The influence of corruption on FDI in high tech and low tech industries

(1)

Master Thesis, June 2018

The influence of corruption on FDI in

high tech and low tech industries

University of Groningen

Faculty of Economics and Business

MSc. International Business and Management

(2)

2 | P a g e

Abstract

1

As there has been a steady rise in foreign direct investment (FDI) over the years, numerous studies investigated the determinants, the flows and obstructions of FDI. These studies discovered that one of the obstructions for FDI is host country corruption. In the more recent years, multidimensional approaches have taken the light and show interesting results on factors moderating the direct relationship. The aim of this study is to further elaborate on that relationship and add a moderating variable. As high tech firms have a bigger investment in R&D and higher value of their intangible assets, they are likely to be more vulnerable than low tech firms. This study contributes to the literature by adding the technological intensity as a moderating variable. A negative binomial model analysis has been conducted on a sample that consists of 63 manufacturing firms with headquarters in The Netherlands, with a total of 3864 subsidiaries. These data is obtained from Orbis and The World Data Bank. Findings suggest indeed that the positive relationship between host country corruption and FDI exists, however it is not confirmed that high tech firms avoid high corrupt host countries more than low tech firms.

Keywords: Foreign direct investment, corruption, corruption perception index, technological intensity, high tech industries

(3)

3 | P a g e

1_{I would like to thank my supervisor Dr. V. Iurkov for his support, guidance and constructive}

feedback throughout the process of writing this thesis.

---

List of Figures & Tables

List of figures

Figure 1 – Corruption perception index of 2016………... 11 Figure 2 – Conceptual Model……….. 16

List of tables

(6)

6 | P a g e

1. INTRODUCTION

As firms grow and the world converges, stakeholders push for internationalization. The goal is to increase potential earnings by either resource, market, efficiency or strategic assets seeking (Cui, Meyer, & Hu, 2014; Franco, Rentocchini, & Marzetti, 2008; Ramamurti & Hashai, 2011). Firms have multiple options to cross borders to establish internationalization, of which foreign direct investment (FDI) is the most popular one. FDI are investments made by firms from their home country into a host country. This is done through acquiring foreign business assets, which can either be full ownership or a controlling interest in the foreign firm.

The economic development of countries is influenced by the FDI. Especially less developed recipient countries are positively affected by this (Cuervo-Cazurra, 2006; Gohou & Soumaré, 2012; Habib & Zurawicki, 2001; Teixeira & Guimarães, 2015). Many scholars have studied how FDI differs from one host country to another and which factors are important in understanding these differences (Bjorvatn & Søreide, 2014; Brouthers, Gao, & McNicol, 2008; Cui, Meyer, & Hu, 2014). A great part of this literature covers market size, labor costs, innovation proficiencies and institutional quality as main determinants for FDI (Blonigen, 2005; Faeth, 2009; Teixeira & Guimarães, 2015), but there is much more to evaluate within these determinants.

(7)

7 | P a g e

implies that the presence of corruption in host country alone does not completely repel firms to invest FDI. Corruption also influences the composition of FDI (Brouthers, Gao, & McNicol, 2008; Smarzynska & Wei, 2000); the change in FDI is not only dependent on level of corruption in the host country (Hakkala, Norbäck, & Svaleryd, 2005), but also depends on corruption in the home country (Cuervo-Cazurra, 2006) and the corruption distance (Godinez & Liu, 2015; Qian & Sandoval-Hernandez, 2016). Corruption distance is the difference in corruption level between home and host country. Besides these observations, there may be other possible variables influencing the FDI in- and outflow besides these observations, such as markets, segments, geographical areas or industries. These variables however are not widely studied yet. It is therefore interesting to investigate which conditions enhance or diminish the effect of corruption on FDI.

As Cuervo-Cazurra (2006) suggested, it can be of high interest to study the effect on corruption and FDI of firms at the industry level. Ramamurti & Hashai (2011) stated that different industries are impacted differently by corruption, but did not elaborate on this topic. It is however imaginable that firms with high technological sophistication for instance are influenced differently by corruption in host country, as these firms are greatly concerned with protecting their intangible assets to hold the competitive advantages over competitor firms (Cohen, 2005; Malone & Rose, 2006; Smarzynska & Wei, 2000). These intangible assets are long term resources of a firm, but have no physical existence such as patents, education levels and business methodologies and are not to be openly shared with other firms. In order to protect their most valuable assets, high technological firms may avoid high corrupt countries and thus influences FDI from high tech firms based on level of corruption in the host country.

(8)

8 | P a g e

lacuna of knowledge on this topic. This study investigates whether the effect of corruption on FDI is differentiated by technological intensity of firms. Therefore, the research question posed in this study is:

Are FDI of high tech firms more influenced by host country corruption than FDI of low tech firms?

In the first hypothesis, the overall impact of host country corruption on FDI will be analyzed. The second hypothesis will investigate the moderating effect of technological intensity on this relationship.

(9)

9 | P a g e

2. LITERATURE REVIEW

2.1 Literature

Foreign direct investment

It is not surprising that stakeholders push for internationalization, since studies found a link between internationalization and improved performance at firm-level (Gattai & Sali, 2016). For example, firms expanding international have been found to have higher returns than firms without these internationalization advantages (Malone & Rose, 2006). Also, other studies show that firms engaging in FDI are more profitable than firms who do not engage in FDI (Gattai & Sali, 2016). A lot of research has been put into FDI trying to interpret these findings (Cui, Meyer, & Hu, 2014; Franco, Rentocchini, & Marzetti, 2008; Gattai & Sali, 2016; Malone & Rose, 2006).

The popularity of internationalization shows greatly in the increase in FDI volume after the liberalization of many FDI recipient countries in 1986 (Habib & Zurawicki, 2002). These countries were very receptive to investments and caused an annually FDI surge of 20% (UNCTAD, 2001). The great worldwide interest in FDI has led to an absolute increase in FDI volume from $196 billion in 1990 to $2400 billion in 2016, with a spike of $3000 billion in 2007 (The World Bank, 2018).

(10)

10 | P a g e

further dug into this factor. They found that corruption itself is not a factor for changes in FDI, but rather corruption distance between countries is the key determinant (Godinez & Liu, 2015). Corruption distance can be defined as the difference in corruption levels between country pairs on bilateral FDI (Habib & Zurawicki, 2002). Corruption can play a big role in the decision whether to invest through FDI in a given country and corruption distance may explain why there is a large amount of FDI going towards countries with a high corruption level. Cuervo-Cazurra (2006) also stated that not only the host country level of corruption matters, but also the home country corruption level.

Although this all sounds promising, participating in FDI is not without risk. Chang & Lu (2011) described failure in product quality control as the main risk in engaging in FDI (Chang & Lu, 2011). While according to Busse & Hefeker (2007) important determinants for outcomes of FDI are government stability, internal and external conflict, ethnic tensions, law and order, democratic accountability of government, quality of bureaucracy and corruption (Busse & Hefeker, 2007). Since there are many factors influencing the outcome of FDI, a thorough risk assessment should be performed prior to investing in FDI to anticipate on possible obstacles.

Corruption

(11)

11 | P a g e

life in underdeveloped countries, often only foreigners living in these countries, entrepreneurs and intellectuals are the ones complaining about corruption (Leff, 1964). Corruption is commonly seen as an additional tax on doing business and a waste of resources. Furthermore, even when the bribes are paid, it does not ensure the enforceability of contracts (Qian & Sandoval-Hernandez, 2016).

Despite the disadvantages of corruption, it can also be used as a tool to speed processes and reduce some costs by cutting through the bureaucratic maze (Leff, 1964; Meon & Weill, 2010). The preservation of corruption is caused by two sides. On the one hand there is the demand side by the public where regulations, tax systems, spending decisions and provision goods and services below market prices are key factors. On the other hand there is the supply side by public officials. This supply side is fueled by tradition & example figures, public sector wages, penalty systems, institutional controls and transparency (Tanzi, 1998).

Although corruption was almost a taboo to write about in the early seventies (Leff, 1964), it is now a widely studied topic. Also numerous regulations and laws are arranged to combat this phenomenon. Examples of the battle against corruption are the recent guidance (EY, 2015) and major settlements (Travieso, 2017) to combat corruption by the US Department of Justice (DOJ) and the US Securities and Exchange Commission (SEC) after they handed out warnings for several years to comply with the Foreign Corrupt Practices Act (FCPA).

(12)

12 | P a g e

challenging. Tanzi (1998) goes even further and states that it is impossible to measure corruption, however it is possible to measure perceptions of corruption. The CPI does exactly that and is based on perceived levels of public sector corruption according to experts and businesspeople. It uses a scale of 0 to 100, where 0 is highly corrupt and 100 is very clean (Transparency International, 2018). The CPI of TIO is the most valid and reliable source available regarding this subject and is widely used (Sanyal & Samanta, 2004). More than two thirds of the countries worldwide score below the threshold 50 and can therefore be labeled as corrupt (Hessler, 2018). The average score of 43 is even lower and thus corruption can still be seen as a huge problem around the world.

Figure 1 Corruption perception index of 2016. Score 0 is highly corrupt and 100 is very clean (Transparency International, 2018)

(13)

13 | P a g e

with laws against bribery have together taken steps against corruption because they see corruption as a negative factor, and in result restrains FDI to corrupt countries (Drabek & Payne, 2001; Habib & Zurawicki, 2002; Teixeira & Guimarães, 2015). Surprisingly, Cuervo-Cazurra (2006) found also that higher host country corruption results in relatively higher FDI from countries with high levels of corruption. Therefore, the research at hand will focus on only one country of origin. A country which is also a member of the OECD, namely The Netherlands.

Cuervo-Cazurra (2006) used worldwide statistics on total FDI per country and source and created countrywide findings for the average firm. As mentioned in his article, homogeneity of industries is assumed and therefore no differences within industries are investigated. Markets, segments, geographical areas or industries were not differentiated for. A more recent study on Swedish firms has analyzed the difference between horizontal and vertical FDI at the firm level and found that the goal of FDI is influenced differently by corruption. (Hakkala, Norbäck, & Svaleryd, 2005). Horizontal investments are made to increase the local sales at the host country and seem much more affected by corruption than vertical investments, which are aimed at lowering costs and exporting back to the home and other countries. They found that not only the type of investment is affected differently by the level of corruption, but it also influences firms different based on the size of a firm.

(14)

14 | P a g e

Industry technology intensity

The average firm has a decline in FDI to high corrupt countries and the average FDI from firms of OECD countries to high corrupt countries is even more negatively moderated by corruption. Ramamurt & Hashai (2011) argued that industries may be impacted differently by corruption. Industries can be subdivided. Low tech industries for example are fully focused on production itself with lower investments towards research and development. Whereas high tech industries are much more complex, with advanced innovations and have a much shorter life cycle of products (Atmer & Thagesson, 2005). High tech industries have invested much more resources into developing the features of their product and thus they are in need of trustworthy parties to do business with. The importance of protecting the firm its intangible assets increases with the technology knowhow of the firm (Malone & Rose, 2006; Smarzynska & Wei, 2000). These intangible assets hold the competitive advantages over competitor firms and thus are of great significance to keep (Cohen, 2005). As high tech firms have much higher R&D investments, they are likely to have higher additional costs of production because of corruption than low tech firms. These intangible assets are higher valued than their low tech counterparts. It is rather important to not lose these advantages and therefore it would be susceptible that high tech firms avoid high corrupt countries. Multinational firms should be cautious when choosing host countries for their subsidiaries, because corruption brings increased uncertainty and additional costs (Kwok & Tadesse, 2006).

(15)

15 | P a g e

moderates the effect of corruption on FDI. The aim of this study will be to investigate if high tech firms are impacted more by host country corruption than low tech firms.

(16)

16 | P a g e

2.2 Theoretical Framework

Some studies have argued that corruption can have a positive impact on FDI by speeding up transactions in heavily regulated countries and enables firms to work through the bureaucratic maze (Leff, 1964; Smarzynska & Wei, 2000). However, more recent studies show that corruption does harm FDI. Brouthers et al (2008) concludes that resource-seeking FDI is heavily impacted by host country corruption. Habib & Zurawicki (2002) state that foreign investors avoid corruption because it creates operational efficiencies. Qian & Sandoval-Hernandez (2016) argues that especially the corruption distance matters in choices of FDI. And Smarzynska & Wei (2000) find that corruption reduces inward FDI.

The research will focus on firms with headquarters based in The Netherlands in order to prevent the country of origin effect. As The Netherlands is highly international, data is widely available and they are a member of the OECD; it is a perfect candidate to perform the research on. To test the validity of the central relationship (H1), the first model tested the main effect without

interference of the moderator.

H1: higher host country corruption lowers FDI towards that country

It is a generally accepted theory that firms from OECD countries have less FDI towards corrupt countries (Cuervo-Cazurra, 2006). In order to protect their intangible assets, it is expected that high tech firms will be even more cautious to enter corrupt countries than low tech firms. The average relationship of the influence of corruption on FDI might therefore be moderated by the firm its technological intensity. Industry technological intensity will therefore be added as a moderator (H2).

H2: For high tech firms, the negative relationship between host country corruption and

(17)

17 | P a g e

The literature has widely agreed that high corruption in the host country reduces the FDI from low corrupt countries towards these countries. With this second hypotheses it can be analyzed if this relationship is stronger for high tech firms.

Figure 2 – Conceptual Model

(18)

18 | P a g e

3. DATA AND METHODOLOGY

3.1 Method

The aim of this research is to give more insight into the influence of level of corruption of the host country on FDI moderated by the technology intensity of firms. Quantitative data are used for statistical analyses in Stata 15. The datasets from Orbis (Orbis, 2018), Transparency International Organization (Transparency International, 2018) and The World Bank (The World Bank, 2018) were combined into one dataset. Cuervo-Cazurra (2006) found that corruption changes the composition of country of origin of FDI. Therefore, this research focusses on only one home country, The Netherlands, to counter this effect. In total, 3864 subsidiaries observations from 63 international manufacturing firms with headquarters located in the Netherlands were selected for data collection. The hypotheses test both for the main influence of corruption on FDI and the moderating effect of technological intensity.

Dependent variable - foreign direct investment

(19)

19 | P a g e

Independent variable – host country corruption level

FDI, as measured in number of foreign subsidiaries, will be tested for influences through corruption. The level of corruption is measured by the host country CPI level. This is in line with Teixeira & Guimarães (2015) their research were they found that a lower level of corruption enhances FDI attraction. Also, Habib & Zurawicki (2002) used the CPI to study the impact of corruption on FDI. The CPI is an index scaling 1-100 where 1 is the highest corrupt and 100 is the lowest corrupt. The Netherlands for example has a CPI of 82 and ranks 8th around the world. The highest CPI in the database is Venezuela with score 18 and the lowest CPI is found in New Zealand with score 89.

Moderator – industrial technology intensity

(20)

20 | P a g e

Control variables

The validity of any research is tested by adding a variety of control variables. The first control variable used in this study is firm age measured in years. The second control variable is firm size measured in number of employees. Thirdly, there will be controlled for GDP growth per capita. The last control variable is total number of subsidiaries per firm.

3.2 Sample Selection

Firms are included in the dataset based on specific requirements. Firstly, to follow the study by Hall and Vopel (1997), only manufacturing firms are included. These are classified within the North American Industry Classification System (NAICS) with codes 31, 32 and 33 (United States Department of Labor, 2018). Secondly, firms need to have their headquarters located in The Netherlands. Thirdly, a mandatory minimum of ten subsidiaries is required per firm. This amount is chosen to reduce influences of specific requirements of smaller firms which may influence choice of country for subsidiaries. Based on these in- and exclusion criteria, 63 firms with a total 3864 subsidiaries abroad are selected for data collection. The full list of firms included in the sample can be found in appendix A, table A2.

(21)

21 | P a g e

from the database. The location of the remaining subsidiaries were then linked to their corresponding CPI host country codes.

3.3 Methodology Empirical model

Based on the compiled high tech SIC codes list by Kile and Philips (2009), firms will be either classified as high (1) or as low (0) tech. This divides the sample into two subgroups and statistical analyses within Stata will determine the different influence the technological intensity moderator has.

The data in the sample will be tested within Stata across multiple tests. In order to get a first indication about the different variables, the descriptive statistics table will be produced. Descriptive statistics are ways to calculate, describe and compile the data from the sample in efficient and clear manner (Vetter, 2017). Thereafter, the correlation matrix will be composed to check for multicollinearity. It measures whether variables are independent from each other and thus do not affect the regression results.

Since count variables are used and the mean of the outcome variable is greater than the standard deviation, a negative binomial regression is performed for three different models. Firstly, the control only variables model will be constructed to test how the control variables influence the dependent variable. In the second model, the variables host country CPI, technology intensity, age of the firm, GDP growth and total subsidiaries per firm are included. The estimates in this second model should give an answer on the relationship between CPI and FDI, as stated in the first hypothesis (H1). Thirdly, the moderation model will incorporate the moderator high and

(22)

22 | P a g e

its technological intensity on the effect of corruption on FDI and thus is able to give an answer on the second hypothesis (H2)

(23)

23 | P a g e

4. RESULTS

4.1 Descriptive Statistics

Table 1 presents the descriptive statistics for the sample used in the study. Technology intensity is measured by classifying firms in either 1 for high tech and 0 for low tech. From the included firms, 67% is defined as a high tech firm and 33% is classified as a low tech firm. The average age of included firms was 56 years old, where the most recently founded firm was 2 years old and the oldest was 141 years old. The average amount of subsidiaries per firm was 229, with a maximum of 901 subsidiaries abroad. The mean number of subsidiaries per country is 3, with a minimum of 1 and a maximum of 95 subsidiaries. The CPI score is measured from 1-100, where 1 is highly corrupt and 100 is not corrupt. The mean CPI score of the host country is 59, which means that the average host country with subsidiaries from manufacturing firms from the Netherlands has a lower level of perceived corruption than the worldwide average level of 43. It also shows that these firms have subsidiaries in high corrupt countries as well as low corrupt countries.

Table 1: Descriptive statistics of the firms included in the sample

Variables Mean Std. Dev. Min Max

Subsidiaries per country 2.96 6.24 1 95

CPI host country 59.15 19.49 18 89

Technological intensity 0.67 0.47 0 1

Firm size 8.01 2.85 2.56 12.38

Firm age 56.05 38.70 2 141

GDP growth per capita 1.54 2.03 -4.38 12.10

(24)

24 | P a g e

4.2 Correlation Matrix

Variables in the sample are tested for multicollinearity (Alin, 2010). When there is a very high correlation among the variables, it might affect the regression results and therefore results may not be reliable. The total number of subsidiaries per firm are slightly correlated with firm age and firm size. This indicates that bigger firms and older firms tend to have more subsidiaries. It is also estimated that there is a moderate correlation between the technology intensity and the size and number of recorder firms. The firms were either classified as low or high tech and therefore it can be stated that high tech firms have more employees and more subsidiaries. None of the other variables have any interesting correlation with another variable. The results can be found in table 2.

Table 2: Correlation matrix

Variables Subsidiaries per country CPI host country Tech. intensity

(25)

25 | P a g e

4.2 Empirical Findings

The controls only model can be seen in table 3. Negative binomial regression show that the number of recorded subsidiaries have a significant relationship with the subsidiaries per country. It shows that for every 1 extra recorded subsidiary, the subsidiaries per country grows with 0.0025. This is found to be statistically significant at the P=0.00 level. The other variables have no significant relationship with the dependent variable.

Table 3: Controls only model (Model 1, with host country fixed-effects). Subsidiaries per country Coefficient Robust Std. Error

Firm size -0.032 0.028

Firm age -0.002 0.001

GDP growth per capita 0.131 0.108 Subsidiaries per firm 0.003*** 0.000 * P < 0,1; ** P < 0,05; *** P < 0,001.

Number of observations = 1,000. R2 = 0.1796. Pseudolikelihood = -1831.4215

Results of the interaction between host country corruption and FDI

The second model, which can be seen in table 4, has added the independent variable CPI in the negative binomial regression in order to test hypothesis H1: higher host country corruption

lowers FDI towards that country. The model shows that there is indeed a positive significant relationship between higher CPI and higher number of subsidiaries present of that country. The coefficient is 0.018 (p = < 0,05), which means that for every increase of host country CPI level by 1, there will be an additionally 0.018 subsidiary present. The hypothesis H1 can be

(26)

26 | P a g e

his study wherein he finds that OECD member countries have lower FDI towards high corrupt countries.

Table 4: Independent variables model (Model 2, with host country fixed-effects) Subsidiaries per country Coefficient Robust Std. Error

CPI host country 0.018** 0.007

Technological intensity -0.288** 0.106

Firm size -0.018 0.190

Firm age -0.003* 0.001

GDP growth per capita -0.208** 0.072 Subsidiaries per firm 0.003*** 0.000 * P < 0,1; ** P < 0,05; *** P < 0,001.

Number of observations = 1,000. R2 = 0.1841. Pseudolikelihood = -1821.468

Moderated effect of high tech industries

In the third model, which can be found in table 5, the moderator variable technological intensity is added. To create the moderating variable, CPI host country levels is multiplied by technological intensity and measures how technological intensity moderates the effect found in hypothesis H1. Table 5 shows the results for hypothesis H2, which was: For high tech firms,

the negative relationship between host country corruption and FDI is stronger. With a coefficient of 0.002 (p = > 0,1), there is no evidence that the relationship tested in hypothesis H1 is stronger for high tech firms. This indicates that the influence of CPI on FDI is not

(27)

27 | P a g e

Table 5: Moderation model (Model 3, with host country fixed-effects)

Subsidiaries per country Coefficient Robust Std. Error

CPI host country 0.018** 0.007

Technological intensity -0.425** 0.175

CPI host country X technological intensity

0.002 0.003

Firm size -0.017 0.019

Firm age -0.03* 0.001

GDP growth per capita -0.229** 0.074 Subsidiaries per firm 0.003*** 0.000 * P < 0,1; ** P < 0,05; *** P < 0,001.

(28)

28 | P a g e

5. CONCLUSION AND DISCUSSION

5.1 Conclusion

The interest from scholars on the effect of corruption on FDI has been expanded in the last decades. Some argue that corruption ‘greases the wheels’ (Leff, 1964) and therefore is efficient, faster and cheaper than going through the bureaucratic maze and poor functioning institutions. However, others contend that these assumptions may only be beneficial in a second best world (Meon & Weill, 2010). Most scholars therefore agreed that corruption is an additional tax on doing business and should be avoided at all times. There are various organizations to create awareness for the negative effects of corruption, such as the TIO and the OECD. There have been conducted multiple studies which show that host country corruption hinders FDI (Cuervo-Cazurra, 2006; Qian & Sandoval-Hernandez, 2016; Tanzi, 1998). In more recent years, scholars have tried to find variances within the established negative effect of corruption on FDI with multidimensional studies by adding a moderator (Hakkala, Norbäck, & Svaleryd, 2005). The aim of this study is to build further on Cuervo-Cazurra (2006) his study and investigate the moderating role of technological intensity.

A sample of 63 manufacturing firms with headquarters in The Netherlands, with 3864 subsidiaries observations, have been gathered. Statistical analysis revealed the following; firstly, a positive relationship between host country corruption and lower FDI has been confirmed. Secondly, the moderating role of technological intensity was examined and the results found no evidence for high tech firms to avoid high corrupt countries more than low tech firms.

Previous studies found that corruption deters FDI. Based on the findings in hypothesis H1, it

(29)

29 | P a g e

that high tech and low tech firms are not impacted differently by host country corruption. They both are negatively impacted by host country corruption, but one not more than the other.

5.2 Discussion

As the literature indicated would happen, host country corruption does indeed lower FDI towards that country. However, the expected moderating variable technological intensity, as suggested by Cuervo-Cazurra (2006), did not have any significant outcomes.

There have been numerous multidimensional studies on corruption examined and there are several positive outcomes. Moderators such as home country corruption level, vertical vs horizontal FDI, wholly owned subsidiary vs joint venture and corruptions distance have been confirmed by other studies. With this research, one can now eliminate the probability that manufacturing companies with headquarters located in The Netherlands are evading high corrupt countries more than low tech firms. If this holds for only The Netherlands, for OECD countries in general or for all countries has yet to be established.

(30)

30 | P a g e

which is rather subjective. The CPI has however been a well-established and widely used indication.

(31)

31 | P a g e

References

Alin, A. (2010). Multicollinearity. WIREs Comp Stat, 2: 370-374.

Atmer, J., & Thagesson, P. (2005). High tech vs Low tech companies- Variations in internationalisation process. Kristianstad: Kristianstad University.

Barbalet, J. (2017). Guanxi as Social Exchange: Emotions, Power and Corruption,. Hong Kong: SAGE Publications Ltd.

Bjorvatn, K., & Søreide, T. (2014). Corruption and competition for resources. Int Tax Public Finance, 21: 997–1011.

Blonigen, B. A. (2005). A review of the empirical literature on FDI determinants (NBER Working Paper No. 11299). Cambridge: National Bureau Of Ecomonmic Research.

Brouthers, L. E., Gao, Y., & McNicol, J. P. (2008). Corruption and Market Attractiveness Influences on Different Types of FDI. Strategic Management Journal, 29(6): 673-680.

Busse, M., & Hefeker, C. (2007). Political risk, institutions and foreign direct investment. European Journal of Political Economy, 23: 397-415.

Chang, P.-L., & Lu, C.-H. (2011). Risk and the technology content of FDI: A dynamic model. Journal of International Economics, 86: 306–317.

Chen, C., Ding, Y., & Kim, C. (2010). High-level Politically Connected Firms, Corruption and Analyst Forecast Accuracy Around the World. Journal of International Business Studies, 1502–1524.

(32)

32 | P a g e

Cuervo-Cazurra, A. (2006). Who cares about corruption? Journal of International Business Studies, 37: 807–822.

Cui, L., Meyer, K. E., & Hu, H. W. (2014). What drives firms’ intent to seek strategic assets by foreign direct investment? A study of emerging economy firms. Journal of World Business, 49: 488–501.

Drabek, Z., & Payne, W. (2001). The Impact of Transparency on Foreign Direct Investment. Geneva: World Trade Organization, Economic Research and Statistics Division. Dunning, J. H., & Lundan, S. (2008). Multinational Enterprises and the Global Economy,

2nd Edition. Cheltenham: Edward Elgar Publishing Ltd.

EY. (2015, December 14). EY Identifies Top Fraud and Corruption Trends for 2016. Retrieved from Cision - PR Newswire: https://www.prnewswire.com/news-releases/ey-identifies-top-fraud-and-corruption-trends-for-2016-300192127.html Faeth, I. (2009). Determinants of Foreign Direct Investment - A Tale of Nine Theoretical

Models. Journal of Economic Surveys, 23(1): 165-196.

Franco, C., Rentocchini, F., & Marzetti, G. V. (2008). Why do firms invest abroad? An analysis of the motives underlying Foreign Direct Investments -. Italy: SSRN.

Gattai, V., & Sali, G. (2016). FDI and heterogeneous performance of European. Econ Polit Ind, 43: 25–65.

Godinez, J. R., & Liu, L. (2015). Corruption distance and FDI flows into Latin America. International Business Review, 24 33-42.

(33)

33 | P a g e

Habib, M., & Zurawicki, L. (2001). Country-level investments and the effect of corruption: some empirical evidence. International Business Review, 10(6): 687–700.

Habib, M., & Zurawicki, L. (2002). Corruption and Foreign Direct Investment. Journal of International Business Studies, 33(2): 291-307.

Hakkala, K., Norbäck, P.-J., & Svaleryd, H. (2005). Asymmetric Effects of Corruption on FDI: Evidence from Swedish Multinational Firms. (Working Paper No. 641), The Research Institute of Industrial Economics.

Hessler, U. (2018, May 18). More than two-thirds of all countries corrupt, TI report says. Retrieved from Deutsche Welle: http://www.dw.com/en/more-than-two-thirds-of-all-countries-corrupt-ti-report-says/a-42682937

Kang, J. (2013). The relationship between corporate diversification and corporate social performance. Strategic Management Journal, 34: 94-109.

Kogut, B., & Chang, S. J. (1991). Technological Capabilities and Japanese Foreign Direct Investment in the United States. Review of Economics and Statistics, 73(3): 401–413. Kwok, C., & Tadesse, S. (2006). The MNC as an agent of change for host-country

institutions: FDI and corruption. Journal of International Business Studies,, 767–785. Lansink, L. (2016). Coopetition in low tech industries versus high tech industries. Enschede:

University of Twente.

Leff, N. H. (1964). Economic Development Through Bureaucratic Corruption. American Behavioral Scientist,, 8(3): 8–14,.

(34)

34 | P a g e

Meon, P., & Weill, L. (2010). Is corruption an efficient grease? World Development, 38(3): 244–259.

Orbis. (2018, May 10). Orbis. Retrieved May 10, 2018, from Orbis: https://orbis2.bvdep.com/

Qian, X., & Sandoval-Hernandez, J. (2016). Corruption Distance and Foreign Direct Investment. Emerging Markets Finance & Trade, 52: 400–419.

Ramamurti, R., & Hashai, N. (2011). The Future of Foreign Direct Investment and the Multinational Enterprise. Bingley, UK: Emerald Group Publishing Limited.

Ramasamy, B., Yeung, M., & Laforet, S. (2012). China’s outward foreign direct investment:. Journal of World Business, 47: 17-25.

Sanyal, R. N., & Samanta, S. K. (2004). Determinants of bribery in international business. Thunderbird International Business Review, 46(2): 133-148.

Smarzynska, B. K., & Wei, S.-J. (2000). Corruption and Composition of Foreign Direct Investment: Firm-Level Evidence. Washington DC: The World Bank.

Tanzi, V. (1998). Corruption Around the World: Causes, Consequences, Scope and Cures. International Monetary Fund.

Teixeira, A. A., & Guimarães, L. (2015). Corruption and FDI: Does the Use of Distinct. Journal of African Business, 16: 1-2, 159-179.

The World Bank. (2018, May 2). Foreign direct investment, net inflows (BoP, current US$). Retrieved from The World Bank:

(35)

35 | P a g e

Transparency International. (2018, May 08). Corruption Perceptions Index 2016. Retrieved from Transparency International - The Global Coalition Against Corruption:

https://www.transparency.org/news/feature/corruption_perceptions_index_2016 Transparency International. (2018, May 08). What Is Corruption. Retrieved from

Transparency International - The Global Coalition Against Corruption: https://www.transparency.org/what-is-corruption#define

Travieso, F. (2017, September 12). How can companies doing business overseas reduce the risk of corrupt practices? Retrieved from The Conversation:

http://theconversation.com/how-can-companies-doing-business-overseas-reduce-the-risk-of-corrupt-practices-83673

UNCTAD. (2001, July). World Investment Report 2001. Retrieved from United Nations Conference on Trade and Development: http://unctad.org/en/Docs/wir2001_en.pdf United States Department of Labor. (2018, May 20). Industries at a Glance - Manufacturing:

NAICS 31-33. Retrieved from Bureau of Labor Statistics: https://www.bls.gov/iag/tgs/iag31-33.htm#about

Vetter, T. R. (2017). Descriptive Statistics: Reporting the Answers to the 5 Basic Questions of Who, What, Why, When, Where, and a Sixth, So What? International Anesthesia Research Society, 125 (5): 1979-1802.

Volkov, M. (2012, March 22). High Tech and Corruption Risks. Retrieved from Corruption, Crime & Compliance: https://blog.volkovlaw.com/2012/03/high-tech-and-corruption-risks/

(36)

36 | P a g e

APPENDICES

Appendix A

Table A1 – List of high tech SIC codes High tech SIC

codes Industry

280 Chemicals & Allied Products 281 Industrial Inorganic Chemicals 282 Plastics Materials and Synthetics

283 Drugs

284 Soap, Cleaners, and Toilet Goods 285 Paints and Allied Product

286 Industrial Organic Chemicals

287 Agricultural Chemicals

289 Miscellaneous Chemical Products 351 Engines & Turbines

352 Farm & Garden Machinery

353 Construction and Related Machinery

354 Metalworking Machinery

355 Special Industry Machinery 356 General Industrial Machinery 357 Computer & Office Equipment 358 Refrigeration & Service Machinery

359 Industrial Machinery, NEC

361 Electric Distribution Equipment 362 Electrical Industrial Apparatus

363 Household Appliances

364 Electric Lighting and Wiring Equipment 365 Household Audio and Video Equipment

366 Communications Equipment

367 Electronic Components and Accessories 369 Misc. Electrical Equipment & Supplies

370 Transportation Equipment

371 Motor Vehicles and Equipment

372 Aircraft and Part

373 Ship and Boat Building and Repairing

374 Railroad Equipment

375 Motorcycles, Bicycles, and Parts 376 Guided Missiles, Space Vehicles, Parts 379 Miscellaneous Transportation Equipment 380 Instruments and Related Products

381 Search and Navigation Equipment 382 Measuring and Controlling Devices 384 Medical Instruments and Supplies

385 Ophthalmic Goods

(37)

37 | P a g e

481 Telephone Communications

482 Telephone & Other Communications

489 Communications Services, NEC

491 Electric Services

492 Gas Production and Distribution 493 Combination Utility Services

737 Computer and Data Processing Services 781 Motion Picture Production & Services

783 Motion Picture Theaters

(38)

38 | P a g e

Table A2 – List of firms included in the sample

Company name Firm Age No of recorded subsidiaries US SIC Primary code 1. ACCELL GROUP NV 20 62 375

2. AEG POWER SOLUTIONS B.V. 14 11 361

3. AIRBUS SE 18 480 372

4. AKZO NOBEL NV 49 615 283

5. AMSTERDAM FERTILIZERS B.V. 36 15 287

6. APOLLO TYRES COOPERATIEF U.A. 9 12 301

7. APOLLO VREDESTEIN B.V. 13 15 301

8. AUSNUTRIA OPERATIONS B.V. 29 17 202

9. AVIKO B.V. 56 12 203

10. BASF NEDERLAND B.V. 64 26 281

11. BAVARIA N.V. 88 25 208

12. BEKAERT COMBUSTION TECHNOLOGY B.V. 52 11 356

13. BOSAL NEDERLAND B.V. 29 42 344

14. BOSE PRODUCTS B.V. 40 12 367

15. BOSTON SCIENTIFIC INTERNATIONAL B.V. 25 18 283

16. BP NEDERLAND HOLDINGS B.V. 49 11 299

17. CIMPRESS N.V. 23 71 275

18. CNH INDUSTRIAL N.V. 6 314 353

19. COOPERATIE KONINKLIJKE COSUN U.A. 48 98 206

20. CORBION N.V. 99 107 206

21. CORDSTRAP B.V. 53 20 229

22. DAF TRUCKS N.V. 25 21 371

23. DE HEUS ANIMAL NUTRITION B.V. 91 27 204

24. DE HOOP TERNEUZEN B.V. 107 68 327

25. DOW AGROSCIENCES B.V. 29 50 287

26. DRAKA COMTEQ B.V. 14 11 366

27. DSM INTERNATIONAL B.V. 14 18 281

28. FERRING B.V. 35 57 283

29. FIAT CHRYSLER AUTOMOBILES N.V. 119 901 371

30.

FRIESLANDCAMPINA INTERNATIONAL

HOLDING B.V. 65 52 202

31. GIBSON INNOVATIONS NETHERLANDS B.V. 5 13 367

32. GREIF INTERNATIONAL HOLDING B.V. 72 22 341

33. HEINEN & HOPMAN ENGINEERING B.V. 42 15 356

34. HOYA HOLDINGS N.V. 29 21 382

35. INALFA ROOF SYSTEMS GROUP B.V. 18 11 371

36. KADANT JOHNSON EUROPE B.V. 61 16 355

37. KONINKLIJKE DSM N.V. 116 378 289

38. KONINKLIJKE PHILIPS N.V. 127 594 363

39. LIFE TECHNOLOGIES EUROPE B.V. 33 11 283

40. LYONDELLBASELL INDUSTRIES N.V. 9 279 286

41. NEDASTRA HOLDING B.V. 5 26 281

42.

NEDERLANDSE RADIATEUREN FABRIEK

B.V. 91 11 371

(39)

39 | P a g e

44. OCE HOLDING B.V. 141 24 357

45. OMRON EUROPE B.V. 29 19 384

46. ORFFA INTERNATIONAL HOLDING B.V. 29 12 209

47. QIAGEN NV 34 80 382 48. REMEHA GROUP B.V. 83 43 343 49. SEALED AIR B.V. 18 11 282 50. SIGNIFY N.V. 2 189 364 51. SMARTRAC N.V. 12 21 367 52. STMICROELECTRONICS N.V. 31 87 367 53. STRAUSS COFFEE B.V. 26 17 209 54. TERBERG GROUP B.V. 48 16 371 55. TIE KINETIX N.V. 31 14 737 56. TKH GROUP N.V. 38 179 335 57. TOMTOM NV 27 63 366 58. UNILEVER NV 91 419 209

59. VANDERLANDE INDUSTRIES HOLDING B.V. 44 20 353

60. VESCOM GROEP B.V. 47 13 239

61. WRIGHT MEDICAL GROUP N.V. 78 70 384

62. YOKOGAWA EUROPE B.V. 71 15 381

63.

ZUIVELCOOPERATIE FRIESLANDCAMPINA

The influence of corruption on FDI in high tech and low tech industries