The Effect of Hofstede’s Fifth and Sixth Dimension on Foreign Direct Investment and Export: a Country Analysis Concerning the United States and Spain

The Effect of Hofstede’s Fifth and Sixth Dimension

on Foreign Direct Investment and Export: a Country

Analysis Concerning the United States and Spain

July 7

, 2011

Master’s Thesis by Iman Haghighirad Student number: 1486438

Supervisor Master Thesis: dr. M.S.S. Krammer Second Supervisor Master Thesis: dr. S.R. Gubbi

Supervisor Methodology: dr. P. Rao Sahib Email: e.haghighirad@student.rug.nl

Contents

Abstract 2

Chapter 1: Introduction 3

Chapter 2: Literature Review 6

2.1 Culture 6

2.2 Cultural distance 6

2.3 Hofstede’s dimensions: from four to six 8

2.3.1 The four-dimensional model 8

2.3.2 The fifth dimension: long-term orientation 10

2.3.3 The sixth dimension: indulgence versus restraint 12

2.3.4 The United States and Spain 13

Chapter 3: Methodology 15

3.1 Sample and data collection 15

3.2 Dependent variables 16

3.3 Independent variables 17

3.4 Control variables 18

3.5 Statistical model 20

Chapter 4: Results 23

Chapter 5: Conclusions and Limitations 24

5.1 Conclusions 26

5.2 Limitations 28

References 28

Abstract

This study investigated the relation of Hofstede’s fifth dimension long-term orientation and his recently launched sixth dimension indulgence versus restraint on outward FDI flows and export flows. Pooled OLS regressions and random effect regression models have been employed to show this relation for respectively the United States and Spain to 89 recipient countries for the years 2002 to 2009. The results are mixed, showing that the fifth dimension is negative and significantly related to FDI and export for Spain, but is positive and significant for the United States. The sixth dimension is found to be negative and significantly related to FDI for both the United States and Spain, however for export the same relation is only found for the United States.

Chapter 1: Introduction

Economic activity between countries has been mainly described through various types of entry modes such as foreign direct investment (FDI) or the classical trade theories. The basic principles of describing these economic activities date back to the 1930s where Richard Coase developed the transaction cost theory. In his book The Nature of the Firm Coase (1937) considers how firms should handle exchange costs (e.g. coordination cost, influence cost and agency cost) in the process of economic activity. For approximately three decades Coase’s article on transaction cost theory did not receive much attention (Klaes, 2000). The renascence of the transaction theory commenced from the 1970s through influential articles by inter alia Arrow (1973) and Williamson (1975). With the insights delivered by these authors, the basics of the transaction cost theory had been extensively researched.

With the birth of Hofstede’s culture indices as a practical tool, new studies were launched in examining the role of cultural distance1 in economic activity (e.g. Kogut and Singh, 1988; Brouthers and Brouthers, 2001; Cho and Padmanabhan, 2005). The majority of these studies attempted to explain how cultural distance could influence the entry mode decision of a firm (i.e. Mergers & Acquisitions (M&A), Greenfield and Joint Ventures (JV)) by utilizing Hofstede’s cultural dimensions. Other studies applied these dimensions to grasp the relationship between cultural distance and export2 (Karunaratna, Johnson and Rao, 2001); however this has been far less covered by research than cultural distance and its effect on FDI.

Notwithstanding, the great amount of studies, Hofstede acknowledged novo gaps for cultural dimensions to investigate. Hitherto, the majority of these studies designed their research around the first four dimensions of Hofstede. However, the collaboration of Hofstede with Michael Bond and Michael Minkov, led to the confirmation of two fairly new dimensions to its reputable dimension matrix. Already in the early stages of his career, Hofstede emphasized that his dimensions provide useful proxies to describe what culture is about, but Hofstede does not perceive his dimensions as being fixed and tangible pillars which demarcate what culture is. Hofstede sees cultures as constructs, which are subject to one’s perception and can be approximated through the usage of his dimensions. With this ideology, Hofstede was able to conveniently extent his work. Hence, Hofstede added respectively the fifth and sixth dimension: long-term orientation (Hofstede and Bond, 1988) and indulgence versus restraint (Minkov, 2007).

These two dimensions are currently representing a workable tool with data from nearly hundred countries. The reason for adducing these fairly novo cultural dimensions can be attributed to the fact that the last two dimensions have not been investigated extensively for their effects on FDI and export. Moreover, Minkov and Hofstede (2011) conclude that “the recent expansion and update of Hofstede‟s doctrine is indebted to the original groundbreaking work of the 1970s yet a key strength of Hofstede‟s work has been its

1

Cultural distance can be perceived as the difference in cultural practice between societies (countries).

2

ability to adapt and remain progressive”. With this powerful finding they imply that these latter two dimensions should have earlier been added since they also play a significant role in explaining cultural differences or similarities. Due to the late adoption of fifth and sixth dimension, they have not been explicitly investigated so far. Therefore, the fifth and sixth dimensions leave gaps to investigate and serve as basic motivation to carry out this study. Hence, the research question of this study is as follows: “Do the cultural dimensions „long-term orientation‟ and „indulgence versus restraint‟ relate to FDI and export?” More specifically, this article examines the relation of long-term orientation distance and indulgence versus restraint distance on FDI and export, controlling for secondary education, common official language, colonial linkage,

geographical distance, regulatory quality, rule of law distance andcontrol of corruption. Moreover, the first four dimension power distance, individualism, masculinity and uncertainty avoidance will be additively used as robustness checks. In doing so, this article contributes to the literature of cultural distance and its relation with economic activity (FDI and export) by researching a novo dimension (indulgence versus restraint) and further investigation on the fifth dimension. Besides the fact of investigating these dimensions, this article is unique in attributing the relation of cultural distance on FDI and export in a single study. Aside from showing these results, the relative influence of the fifth and sixth dimension will be shown for FDI and for export. The results will be given using pooled OLS regressions and random effect regression models from 2002 to 2009, by taking United States and Spain as the countries of origin. These two countries will be used to investigate the relationship of the fifth and sixth dimension on outward FDI flows and export flows to 89 recipient countries around the world.

Chapter 2: Literature Review

2.1 Culture

Before Hofstede’s cultural dimensions will be considered, this article advances forward by providing an explanation of what culture is about. The word culture stems from the Latin word Cultura, which refers to the “the tilling of land” or “cultivating, agriculture” (Harper, Douglas, 2001). It is hardly possible to precisely define culture. The Concise Oxford English Dictionary (2008) offers the next translation of culture: “The customs, ideas and social behaviour of a particular people or group.” Next to this, many scholars have endeavored to provide semantics for culture, which tend to be varying per source. According to Linton (1945) "A culture is a configuration of learned behaviors and results of behavior whose component elements are shared and transmitted by the members of a particular society" (p. 32). Another cultural anthropological view defines culture “as that complex whole which includes knowledge, beliefs, arts, morals, law, custom, and any other capabilities and habits acquired by man as a member of society” (Tylor 1871/1958 p. 1) in Kottak (2006). Apart from these semantics there is a wide scale of other scholars who provide their peculiar definitions. Thus, one can say there is no straightforward consensus regarding the exact semantics of culture. The main reason for this is that culture is to a certain degree a matter of perception, since perceptions differ on an individual basis, semantics concerning culture become intangible and arduous to define demarcated. On the other hand, the core purport of the semantics originating from different sources roughly point out to the same direction. However, for culture to be a practical tool to work with, an apparatus is required to propound the influence of culture.

2.2 Cultural distance

source, and therefore gain the preference above other possible sources3. Moreover, the content of the same dimensions by GLOBE differ substantially from that of Hofstede’s work (Hofstede, 2010).

Even though Hofstede’s cultural dimensions have been widely used by scholars, Hofstede’s dimensions have been under scrutiny since he developed his model. Du, Lu and Tao (2006) argue that the cultural dimensions of Hofstede being rigid since they do not change over time. In contrast, Kotter and Heskett (1992; p. 12) perceive culture as a concept which does not change over short timeframes, but takes on a more steady process, which matches this article’s point of view. This has been empirically supported by Barkema and Vermeulen (1997) in where they test that Hofstede’s dimensions are still valid. Furthermore, one can question the accuracy of comparing cultures with each other (Frenkel, 2008).However, I find myself in line with Hofstede’s vision on cultures; where Hofstede explains that cultures belong to the same type of food like fruit, but can be divided in different sorts e.g. apples and oranges (Hofstede, 1980). This basically means that the world exists of humans with different cultures, but that these cultures can actually be compared with each other. With these comprehensions, studies have used Hofstede’s dimensions to test for its relation on FDI and export.

The studies involving Hofstede’s dimensions have been generally used for two distinct purposes: psychic distance and cultural distance. Psychic distance captures a holistic term of which cultural distance is part of (Swift, 1998). Johanson and Wiedersheim-Paul (1975) and Johanson and Vahlne (1977) have constructed the building blocks for psychic distance. According to them, psychic distance exists when certain factors can impede the exchange of information flows between firm and market. They allocate components like education, language, economic situation, political and legal regulation to form the phenomena psychic distance. Evans (2002) also describes that psychic is a complex term containing facets of cultural and business differences formed by a thought process depending on one’s perception. In other words, psychic distance is a broad term which

3

depends also on one’s notion about what psychic distance can be. This article instead utilizes the variables like education, language and institutional proxies as control variables, which allows this research to deliver independent results for the fifth and sixth dimension.

Cultural distance as opposed to psychic distance captures specific differences in cultural dimensions between societies (Benito and Gripsgud, 1992; Kogut and Singh, 1988). In the course of the studies on cultural distance there have been generally two modes of measuring cultural distance. Kogut and Singh (1988) developed a mechanism to capture the four dimensions of Hofstede into a composite variable of cultural distance. The other type of measurement focuses on isolated dimensions of cultural distance, for example uncertainty avoidance or power distance. These latter two dimensions have been used to justify the individual importance of the single dimensions on the entry mode decision of a firm (Eramilli, 1996; Evans, 2002; Quer, Claver and Rienda, 2007; Shenkar, 2001). This article will not utilize Kogut and Singh’s composite index, because cross-cultural studies that deal with culture as a package, including numerous variables, makes it arduous to observe which of the variables in the package might be explaining a given outcome (Singelis, Bond, Sharkey and Lai, (1999). Hence, this study rather utilizes the differences between Hofstede’s single dimensions to explain cultural distance. By doing this, the isolated relations of long-term orientation distance and indulgence versus restraint distance dimensions can be properly mapped.

2.3 Hofstede’s dimensions: from four to six

2.3.1 The four-dimensional model

assigned. The scores provide an indication of the relative differences in culture between countries (societies) specified per dimension (and ranging from zero to hundred). The first four cultural dimensions have been defined in his book Culture‟s Consequences (1980), and connote:

1. Power distance: “the extent to which the less powerful members of institutions and organizations accept that power is distributed unequally”.

2. Uncertainty avoidance: “the extent to which people feel threatened by ambiguous situations, and have created beliefs and institutions that try to avoid these”

3. Individualism versus collectivism: individualism can be described as “a situation in which people are supposed to look after themselves and their immediate family only”. Collectivism explains “a situation in which people belong to in-groups or collectivities which are supposed to look after them in exchange for loyalty”.

4. Masculinity versus femininity: Masculinity is “a situation in which the dominant values in society are success, money and things”. Femininity is the opposite and means “a situation in which the dominant values in society are caring for others and the quality of life”.

distance) are factors that impede efficient negotiations and contracts. This idea is supported by Kayak and Kothari (1984), where they mention that cultural barriers form problems in the decision making of the exporter to export their product or services. Moreover, they concluded that cultural distance between countries affects the drive of the exporter’s country positively but has negative effects on the performance of it. In a study by Shoham and Albaum (1995) on export barriers, they find that cultural distance is the strongest variable in explaining the significance of export barriers. Guo (2004) investigates how culture influences foreign trade between United States and China, and finds that cultural differences matter significantly in explaining trade between countries. Guo (2004) in particular uses religion and language as proxies for culture. Thus, cultural distance is a characteristic that influences the successfulness for both FDI and export.

The majority of the research has been focused on the four-dimensional model by Hofstede, however roughly a decade after Hofstede published his four-dimensional model, the fifth cultural dimension was on its way to be introduced followed by a sixth dimension two decades later.

2.3.2 The fifth dimension: long-term orientation

world. The result was that three dimensions of the CVS again appeared to be significant; however the fourth dimension did not correlate with the IBM dimension. This dimension, uncertainty avoidance was seemingly different from the IBM dimensions, where values concerning future, past and present orientation became present. Hofstede gathered the pieces together and concluded with the establishment of the fifth dimension, which he called:

5. Long-term orientation: “Long-term oriented societies foster pragmatic virtues oriented towards future rewards, in particular saving, persistence, and adapting to changing circumstances. Short-term oriented societies foster virtues related to the past and present such as national pride, respect for tradition, preservation of "face", and fulfilling social obligations” (Hofstede and Bond, 1988).

However, the study resembled merely 23 countries. Approximately sixteen years later, Michael Minkov continued the study and brought forward a new analysis for the fifth dimension. His research was based upon the World Values Survey (WVS) data, where he utilized representative national population samples of 93 countries around the world. Minkov concluded that long-term orientation is correlated with “certain family values, with school results, with business values, with environmental values and, under favorable historical conditions and with economic growth” (Hofstede and Minkov, 2010).

cultural distant countries in terms of the dimension long-term orientation distance are negatively related to financial performance and strategic effectiveness. Furthermore, Jiang, Chen and Liu (2009) invoked upon the research done by the GLOBE to investigate if FDI affects the Chinese culture, which is a reverse study compared to this study. Jiang et al. (2009) used GLOBE which on their turn extended Hofstede’s work of the dimension long-term orientation. In doing this, they divided long-term orientation into three different pillars, which are: human orientation, future orientation and performance orientation. Mixed results have been found for the effect of FDI on the Chinese culture. However, their questionnaires are based on different items and therefore capture different outcomes which can not be one-on-one compared with Hofstede (Minkov and Hofstede, 2011). As regard to export, Zhang, Tamer Cavusgil and Roath (2003) discourse that long-term orientation distacnce as a cultural dimension is negatively related to export, but do not test the direct relationship of this dimension with export since their research is focused on relational norms. In contrast, Karunaratna, Johnson and Rao (2001) discuss that reducing inefficiencies in the dimensions long-term orientation distance will be beneficial for economic activity. To reduce inefficiencies the exporter has to grasp the importance of cultural differences between nations, however they did not test their hypotheses. All in all, it is fairly difficult to find articles which used Hofstede’s fifth dimension in a direct relationship with both FDI and export.

2.3.3 The sixth dimension: indulgence versus restraint

Organizations: Software of the Mind, Hofstede and Minkov established scores for 93 countries of which the scores have no equivalence with the other five dimensions (Minkov and Hofstede, 2011). Hence, the sixth dimension was realized and called:

6. Indulgence versus restraint: “Indulgence stands for a society that allows relatively free gratification of basic and natural human drives related to enjoying life and having fun. Restraint stands for a society that suppresses gratification of needs and regulates it by means of strict social norms” (Minkov and Hofstede, 2011).

High scores on this dimension means that societies are more “indulgent” and find aspects related to enjoyment and happiness important, whereas for “restraint” societies the opposite holds. Since, Hofstede has recently launched the sixth dimension; I have not been able to find articles that address this particular dimension. There are articles which utilize certain aspects that relate to the sixth dimension, e.g. Tadesse and White (2010). They utilize measures to compare cultural differences (between the United States and 75 receiving countries) in norms and values and find that increasing cultural differences diminish state-level export to the receiving country. They use two main proxies for culture, which are secular-rational authority and survival versus self-expression values. Especially, the latter component is interesting, since it is a component that is related to indulgence versus restraint. However, studies utilizing Hofstede’s exact sixth dimension are not present to my knowledge.

2.3.4 The United States and Spain

reasonably high on the dimension indulgence versus restraint. The United States can be seen as a country that perceives issues related to the past and the present as being important and therefore are relatively more short-term orientated. Furthermore, the United States is a society that puts high priority to natural human drives related to happiness and hence can be perceived as an “indulgent” society.

Table 1

Hofstede’s scores of the fifth and sixth dimension concerning the United States, Spain and total sample average. Long-term versus short-term orientation Indulgence versus restraint United States 26 68 Spain 48 44

Average of total sample 46 45

Spain, by contrast scores for both dimensions just under the 50, which makes Spain standing somewhere between long-term and short-term orientation, and between indulgence and restraint. However, since Spain scores under the 50 they nag more towards being a short-term orientated country and based on their score Spain is slightly more “restrained” than “indulgent”. Appendix 1 shows Hofstede’s scores on all his six dimensions for the concerning 89 countries. When comparing for example the scores of Spain with that of Mexico, one can see that they differ substantially on each dimension. On the other side, the United States seems to be rather in line with Great Britain’s culture. Knowing that the United States and Spain have common grounds with these countries in language and are linked colonial wise with other countries, these examples make the these countries even more intriguing from a “country point of view”. Therefore, the next hypotheses have been constructed to precisely define the expected outcome, and are as follows:

Hypothesis 1: Long-term orientation distance is negatively related to outward FDI flows.

Hypothesis 2: Long-term orientation distance is negatively related to export flows.

Hypothesis 4: Indulgence versus restraint distance is negatively related to export flows.

The reason for expecting these outcomes is that countries with the similar scores on the particular dimension are expected to be favorable countries for FDI and export. A society that is more long-term orientated rather than short-term orientated can find problems in terms of working together. The long-term orientated culture might be willing to invest more in a certain project to achieve in a later stadium more profits, where the short-term orientated society rather wants short-term investments to achieve quick profits. A similar rational holds for indulgence versus restraint distance. A restraint society that suppresses room for enjoying life, might ask more effort of an employer or employee in terms of work hours than an “indulgent” society, which can lead to tensions between firms. Large differences in culture can therefore create irregularities between cultures (Barkema and Vermeulen, 1997).

Chapter 3: Methodology 3.1 Sample and data collection

To investigate the hypotheses, country variables have been used in pooled OLS regressions and random effect estimator regressions from 2002 to 2009. Making use of this particular time constraint, allows this research to have a strongly balanced longitudinal data set. The current dimension matrix of Hofstede provides data for 111 countries; however not for each country all six dimensions have fully assigned scores. Some countries have merely data for two dimensions, whereas others provide data four or more dimensions. The data set for the fifth dimension and the sixth dimension covers 89 countries4. The scores assigned per dimension for each country can be found in Appendix 1. Next to the cultural dimensions of Hostede, data on FDI and export will be measured by using respectively outward FDI flows and export flows from the database of the

4 _{The countries involved in this research are: Albania, Algeria, Andorra, Argentina, Armenia, Australia, Austria, Azerbaijan,}

Organisation for Economic Co-operation and Development (OECD). Furthermore, gross domestic product (GDP) can be captured through the World Development Indicators (WDI). Subsequently, outward FDI flows and export flows will be calculated as a proportion of GDP (FDI/GDP and EXP/GDP). The effects of long-term orientation distance (LTO) and indulgence versus restraint distance (IVR) on FDI/GDP and EXP/GDP are then controlled for by using the following control variables: secondary education (SECED), common official language (CLANG), colonial linkage (COLON), geographical distance (GEOD), and proxies for institutional distance. Data on geographical distance (GEOD), common official language (CLANG) and colonial linkage (COLON) is made available through the Centre d'Etudes Prospectives et d'Informations Internationales (CEPII). Next to GDP, the data concerning secondary education (SECED) can be found on WDI. The proxies for institutional distance consist of regulatory quality (RQ), rule of law distance (RL), and control of corruption (CC) and can be acquired through the World Governance Indicators (WGI). Finally, robustness checks will be conducted by adding Hofstede’s first four dimensions power distance (PDI), individualism (IDV), masculinity (MAS) and uncertainty avoidance (UAI) to the main regressions as control variable5.

3.2 Dependent variables

There are two dependent variables: outward FDI flows divided by GDP (FDI/GDP) and export flows divided by GDP (EXP/GDP). By using ratios, it can be determined to what extent outward FDI flows and export flows take on a significant part of the economy size (Navaretti and Venables, 2004). This is clearly illustrated by Table 2, where the United States has relatively higher FDI and export ratios compared to Spain. Most articles partition FDI into different types e.g. M&A, Greenfield and JV (Kogut and Singh 1988; Cho and Padmanabhan, 2005; Quer et al., 2007). However, this article seeks to comprehend whether there is a significant relationship present of the fifth and sixth dimensions on FDI and export as a whole.

5

3.3 Independent variables

The independent variables for this research are long-term orientation distance (LTO) and indulgence versus restraint distance (IVR). In line with Shenkar’s study in 2001 on the usage of isolated dimensions, I will consider the fifth and sixth dimensions separately. The distance has been depicted by calculating the absolute difference between the countries of origin (the United States and Spain) and the 89 recipient countries. The independent variables are scores deducted from Hofstede’s dimension matrix. The scores used by Hofstede are relative scores between societies. A change in the culture of one’s society automatically forces to change the level of the scores of the other societies due to their relative relationship. The scores in Hofstede’s dimension matrix have been thoroughly investigated and progress reliable and stable outcomes (Kogut and Singh, 1988). Table 2 provides next to the mean, standard deviation, minimum and maximum value also the expected signs for the relations between long-term orientation distance

Table 2

Descriptive statistics of the United States and Spain for both FDI and export.

Variable Expected Signs Mean Std. Dev. Min Max

(LTO) and indulgence versus restraint distance (IVR) on FDI and export. The expected signs contain the expected type of relations (positive or negative) between the variables and FDI and export.

3.4 Control variables

“healthy” a country can be perceived, which is more inviting for countries to start a FDI or export to that particular country. The main difference of the institutional proxies with Hofstede’s cultural dimensions is that WGI has been capturing and assigning annually specific scores for each of the institutional variables, and will be used as such. Finally, the additional control variables power distance (PDI), individualism (IDV), masculinity (MAS) and uncertainty avoidance (UAI) serve as robustness checks. All the elements mentioned above will added in the next section to form the statistical model.

3.5 Statistical model

The purpose of this study is to show whether there exists a relationship between the dimensions long-term orientation distance (LTO) and indulgence versus restraint distance (IVR) and the dependent variables outward FDI flows and export flows. In order to obtain the results, the next log-linear multiple regression will be used:

Log (Economic Activityxijt / GDPit) = β0 + β1 *Long-term Orientation Distanceij + β2 *

Indulgence Versus Restraint Distanceij + β3 * Geographical Distanceij + β4 * Secondary

Educationit + β5 * Common Official Languageij + β6 * Common Colonizerij + β7 *

Regulatory Qualityit + β8 * Rule of Law Distanceit + β9 * Control of Corruptionit +

ε

(1)

Where i stands for the recipient country, j for either United States or Spain, x stands for the type of economic activity (FDI or export) and t depicts whether a time dimension is involved. Since economic activityxijt consists of FDI or export, equation 1 represents four

effects in this scenario, the variables long-term orientation distance (LTO), indulgence versus restraint distance (IVR), secondary education (SECED), common official language (CLANG), colonial linkage (COLON) and geographical distance (GEOD) will be dropped out of the model. Therefore, in the latter case random effects will as well be utilized, because it still approaches the results as close as possible even when being subject to some biasness. Therefore, the usage of the pooled OLS regression can serve as a useful benchmark and an extra robustness check. Moreover, the Breusch-Pagan test for random effects confirmed that for each regression random effects are present. Intuitively this makes sense, since Hofstede’s scores for each dimension contain fixed values that do not change over time (Wilkingson et al. 2008). Nevertheless, when the random effect regression model is shown to be best estimator, it should be seen as the most appropriate regression form and has to be considered as the leading regression to answer the hypotheses. Otherwise, the pooled OLS regression is a good benchmark for the outcome. However, before the results are going to be provided, the data will be discussed on its statistical validity. First of all, the data has been checked on its distribution. The Jarque-Bera test for normality showed that the dependent variables were subject to skewness6. To overcome this problem the independent variable has been transformed by a logarithm, which makes the model log-linear and provides a smoother distribution. Consequently, the variances have been observed for hetereoskedasticity by using a Breush-Pagan and Cook-Weisberg test. For each regression the results are convincingly robust. Finally, Table 2 contains the correlation matrix of the variables concerning the regressions. The results of the correlation matrix reveal a few high correlations taking place within the regressions. The outcome is not surprising considering the fact that three proxies for institutions have been used. WGI itself puts forward these variables with the notion that the in between correlation of the variables is very much present. The proximity of questionnaires for each of these proxies is quite high, which causes partly the high correlated errors. As regards to the rest of the correlation, these are seemingly weaker correlated. For example, it is intuitively likely that common official language (CLANG) and colonial linkage (COLON) are to certain extent correlated with each other.

Ultimately, the variables used in these regressions are variables that are based on large samples, which contain a widespread representation of diverse topics. According to Hofstede, long-term orientation distance (LTO) correlates with educational results (which is confirmed) and indulgence versus restraint distance (IVR) correlates with human rights (e.g. freedom to express one’s opinion). Between long-term orientation distance (LTO) and indulgence versus restraint distance (IVR) there is some correlation detectable, but the correlation is fairly low. As regards to the rest of Hofstede’s variables, the correlation between power distance (PDI) and individualism (IDV) separates itself from the rest with a correlation of 0.567 (significant at a 1% level). Even though there are correlations detectable, and on forehand to be expected, I find it highly important to include these specific variables in order to provide a broad picture of the relationship between Hofstede’s fifth and sixth dimension and the corresponding dependent variables.

Chapter 4: Results

contrast, finds only significant variables for regulatory quality (RQ) and rule of law distance (RL) at a 10% level. As regards to Spain (model 2a and 2b) the results are different compared to the United States. The coefficient of long-term orientation distance (LTO) is negative and significant for both the pooled OLS regression as the random effect regression model, but for the variable indulgence versus restraint distance (IVR) it is not found to be significant. Moreover, geographical distance (GEOD) appears to be highly significant in both model 2a and 2b, but has a low negative effect on FDI/GDP (-0.000). Other significant variables are colonial linkage (COLON) and regulatory quality (RQ) for model 2a and for model 2b only regulatory quality (RQ). In addition, it has to be

Table 4

Regression results: the relation of long-term orientation distance and indulgence versus restraint distance on FDI/GDP. Model 1a OLS US Model 1b RE US Model 2a OLS Spain Model 2b RE Spain Dependent variable: FDI/GDP

Constant -3.233*** (0.658) -4.108*** (1.312) -20.07*** (0.931) -20.56*** (1.853) LTO 0.008 (0.005) 0.011 (0.012) -0.031*** (0.005) -0.032*** (0.011) IVR -0.246*** (0.007) -0.031** (0.015) 0.004 (0.006) 0.101 (0.118) SECED -0.313*** (0.095) -0.220 (0.189) 0.164 (0.129) 0.207 (0.252) CLANG 0.506** (0.220) 0.650 (0.489) 0.456 (0.629) 0.063 (0.140) COLON -0.413 (0.358) -0.216 (0.839) 1.373** (0.606) 1.253 (1.101) GEOD -0.000*** (0.000) -0.000 (0.000) -0.000*** (0.000) -0.000*** (0.000) RQ 0.915*** (0.264) 0.651* (0.361) 0.735** (0.374) 1.055* (0.561) RL -0.423 (0.366) -0.819* (0.448) 0.135 (0.519) -0.028 (0.665) CC -0.068 (0.309) 0.398 (0.361) 0.384 (0.438) -0.637 (1.853) n 439 439 286 286 df 9 9 9 9 Chi Sqc 25.64 30.26 Signif Chi Sq 0.002 0.000 R2 _{0.208 0.304} F 13.65 13.99 P 0.000 0.000 Adj r2 _{0.206 0.291}

a _{The results are from the pooled OLS regressions with robust standard errors parentheses beneath the}

coefficients.

b _{The results are from the Gaussian random effect estimator with robust standard errors in parentheses}

beneath the coefficient.

c

mentioned that all coefficients of model 1and 2 which appear to be significant in Table 4, are in accordance with the expected signs of Table 2.

Table 5 provides the results for EXP/GDP. Model 3a illustrates the pooled OLS regression results and model 3b the random effect regression model for the United States. Unlike the results of model 1a, long-term orientation distance (LTO) in model 3a does find significant results for export for the United States. However, the sign of the coefficient is positive. Model 3b (the random effect regression model), does not find enough evidence to support that long-term orientation distance relates significantly to

Table 5

Regression results: the relation of long-term orientation distance and indulgence versus restraint distance on EXP/GDP. Model 3a OLS US Model 3b RE US Model 4a OLS Spain Model 4b RE Spain Dependent variable: EXP/GDP

Constant -0.482* (0.280) -2.605*** (0.480) -4.679*** (0.250) -5.155*** (0.462) LTO 0.005** (0.002) 0.004 (0.006) -0.013*** (0.002) -0.006 (0.004) IVR -0.011*** (0.003) -0.019*** (0.007) -0.008*** (0.002) -0.004 (0.004) SECED -0.523*** (0.040) -0.145*** (0.054) 0.091*** (0.034) 0.115** (0.051) CLANG 0.329*** (0.092) 0.491* (0.267) 0.806*** (0.162) 0.816* (0.465) COLON -0.469*** (0.161) -0.251 (0.470) 0.276* (0.153) 0.273 (0.444) GEOD -0.000*** (0.000) -0.000 (0.000) -0.000*** (0.000) -0.000*** (0.000) RQ 0.895*** (0.104) 0.205** (0.088) 0.399*** (0.087) -0.141* (0.084) RL -0.579*** (0.147) -0.285*** (0.106) 0.311** (0.125) 0.321*** (0.099) CC -0.082 (0.124) 0.058 (0.087) -0.183* (0.100) 0.089 (0.080) n 683 683 684 684 df 9 9 9 9 Chi Sq 35.62 107.49 Signif Chi Sq 0.000 0.000 R2 _{0.271 0.531} F 48.03 102.83 P 0.000 0.000 Adj r2 0.383 0.573

a _{The results are from the pooled OLS regressions with robust standard errors parentheses beneath the}

coefficients.

b _{The results are from the Gaussian random effect estimator with robust standard errors in parentheses}

beneath the coefficient.

c

EXP/GDP. Furthermore, all variables in model 3a except control of corruption (CC) are significant. Model 3b finds significant results for secondary education (SECED), regulatory quality (RQ) and rule of law distance (RL). Model 4 contains the results for Spain on EXP/GDP. The relation of long-term orientation distance (LTO) and indulgence versus restraint distance (IVR) with EXP/GDP is negative and significantly supported by model 4a.Model 4b does not find significant evidence to support these relationships. Furthermore, all other single variables in model 4a appear to be significant. In the case of the random effect regression model (4b), the variables that are significant are: secondary education (SECED), common official language (CLANG), geographical distance (GEOD), regulatory quality (RQ) and rule of law distance (RL). Again, the sign of the coefficients are in accordance with the expected signs of Table 2. Finally, robustness checks including all six dimensions of Hofstede can be found in appendix 3 and 4. In terms of coefficient signs and significance levels, Appendices 3 and 4 resemble fairly similar results with Table 4 and 5.

Chapter 5: Conclusions and Limitations

5.1 Conclusions

Spain for both FDI and export, but the opposite holds for the United States. Taking into account the robustness checks of Appendices 3 and 4 for the United States there is enough evidence to say that the dimension long-term orientation distance of the United States is positively related to FDI and export. Multiple factors can account for the positive relation between long-term orientation distance and FDI and export. First of all, FDI is taken on an aggregate scale, where the distinction between M&A, Greenfield (horizontal or vertical FDI) is not defined. Previous studies found different results when distinguishing between different types of FDI (Barkema and Vermeulen 1997; López-Duarte and Vidal-Suárez, 2010; Siegel et al., 2010). Furthermore, Shenkar and Zeira (1992), describe in their article that Hofstede’s cultural dimensions can be complementary to each other. Empirically this is traced back when looking at the results on the first four dimensions from articles by Barkema and Vermeulen, 1997 and Siegel et al. 2010. The individual attitude of one society can be accessory to the collectivistic attitude of the other society. In terms of long-term orientation distance, it might be possible, that the United States rather engages with countries that are culturally different from the United States in term orientation. The score of the United States on long-term orientation is 26 against the average sample score of 46, meaning that the United States is fairly short-term orientated compared to the rest of the sample. Being short-term orientated as a country, means caring about norms and values related to the past and the present, for instance respect for traditional habits. It can be the case that firms in the United States might feel triggered to collaborate with firms of other societies that are more long-term orientated, which are societies that consider norms and values related to the future, like savings and a flexible attitude as important aspects. These aspects can be inviting for the United States in establishing a plant or engaging in export with another country. Hence, the results are mixed for hypothesis 1 and 2. However, the results show that there is more evidence to support hypotheses 1 and 2 for Spain than for the United States, because the p-values for Spain are all significant at a 1% level, and for the United States at a 5% level. When not looking at the signs, it can be concluded that there is evidence found that long-term orientation distance is related to FDI and export for both the United States and Spain.

As regards to the sixth dimension, hypotheses 3 and 4 have been found to be significantly supported for the United States and partially for Spain. For the United States, indulgence versus restraint distance is negatively related to both FDI and export. For Spain indulgence versus restraint distance is found to have a significant and negative relation with export (the same holds for the robustness checks). For FDI, there is not enough evidence found to say that there is a relation between indulgence versus restraint distance and FDI. The fact that no significant results were found might be due the fact that FDI was not defined in different components as has been explained above. Hence, hypothesis 3 has been significantly supported for the United States, whereas hypothesis 4 has been found to be significant for both the United States and Spain.

Conclusively, the answer to the main question of this research is that long-term orientation distance relates to FDI and export and that for indulgence versus restraint distance partial evidence has been found that this dimension relates to FDI and export. Even though the results are mixed for this fifth dimension in terms of their coefficient signs, there is evidence found that there is a significant relation between long-term orientation distance and FDI and export. In the case of indulgence versus restraint distance, the results for both the United States and Spain point out a negative and significant relation with export. For FDI, significant results have been only found the United States. Interestingly, the variables long-term orientation distance and indulgence versus restraint distance are on average stronger related to FDI and export than Hofstede’s primary dimensions, due to higher p-values and stronger coefficients.

5.2 Limitations

outward FDI flows and export. It might be interesting to also look at inward FDI flows and import. This study aimed at investigating whether the fifth and sixth dimension in fact relate to FDI and export. With the knowledge from this study, research can be extended in different forms.

References

Arrow, K.J., 1973. Information and Economic Behavior. MA: Harvard University Press (1984), 4, 386-405.

Barkema, H.G. and F. Vermeulen., 1997. What Differences in the Cultural Backgrounds of Partners Are Detrimental for International Joint Ventures? Journal of International Business Studies, 28(4), 845-864.

Benito, G.B.R. and G. Gripsrud., 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.

Blomstrom, M., R. Lipsey, and M. Zejan, 1992, What explains growth in developing countries? NBER Working Paper, 4132, 1-31.

Borensztein, E., J. De Gregorio and J-W. Lee., 1998. How does foreign direct investment affect economic growth? Journal of International Economics, 45, 115–135.

Brakman, S. and C. van Marrewijk., 2008. It’s a big world after all: on the economic impact of location and distance. Cambridge Journal of Regions, Economy and Society, 1, 411–437.

Brouthers, K., and E. Brouthers, 2001. Explaining the national cultural distance paradox. Journal of International Business Studies, 32(1), 177–189.

Cho, K. R., and P. Padmanabhan, 2005. Revisiting the role of cultural distance in MNC’s foreign ownership mode choice: The moderating effect of experience attributes.

International Business Review, 14(3), 307–324.

Coase, R.H., 1937. The Nature of the firm. Economica N.S, 4, 386-405.

Concise Oxford English Dictionary, 2008. In: Oxford University Press. 11th ed. Dictionary. United Kingdom.

Eichengreen, B., Y. Rhee and H. Tong., 2004. The Impact of China on the Exports of Other Asian Countries. NBER Working Paper Series, 10768, 1-34.

Evans, J., and F.T. Mavondo, 2002. Psychic Distance and Organizational Performance: An Empirical Examination of International Retailing Operations. Journal of International Business Studies, 33(3), 515-532.

Frenkel, M., 2008. The Multinational Corporation as a third space: Rethinking international management discourse on knowledge transfer through Homi Bhabha. Academy of Management Review, 33(4), 924-942.

Guo, R., 2004. How culture influences foreign trade: evidence from the U.S. and China. The Journal of Socio-Economics, 33, 785–812.

Harper, D., 2010. Online Etymology Dictionary. Carlisle: Dickinson College. Available from: http://www.etymonline.com [Accessed 2001]

Hofstede, G., 1980. Geert Hofstede Personnel Website. Available from: http://www.geerthofstede.nl/

Hofstede, G., 1980. Culture’s Consequences: International Differences in Work Related Values. Sage Publications, 5, 1-327.

Hofstede, G., 1991, Cultures and Organizations: Software of the Mind. McGraw-Hill, NewYork, NY, 1-279.

Hofstede, G., 2010. The GLOBE debate: Back to relevance. Journal of International Business Studies, 41(8), 1339-46.

Hofstede, G. and M. Minkov, 2010. Long- versus short-term orientation: new perspectives. Asia Pacific Business Review, 16(4), 493-504.

Hofstede, G. and M.H. Bond, 1988.The Confucius Connection: From Cultural Roots to Economic Growth. Organizational Dynamics, 16 (4), 4-21.

Inglehart, R. and W.E. Baker, 2000). Modernization, cultural change, and the persistence of traditional values, American Sociological Review,65(1), 19-51.

Jiang, L., Q. Chen. and Y. Liu, 2009. FDI and Change of the Chinese Culture. International Journal of Social Economics, 37 (2), 101-118.

Johanson, J. and F. Wiedersheim-Paul., 1975. The Internationalization of the Firm: Four Swedish Cases. Journal of Management Studies, 12(3), 305-322.

Karunaratna, A.M., L. W. Johnson and C.P. Rao., 2001. The Exporter - Import Agent Contract and the Influence of Cultural Dimensions. Journal of Marketing Management, 17,137-158.

Kaufmann, D., A. Kraay and M. Mastruzzi., 2010. The Worldwide Governance

Indicators : A Summary of Methodology, Data and Analytical Issues. World Bank Policy Research. 5430, 1-28.

Kaynak, E. and V. Kothari., 1984. Export behavior of small manufacturers: a

comparative study of American and Canadian firms. European Management Journal, 41-47.

Klaes, M., 2000. The History of the Concept of Transaction: Neglected Aspects. Journal of the History of Economic Thought, 22 (2), 192-216.

Kogut, B. and H. Singh., 1988. The Effect of National Culture on the Choice of Entry Mode. Journal of International Business Studies, 19 (3), 411-432.

Kostova, T. and K. Roth., 2002. Adoption of an Organizational Practice by Subsidiaries of Multinational Corporations: Institutional and Relational Effects. The Academy of Management Journal, 45(1), 215-233.

Kottak, C.P., 2006. Athropology: The Exploration of Human Diversity. New York: McGraw-Hill, 11, 1-271.

Kotter, J.P. and J.L. Heskett., 1992. Corporate Culture and Performance. The Free Press,(12)1-205.

Linton, R., 1945. The Cultural Background of Personality. Appleton Century Crofts Inc, 1, 157.

Lohmann, J., 2010. Do language barriers affect trade? Economics Letters, 110, 159-162. López-Duarte, C. and M.M. Vidal-Suárez., 2010. External uncertainty and entry mode choice: Cultural distance, political risk and language diversity. International Business Review, 19, 575–588.

Mayer, T. and S. Zignago., 2006. Notes on CEPII’s distances measures.

Minkov, M. and G. Hofstede., 2011 The evolution of Hofstede's doctrine. Cross Cultural Management: An International Journal, 18(1), 10-20.

Navaretti, G.B. and A.J. Venables, 2004. Multinational Firms in the World Economy. Princeton University Press, 1-325.

Quer, D., E. Claver and L. Rienda, 2007. The impact of country risk and cultural distance on entry mode choice: An integrated approach. Cross Cultural Management: An

International Journal, 14(1), 74-87.

Shenkar, O., 2001. Cultural Distance Revisited: Towards a More Rigorous

Conceptualization and Measurement of Cultural Differences. Journal of International Business Studies, 32 (3), 519-535.

Shenkar, O and Y. Zeira., 1992. Role conflict and role ambiguity of chief executive officers in international joint ventures. Journal of International Business Studies, 23(1), 55-75.

Shoham A., G.S. Albaum, 1995. Reducing the Impact of Barriers to Exporting: A Managerial Perspective. Journal of International Marketing, 3(4), 85-105.

Siegel, J.I., A.N. Licht and S.H. Schwartz., 2010. Egalitarianism, Cultural Distance, and FDI: A New Approach. The Berkeley Electronic Press, 18 (133), 1-23.

Singelis, T.M., M.H. Bond, W.F. Sharkey, and C.S.L. Lai, 1999, Unpackaging culture’s influence on self-esteem and embarassability: the role of self-construals. Journal of Cross-cultural Psychology, 30(3), 315-341.

Suárez-Ortega, S., 2003. Export Barriers: Insights from Small and Medium-sized Firms. International Small Business Journal, 21 (4), 403-419.

Swift, J.S., 1998. Cultural Closeness as a Facet of Cultural Affinity: A Contribution to the Theory of Psychic Distance. International Marketing Review, 16(3), 182-201. Tadesse, B., and R. White, 2010. Cultural distance as a determinant of bilateral trade flows: do immigrants counter the effect of cultural differences? Applied Economics Letters, 17(2), 147-152.

Wilkinson, T.J., G.Z. Peng., L.E. Brouthers. and P.W. Beamish, 2008. The diminishing Effect of Cultural Distance on Subsidiary Control. Journal of International Management, 14, 93-107.

Zhang, C., S. Tamer Cavusgil. and A.S. Roath, 2003. Manufacturer Governance of Foreign Distributor Relationships: Do Relational Norms Enhance Competitiveness in the Export Market? Journal of International Business Studies, 34(6), 550-566.

Appendices

Appendix 1

Hofstede’s scores on his six dimensions

Countries Power

distance

Iraq 25 17 Ireland 28 70 68 35 24 65 Israel 13 54 47 81 38 Italy 50 76 70 75 61 30 Japan 54 46 95 92 88 42 Jordan 16 43 Korea South 60 18 39 85 100 29 Kyrgyz Rep 66 39 Latvia 44 70 9 63 69 13 Lithuania 42 60 19 65 82 16 Luxembourg 40 60 50 70 64 56 Macedonia Rep 62 35 Malaysia 104 26 50 36 41 57 Mali 20 43 Malta 56 59 47 96 47 66 Mexico 81 30 69 82 24 97 Moldova 71 19 Montenegro 75 20 Morocco 70 46 53 68 14 25 Netherlands 38 80 14 53 67 68 New Zealand 22 79 58 49 33 75 Nigeria 13 84 Norway 31 69 8 50 35 55 Pakistan 55 14 50 70 50 0 Peru 64 16 42 87 25 46 Philippines 94 32 64 44 27 42 Poland 68 60 64 93 38 29 Portugal 63 27 31 104 28 33 Romania 90 30 42 90 52 20 Russia 93 39 36 95 81 20 Rwanda 18 37 Saudi Arabia 36 52 Serbia 86 25 43 92 52 28 Singapore 74 20 48 8 72 46 Slovak Rep 104 52 110 51 77 28 Slovenia 71 27 19 88 49 48 South Africa 34 63 Spain 57 51 42 86 48 44 Sweden 31 71 5 29 53 78 Switzerland 34 68 70 58 74 66 Tanzania 34 38 Thailand 64 20 34 64 32 45

Trinidad and Tobago 47 16 58 55 13 80

Zambia 30 42

Zimbabwe 15 28

Source: http://www.geerthofstede.nl

Appendix 2

Hausman test and Breush for random effects versus fixed effects.

Hypothesis 1 and 3 Hypothesis 2 and 4 Hausman test p-values United States: 0.550

Spain: 0.800

United States: 0.000 Spain: 0.007 Breush Pagan test p-values United States: 0.000

Spain: 0.000

United States: 0.000 Spain: 0.000

Appendix 3

Regression results: the effect of long-term orientation distance and indulgence versus restraint distance on FDI/GDP. Model 5a OLS US Model 5b RE US Model 6a OLS Spain Model 6b RE Spain Dependent variable: FDI/GDP

Constant -3.598*** (0.811) -4.554*** (1.595) -21.30*** (1.052) -22.35*** (2.118) LTO 0.014*** (0.005) 0.014 (0.115) -0.026*** (0.006) -0.024** (0.012) IVR -0.042*** (0.008) -0.045*** (0.015) 0.009 (0.008) 0.018 (0.015) PDI 0.003 (0.007) 0.008 (0.015) -0.043*** (0.011) -0.045** (0.023) IDV 0.012** (0.006) 0.018 (0.013) -0.021* (0.011) -0.028 (0.022) MAS -0.015** (0.007) -0.020 (0.014) -0.019** (0.008) -0.025 (0.017) UAI -0.010* (0.006) -0.008 (0.013) 0.010 (0.008) 0.532 (0.280) SECED -0.250** (0.100) -0.156 (0.191) 0.431*** (0.145) 0.146* (1.644) CLANG 0.673** (0.286) 0.774 (0.598) 0.397 (0.873) 0.532 (0.279) COLON -0.576* (0.324) -0.514 (0.743) 1.820** (0.820) 1.959 (1.510) GEOD -0.000*** (0.000) -0.000 (0.000) -0.000*** (0.000) -0.000*** (0.000) RQ 1.350*** (0.283) 1.148*** (0.414) 1.016** (0.413) 1.291** (0.608) RL -0.932** (0.379) -0.750 (0.500) -0.259 (0.566) -0.295 (0.723) CC 0.183 (0.304) 0.280 (0.390) -0.081 (0.468) -0.385 (0.591) n 353 353 267 267 df 13 13 13 13 Chi Sqc 51.14 42.10 Signif Chi Sq 0.000 0.000 R2 _{0.379 0.389} F 16.42 12.74 P 0.000 0.000 Adj r2 _{0.363 0.365} a

The results are from the pooled OLS regressions with robust standard errors parentheses beneath the coefficients.

b _{The results are from the Gaussian random effect estimator with robust standard errors in parentheses}

beneath the coefficient.

c _{The Wald Chi square has been used for this statistic.}

Appendix 4

Regression results: the effect of long-term orientation distance and indulgence versus restraint distance on EXP/GDP. Model 7a OLS US Model 7b RE US Model 8a OLS Spain Model 8b RE Spain Dependent variable: EXP/GDP

Constant -1.569*** (0.321) -3.396*** (0.641) -4.141*** (0.345) -4.201*** (0.625) LTO 0.011*** (0.002) 0.013** (0.006) -0.014*** (0.002) -0.010* (0.006) IVR -0.022*** (0.003) -0.030*** (0.007) -0.009*** (0.002) -0.004 (0.006) PDI 0.002 (0.003) -0.003 (0.008) -0.003 (0.003) -0.005 (0.009) IDV 0.014*** (0.002) 0.020*** (0.006) -0.004 (0.004) -0.006 (0.010) MAS -0.012*** (0.003) -0.007 (0.007) -0.012*** (0.003) -0.009 (0.007) UAI -0.009*** (0.002) -0.006 (0.007) -0.002 (0.002) -0.001 (0.007) SECED -0.384*** (0.039) -0.080 (0.056) 0.078* (0.042) 0.061 (0.054) CLANG 0.724*** (0.113) 1.160*** (0.322) 0.494** (0.220) 0.373 (0.640) COLON -0.549*** (0.140) -0.312 (0.419) 0.425** (0.193) 0.313 (0.555) GEOD -0.000*** (0.000) -0.000* (0.000) -0.000*** (0.000) -0.000*** (0.000) RQ 1.108*** (0.112) 0.090 (0.104) 0.467*** (0.107) 0.261*** (0.097) RL -0.733*** (0.157) -0.183 (0.116) 0.232 (0.166) 0.069 (0.106) CC -0.073 (0.125) 0.058 (0.092) -0.149 (0.125) -0.021 (0.086) n 473 473 474 474 df 13 13 13 13 Chi Sqc 58.28 91.64 Signif Chi Sq 0.000 0.000 R2 _{0.466 0.625} F 54.04 63.81 P 0.000 0.000 Adj r2 _{0.594 0.633} a

The results are from the pooled OLS regressions with robust standard errors parentheses beneath the coefficients.

b _{The results are from the Gaussian random effect estimator with robust standard errors in parentheses}

beneath the coefficient.

c _{The Wald Chi square has been used for this statistic.}