CORRUPTION, AN INCENTIVE OR DISINCENTIVE FOR FOREIGN DIRECT INVESTMENT? AN EXTENSIVE LITERATURE REVIEW AND QUANTITATIVE EMPIRICAL ANALYSIS

(1)

CORRUPTION, AN INCENTIVE OR DISINCENTIVE

FOR FOREIGN DIRECT INVESTMENT? AN

EXTENSIVE LITERATURE REVIEW AND

QUANTITATIVE EMPIRICAL ANALYSIS

by

LAURA LIETMEIJER

University of Groningen

Faculty of Economics and Business

Master Thesis

MSc International Business & Management

June 24, 2015

l.lietmeijer@student.rug.nl Student number 2186608

(2)

2

Preface

During my Bachelor in Business Administration and Masters in International Business and Management and Business Administration (variant: Organizational and Management Control) I have always been interested in the ethical side of doing business. Therefore, subjects like Corporate Social Responsibility (CSR), corruption, bribery, and responsibility along the Global Value Chain (GVC) have really interested me.

When I was assigned the theme “Multinational Corporations and Internationalization” I saw the great opportunity to develop my knowledge in a subject that suited me. Therefore, I decided to investigate the role of corruption in doing business, and more specifically, the relationship between corruption and Foreign Direct Investment. Following ‘Chapter 5: Conclusion’ is an additional short chapter ‘Self-reflection’ in which I will critically reflect on my way of working during this Master Thesis project.

(3)

3

Abstract

This Master Thesis focuses on the main research question: “What is the relationship between

corruption and inward FDI, and how is this relationship influenced by cultural distance between countries?”. First, the relationship between corruption and Foreign Direct

Investment (FDI) is investigated by means of an extensive literature review. It is found that there are contradictory findings in the literature regarding this relationship. While some find a positive relationship, others find a negative or no relationship at all. The relationship between corruption and inward FDI is investigated by means of data analysis of secondary data and it is found, that a negative relationship exists between corruption and inward FDI. This means that countries are less inclined to invest in host countries that are corrupt.

Also, a gap in the literature was found regarding the possible moderating effect of cultural distance between the investing countries and the receiving country on this

relationship. The possible moderator ‘cultural distance’ has also been investigated by means of data analysis of secondary data, this however, did not yield any significant results due to a lack of data. Because it is expected that cultural distance does provide a moderating effect, propositions are made about a possible positive moderating effect of cultural distance on the negative relationship between corruption and inward FDI.

(4)

4

List of definitions

This list provides the definitions of the three main concepts of this Master Thesis: corruption, cultural distance and inward foreign direct investment. In the main text the concepts in

question are indicated with (an) asterisk(s) so that the reader can easily refer to this list for the extensive definition.

Definitions

Corruption* “The abuse of entrusted power for private gain” (Transparency International, 2015a). It is possible to classify corruption into three distinct classes, depending on the quantities of money lost and also the sector in which it occurs.

1) The first class, grand corruption, consists of acts committed at a high level in the government that distort the central functioning or policies of the state by enabling certain leaders to benefit at the expense of the public good.

2) The second class, petty corruption, has to do with the everyday abuse of entrusted power by low- and mid-level public officials in their interactions with common citizens, who try to gain access to basic services or goods in places such as schools, police

departments, hospitals and other agencies.

3) The final class, political corruption refers to manipulation of institutions, policies and rules of procedure while allocating resources and financing by political decision makers who abuse their position to maintain their status, wealth and power

(Transparency International, 2015a).

(8)

8 Cultural distance** Cultural distance is defined as the distance between countries in terms

of Hofstede’s six dimensions: power distance, individualism versus collectivism, masculinity versus femininity, uncertainty avoidance, long-term orientation versus short-term orientation, and indulgence versus restraint (Hofstede & Hofstede, 2015a). Kogut and Singh’s (1988) index will be used to measure cultural distance, however, it will be adapted as to include all six dimensions instead of only four. A short description of each dimensions follows (Hofstede & Hofstede, 2015a):

1) Power distance: the degree to which less powerful members of institutions (e.g. family) and organizations expect and accept that power is unequally distributed.

2) Uncertainty avoidance: this terms refers to a society’s tolerance for uncertainty.

3) Individualism versus collectivism: the degree to which the individual is integrated in a group.

4) Masculinity versus femininity: the emotional role distribution between men and women which is an essential issue for societies to which a range of solutions are created.

5) Long-term orientation versus short-term orientation: the degree to which society focuses on pragmatic virtues oriented towards future rewards versus the degree to which societies value virtues relating to past and present (e.g. tradition).

6) Indulgence versus restraint: the extent to which society allows relatively free gratification of natural and basic human drives regarding having fun and enjoying life versus the extent to which a society prefers suppression of gratification of needs and also favours regulation by means of strict social norms.

Inward foreign The World Bank (2015) defines inward foreign direct investment as direct investment*** ‘the net inflows of investment to acquire a lasting management interest

(9)

9

Introduction

‘Hundreds of UK companies face tougher anti-corruption rules from next year under European efforts to shed more light on payments to governments by the oil, gas and mining

sectors’.

(Wilson, 2014)

The attention that has been paid to corruption* (please refer to back to the ‘List of definitions’ on page 7, for the definitions of the main concepts) around the world is ever-increasing and has also led to increasing anti-corruption regulation as described above by Wilson (2014). Besides conventional regulation, there are also self-regulatory global initiatives such as the United Nations Global Compact (UNGC) (Voegtlin & Pless, 2014). The UNGC has grown exponentially in number of participants from business and civil society since its foundation in 2000. One of the ten UNGC principles revolves around anti-corruption, and states:

‘Businesses should work against corruption in all its forms, including extortion and bribery’ (Voegtlin & Pless, 2014).

On the other hand, De Jong, Tu and van Ees (2012) state that researchers from existing research have the tendency to ‘over-moralize’ bribery and, consequently, do not accurately account for the possible advantages of bribery. De Jong et al. (2012) acknowledge the ethical issues concerned with corruption but emphasize that, at least at a micro level, there are advantages and disadvantages to bribery. Advantages of bribery in emerging economies are the creation of a network of informal relationships with public officials, and the associated benefits that come from this network (De Jong et al., 2012).

From the above, it can be concluded that corruption receives increasingly more

attention, anti-corruption regulation is growing (whether we talk about laws or self-regulating initiatives), and there are advantages and disadvantages to corruption.

(10)

10 cause of stopping corruption. The organization is financed mostly by government

development agency budgets and foundations. Several other sources include donations from individuals and private sector companies and project funds from international organizations.

Now that the increasing attention for corruption in today’s business world has been clearly indicated, it is important to turn to the specific problem at hand: the lack of clarity about the relationship between corruption and foreign direct investment (FDI) in existing literature. The relationship between these two concepts has been investigated by many researchers (e.g. Blonigen, 2005; Egger & Winner, 2005; Brouthers, Gao & McNicol, 2008; Javorcik & Wei, 2009; Curtis, Rhoades & Griffin, 2013; Melo & Quinn, 2015; Quazi, 2014). However, these researchers have found contradicting results about the relationship between corruption and FDI. According to Curtis et al. (2013) the relation between FDI inflows and corruption is not well defined because some studies find a negative relationship and others find a positive or no relationship at all.

It is interesting to investigate this relationship and also its possible moderators and mediators because there do in fact exist contradictory findings in the literature. From the extensive literature review below (see chapter 1) it will become clear that cultural distance** is a possible moderator that has not been investigated yet. Also, because of the availability of data, the subsequent quantitative analysis in the third chapter will focus specifically on inward FDI***. Therefore, the main research question of this Master Thesis will be:

What is the relationship between corruption and inward FDI, and how is this relationship influenced by cultural distance between countries?

This Master Thesis has a specific set-up which is graphically depicted in figure 1 to clearly indicate each subsequent step. In the first chapter of this Master Thesis an extensive literature review will be carried out. The literature review is presented before the

methodology section in order to clearly indicate the gap of cultural distance as a possible moderator. The literature review will discuss every relevant study regarding the relationship between corruption and FDI. The goal is to visualize the previous research in an organized framework. Explanations are sought for the contradictory results. This is achieved by

(11)

11 The second chapter explains the methodology of this Master Thesis. Based on the extensive discussion of the existing literature, the main research question and sub-questions are further formulated regarding the ‘gap’ that will be investigated. Also, a conceptual model and specific details on the research method, scope, data sources, reliability and validity are provided. This Master Thesis is another step in the direction of understanding the relationship between corruption and FDI and essentially provides one extra piece in completing the ‘puzzle’.

The third chapter ‘Empirical results’ provides the detailed results of the quantitative analyses. Consequently, the fourth chapter ‘Discussion’ will extensively discuss the results regarding the relationship between corruption and FDI. Also, the possible moderating effect of cultural distance is discussed. Finally, the main conclusions and recommendation are summarized in the fifth chapter together with the limitations of this research and possibilities for future research.

(12)

12

Chapter 1: Extensive literature review

In order to determine possible gaps and inconsistencies in the existing literature, this chapter is divided into two parts (see figure 2). First, a comparison and discussion of existing studies is carried out. The relevant studies are divided in three sections, namely, studies that find a positive relationship between corruption and FDI, studies that find a negative

relationship and studies that find no relationship. The subsequent fourth section compares the contradicting findings from the first, second and third section. In this first part of the literature review the inconsistencies will become clear. Second, the findings from the

literature review are presented in an organized framework. The added value of this framework is the fact that the ‘complicated web’ of studies regarding the subject at hand is presented in an organized way, instead of it remaining a ‘collection of loose elements’.

The main studies are summarized in chronological order in table 3 in Appendix A. This table will be referred to in paragraph ‘1.4 comparing the contradicting findings’. See Appendix A for an explanation of how the table will be referred to in the main text and also how the most relevant studies are selected for the extensive literature review.

Finally, studies will be used in the extensive literature review that discuss the relationship between corruption and FDI at different levels (e.g. firm-, industry-, country level). Therefore, this Master Thesis uses an all-inclusive definition of corruption* which includes corruption in the private and the public sector of a country. This reduces the limits of the studies that can be included in the extensive literature review.

The chapter will now turn to the comparison and discussion of existing studies in order to clearly indicate the contradicting results that have been found.

(13)

13

1.1 Studies that find a positive relationship

According to Quazi (2014), many recent studies have investigated the impact of corruption in host countries on their subsequent FDI inflows from investing countries abroad. Corruption either reduces FDI inflows by raising transaction costs and uncertainty or it facilitates FDI by ‘greasing’ the wheels of commerce in the presence of a regulatory environment that is weak. Quazi (2014) found that corruption facilitates FDI inflows to Africa, that the overall

regulatory environment in Africa is weak, and that Sub-Saharan Africa suffers from a locational disadvantage in attracting FDI compared to the rest of Africa.

This positive moderating effect of a weak regulatory environment on the positive relationship of the level of corruption in the host country on FDI in the host country by investing countries abroad is graphically represented in figure 3 below.

Figure 3 The positive moderating effect of a weak regulatory environment (Quazi, 2014)

Egger and Winner (2005) executed an empirical analysis with the help of a sample of 73 developed and less developed countries over the time period 1995-1999 to assess the

relationship between corruption and inward FDI (i.e. direct investments made by foreign investing countries, from the perspective of the receiving host countries). They found that there is an apparent positive relationship between corruption and inward FDI. This means that corruption is a stimulus for FDI. They argue that this confirms Leff’s (1964) position that corruption can be an advantage with regard to circumventing administrative and regulatory restrictions.

(14)

14 which means that profits of firms are reduced and, consequently, the incentive for firms to invest abroad is lowered. Also, corruption decreases a country’s locational attractiveness to foreign investors because it reduces the productivity of public inputs such as infrastructure (Bardhan, 1997; Rose-Ackermann, 1999; Lambsdorff, 2003). The costs of a firm’s foreign investment are increased in the short run because:

(a) it has to pay bribes;

(b) it is engaged in rent seeking activities during which resources are wasted (Applebaum & Katz, 1987; Murphy, Shleifer & Vishny, 1991; Shleifer & Vishny, 1993); and (c) it has to incur additional contract-related risks since corruption contracts are not

enforceable in a court (Boycko, Shleifer & Vishny, 1995).

On the other hand, corruption can also be observed as a ‘helping hand’, which increases profits of the foreign investor. A firm may be willing to pay bribes so it can:

(a) speed up the bureaucratic process in order to receive legal permission for setting up a foreign plant (see Lui, 1985); and

(b) access publicly funded projects (Tanzi & Davoodi, 2000).

The positive relationship between corruption and FDI found by Egger and Winner (2005) is graphically represented in figure 4 below. The figure also shows the positive mediating effect of the value of a local partner (in the corrupt host country) that can cut through the bureaucratic maze, from the point of view of the foreign investor, as argued by Javorcik and Wei (2009).

Figure 4 The results of the studies from Egger and Winner (2005) and Javorcik and Wei (2009)

(15)

15 this paragraph and paragraph ‘1.2 Studies that find a negative relationship’ because they found that a U-shaped relationship exists between corruption levels in the host country and FDI in the host country by investing countries abroad. This provides evidence of the ‘grabbing hand’ view at corruption levels that are low to moderate, and supporting the ‘helping hand’ view at high corruption levels. They also found that market-seeking motives have a positive moderating effect on this relationship. A very basic graphical representation of the relationship that Petrou and Thanos (2014) found between corruption and FDI is presented in figure 5.

Figure 5 The U-shaped relationship between corruption and FDI (Petrou & Thanos, 2014)

(16)

16 Figure 6 The positive relationship between corruption and FDI (Cuervo-Cazurra, 2006)

1.2 Studies that find a negative relationship

Wei (2000) conducted a study regarding FDI from twelve investing countries to 45 host countries and found that an increase in corruption level in the host governments reduces the inward FDI. Wei (2000) also found that, even though American investors are averse to host country corruption, they are no more averse to it than other investors, despite America’s unique Foreign Corrupt Practices Act (FCPA) that penalizes multinationals or their officers with jail terms or fines when they bribe foreign government officials. Up until February 1999 the United States was the only source country that had such an Act.

Cuervo-Cazurra (2006), on the other hand, found that corruption does not impact equally on all foreign investors since there is variability in the costs of engaging in bribery abroad (as discussed in the previous paragraph). Countries with high levels of corruption also seek out other countries with high levels of corruption to invest in. However, laws against bribery abroad increase the cost of engaging in bribery. Therefore, investors from countries with these type of laws are very likely to limit their FDI further in corrupt countries. Hines (1995) found a similar effect of laws against bribery. Corruption generates challenges for investors, since it increases risk and uncertainty involved but it also increases the cost of operating abroad.

Blonigen (2005) argues that the quality of institutions is very likely to be a major determinant of FDI activity, especially for less-developed nations. For example, corruption increases the cost of doing business and therefore, decreases FDI activity.

(17)

17 of technological know-how. These trade-offs lead to the fact that when corruption levels are sufficiently high, no foreign investment in any sort of ownership form will take place.

The findings from the previous studies are combined and graphically represented in figure 7 below.

Figure 7 Combined findings of Cuervo-Cazurra (2006), Blonigen (2005), and Javorcik and Wei (2009)

Habib and Zurawicki (2001) investigated the impact of corruption on foreign and local direct investments and found that corruption has a negative effect on investments. They emphasize a crucial but so far overlooked distinction, namely, the effect of corruption on local direct investments is significantly weaker than the effect of corruption on foreign direct investment. Also, the political stability of the host market and the degree of international openness negatively moderate (weaken) the negative effect of corruption on FDI.

(18)

18 Brouthers et al. (2008) found that the negative impact of corruption on market-seeking investment is mitigated by greater market attractiveness. However, as corruption levels

increase the ability of market attractiveness to mitigate the negative influence of corruption on resource-seeking FDI disappears. The findings of Brouthers et al. (2008), but also of Habib and Zurawicki (2001; 2002) are graphically represented in figure 8 below.

Figure 8 Combined findings of Habib and Zurawicki (2001; 2002) and Brouthers et al. (2008)

(19)

19 Voyer and Beamish (2004) hypothesized that investors tend to avoid countries where a high degree of corruption is present. They investigated this by examining the relationship between levels of perceived corruption and Japanese FDI in industrialized and emerging economies. Voyer and Beamish (2004) find that in emerging nations, where comprehensive legal and regulatory frameworks are lacking, corruption reduces FDI. Therefore, it is

important that managers consider the level of perceived corruption when assessing potential markets for investment. Voyer and Beamish (2004) refer to Husted (1999), who found that certain cultural dimensions (specifically masculinity, power distance, and uncertainty avoidance) increase a country’s overall level of corruption. And also that economic wealth plays an important role in a country’s overall level of corruption (i.e. corruption and poverty seem to go hand in hand).

Doh, Rodriguez, Uhlenbruck, Collins and Eden (2003) explain that the impact of government corruption on FDI by investing countries to receiving developing countries has received too little attention even though it is a pervasive characteristic of the international business environment with its damaging effects on firms, governments and the broader society. Doh et al. (2003) theorize that higher arbitrariness and pervasiveness of corruption both reduce FDI and total investment, where pervasiveness of corruption is the level of corruption and arbitrariness of corruption is the uncertainty surrounding corruption (what to pay, who to pay, whether payment will result in delivery of promised services or goods). The findings of the three previous studies are graphically represented in figure 9 below.

(20)

20 Finally, Egger and Winner (2006) studied the impact of corruption in a panel of bilateral outward FDI stocks of OECD economies and non-OECD economies. First, they found a negative impact of corruption on FDI. Second, corruption is not important for extra-OECD FDI, but it is important for intra-OECD FDI. They found that the impact of corruption has diminished over the years which indicates that alternative factors (such as market growth) have become relatively more important.

1.3 Studies that find no relationship

Wheeler and Mody (1992) investigated the relationship between RISK (composite index for relevant socio-political conditions, including corruption) and FDI and found no significant relationship. It appears that, for developing countries, there are several other more important factors such as rapid industrial growth, stable international relations, an expanding domestic market, and infrastructure development.

Hines (1995) performed a quantitative empirical analysis including corruption and US FDI data of the time periods 1966-1977 (before the passage of the Foreign Corrupt Practices Act (FCPA) of 1977) and 1977-1982. He found that corruption levels in the host countries in which the US invested did not affect their level of total inward FDI. However, corruption levels in host countries in which the US invested did have a negative effect on the growth of FDI after passing the FCPA in 1977. Hines (1995) therefore concludes that American

legislation does significantly decrease US FDI in host countries in which government officials regularly receive bribes (as indicated in the previous paragraph).

Henisz (2000) conducted a quantitative analysis including data from the time period 1980-1992 and found that investing US firms tend not to be affected by corruption in their investments. Therefore, Henisz’s (2000) study is placed in this paragraph ‘1.3 Studies that find no relationship’. However, it might also be placed in paragraph ‘1.1 Studies that find a positive relationship’ because Henisz (2000), in some cases, found that corruption increases the probability of investment in the foreign country by US firms. But these cases were few in number.

(21)

21 quality of institutions), the negative effects of corruption on FDI inflows disappear and sometimes even become positive but statistically insignificant.

Finally, by performing a panel data as well as cross-sectional data analysis, Abed and Davoodi (2000), have investigated the effects of corruption levels in host countries on per capita FDI inflows to transition economies. It was found that countries with a low corruption level receive more per capita FDI. However, corruption becomes insignificant, when

controlled for the structural reform factor. They conclude that the corruption level is less important than structural reform in attracting FDI. With structural reforms they mean fundamental economic reforms (e.g. reforming tax and customs administrations and simplifying the tax system) whose goal is to eliminate the conditions that make corruption possible.

1.4 Comparing the contradicting findings

Since a representative part of the existing literature is used for this review, it immediately becomes clear that the majority of the studies find a negative relationship. Studies that find a positive relationship or no relationship are significantly smaller in number. This may indicate that corruption in the host country indeed reduces the amount of FDI it receives from

investing countries abroad (as will be confirmed in ‘Chapter 3: Empirical results’ by means of quantitative analysis). The comparison below may shed some light on the situation as to why the results are contradicting.

1.4.1 Reviewing the studies in chronological order

(22)

22 Figure 10 The timeline of the studies included in the extensive literature review

1.4.2 Determination of concepts

The way concepts are defined or measured may also be a cause of the contradicting findings. If the concepts used by the studies (Appendix A, column a) are taken into account, it becomes clear that a small majority (ten out of nineteen studies) use the CPI of Transparency

International to measure corruption (Appendix A, column a, rows 5-7, 9, 11, 13-14, 17-19). However, when comparing the main findings of these studies it becomes clear that the results are mixed (U-shaped, positive, negative, and no relationship are found by these studies) (Appendix A, column d, rows 5-7, 9, 11, 13-14, 17-19).

1.4.3 Scope of the studies

(23)

23 other way around. It could also be the case that the forces that ensure that corruption acts as a ‘grabbing hand’ and ‘helping hand’ are equal in strength. Therefore, the contradicting results could be the result of the scope of the study (i.e. choice of countries included) (Appendix A, column c).

It is also interesting that Egger and Winner (2005) find a positive relationship between corruption in the host country and FDI by foreign investing countries (Appendix A, column d, row 11) because they explain that prior empirical research has focused on analyzing cross-section data (an implicit focus on long run effects of corruption) and that these studies tend to find a negative long run effect of corruption on FDI. It thus might be the case that the length of the time period from which data is used can influence the results because Egger and Winner use data from a time period of five years instead of searching for long run effects (Appendix A, column c, row 11). However, when comparing this to the study of Wei (2000), it becomes clear that data was used from one year (Appendix A, column c, row 5) and a clear

negative relationship was found between corruption and FDI (Appendix A, column d, row 5).

It appears that comparing studies on short run and long run effects of corruption on FDI leads to mixed results.

1.4.4 Defining institutional quality

Quazi (2014) and Blonigen (2005) both, either explicit or implicit, introduce the quality of institutions as an important factor influencing the relationship between corruption and FDI (Appendix A, column d, rows 10 and 18). However, they both use this argument in a different manner.

On the one hand, Quazi (2014) indicates that corruption in the presence of a weak regulatory environment facilitates FDI by ‘greasing’ the wheels of commerce. Quazi (2014) thus implies that lower institutional quality (i.e. weak regulatory environment) facilitates FDI. On the other hand, Blonigen (2005) explains that quality of institutions is very likely to be a determinant of FDI activity, for example, corruption increases the cost of doing business and therefore, decreases FDI activity. Blonigen (2005) thus views corruption as an indicator for institutional quality and claims that lower institutional quality (i.e. corruption) decreases FDI activity. This is the exact opposite of Quazi’s (2014) findings.

(24)

24 impact at all. It could depend on the indicators included in the study, whether a negative, positive or no relationship is found.

1.4.5 Variables included in the research

Something that stands out in paragraph ‘1.3 Studies that find no relationship’ is that two studies (Abed and Davoodi, 2000; Al-Sadig 2009) initially find a negative relationship, however, after including more variables (e.g. structural reforms and host country

characteristics such as the quality of institutions) in the research, this relationship disappears. Wheeler and Mody (1992) also state that there are several other more important factors. The sole focus on corruption and the omission of other variables could skew the results.

1.4.6 Gap in the literature

The extensive literature review indicates that numerous researchers have investigated the impact of corruption on FDI combined with many additional variables such as industrial, economic, and institutional variables. What seems to be lacking is the cultural aspect. For example, as explained earlier, Voyer and Beamish (2004) only touch upon this subject very briefly by stating that Husted (1999) found that certain cultural dimensions play an important role in a country’s overall level of corruption. This Master Thesis will investigate the impact of cultural distance between the investing country and the host country on the relationship between the level of corruption in the host country and the level of FDI in the host country by the investing country. This will be done by means of a quantitative empirical analysis.

1.5 Organized framework

On the next page (see figure 11) the organized framework can be found that emerged from the extensive literature review. The framework is divided in roughly two halves by a dotted demarcation line. The findings from the studies that found a positive relationship between corruption and FDI can be found in the upper half of the figure and the studies that found a negative relationship are represented in the bottom half.

(25)

(26)

26

Chapter 2: Methodology

This chapter discusses the main research question, its sub-questions and the subsequent conceptual model, which all three resulted from the preceding literature review. Also, the research method of this Master Thesis is discussed and graphically depicted.

Following paragraph ‘2.2 Research method’ is the paragraph about the scope of the research. This section provides details about the countries included in the research and also the time period from which data is used.

The fourth paragraph discusses the data sources that are used for the quantitative analysis. And finally, the reliability and validity of the research are elaborated upon to clearly indicate how the main research question will be answered. The overview of this chapter can be found in figure 12 below.

Figure 12 Overview of ‘Chapter 2: Methodology’

The following paragraph will first elaborate upon the main research question, its

(27)

27

2.1 Main research question, sub-questions and conceptual model

As mentioned in the ‘Introduction’, the main research question is based on the extensive literature review and is as follows:

What is the relationship between corruption and inward FDI, and how is this relationship influenced by cultural distance between countries?

Because the main research question essentially consists of two parts, it will be split up in two sub-questions in order to provide a clear answer. The sub-questions are:

a: What is the relationship between the level of corruption in country A and the total level of inward foreign direct investment it receives from investing countries? b: How does the cultural distance between country A and B influence (i.e. possibly

moderate) the relationship between the level of corruption in country A and the level of inward foreign direct investment by country B?

The sub-questions are graphically represented in the form of a conceptual model in figure 13.

Figure 13 Conceptual model of the Master Thesis

2.2 Research method

First, the extensive literature review has already been carried out in the second chapter and has functioned as the input for further quantitative analysis. Qualitative scientific sources have been used to assure the quality of the research.

Quantitative analysis (using SPSS) will be performed in order to determine the relationship between corruption and inward FDI. A novel contribution to the existing

literature is provided by adding the variable ‘cultural distance’ and determining its (possible) moderating effect on the relationship between corruption and inward FDI.

(28)

28 help of two separate data analyses based on two different data sets. Data analysis 1 will investigate the relationship between corruption and inward FDI and thus contribute to answering sub-question a. Data analysis 2 will investigate the possible moderating effect of cultural distance and thus contribute to answering sub-question b. This research method is graphically depicted in figure 14.

The step(s) of data analysis 1 (correlation and regression) and data analysis 2 (moderation) will be explained first before turning to the scope, data sources and reliability and validity of the research.

2.2.1 Data analysis 1 - Correlation analysis and regression analysis

First, a correlation analysis with regard to corruption and inward FDI will determine the extent to which these variables vary in unison. There can be a positive correlation (as one variable increases so does the other), negative correlation (as one variable decreases, the other increases) or no correlation (the variables are independent or unrelated) (Thomas, 2004: 210). Correlation analyses are an important first step. They are merely an indication as to whether the variables are related and not the ultimate test of the thesis.

Second, a regression analysis with regard to corruption and inward FDI will determine whether there is a linear relationship between the variables. The difference between the correlation analysis and the regression analysis is that the regression analysis assumes a

causal relationship (Huizingh, 2008: 276).

2.2.2 Data analysis 2 - Moderation analysis

A moderation analysis with regard to the influence of cultural distance on the relationship between corruption and inward FDI will determine if the ‘cultural distance’ between country A and country B is a moderator variable and, consequently, if it strengthens or weakens the relationship (Goldsby, Knemeyer, Miller & Wallenburg, 2013). If no significant results are

(29)

29 found or if no conclusions can be drawn (as will be the case), in ‘Chapter 4: Discussion’, propositions regarding the possible moderating effect of cultural distance on the relationship between corruption and inward FDI will be developed with the help of arguments from literature. The reason for this is because a moderating effect of cultural distance in expected by the researcher. With the help of these propositions, a preliminary answer (before any further future empirical research) can be provided to sub-question b.

2.3 Scope

Keeping in mind the available data, time and resources it has been decided to perform two separate data analyses to answer each sub-question, as noted in the paragraph ‘Research method’. The reason for using two different secondary databases will be discussed in detail in the next paragraph ‘Data sources’. This paragraph will discuss the scope of each data analysis.

The first data analysis that investigates the relationship between corruption and inward FDI includes 154 countries. A list of these countries can be found in table 4 in Appendix B. The countries are sorted per continent (Asia, Africa, Europe, North America, South America, and Oceania) since the continent will be used as a control variable in this quantitative analysis. The most recent data regarding inward FDI will be used, which is data from 2013 (World Bank, 2015).

The second data analysis that investigates the possible moderating effect of cultural distance on the relationship between corruption and inward FDI includes twenty-one

countries. A list of the twenty-one countries can be found in table 5 in Appendix B. The most recent data regarding inward FDI will be used, which is data from 2012 (OECD, 2014). From table 5 in Appendix B it becomes clear that EU-countries by far comprise the biggest part of the list (16 out of 21 countries are EU-countries).

2.4 Data sources

The secondary databases include the World Bank (2015), Transparency International (2015e), OECD.StatExtracts (OECD, 2014), and Hofstede and Hofstede (2015b).

2.4.1 Data sources for data analysis 1

(30)

30 Corruption - Corruption will be measured with the help of the Corruption Perceptions Index (CPI) (Transparency International, 2015e). It is the most widely used indicator of corruption in the world and scores and ranks countries based on the perception of how corrupt a certain country’s public sector is. Through a combination of surveys and assessments of corruption (collected by various reputable institutions) this composite index is created (Transparency International, 2015f). The CPI gives a score between 0 (highly corrupt) and 100 (very clean). In the first analysis the CPI of 2013 of the receiving country will be used. A list of the 13 data sources that were used to construct the CPI of 2013 is provided in table 6 in Appendix C.

2.4.2 Data sources for data analysis 2

Inward FDI - In order to determine if the variable ‘cultural distance’ has a moderating effect on the relationship between corruption and inward FDI, it is necessary to know each inward FDI flow from every investing country to the receiving country. This way it can be

investigated whether the cultural distance between country pairs moderates the relationship between corruption and inward FDI. Because the World Bank (2015) provides the total inward FDI from all the investing countries to the receiving country, the data cannot be used to determine the possible moderating effect of cultural distance.

The OECD database (OECD, 2014), on the other hand, reports each inward FDI flow of 2012 from each investing country to the receiving country. Therefore, this database is best suited for the analysis in question. Inward FDI is defined (similarly to the World Bank

(2015)) as an unincorporated or incorporated enterprise in which a foreign investor owns 10% or more of the voting power or ordinary shares of the equivalent of an unincorporated

enterprise or an incorporated enterprise (OECD, 2014).

(31)

31 Corruption – Corruption, again, will be measured with the help of the CPI

(Transparency International, 2015e). However, now the CPI of 2012 will be used since the data on inward FDI flows is also from 2012 (OECD, 2014). A list of the 13 data sources that were used to construct the CPI of 2012 is provided in table 7 in Appendix C.

Cultural distance - Hofstede (1980: 25) defines culture as ‘the collective programming of the mind which distinguishes the members of one human group from another’.

Kogut and Singh (1988) created a composite index to calculate cultural distance based on the deviation along each of the four cultural dimensions (uncertainty avoidance, power distance, masculinity/femininity, and individualism/collectivism) using Hofstede’s indices. Formula 1 is as follows:

(1)

Kogut and Singh’s (1988) framework is not without criticism. According to Shenkar (2001), the cultural distance concept includes methodological and/or conceptual properties that are presented in the form of hidden assumptions that mostly go unnoticed but are not supported by logic or empirical evidence (e.g. illusion of symmetry, illusion of stability, illusion of linearity, illusion of causality, illusion of discordance, assumption of corporate homogeneity, assumption of spatial homogeneity, and assumption of equivalence).

It is indeed a very difficult endeavour to measure cultural distance since there are so many variables to consider. Nevertheless, Kogut and Singh’s (1988) index of cultural distance is widely cited and used (e.g. Brouthers, 2013; Harzing, 2003; Shenkar, 2001; Tihanyi,

Griffith & Russell, 2005). Therefore, Kogut and Singh’s (1988) index will be used to measure cultural distance, however, it will be adapted. Currently it only includes four dimensions of Hofstede, but it will be adapted in such a way that it will include all six dimensions of Hofstede, namely: power distance, individualism versus collectivism, masculinity versus femininity, uncertainty avoidance, long-term orientation versus short-term orientation, and indulgence versus restraint (Hofstede & Hofstede, 2015a). The data of the 21 countries will be retrieved from the website of Hofstede and Hofstede (2015b). Formula 2 will be as follows:

(32)

32

2.5 Reliability and validity

Reliability - Each step of the research strategy will be executed meticulously so that if the research method will be applied to the same phenomenon over and over again this will yield the same results (Thomas, 2004: 30-31). Also, qualitative scientific journals are used and the objectivity of the researcher is ensured by being flexible with regard to contradicting findings and results. A third party, the supervisor, will provide objective feedback throughout the project. This helps in preventing tunnel vision.

Internal validity - Internal validity refers to the extent to which the researcher really measures the concepts that are intended to be measured (Thomas, 2004: 31). CPI is the most widely used indicator of corruption worldwide (Transparency International, 2015f) and the measurement of inward FDI (World Bank, 2015; OECD, 2014) is very straightforward. The cultural distance index (Kogut & Singh, 1988) can be criticized on several points as

mentioned, however, at the present time it is the most appropriate measure and the missing dimensions in the original formula have been accounted for by the adaptation that has been made.

(33)

33

Chapter 3: Empirical results

This chapter reports the empirical results regarding the data analyses (for more detailed results of the data analyses, see tables 8-12 in Appendix D). Data analysis 1 is carried out in order to seek an answer to sub-question a: “What is the relationship between the level of corruption in

country A and the total level of inward foreign direct investment it receives from investing countries?”. And data analysis 2 is carried out in order to seek an answer to sub-question b: “How does the cultural distance between country A and B influence (i.e. possibly moderate) the relationship between the level of corruption in country A and the level of inward foreign direct investment by country B?”, as mentioned in paragraph ‘2.2 Research method’.

3.1 Data analysis 1

3.1.1 Descriptive statistics

The dataset comprises 154 countries (N=154) from Asia (N=40), Africa (N=52), Europe (N=31), North America (N=17), South America (N=12), and Oceania (N=2) (see Appendix B, table 4). The percentage of countries per continent is graphically depicted in figure 15.

Table 1 provides the descriptive statistics of the independent variable ‘CPI of 2013’ and the dependent variable ‘inward FDI of 2013’.

Table 1 Descriptive statistics of the data for data analysis 1 CPI 2013 Inward FDI 2013

Mean 41.35 $11,018,611,818.94

Median 37.00 $1,326,022,125.90

Standard Deviation 18.14 $38,788,318,682.81

Minimum 8 $800,000.00

Maximum 91 $347,848,740,396.86

(34)

34 The variable ‘inward FDI 2013’ (labeled in figure 16 as ‘FDI 2013’) ‘ has a log-normal distribution, which means that after log-transformation, the values are normally distributed. Therefore, the correlation and regression analyses will be carried out with the log of the variable ‘inward FDI 2013’. The new resulting variable is labeled ‘LN FDI 2013’. Figure 16 shows the histogram of the log-normal distribution on the left side and the histogram after logarithmic transformation on the right side.

3.1.2 Correlation analysis

A correlation analysis has shown that the CPI of 2013 and the log of inward FDI of 2013 correlate significantly with each other (r = 0.331, p < .001). It is important to keep in mind the meaning of the values of the CPI. A value of 0 means highly corrupt and a value of 100 means no corruption at all (as explained in the paragraph ‘data sources’ in ‘Chapter 3: Methodology’). Therefore, the less corrupt a country is, the more FDI it receives from other investing countries.

3.1.3 Regression analysis

In order to analyse whether the degree of corruption in country A affects the amount of inward FDI it receives, a linear regression analysis has been carried out of the CPI of 2013 on the log of inward FDI of 2013. This regression analysis was significant, R2 = 0.110, F(1,152) = 18.70, p < .001. Again, it is important to keep in mind the meaning of the values of the CPI. Therefore, the degree of corruption in country A negatively affects the amount of inward FDI it receives from all the investing countries, B = 0.040, p < .001.

(35)

35 country in the dataset has received a code per continent. For example, a country received a ‘0’ if it did not reside in the particular continent at hand, and a ‘1’ if it in fact did reside in the continent at hand. After entering the dummy variables (continents) in the analysis, it can be concluded that the particular continent in which the countries reside is not of any influence with regard to the negative relationship between corruption and inward FDI. No significant results have been found at the 5% significance level, for any of the continents, which becomes clear because of the insignificant p-values (p = 0.150; p = 0.194; p = 0.086; p = 0.159; p = 0.743).

In conclusion, to answer sub-question a, “What is the relationship between the level of

corruption in country A and the total level of inward foreign direct investment it receives from investing countries?”, it was found that a negative relationship exists between corruption and

inward FDI. This means that investing countries are less inclined to invest in countries that are perceived to be more corrupt.

3.2 Data analysis 2

3.2.1 Descriptive statistics

The dataset comprises 21 countries (N=21) from Asia (N=2), Europe (N=16), North America (N=2), and South America (N=1) (see Appendix B, table 5). Table 2 provides the descriptive statistics of the independent variable ‘CPI of 2012’, the dependent variable ‘inward FDI of 2012’ (taken as a total per country; labelled ‘Total Inward FDI 2012’), and the possible moderator ‘cultural distance’ (the average of all the cultural distances between country A and the 20 investing countries, the average of all the cultural distances between country B and the 20 investing countries, etcetera, so that each of the 21 countries is represented; labelled ‘Average Cultural Distance’). See Appendix E for the individual cultural distances between the twenty-one countries (please note: the distances are mirrored along the diagonal).

CPI 2012 Total Inward FDI 2012 Average Cultural Distance Mean 66.67 $23,017,535,580 2.1 Median 69 $5,672,666,667 2.024 Standard deviation 16.06 $50,223,141,050 0.625 Minimum 34 $608,641,053 1.151 Maximum 90 $223,878,534,000 3.680

(36)

36

3.2.2 Moderation analysis

To analyse whether cultural distance has a moderating influence on the relationship between corruption and inward FDI, a regression was carried out with the mean centered variable ‘CPI of 2012’, the mean centered variable ‘Average Cultural Distance’, and the mean centered variable ‘CPI of 2012’ multiplied by the mean centered variable ‘Average Cultural Distance’. ‘Total Inward FDI of 2012’ was entered into the analysis as the dependent variable. This regression was not significant, R2 = 0,180, F(3,17) = 1.242, p = 0.325.

In conclusion, to answer sub-question b, “How does the cultural distance between

country A and B influence (i.e. possibly moderate) the relationship between the level of corruption in country A and the level of inward foreign direct investment by country B?”, it

was found that the model for data analysis 2 was not significant. This means that no

(37)

37

Chapter 4: Discussion

This chapter starts with a short introduction of the results that emerged from ‘Chapter 3: Empirical results’. Subsequently, it will be divided into the two paragraphs ‘Corruption and inward FDI’ and ‘Cultural distance’.

The first paragraph discusses the arguments that predict a negative relationship between corruption and FDI, as found in ‘Chapter 1: Extensive literature review’. This will help in providing a detailed answer to sub-question a. The main arguments from the extensive literature review will be used in such a way that the most important elements are emphasized once again, without repeating all the results from paragraph ‘1.2 Studies that find a negative relationship’. Therefore, it is essentially a short recap of the main arguments found in the first chapter.

The second paragraph discusses the possible moderating effect of cultural distance. With the help of arguments from literature, two propositions are developed that predict the moderating effect of cultural distance on the negative relationship between corruption and inward FDI. In order to hold a substantiated discussion it is crucial to also review the counterarguments that can be given regarding these propositions. Finally, based on the discussion of these two propositions and its counterarguments, a preliminary answer can be given to sub-question b. The next chapter, ‘Chapter 5: Conclusion’, provides an answer to the main research question with the help of the answers to the two sub-questions from this

chapter. An overview of this chapter can be found in figure 17 below.

(38)

38 Based on the empirical results provided in chapter 3, the conceptual model (see figure 13) can now be represented as a causal model (see figure 18). It has become clear that a negative relationship exists between the level of corruption in a country and its level of inward FDI. This means that, the more corrupt a country is, the less inclined other countries are to invest in that particular country.

The second analysis regarding the moderator of cultural distance did not yield any significant results. The moderator is indicated with dotted lines because a moderating influence is however still expected. An analysis that includes data of more countries would perhaps yield very different results.

Figure 18 Causal model based on empirical evidence

The results of the analyses from ‘Chapter 3: Empirical results’ will be discussed next. First, the negative relationship between corruption and inward FDI is discussed. Second, the possible moderating effect of cultural distance is discussed.

4.1 Corruption and inward FDI

4.1.1 Short recap

This sub-paragraph will give a short recap of the reasons that were found in the extensive literature review (see the first chapter) as to why a negative relationship exists between corruption and FDI, and then inward FDI in particular. Sub-paragraph ‘4.1.2 Answering sub-question a’, will provide an answer to the first sub-sub-question based on these findings in the literature.

(39)

39 is despite the America’s Foreign Corrupt Practices Act (FCPA) and the fact that the United States was the only source country, up until February 1999, that had such an Act.

Habib and Zurawicki (2002) explain that the negative relationship between corruption and FDI indicates that foreign investors want to avoid corruption because it is considered to be very wrong. It can also indicate that foreign investors avoid corruption because it can create operational inefficiencies.

Another effect of corruption, as explained by Blonigen (2005), is that it can increase the cost of doing business and therefore decrease FDI activity. Javorcik and Wei (2009) come to a similar conclusion and find that corruption decreases the transparency of local

bureaucracy. This decrease in transparency in turn leads to increased costs of doing business. Javorcik and Wei (2009) also point out that that foreign investors with sophisticated

technology, in case of corruption, worry about its misuse by joint venture partners or leakage of technological know-how. This means that, when corruption levels are high enough, no foreign investment will take place.

Also, anti-corruption regulation (i.e. laws against bribery abroad) have been found to negatively impact FDI in host countries by investing countries abroad (Hines, 1995; Cuervo-Cazurra, 2006).

Finally, Melo and Quinn (2015) found that a reinforcing relationship exists between corruption and FDI, which may very well lead to positive or negative spirals in institutional quality. They have emphasized that organizations have to consider the negative institutional side-effects from investing in oil rich countries or when dealing with poor governments.

4.1.2 Answering sub-question a

Sub-question a, “What is the relationship between the level of corruption in country A and the

total level of inward foreign direct investment it receives from investing countries?”, can now

be answered with the help of the empirical results of the third chapter and also the discussion above. Based on the results of the first quantitative analysis in ‘Chapter 3: Empirical results’, it has become clear that a strong (highly significant) negative relationship exists between the level of corruption in country A and its total level of inward FDI.

(40)

40 However, the reasons do not necessarily have to be about morality. For example, the negative relationship between corruption and inward FDI can also be because it creates operational inefficiencies (Habib & Zurawicki, 2002) or increases the cost of doing business (Blonigen, 2005), perhaps caused by a decrease in transparency of local bureaucracy

(Javorcik & Wei, 2009). Moreover, the presence of anti-corruption regulation could have deterred investing countries (Hines, 1995; Cuervo-Cazurra, 2006). Misuse of sophisticated technology and leakage of technological know-how (Javorcik & Wei, 2009) by companies in country A, could also have been the reason for lower investment connected to corruption. Finally, investing countries may have considered the negative institutional side-effects (Melo & Quinn, 2015) for country A, while deciding whether or not to invest.

4.2 Cultural distance

From the second data analysis in ‘Chapter 3: Empirical results’, the conclusion had to be drawn that no definitive answer could be given with regard to the possible moderating effect of cultural distance on the negative relationship between corruption and inward FDI.

However, with the help of findings and arguments from scientific literature it is possible to develop propositions about the possible moderating effect of cultural distance on the negative relationship between corruption and inward FDI.

In this paragraph findings and arguments from scientific literature pertaining FDI in general will be used, since inward and outward FDI, in this case, are essentially two sides of the same coin. For example, if investing countries (outward FDI) would be more inclined to invest in countries that are culturally similar, this means that the receiving countries (inward FDI) would receive more investments in the case of cultural proximity.

4.2.1 Cultural proximity

Buckley, Clegg, Cross, Liu, Voss and Zheng (2007) performed a data analysis on Chinese outward direct investment data, collected between 1984 and 2001, and found that Chinese outward FDI is associated with cultural proximity to host countries. That cultural proximity is a significant factor, indicates that network effects and reduced transaction costs are crucial in attracting Chinese investors, and also, relational assets represent a special ownership

advantage (even for state-owned firms) (Buckley et al., 2007). The validity of this study can be put into question, since the data used for the analysis is relatively old (1984-2001), however, below studies will follow that have similar findings to Buckley et al. (2007).

(41)

41 FDI in the context of Bosnia and Herzegovina. He found that cultural proximity plays a strong role in facilitating FDI. Using a relational approach that centers around cultural, economic, institutional and political connections that function as channels for transactions, Bandelj (2002) suggests that such ties are specifically important in high risk conditions. She finds, in the context of Central and Eastern Europe, that cultural ties are expected to facilitate FDI.

Passakonjaras (2012) investigated outward foreign direct investment (OFDI) of the Thai garment industry and, by means of interviewing and surveying, she explored the differences between firms with and without OFDI. Passakonjaras (2012) finds that cultural proximity (for example, no language barriers) is one of the most important criteria in investment location selection. For example, many South Asian nations such as Pakistan, India, Sri Lanka, and Bangladesh have an abundance of raw materials and relatively low labour costs, however these countries are not attractive to Thai investors. One explanation is the cultural distance that covers up these countries as potential investment locations.

Mariscal, Zhang and Pascual (2012) investigated the different factors that influence the FDI decisions of multinational banks. By using eclectic theory, an estimation model is set by Mariscal et al. (2012) with panel data from seven Latin American countries in order to test hypotheses. Dunning’s (1988) eclectic theory is a popular theoretical framework of firm’s internationalization. It emphasizes the importance of ownership-specific, internationalization-specific and location-internationalization-specific factors in the FDI decisions of multinational banks (Williams, 1997). Mariscal et al. (2012) find that cultural proximity has a significant impact on banking FDI. Specifically, cultural proximity positively and significantly affects banking FDI.

Based on the studies above, a proposition can be developed regarding the moderating effect of cultural distance. The presence of cultural proximity between countries is very likely to positively influence the amount of FDI they receive. Therefore, the more distant two countries are, in terms of culture, the less they will invest in each other. This would mean that cultural distance positively moderates (strengthens) the negative relationship between

corruption and FDI, as stated in proposition 1.

Proposition 1: Cultural distance positively moderates the negative relationship between corruption and FDI.

(42)

42 inclined investing countries are to invest in host countries given the same level of corruption. Since the moderating effect of cultural distance is currently a ‘gap’ in the existing literature, more research is needed regarding its influence on the specific relationship between

corruption and FDI before definitive statements can be made.

4.2.2 Internationalization processes

Existing theory has suggested that early investments of organizations often occur in countries with a similar cultural background to the home country (Johanson & Vahlne, 1977) or where the relational assets in the form of familial or ethnic ties with a certain minority population in a host country could be exploited (Lecraw, 1977; Wells, 1983; Lau, 2003).

Buckley et al. (2007), in their investigation regarding Chinese FDI, initially find that the cultural proximity variable does not change as time passes. However, they explain that this suggests that Chinese FDI is still in an early development stage and cultures in host countries that are more familiar continuously help in promoting Chinese inward investment. Buckley et al. (2007) suggest that their findings call for further research of a longitudinal nature.

Johanson and Vahlne (1977) developed a model (the Uppsala internationalization process model), based on empirical research, of the internationalization process of firms. The model concentrates on the gradual acquisition, integration and usage of knowledge about foreign operations and markets, and also on the incrementally increasing commitments to foreign markets. The model is based on empirical observations that have shown that Swedish organizations usually develop their international operations in subsequent small steps, instead of making large foreign production investments at a certain point in time.

(43)

43 Barkema and Drogendijk (2007) argue that firms may enter foreign environments either incrementally (as suggested by long-established internationalization strategy theory such as the Uppsala model) or by taking bigger steps that might result in lower initial

performance, but through experience and learning, lead to higher performance of expansions in the future. Expansion steps might also be too large and limit the exploration of foreign environments. Barkema and Drogendijk (2007) conclude that sequential internationalization processes do still matter, maybe even more so today, if the increasing pace of innovation and globalization (Brown & Eisenhardt, 1997; Hitt, Keats & DeMarie, 1998; Murtha, Lenway & Hart, 2001) makes a strategy of exploration (e.g. entering a new cultural bloc) more relevant.

Barkema, Bell and Pennings (1996) measure cultural distance in several ways in their research in order to reduce the extent to which their findings are method-bound. They find that the resulting evidence confirms the previous results on key notions of the Uppsala internationalization process model, namely, cultural barriers are very relevant in the foreign entry process and companies learn about cultural barriers through time from their earlier expansions.

Based on these findings, it can be concluded that countries initially appear to invest in countries that are less ‘psychically distant’ (e.g. in terms of culture and language) and as time passes, they will invest more and more in countries that are more ‘psychically distant’. Therefore, the positive moderating effect of cultural distance will diminish over time, since this variable will become less relevant in investment decisions as time passes. This is stated in proposition 2 below.

Proposition 2: Because companies gradually enter other markets that are further away in psychic distance terms, the positive moderating effect of cultural distance on the negative relationship between corruption and FDI, will diminish as companies develop along their internationalization process.

As explained above, the larger the cultural distance, the less inclined investing countries are to invest in host countries given a certain level of corruption. And the smaller the cultural

(44)

44

4.2.3 Counterarguments

This sub-paragraph discusses some of the counterarguments that could be given regarding the two propositions. Some crucial elements come to light that have to be kept in mind (e.g. proper measurement of the variable ‘cultural proximity’), but also some counterarguments that in itself can be disputed, as will become clear.

Measurement of cultural proximity – Some counterarguments can be made against the propositions above. For example, Mariscal et al. (2012), in their study regarding cultural proximity and banking FDI, question their measure of cultural proximity. They state that it is arguable whether ‘use of the same language’, as a proxy for cultural proximity, fully

represents the concept at hand.

It is important to make sure that the measurement of cultural proximity really captures what one intends to capture in order to determine whether cultural distance positively

moderates the negative relationship between corruption and inward FDI. This means that it is crucial to ensure the internal validity of the research, which refers to (as mentioned in the second chapter regarding the methodology), the extent to which the researcher really measures the concepts that are intended to be measured (Thomas, 2004: 31).

Mariscal et al. (2012) explain that, in the field of cross-cultural studies, previous studies have developed a better measure than the measure using only language (which simplifies the relationship). These other studies also add factors of communication, cultural values and other cultural dimensions.

Importance of psychic distance – Because the original Uppsala internationalization process model is relatively old (Johanson & Vahlne, 1977), it has been revisited in the light of theoretical advances that have been made and also the changes in business practices since 1977 (Johanson & Vahlne, 2009). Johanson and Vahlne (2009) discuss that the business environment is now considered to be a network (a web of relationships), instead of a neoclassical market consisting of independent suppliers and customers. They explain that outsidership, in relation to the specific network at hand, even more than psychic distance, is the cause of uncertainty. A company that does not hold a position in the relevant network is considered to be an ‘outsider’.

CORRUPTION, AN INCENTIVE OR DISINCENTIVE FOR FOREIGN DIRECT INVESTMENT? AN EXTENSIVE LITERATURE REVIEW AND QUANTITATIVE EMPIRICAL ANALYSIS