, 1997; Craig, Green & Douglas, 2005; Li & Guisinger, 1992). Cultural distance is therefore considered as an important source of costs and challenges firms for operating internationally (Barkema, Bell & Pennings, 1996; Beugelsdijk, Slangen, Maseland & Onrust, 2014).
Page 7 of 47
the content and context can be interpreted differently among cultures, which can lead to inaccurate and invalid data (Hofstede, 1980; Shenkar, 2001). Sousa & Bradley (2006) therefore propose to use psychic distance instead. This because the concept of psychic distance includes a broader range of factors that may lead to distance between countries. Psychic distance is a multidimensional instrument designed by Dow & Karunarantna (2006), and defines the barrier that can exist between MNEs, including cultural factors (Sousa & Bradley, 2006). The concept of psychic distance has already been applied to several research areas, ranging from FDI to firm
performance (Benito & Gripsrud, 1992). Therefore, the main focus in this study is on psychic distance instead of culture.
Johanson & Vahlne (1977) and Johanson & Wiedersheim-Paul (1975) introduced the concept of psychic distance. They did this by studying the classic Uppsala internationalization process of firms (Johanson & Vahlne, 1977). Countries which are psychically more close to each other offer a more familiar operating
environment and are therefore more easily understood than countries which are less close to each other (Evans, Treadgold & Mavondo, 2000; O’Grady & Lane, 1996). Psychic distance can therefore explain the flow of information between firms (Håkanson, 2014). The more the costs of transferring and interpreting information increase, the more the psychic distance increases, which will in turn have a negative effect on the FDI flows between countries (Håkanson, 2014; Johanson &
Wiedersheim-Paul, 1975; Nordström & Vahlne, 1992). There have been several studies to describe the role of psychic distance as an explanatory factor for internationalization of MNEs (Barkema, Bell & Pennings, 1996; Cavusgil, 1980; Johanson & Vahlne, 1977). Barkema, Bell & Pennings (1996) suggest that further research should examine whether their conclusions are robust for using data on expanding firms from other home countries.
Page 8 of 47
and outside the EU, focused on developed countries (Schaap, 2015; Braconier, Ekholm & Knarvik, 2001; Pottelsberghe & Lichtenberg, 2001; Barkema, Bell & Pennink, 1996). There are numerous studies about the trends and drivers in OFDI. However, only a few studies empirically estimated the impact of these drivers on OFDI from developing countries (Saad, Noor & Nor, 2014; Blomkvist & Drogendijk, 2013). According to the conventional internationalization process firms from
developing markets enter new markets with successively greater psychic distance (Luo & Tung, 2007). Whereas firms from developed markets tend to start
internationalization in those markets they can most easily understand and where the perceived liabilities of foreignness are low (Johanson & Vahlne, 1977; Davidson, 1980). The explanatory value of psychic distance for understanding OFDI from developing countries is still unclear (Blomkvist & Drogendijk, 2013). Perhaps psychic distance may have been weakened because MNEs from developing countries are likely to differ to some extent in their internationalization process from developed countries (Child & Rodrigues, 2005; Cuervo-Cazurra & Genc, 2008). Sofar, no comparative study has been performed about the effect of psychic distance on OFDI in developing markets within the EU (Håkanson, 2014). Therefore, an interesting topic for research would be to examine the effect of psychic distance on OFDI in developing markets within the EU.
To gain insight on the effect psychic distance has on OFDI in developing countries in the EU, the focus of this research is on the effect of psychic distance on OFDI in developing countries in the EU. The aim of this research is to explore the effect of psychic distance on the internationalization process of MNEs in developing countries within the EU. And in turn how psychic distance influences the OFDI in these countries. This aim of this research can be summarized in the following research question:
‘What is the influence of psychic distance on OFDI from developing countries within the EU?’
Page 9 of 47
limitations, and implications for future research. Whereby an answer is given to the main research question and the sub research question. It also provides the
discussion based on the literature review and the data collection.
2. Theoretical Background
This chapter gives an overview of the relevant literature regarding the research question. Based on the literature about psychic distance and OFDI several
hypotheses are drawn to give an expectation of the psychic distance effect on OFDI.
2.1 Psychic Distance and OFDI
FDI is defined as “FDI is defined as investment that is made to acquire a lasting management interest (usually 10 percent of voting stock) in an enterprise operating in a country other than that of the investor (defined according to residency), the
investor’s purpose being an effective voice in the management of the enterprise. It is the sum of equity capital, reinvestment of earnings, other long-term capital, and
short-term capital as shown in the balance of payments” (Al-Sadig, 2009, p 289). The
direct investor's purpose is to exert a significant degree of influence on the
management of the enterprise resident in the other economy (International Monetary Fund, 1993). Direct investments comprises the initial transaction, establishing the relationship between the investor and the enterprise. It also comprises all subsequent transactions between the investor and the enterprise and among affiliated
enterprises, both incorporated and unincorporated (International Monetary Fund, 1993). Proponents of outward investments point out that OFDI enables firms to enter new markets, to import goods from foreign associates at lower costs (Johanson & Vahlne, 1977; Johanson & Wiedersheim-Paul, 1975; Weisfelder, 2001). Also firms are enabled to access foreign technology while the entire domestic economy benefits from OFDI due to the increased competitiveness of the investing companies and associated productivity spill overs to local firms (Johanson & Vahlne, 1977; Johanson & Wiedersheim-Paul, 1975; Weisfelder, 2001). Outward investments enables firms to enter new markets, import intermediate products at a lower cost and provide
Page 10 of 47
countries OFDI wave originated from 15 countries that contributed 81% of all OFDI from developing countries in 1980. OFDI stock from Brazil and Argentina were 13.4% and 20.4% respectively followed by the Asian new industrialised countries (Korea, Singapore, Taiwan, and Hong Kong) which together accounted for almost 22%. Other significant players are Malaysia, Brazil, Argentina, India and China (Dunning et al 1998; Dunning & Gugler, 2008).
The current literature on FDI from developing economies is based on theories and studies that aim to explain the motives and patterns of FDI in general (Saad, Noor & Nor, 2014). Dunning & Lundan (2008) and Buckley & Casson (1976) are explaining FDI by market imperfections. They proposed that multinationals must possess specific advantages over local firms in order to make up for higher
transactional costs and subsequently prosper in foreign markets. According to Kogut & Singh (1988) companies are increasingly involved in FDI and need to learn to familiarize to the idiosyncratic milieus of foreign markets places. Due to a lack of knowledge about foreign countries and to propensity to avoid uncertainty, companies start investing in countries that are comparatively similar and well-known with regard to business practices (Johanson & Wiedersheim-Paul, 1975).
FDI is an important source of economic growth and could be one of the connections between economic growth and business cycles (Borensztein, De Gregorio & Jong, 1998). However cultural differences start playing a role when companies are involved in FDI (Hofstede, 2001).For firms who get involved in FDI differences in culture become more important (Björkman & Forsgren, 1997). The most characteristic feature, for firms that get involved in FDI, is that they consist of subsidiaries located in foreign countries. Thus, these firms are located in different cultural milieus and people with different nationalities have to get along with each other (Hofstede, 1980). According to Adler (1986) therefore, misunderstandings are more likely to occur when people from different cultures work together. Cultural differences may become one of the most crucial problems when managing international firms (Hofstede, 1983).
Page 11 of 47
conceptualization and measurement (Shenkar, 2001). Cultural distance has unsupported hidden assumptions to questionable methodological properties, undermining the validity of the construct and challenging its theoretical role and application (Shenkar, 2001). Johanson & Wiedersheim-Paul (1975) propose that cultural distance is not the only exploratory factor that explains FDI, but they argue that psychic distance is a better proxy to explain FDI flows.
The concept of psychic distance in the internationalization literature has been described by the Uppsala internationalization theory. This theory proposes that companies choose markets based on psychic proximity (Brewer, 2007a). The Uppsala internationalisation theory is a theory that clarifies how firms gradually intensify their activities in markets outside their home market (Johanson &
Wiedersheim-Paul, 1975; Johanson & Vahlne, 1977). The key feature of the model is that firms first gain experience from the domestic market before they move to foreign markets. Firms start their foreign operations from culturally and/or geographically close countries and move gradually to culturally and geographically more distant countries (Johanson & Vahlne, 1990). Research has provided empirical support for both, developed and developing, markets for this internationalization theory
(Björkman & Forsgren, 1997; Medinets, Muchai & Odiyo, 2011). Besides that, Schlegelmilch & Stottinger (1998) and Petersen & Pedersen (1996) indicate that psychic distance is a result of supposed business differences between the firms’ home market and the market of the foreign country market. When the perceived differences are larger a country will be less likely to be selected. Therefore, psychic distance is a significant determinant to market selection. However, there exists a view of mixed evidence about the concept of psychic distance and its impact on OFDI. As Dow & Karunaratna (2006, p. 589) stated: ‘within the realm of International Business Research, psychic distance is one of the most commonly cited, yet vaguely
Page 12 of 47
2.2 European Union and the Balkan Countries
Benito, Grøgaard & Narula (2003) found concrete evidence that there are substantial benefits for MNEs operating within the EU, because they share standards1,
compared to those operating outside it. Being outside of the EU will carry the price of becoming less attractive to MNE activity (Benito et al., 2003). Since trade between EU members is free of tariffs and quotas (Head & Mayer, 2000), MNE activity is more likely to take place between countries within the EU than in countries operating
outside the EU (Benito et al., 2003). However, a member of the EU can experience a decline of its locational advantages (Benito et al., 2003). This is related with deep integration schemes, since the state must reorient its economy to the supra-regional norms. However, according to Benito et al. (2003) this is presumed to be
counterweighted by an industrial redistribution within an area, based on potential access to a larger unified market and a comparative advantage.
The Balkan countries (Greece, Albania, Macedonia, Bulgaria, Romania, Serbia, Montenegro, and Bosnia) experienced complex post-communist
transformations from the beginning of the last era of the past century (Bogomilova, 2005). These countries have many common features as a result of their shared history and similar transition experiences (Bogomilova, 2005). Most Balkan countries experienced high political and economic instability in the nineties, while economic recovery and transition related economic reforms have been generally slower than in Central Eastern Europe (Estrin & Uvalic, 2013). The countries evaluated in this study are Bulgaria and Serbia. Bulgaria joined the EU in 2007 whereas Serbia is not in the EU.
Bulgaria and Serbia share a rich history with each other (Forbes, 1915). There was a constant conflict between these countries about which part of their countries actually belonged to the other country (Detrez, 1996). Beside that Macedonia was also in both countries’ interests (Detrez, 1996). Hence, it is not strange that the Macedonian language is a mix of both languages (Lampe, 2000). Although both countries have a different language, they both share the South Slavic dialect (Lampe, 2000). Beside the share of the South Slavic dialect they also share similarities in
1
European Neighborhood Policies, information retrieved on 26 October 2015 from:
Page 13 of 47
religion. The majority of people from both countries are Orthodox Christians
(Bogomilova, 2005). Also according to Rovny (2014) Bulgaria and Serbia both had the same communist (patrimonial) regime type. Rovny (2014) found that both countries accept social liberal views and the economic left maintains populist-nationalist stances.
The countries are also quite similar to each other when comparing the
economic country characteristics with each other (table 1). This suggests that there is no clear hierarchy between those countries. However, Bulgaria has a locational advantage, regarding the other Balkan countries, because Bulgaria differs in the way in which their economical and industrial structures have been converged due to deep integration taking place within the EU. Bulgaria has been granted a EU membership status since 2007, whereas Serbia has not a EU membership status. However, Serbia is since March 2012 a candidate to become a member of the EU. Serbia is connected to the EU through the European Economic Area, which is an agreement between the EU and a few non-European countries2.
Table 1: Country characteristics Bulgaria and Serbia
2
Information retrieved from: http://www.efta.int/eea, accessed on 03-01-2016
a
CIA Fact Book (2015)
b UNCTAD database (2015) Bulgaria Serbia Population 2015 (million)a 7,18 7,18 Religion Orthodox 59.4%, Muslim 7.8%, Other 1.7% Orthodox 84.6% Catholic 5% Muslim 3.1% Other 2.9%
GDP per capita (PPP) in USD 2014a 17900 13300
Average OFDI flow 2007-2012 (million)ᵇ 282 279
Page 14 of 47
2.3 Hypothesis Development
Dow & Karunarantna (2006) developed a multidimensional instrument to measure psychic distance. They developed seven different psychic distant stimuli; culture, education, languages, industrial development, political systems, religion, and time zone. These stimuli are macro-level factors and they influence the manner in which people communicate and interpret information. The motive for these seven
dimensions will be explained below and hypotheses will be given.
As mentioned before, differences in national culture are the most widely acknowledged form of psychic distance stimulus (Boyacigiller, 1990; Johanson & Vahlne, 1977; Evans et al., 2000). Through the use of Hofstede's scales it has dominated empirical investigations for fifteen years (Kogut & Singh, 1988). The culture a person grew up in influences not only how he/she behaves, but also how he/she interprets information (Carlson, 1974). Large cultural distances between two groups of people will increase the cost of interpreting information flows between the groups (Boyacigiller, 1990). This will increase the risk of misunderstanding each other and thus increase the transaction costs (Boyacigiller, 1990). Johanson & Wiedersheim-Paul (1975) propose that these increases in transaction costs, will in turn influence a company’s perception of the attractiveness of doing business with a group of individuals. As such, cultural distance is expected to be an important
determinant of psychic distance. On the basis of the preceding discussion, it is argued that there is a positive relationship between cultural distance and psychic distance (Sousa & Bradley, 2006). It is on this basis that large cultural differences among countries are predicted to negatively influence FDI (Kogut & Singh, 1988; Davidson, 1980). Thus, hypothesis 1 is proposed as follows:
Hypothesis 1: The greater the differences in culture between countries, the lower the intensity of OFDI will be between these countries.
Page 15 of 47
levels in their empirical analyses is small (Vahlne & Wiedersheim-Paul, 1977; Dow & Karunaratha, 2006; Davidson and McFetridge, 1985; Kobrin, 1976). However, FDI levels are significantly higher in countries with higher levels of education (Akin & Vlad, 2011). Also, countries with low education levels do not have benefits from FDI investments (Borensztein et al., 1998). Akin & Vlad (2011) state that lower education is positively related to lower wages for unskilled workers and that this indicates that countries with low education levels have a lower level of competitiveness on the FDI market. Therefore it is assumed that differences in education levels will cause a lower OFDI. Thus, a hypothesis 2 is proposed as follows:
Hypothesis 2: The greater the differences in educational levels between countries, the lower the intensity of OFDI will be between these countries.
Language differences among countries are another psychic distance stimulus (Dow & Karunaratha, 2006). Efficiencies in communication are presented by
similarities in languages (Tushman, 1978). Differences in languages between markets tend to increase both the costs and the risks of a transaction (Dow & Karunaratna, 2006). As a result, language is a key component of psychic distance and influences FDI. Therefore, language differences influence international
expansion patterns and are a significant component of psychic distance (Welch et al., 2001). Thus, hypothesis 3 is proposed as follows:
Hypothesis 3: The greater the differences in languages between countries, the lower the intensity of OFDI will be between these countries.
Differences in the level of industrial development have a similar history to differences in languages within the psychic distance literature (Dow & Karunaratna, 2006). Technology diffusion plays a central role in the process of economic
development (Borensztein, De Gregorio & Jong, 1998). Business and communication norms in a developing economy are likely to be dramatically different from those of a highly industrialized economy (Medinets, Muchai & Odiyo, 2011). These differences result in extra uncertainties and costs into transactions and thus are likely to influence market selection decisions (Kobrin, 1976; Vahlne & Wiedersheim-Paul, 1977;
Page 16 of 47
companies in a foreign country will be affected by these differences in business practices and cultures (Dow & Karunaratna, 2006). However, in contrast to developed countries, Dunning, Hoesel & Narula (1997) found that developing countries do not seem to shy away from investing in higher industrial countries. Therefore, in contrast with previous psychic distance studies, it will be expected that differences in industrial development will positively impact the OFDI between
countries. Thus, hypothesis 4 is proposed as follows:
Hypothesis 4: The greater the differences in the degree of industrial development between countries, the greater the intensity of OFDI will be between these countries.
Differences in political systems represent barriers to the international transfer of information (Carlson, 1974). Thus, political systems have been selected as a potential psychic distance stimulus (Dow & Karunaratna, 2006). Differences in
political systems can possibly impact on managers at two levels (Dow & Karunaratna, 2006). First of all, most industries involve a substantial amount of
business-to-government and business-to-government-to-business communication (Dow & Karunaratna, 2006). The costs of uncertainty will be increased by differences in political systems (Dow & Karunaratna, 2006). Secondly, governments also play a key role in policing various business-to-business and business-to-consumer interactions, such as the enforcement of contracts and the monitoring of anti-competitive behaviour (Dow & Karunaratna, 2006). Both of these phenomena influence the choice of market selection because they increase the costs and risks of doing business in a country abroad. (Dow & Karunaratna, 2006). Therefore, when larger differences in political systems occur less FDI is expected. Thus, hypothesis 5 is proposed as follows:
Hypothesis 5: The greater the differences in political systems between countries, the lower the intensity of OFDI will be between these countries.
Differences in religion relate closely to cultural differences and are therefore a relevant determinant of psychic distance (Blomkvist & Drogendijk, 2013; Ronen & Shenkar, 1985). Differences in religion can lead to communication problems, as differences in cultural values do, and therefore increase the risks and costs of
Page 17 of 47
differences in points of opinion (Triandis, 2000). According to Blomkvist & Drogendijk (2013) religion can have a major influence on culture throughout history and
continues to affect people’s values, norms and behaviour today. Differences in religion will therefore increase the risk of misunderstandings and the cost of a
transaction, and are likely to reduce the trading intensity between countries (Triandis, 2000). According to Helbe (2007), Guiso, Sapienze & Zingales (2009) partners from the same religion communities are more likely to cooperate, therefore trade between countries is more likely when there are more religious similarities. Thus, it is expected that the more differences in religion between countries, the less the intensity in
outflow of FDI will be. Thus, hypothesis 6 is proposed as follows:
Hypothesis 6: The greater the differences in religions between countries, the lower the intensity of OFDI will be between these countries.
The last psychic distance stimulus is differences in time zones (Dow &
Karunaratna, 2006). Although advances in communication technology have reduced the costs of communication with people in other continents, the time zone differences remains a problem for managers attempting to span such regions. Unlike the
previous six factors, differences in time zones are not likely to disrupt the
interpretation of information, but they do create uncertainty about the ability for rapid communication (i.e., resolving an urgent problem), if and when it is needed (Dow & Karunaratna, 2006). Stein & Daude (2007) found that differences in time zones have a significant and negative effect on the decision for locating FDI. Therefore,
transaction costs are associated with time zone differences (Stein & Daude, 2007). Because of this Dow & Karunaratna (2006) included time zone differences in their model. Differences in time zones will increase the transaction costs and therefore less outflow of FDI is expected when the difference in time increases. Thus, hypothesis 7 is proposed as follows:
Hypothesis 7: The greater the differences in time zones between countries, the lower the intensity of OFDI will be between these countries.
Page 18 of 47
& Odiyo (2011) found that the same can be applied to developing countries. The greater the psychic distance between countries the more barriers for understanding information are implied (Dow & Karunaratna, 2006). Child et al. (2002) found that OFDI is negatively related to an increase in psychic distance. Therefore it is expected that the smaller the psychic distance between countries, the more likely the host country will select that market for investments and thus the OFDI will increase. Thus, hypothesis 8 is proposed as follows:
Hypothesis 8: The greater the differences in the aggregate of psychic distance between countries, the lower the intensity of OFDI will be between these countries.
3. Data Collection and Sample
The data for this study has been drawn from secondary data sources: United Nations Conference on Trade and Development (UNCTAD), The World Bank, and the
psychic distance scores from Dow & Karunaratna (2006). Analysing change requires at least two reference points in time (Saldana, 2003). Studies suggest that the most impropriate timeframe to collect data should be at least 10 years (Church, 2001; Green, Tull & Albaum, 1993). However, from the current existing developing countries within the EU, Bulgaria (and Romania) joined the EU first in 20073. As OFDI data is only available till 2012, the time span ranges from 2007 till 2012. Furthermore, data from Bulgaria/Serbia has been associated with 56 countries, for which almost full data is available at Dow (2011)4.
3.1 Dependent Variable: Outflow FDI Bulgaria and Serbia
The dependent variables in this study are the Bulgarian OFDI and the Serbian OFDI. Flows of FDI has been chosen because FDI flows, measure an interval of time, representing more reliability (Beugelsdijk et al., 2014). The mean of the OFDI from
3
Romania and Bulgaria are labeled as developing according to The World Bank. Information retrieved from http://data.worldbank.org/region/ECA, accessed on 01-11-2015.
4
Page 19 of 47
2007-2012 has been calculated which is, according to previous studies, more appropriate because then it is more likely to get a more robust variable (Slemrod, 1990; Blomkvist & Drogendijk, 2013). The data has been collected from UNCTAD.
3.2 Independent Variable: Psychic Distance
Psychic distance is the independent variable in this study. Numerous attempts have been made to develop a measurement of psychic distance between countries (Dow & Karunaratna, 2006; Clark & Pugh, 2001; Brewer, 2007b). Clark & Pugh (2001) made a four-single item independent indicator to explain the priority of countries. The level of economic development appears to be the most important driver of country priority. This instead of the market size, cultural distance, and geographical distance. These results confirm earlier findings by for example Nordström (1991). These studies however, included very simple measurements. It did also include a limited number of factors that influences psychic distance. More relevant, in this aspect, is the work by Dow & Karunaratna (2006).
For the practical use of psychic distance the multidimensional instrument of Dow & Karunaratna (2006) has been used. They developed this instrument based on seven different psychic distance stimuli: cultural distance, educational distance, language distance, industrialization distance, political distance, religion distance, and time zone difference. An overview of these stimuli can be found in appendix I.
To measure cultural distance the scores of Hofstede (1980) have been used. He developed six dimensions of national culture. However, because only full data is available for the four original dimensions (power distance, uncertainty avoidance, masculinity, and individuality), only these dimensions have been included in this study. The scores were available for 50 out of 56 countries, four scores (Congo, Algeria, Macedonia, and Bosnia) were retrieved by extending the scores of the available neighbour countries and using the mean of these scores. The scores for Kazakhstan and Uzbekistan were retrieved by earlier empirical work by Suanet & Van de Vijver (2009), who proposed to use Russia’s scores for Kazakhstan and Uzbekistan. The cultural dimension is tested twice, once with each of the four
Page 20 of 47
differences in the variance of each dimension and then arithmetically (Benito & Gripsrud, 1992). Algebraically, the Kogut & Singh (1988) index for cultural distance CDjis as follows
The next five stimuli (education, languages, industrial development, political systems, and religions) are measured by using 5 point Likert-scales. Whereas 5 stands for a large difference between the countries and 1 for hardly no difference. For all the five stimuli, the preceding indicators have been reduced to a single factor using confirmatory factor analysis (Dow, 2011).
The absolute differences were used for this single factor with regard to
education, industrialization and political systems (Dow, 2011). The measurements for these stimuli can be found in appendix I. The variables measured for educational differences and industrial differences are based on the data from United Nations (1995a, b). Furthermore, the political differences were measured based on five indicators, representing two different dimensions (Dow & Karunaratna, 2006). The first dimension is the difference of the degree of democracy between countries, which is represented by four variables. The first variable is the Political Constraint Index Dataset (POLCON) scale between countries. This data had been derived from Henisz’s (2000). The second variable is the difference in the Polity IV scale, which is derived from Gleditsch (2003). The final two variables (political rights and civil
liberties) are derived from Freedom House (2000).The second dimension are right-centre-left scales, which measure the ideological leanings or policy preferences of the decision-makers’ (Beck, Clarke, Groff, Keefer and Walsh’s, 2001).
Page 21 of 47
(2005) and data for the religion differences have been derived from Barrett (1982). Again the absolute differences were used for the single factor analysis (Dow, 2011).
For the last variable Dow & Karunaratna (2006) created a new variable by replacing geographic distance into time zone differences, thereby retaining the standardized residuals. The residuals represent a component of differences in time zones that is independent of distance. In order to distinguish the time zone effect from potentially related geographic distance, the method of Dow & Karunaratna (2006) has been followed. The time zone differences have been acquired by
calculating the difference between countries5 in terms of coordinated universal time (UTC). For the factor analysis absolute differences in UTC have been used (Dow, 2011). Finally, the psychic distance concept was calculated by correcting the individual seven stimuli scores with the Kogut & Singh (1988) formula for cultural distance (Blomkvist & Drogendijk, 2013).
3.3 Control Variables
The control variables used for this research are variables which are cited by scholars as important contributory factors for FDI (Slangen & Beugelsdijk, 2010; Hirvensalo & Hazley, 1998; Mayer, 1998; Ziacik, 2000). These control variables are: market size, trade openness, inflation, gross domestic product (GDP) growth, exchange rates, corruption, and human development index (HDI).
Following the work by Blomkvist & Drogendijk (2013) the control variables represent different motives or different types of OFDI: market asset seeking and resource seeking. To measure market asset seeking the market size in the host country has been used (Buckley et al. 2007). For the market size the population rather than the GDP of countries have been compared, because especially developing countries have relatively large populations compared with their GDP (Slangen & Beugelsdijk, 2010). For foreign MNEs the huge populations of these countries are an important reason to undertake horizontal and vertical activity (Khanna, 2007), making population size a better measure for host-market size than GDP. Therefore, the market size in this study has been controlled through the population size of each host country. This data has been collected from the World Bank Development Indicator for the years 2007–2012, using the mean of these years
5
Page 22 of 47
(Blomkvist & Drogendijk, 2013; Sethi et al., 2003; Habib & Zurawicki, 2002). In order to control for resource seeking FDI, the ratio of imports and exports between 2007 and 2012 has been used. The degree of openness of trade is controlled for each host countries by the sum of exports and imports as a percentage of GDP. Greater
openness of trade results in increased FDI activity (Blonigen & Davies, 2004).
Therefore, it is expected that the higher the openness of trade the more it will have a positive effect on OFDI (Al-Sadig, 2009). Data has been collected from the World Bank Development Indicator for the years 2007-2012, again using the mean over these years (Buckley et al. 2007; Blomkvist & Drogendijk, 2013).
OFDI has also been controlled for the effects of inflation, exchange rates, GDP growth, corruption, and HDI. Data for inflation was collected from IMF’s World
Economic Outlook, and data for exchange rates was collected from the World Bank Development Indicator6. The mean fluctuation was for both, inflation- and exchange rates, calculated to give a more robust value controlling for fluctuations (Blomkvist & Drogendijk, 2013). GDP growth is recognized as a significant determinant of FDI flows (Al-Sadig, 2009). The GDP growth rate in the host countries serves as a proxy for market growth to control the market potential and the host country’s market size (Al-Sadig, 2009; Seyoum, 2011). Data for GDP growth has been retrieved from the World Bank Development Indicator (Al-Sadig, 2009; Seyoum, 2011).
Al-Sadig (2009) found that corruption in the host country has a significant negative effect on FDI flows. The data for the corruption of each country has been retrieved from the Transparency International Corruption Perception Indicator (CPI). Although the CPI is not designed to allow for country scores to be compared over time7, it is the only source available which offers complete data for every country within the measured timeframe. Again all the data was collected for the years 2007– 2012, the mean was calculated to give a more robust value controlling for fluctuations (Blomkvist & Drogendijk, 2013).
6 The exchange rate is the country’s currency expressed in terms of the U.S. dollar. For the Slovakian
Republic and Estonia the average Euro/USD rate has been taken. Both countries switched currencies in relatively 2009 and 2010.
7 This is because the index draws on a country’s rank in the original data sources, rather than its
Page 23 of 47
According to Globerman & Shapiro (2003) and Peterson, Malhotra & Wagner (1999) the HDI is a significant determinant of FDI flows. Increased levels of physical infrastructure and human capital are positively related to FDI flows (Modi &
Srinivasan, 1998). Therefore, it is assumed that higher levels of HDI will attract more FDI. Data has been retrieved from the United Nations Development Programme (Seyoum, 2011). The data is scaled from 0 to 1, whereas 0 stands for a very low HDI and 1 for a very high HDI. There is only data available for the years 2008, 2010, 2011, and 2012, therefore the mean has been calculated over these years to get a more robust variable (Blomkvist & Drogendijk, 2013).
4 Results
This chapter provides an overview of the descriptive statistics. This will be followed by several robustness checks, a correlation matrix, and the results of the regression analysis.
4.1 Descriptive Statistics
Table 2, 3 and 4 represent the descriptive statistics for this study. Table 2 represents the ten countries with the most FDI flows. As can be seen, in the table, there is a more equal distribution of OFDI flows for Bulgaria than for Serbia. Table 3 (Bulgaria) and table 4 (Serbia) provides the descriptive statistics of this study, including the sample size, the mean, the minimum, the maximum, and the standard deviation. The sample consists of 56 countries, as OFDI data for more countries is not available. The selected countries can be found in appendix II. The mean exchange rate for Bulgaria is 0,97 while the mean for Serbia is 0,87, this indicates that Serbia’s
Page 24 of 47
Table 2: OFDI Flows Bulgaria/Serbia
Top ten countries by OFDI*
Bulgaria Serbia
1 Italy 37.04 Bosnia 174.22
2 Romania 31.96 Slovenia 28.52
3 Luxembourg 20.18 Macedonia 10.97
4 Germany 16.42 Netherlands 9.24
5 Serbia 14.35 United Kingdom 8.77
6 United States of America 12.64 Turkey 6.80
7 Austria 11.76 Bulgaria 5.07
8 United Kingdom 10.35 Croatia 4.33
9 Turkey 10.29 United States of America 3.89
10 Macedonia 10.06 Belgium 3.74
* Average outflow of FDI (in millions) over 2007-2012
Table 3: Descriptive Statistics Bulgaria
Sample Mean Minimum Maximum Std. Deviation
Dependent Variable OFDI Bulgaria 56 2.68 -20.99 37.04 9.65 Independent Variables Psychic Distance 56 3.08 0.46 7.55 1.57 Cultural Distance 56 1.31 0.02 4.84 1.21 Education Distance 56 0.50 0.03 1.66 0.35 Language Distance 56 0.28 0.00 0.53 0.15
Industrial Development Distance 56 0.54 0.01 1.45 0.38
Political System Distance 56 0.53 0.01 1.77 0.54
Page 25 of 47
Table 4: Descriptive Statistics Serbia
Sample Mean Minimum Maximum Std. Deviation
Dependent Variable OFDI Serbia 56 4.61 -6.14 174.22 23.53 Independent Variables Psychic Distance 56 3.92 0.83 9.97 1.86 Cultural Distance 56 1.81 0.02 6.67 1.66 Education Distance 56 0.55 0.01 1.45 0.35 Language Distance 56 0.27 0.00 0.53 0.15
Industrial Development Distance 56 0.65 0.01 1.91 0.49
Political System Distance 56 1.31 0.05 1.88 0.57
Religion Distance 56 0.65 0.01 1.27 0.26 Timezone Distance 56 2.32 0.00 10.00 2.79 Control Variables Market Size (x1000) 56 62812.27 504.00 1222160.00 167602.51 Openness of Trade 56 94.52 24.37 335.48 52.05 Inflation 56 5.01 -0.16 20.34 4.00 GDP Growth 56 2.47 -3.70 8.68 2.65 Exchange Rate 56 0.87 -5.96 12.80 3.48 Corruption 56 5.14 1.57 9.28 2.16 HDI 56 0.79 0.36 0.94 0.12
4.2 Robustness Checks
Several assumptions must be met before conclusions can be drawn from the
regression analysis (Berry, 1993). The first requirement is that the dependent and the independent variables should be measured at the continuous level, this requirement has been met. The second assumption is that a linear relationship between the dependent and independent variables is required. A scatterplot diagram has been composed to control for this linear relationship. As can be seen in appendix III, also this assumption has been met for both countries. The third assumption is that there should not be any significant outliers. To control for outliers a P-P plot has been made for both countries (see appendix III). The P-P plot for Serbia shows a slight S-shape distribution for this variable. This indicates that the normality assumption for psychic distance is not perfectly met. The most logical step in this case is to
transform the non-normal data. Though, no support could be found in the literature to transform the performance measure psychic distance, so the data limitation has been accepted. Furthermore, no significant outliers have been found.
Page 26 of 47
less than one or more than three indicates that there is a correlation between the residual terms. If the Durbin-Watson value is close to the mean value (2) the residuals are uncorrelated. The Durbin-Watson value for the OFDI of Bulgaria is 1.847 and for Serbia 2.033, which indicates that both residual terms are uncorrelated. The Durbin-Watson test can be found in appendix III. The fifth assumption is that the data should not show heteroscedasticity, but homoscedasticity instead. To check for homoscedasticity a scatterplot between standardized residuals and standardized predicted values has been used (appendix III). The scatterplot does not show clear homoscedasticity, but a clear pattern can be found.
The final assumption is that residuals should be normally distributed. To check for normal distribution, two histograms (appendix III) have been made. The histogram of Bulgaria is normally distributed, however the histogram of Serbia shows an edge peak distribution. According to Kumar & Phrommathed (2005) the outputs of many processes, perhaps even a majority of them, do not form normal distributions.
However, it does not advocate something is wrong with those processes, so this data limitation has been accepted.
To control for multicollinearity the variance inflation factors (VIF) have been calculated. According to Tabachnik & Fidell (2007) a common cut off point for VIF is five. The variables are tested on an individual level as well as together. The results showed that collinearity was present. In order to solve this, the control variables inflation, corruption, and HDI have been left out, the results can be found in Table 5. The highest VIF is 2.626, therefore it is not likely that multicollinearity is present. Since all criteria have been met, a linear regression analysis can be accumulated.
4.3 Correlations
The correlation matrix shows the correlation between the dependent, independent, and control variables. A Pearson correlation test has been performed to detect whether there is correlation among the implemented variables. The Pearson correlations can be found in the tables 6 and 7.
For Bulgaria, the results indicate that most variables show a negative
Page 27 of 47
Note: Dependent variable is the OFDI Bulgaria/Serbia
The results also indicate that all the individual psychic distance stimuli, except culture, have a significant positive correlation with psychic distance. This positive relationship is in line with previous research done by Blomkvist and Drogendijk (2013), who state that higher correlations are logically found between the psychic distance stimuli and the psychic distance index.
For Serbia, the psychic distance stimuli do not show a significant correlation with the OFDI. All individual psychic distance stimuli show a positive relationship with psychic distance. All the stimuli are significant positive correlated at a 1% confidence level. Again this result is in line with previous research done by Blomkvist and
Drogendijk (2013).
Table 5: Multicollinearity statistics for models 2 and 3
Model 2 Model 3
Bulgaria Serbia Bulgaria Serbia
Market Size (x1000) 1.262 1.278 1.552 1.598 Openness of Trade 1.099 1.099 1.224 1.358 GDP Growth 1.249 1.193 1.886 2.256 Exchange Rate 1.129 1.302 1.545 1.585 Psychic Distance 1.308 1.361 Cultural Distance 1.724 1.954 Education Distance 1.535 1.893 Language Distance 1.399 1.456
Industrial Development Distance 2.269 2.518
Political System Distance
1.761 2.626
Religion Distance 1.411 1.548
Page 28 of 47
Table 6: Pearson's Correlations Bulgaria
Bulgaria 1 2 3 4 5 6 7 8 9 1. OFDI 1 2. Psychic Distance -.212 1 3. Culture -.030 .152 1 4. Language -.130 .644** -.043 1 5. Religion .076 .496** .008 .105 1 6. Industrial Development -.047 .684 ** .369** .318* .226 1 7. Education -.182 .473** -.048 .206 -.059 .383** 1 8. Political System -.068 .278* -.339* .291* -.049 .091 .077 1 9. Time zone -.300* .480** .062 .288* .009 .366** .086 -.013 1
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
4.4 Regression Analysis
Multiple linear regressions have been carried out to find causal relationships with the OFDI of Bulgaria/Serbia. These regressions show whether the hypotheses can be supported. For both countries three models have been applied. Model one
represents the control variables, model two the control variables including psychic distance, and model three the control variables including the different psychic distance stimuli.
Table 7: Pearson's Correlations Serbia
Serbia 1 2 3 4 5 6 7 8 9 1. OFDI 1 2. Psychic Distance -.095 1 3. Culture -.133 .596** 1 4. Language .000 .410** -.124 1 5. Religion .053 .405** .031 .128 1 6. Industrial Development -.060 .727 ** .659** .004 .065 1 7. Education -.082 .368** .166 .063 -.236 .384** 1 8. Political System .083 .404** .501** -.298* .133 .509** .102 1 9. Time zone -.156 .360** -.056 .371** .048 -.053 -.076 -.172 1
Page 29 of 47
Table 8 (Bulgaria) and table 9 (Serbia) show the relationships between the different determinants and the OFDI flows. In model three the adjusted R2 for
Bulgaria is 5,1% which indicates a small increase in comparison with the R2 of model one (4,6%). Generally this would imply that the model predicts 5.1% of the variance of Bulgarian OFDI, however the F-value is not statistically significant. Therefore, no assumptions of the relation to the variance can be made. This could be explained by the relatively small sample size8. Model three of Serbia has a R2 value of -14.1%, which can be interpreted as 0% (Tabechnik & Fidell, 2007). The R2 value is again not significant.
For Bulgaria, model one shows that GDP growth is significant negatively (-0.362, p < 0,05) related to OFDI. The other control variables do not show significant relationships with OFDI. The GDP growth effect on OFDI is also supported in the other two models. In model two, the psychic distance variable shows a negative relationship to OFDI (-0.152), which is in line with hypothesis 8. However, the variable is not significant and therefore hypothesis 8 is not supported. Model three indicates that education distance (-0.304, p < 0,10) and industrial development distance (0.351, p < 0,10) have significant effects on OFDI. Both results are in line with hypothesis 2 and 4, therefore hypotheses 2 and 4 are supported. The other psychic distance stimuli variables do not show a significant relationship with OFDI, thus, hypotheses 1, 3, 5, 6, and 7 are not supported for Bulgaria.
For Serbia, model one do not indicate significant statistics for the control variables. In model two the psychic distance variable again shows a negative relationship (-0.145 ) to OFDI. However, the variable is not significant and therefore hypothesis 8 is not supported. Model three do not show significant relationships between the psychic distance stimuli and OFDI, therefore hypotheses 1, 2, 3, 4, 5, 6, and 7 are not supported for Serbia.
8
Page 30 of 47
Table 8: determinants of Bulgarian OFDI over 2007-2012 Bulgaria Model 1 Model 2 Model 3 Market Size (x1000) 0.071 (0.504) 0.119 (0.798) 0.151 (0.926) Openness of Trade 0.026 (0.187) 0.009 (0.066) -0.046 (-0.318) GDP Growth -0.362 (-2.568)* -0.319 (-2.171)* -0.374 (-2.074)* Exchange Rate 0.122 (0.880) 0.100 (0.718) -0.039 (-0.238) Psychic Distance -0.152 (-1.010) Cultural Distance -0.248 (-1.438) Education Distance -0.304 (-1.866)† Language Distance -0.049 (-0.318) Industrial Development Distance 0.351 (1.774)†
Political System Distance 0.044 (0.252)
Religion Distance -0.084 (-0.537)
Time zone Distance -0.282 (-1.652)
N 56 56 56
F-Value 1.664 1.536 1.267
Adjusted R2 0.046 0.046 0.051
Note: T-values in parentheses ; † p < 0.10; * p < 0.05
Table 9: determinants of Serbian OFDI over 2007-2012
Serbia Model 1 Model 2 Model 3
Market Size (x1000) -0.051 (-0.346) -0.005 (-0.031) -0.022 (-0.121) Openness of Trade -0.027 (-0.183) -0.041 (-0.283) -0.091 (-0.545) GDP Growth -0.061 (-0.410) -0.086 (-0.562) 0.037 (0.172) Exchange Rate -0.027 (-0.181) -0.080 (-0.505) -0.021 (-0.117) Psychic Distance -0.145 (-0.888) Cultural Distance -0.226 (-1.120) Education Distance -0.119 (-0.602) Language Distance 0.104 (0.599) Industrial Development Distance 0.037 (0.164)
Political System Distance 0.205 (0.878)
Religion Distance -0.003 (-0.014)
Time zone Distance -0.212 (-1.169)
N 56 56 56
F-Value 0.123 0.256 0.383
Adjusted R2 -0.068 -0.073 -0.141
Page 31 of 47
5 Discussion & Conclusion
This study examined the effects of psychic distance and psychic distance stimuli on OFDI in Europe from developing countries. The effect of psychic distance on OFDI has been compared between a EU-member country (Bulgaria) and a non-EU country (Serbia). The difference with previous studies is that both countries are from a
developing character (World Bank).
The aim of this study was to investigate the effect of psychic distance on the internationalization process of Bulgarian and Serbian MNEs and in turn influencing the OFDI of both countries. Former studies investigated, in a broad context, the psychic distance effect on the OFDI of developed countries. However, limited research has been conducted for the relationship between psychic distance and OFDI in developing countries. Blomkvist & Drogendijk (2013) researched the effect of psychic distance on the OFDI of China, but no research has been done about the effect of psychic distance on OFDI of developing European countries. Schaap (2015) found that the effect of psychic distance on OFDI significantly varies between
developed EU member countries and developed non-EU member countries. Again no research has been conducted on the difference between the effect of psychic distance on developing EU member countries and developing non-EU countries. This research made an effort to contribute to the literature by investigating the effect of psychic distance on the OFDI of a developing EU country (Bulgaria) and the OFDI of a developing non-EU country (Serbia). By investigating this, the following main research question can be answered:
‘What is the influence of psychic distance on OFDI from developing countries within the EU?’
A negative relationship between psychic distance and OFDI has been found for both countries. Moreover the EU country (Bulgaria) seems to be more affected by psychic distance than the non-EU country (Serbia). This result is line with the
Page 32 of 47
for Serbia, so no conclusions can be drawn. However, Blomkvist & Drogendijk (2013) found in their research that a statistically negative relationship between psychic distance and OFDI exist in the case of China. A larger sample size might result in statistically significant results for both countries. If more information comes available, future research should examine whether the effect of psychic distance is greater on OFDI of developing EU countries than on OFDI of developing non-EU countries.
The data also suggests numerous other conclusions about the effects of psychic distance on OFDI in developing countries. Most psychic distance stimuli factors do not indicate significant direct relationships to OFDI. Although language differences and political differences do not show a significant direct relationship to OFDI, the results are surprising. As proposed in hypotheses 3 and 5, a negative relationship between language differences and political differences, and OFDI was expected. This result could be in line with the findings of Dunning, Hoesel, & Narula (1997) who found that other developing countries (respectively Korea and Taiwan) do not seem to shy away from investing in developed countries which are not relatively close to their countries. Blomkvist & Drogendijk (2013) also found that, in China, language differences do not have a negative effect on OFDI. They suggested to explore the role of English, as a stimulus in host countries, in relationship with FDI. Since, in practice, English is used more as a business language in cross-border communications than local national languages. This avenue might be interesting for future research.
5.1 Limitations
Page 33 of 47
whereas Church (2001), and Green, Tull & Albaum (1993) suggest to use a ten year time period. Because there was only data available till 2012 and Bulgaria joined the EU since 2007, no bigger time frame could be taken to control for differences between developing EU-countries and developing non EU-countries. Though, by extending the time period from 2003 to 2012, the results on the influence of psychic distance on OFDI in developing countries could have been improved.
Thirdly, 23% of the total Bulgarian OFDI flows are unspecified, indicating that there is no data available to which countries these flows have gone to. Including this data might lead to more robust results. Fourthly, Tabachnik & Fidell (2007)
recommend, as a rule of thumb, that a regression model with “m” predictors requires a sample size greater than 50 + 8 * m, for testing the overall model. When applying this rule of thumb to this research, a minimum sample size of 122 was needed. This research does not even have half of this sample size. However, more data was not available, which is in line with the general issue data issue of developing countries. Developing countries have in general less data available than developed countries (Harrison, 1996). A bigger sample size could have increased the validity of the
results. Finally, previous research has discussed that national cultures are not stable and may change over time (Shenkar, 2001; Tung & Verbeke, 2010), this could be a limitation of this study as well. However, Van Hoorn, Maseland & Beugelsdijk (2013) debated that if national cultures change, they change in the same direction. Thus, cultural change is assumed not to influence the findings of this study.
5.2 Future Research
There are several issues that could be addressed in future research. First of all, as mentioned in the limitations, a specified OFDI stock per country is for Serbia not available yet. When this information becomes available it would be interesting to take this as a dependent variable rather than FDI flows to get less spurious results
Page 34 of 47
(China) language differences had a significant positive effect on OFDI (Blomkvist & Drogendijk, 2013). This is in contradiction with the theory of Dow & Karunaratna (2006), who state that language differences should have a negative effect of OFDI. It would be interesting to explore the role of English as a stimulus in host countries in relationship with FDI, since in practice English is used more as a business language in cross-border communications than local national languages (Blomkvist &
Page 35 of 47
References
Adler, N. (1986). International Dimension of Organizational Behavior. Kent Publications, Boston, MA.
Al-Sadig, A. (2009). The Effects of Corruption on FDI Inflows. Cato Journal, 29(2). Akin, M.S. & Vlad, V. (2011). The Relationship between Education and Foreign Direct
Investment: Testing the Inverse U Shape. European Journal of Economic and Political
Studies, 4(1).
Barkema, H. G., Bell, J. H. & Pennings, J. M. (1996). Foreign entry, cultural barriers, and learning. Strategic Management Journal, 17(2): 151-166.
Barrett, D. B (2002)., World Christian Encyclopaedia, CIA Fact book
Beck, T., Clarke, G., Groff, A., Keefer, P., & Walsh, P. (2001). New tools in comparative political economy: The Database of Political Institutions. the world bank economic review, 15(1), 165-176.
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): 461-476.
Benito, G. R. (1997). Divestment of foreign production operations. Applied Economics,
29(10): 1365-1377.
Benito, G.R.G., Grøgaard, B. & Narula, R. (2003). Environmental Influences on MNE Subsidiary Roles: Economic Integration and the Nordic Countries. Journal of International Business Studies, 34(5): 443-456.
Berry, W.D. (1993). Understanding regression assumptions. Sage university paper series on quantitative applications in the social sciences, 07-092. Newbury Park, CA: Sage. Beugelsdijk, S., Slangen, A., Maseland, R. & Onrust, M. (2014). The impact of home-host
cultural distance of foreign affiliate sales: The moderating role of cultural variation within host countries. Journal of Business Research, 26: 1638-1646.
Björkman, I. & Forsgren, M. (1997). The Nature of the International Firm: Nordic
Contributions to International Business Research. Copenhagen: Handelshøjskolens forlag.
Blomkvist, K., & Drogendijk, R. (2013). The impact of psychic distance on Chinese outward foreign direct investments. Management International Review,53(5), 659-686.
Page 36 of 47
Bogomilova, N. (2005). The religious situation in contemporary Bulgaria, and in Serbia and Montenegro: differences and similarities. Occasional Papers on Religion in Eastern Europe, 25(4), 1.
Borensztein, E., De Gregorio, J. & Jong, W.L. (1998). ‘How does Foreign Direct Investment affect Economic Growth?’, Journal of International Economics, 45:115-35.
Boyacigiller, N. (1990). ‘The role of expatriates in the management of interdependence, complexity and risk in multinational corporations’. Journal of International Business Studies 21(3): 357-381.
Braconier, H., Ekholm, K. & Knarvik, K. M. (2001). ‘Does FDI Work as a Channel for R&D Spillovers? Evidence based on Swedish Data’, Review of World Economics, 137(4): 644-65.
Brewer, P. A. (2007a). Psychic distance and Australian export market selection. Australian Journal of Management, 32(1): 73-94.
Brewer, P. A. (2007b), "Operationalizing Psychic Distance: A Revised Approach", Journal of International Marketing, Vol.15, No. 1, pp. 44-66.
Buckley P J and Casson M (1976) The Future of the Multinational Enterprise, London: Buckley, P. J., Clegg, L. J., Cross, A. R., Liu, X., Voss H., & Zheng, P. (2007). The
determinants of Chinese outward foreign direct investment. Journal of International Business Studies, 38(4), 499–518.
Carlson, S. (1974) 'International transmission of information and the business firm', Annals of the American Academy of Political and Social Science 412: 55-63.
Cavusgil, S.T. (1980). ‘On the internationalization process of firms’. European Research. 8: 273-281.
Child, J., & Rodrigues, S. (2005). The internationalization of Chinese firms: A case for theoretical revision? Management and Organization Review, 1(3), 381–410.
Church, R.M. (2001). The Effective Use of Secondary Data. Learning and Motivation, 33: 32-45.
Clark, T. and Pugh, D.S. (2001), “Foreign country priorities in the internationalization process: a measure and an exploratory test on British firms”, International Business Review, 10 (3), 285–303.
Page 37 of 47
Cuervo-Cazurra, A., & Genc, M. (2008). Transforming disadvantages into advantages: Developing-country MNEs in the least developed countries. Journal of International Business Studies, 39(6), 957–979.
Culem, C.G. (1988). ‘The locational determinants of direct investments among industrialized countries’, European Economic Review 32: 885-904.
Davidson, W.H. (1980). ‘The location of foreign direct investment activity: country
characteristics and experience effects’. Journal of International Business Studies 11(2): 9-22.
Davidson, W.H. & McFetridge, D.G. (1985). ‘Key characteristics of the choice of international technology transfer mode’. Journal of International Business Studies. 16(2): 5-22. Detrez, R. (1996). De sloop van Joegoslavië. Relaas van een boedelscheiding. Publisher
Hadewijch
Dow, D. & Karunaratna, A. (2006). Developing a multidimensional instrument to measure psychic distance stimuli. Journal of International Business Studies, 37(5): 578-602. Drogendijk, R., & Martin, O. M. (2008). Country Distance: an Objective Measure and its Impact on International Market Selection. Academy of International Business 2008 Dunning, J.H. (1988). Explaining International Production, Unwin Hyman: London.
Dunning, J. H. (1999). Globalization and the Theory of MNE Activity. University of Reading, Department of Economics.
Dunning, J. H., van Hoesel, R., & Narula, R. (1996). Explaining the ‘new’ wave of outward FDI from developing countries: the case of Taiwan and Korea’.Research Memoranda, 9, 1-25.
Dunning, J.H. & Lundan, S. (2008). Multinational Enterprises and the Global Economy. 2nd ed. Cheltenham: Edward Elgar.
Estrin, S., & Uvalic, M. (2013). Foreign direct investment into transition economies: Are the Balkans different?. LEQS Paper, (64).
Evans, J., Treadgold, A. and Mavondo, F. (2000) 'Explaining export development through psychic distance', International Marketing Review 17(2): 164-168.
Forbes, N. (1915). The Balkans: A History of Bulgaria, Serbia, Greece, Rumania, Turkey. Clarendon Press.
Page 38 of 47
Ghemawat. P. (2001). Distance still matters: The hard reality of global expansion. Harvard Business Review, 79(8), 137–146.
Gordon, R. G. (ed), Ethnologue: Languages of the World, 2005 Gleditsch, K. S. (2003) 'Modified Polity P4 and P4D Data',
http://www.systemicpeace.org/polity/polity4.htm (accessed 16 November 2015). Habib, M., & Zurawicki, L. 2002. Corruption and foreign direct investment. Journal of
International Business Studies, 33(2): 291–307.
Harrison, A. (1996). Openness and growth: A time-series, cross-country analysis for developing countries. Journal of development Economics, 48(2), 419-447.
Håkanson, L. (2014). The role of psychic distance in international trade: a longitudinal analysis. International Marketing Review, 31(3): 210-236.
Head, K., & Mayer, T. (2000). Non-Europe: the magnitude and causes of market fragmentation in the EU. Weltwirtschaftliches Archiv, 136(2), 284-314.
Henisz, W. J. (2000) 'The Institutional environment for Economic Growth', Economics and Politics 12(1): 1-31.
Hirvensalo, I. – Hazley, C. (1998). Barriers to Foreign Direct Investments in the Baltic Sea Region. The Research Institute of the Finnish Economy, Helsinki.
Hofstede, G. (1980). Cultures’ Consequences: International Differences in Work-Related Values. Beverly Hills, CA: Sage Publications.
Hofstede, G. (2001). Cultures consequences: Comparing values, behaviors, institutions and organizations across nations (2nd ed.). Thousand Oaks: Sage Publications.
Hofstede, G. (1983). The Cultural Relativity of Organizational Practices an Theories. Journal of International Business Studies: 75-89.
House R.J., Hanges, P.J., Javidan, M., Dorfman, P.W., Gupta, V. (2004) Culture,
Leadership, and Organizations: The GLOBE Study of 62 Societies. Thousand Oaks, CA: Sage.
International Monetary Fund (1993). Balance of Payments Manual 5th Edition, Washington DC.
Jackson, S. & Markowski, S. (1996). ‘The attractiveness of countries to foreign direct investors’, Australian Journal of Management, 21(2): 113-138.
Johanson, J. & Vahlne, J. E. (1977). The internationalization process of the firm: A model of knowledge development and increasing foreign market commitments. Journal of
