Elephants or Donkeys* as head of the MNE

Elephants or Donkeys* as head of the MNE

Their effect on cross-border acquisition investment

commitment in unfamiliar environments

Course: MSc Thesis IB&M

Version: Finalized (on 16-06-2017) Author: Pim Boderie

Student number: S3032604 Supervisor: dr. R.W. de Vries Co-assessor: dr. M.H.F. Ridder

de van der Schueren

Word count: 11,225 (excl. references and appendices)

3

ABSTRACT

In a firm-level study, 215 US based firms have been analysed to test what effect political ideologies of CEOs have on cross-border acquisition investment commitments in unfamiliar environments. The thesis hypothesizes, by following the incremental internationalization literature, that a greater psychic (here: cultural and institutional) distance between the home and host country leads to lower cross-border investment commitments. The results of the empirical research proved that both hypotheses can be confirmed.

With a sample size of 122 American CEOs, the moderating effect of the executive’s political ideology has been tested. With the leading argument that conservative leaning individuals follow a risk averse strategy the thesis proposes that investment commitments made by firms that are led by a conservative CEO are lower in culturally distant countries, than the investment commitment of a liberal led firm would have been. However, conservative individuals have a strong believe in free markets with low regulatory pressures. Since the US has a developed institutional environment and because the paper only incorporates US based firms, large institutional distances do in this paper only occur in weaker institutionalized host countries. Therefore, one of the hypothesis argues that investment commitments made by US based firms that are led by a conservative CEO are higher in institutionally distant countries, than the investment commitment of a liberal led firm would have been. After running an empirical test, not enough evidence was found to prove the effect of political ideology of executives on investment commitment.

Even though not all hypotheses could be confirmed, this research contributes significantly to the IB literature by connecting relevant theories and manifesting interesting outcomes. In the discussion section, numerous research opportunities have been listed in order to improve future investigations towards this field of research.

Key words: Cultural distance, Institutional distance, FDI, Cross-border acquisitions, Political

5

ACKNOWLEDGEMENTS

Coming towards the end of this master thesis, I would like to express my gratitude towards the persons who helped me during this semester to complete my thesis successfully. First, I would like to thank my supervisor, dr. R.W. (Rudi) de Vries, for all the time and feedback that he provided to me throughout the entire semester. Furthermore, I thank all my close (study)friends for all the constructive conversations, discussions and revisions on my work. Concluding, also the moral support of my family helped me to successfully complete my final thesis at the University of Groningen.

Pim Boderie – 01-06-2017

ABBREVIATION LIST

Translation of the most applied abbreviations in this thesis:

CBA = Cross-Border Acquisition IC = Investment Commitment CEO = Chief Executive Officer MNE = Multinational Enterprise CSR = Corporate Social Responsibility UET = Upper Echelons Theory

-7

TABLE OF CONTENT

2.2 Host country specific factors ……… 11

2.2.1 Uppsala model ……….. 11

2.2.2 Cultural distance ……… 12

2.2.3 Institutional distance ………. 13

2.3 Political ideology ………... 15

2.3.1 Upper Echelons Theory ………. 15

2.3.2 Liberalism vs. Conservatism ………. 16

3. Methodology and descriptive statistics ………... 19

3.1 Sample and variable description ……… 19

3.2 Research methods ……….. 26

4. Results ... 29

4.1 Hypotheses testing ………. 30

4.2 Robustness checks ………. 33

5. Discussion and conclusion ... 35

4.2 Discussion and limitations ………. 35

4.2 Future research opportunities ……… 38

4.2 Conclusion ………. 39

Bibliography ... 40

Appendix 1 - Variable overview ... 46

Appendix 2A – Robustness tests cultural distance ... 48

Appendix 2B – Robustness tests Institutional distance ... 50

8

1. INTRODUCTION

“People have accused me of being in favour of globalisation. This is equivalent to accusing me of being in favour of the sun rising in the morning.”

Clare Short, in October 2000

(Former British Secretary of State for International Development)

The statement above from the British Labour politician Clare Short is just one of the countless indications that illustrate the vast amount of ongoing internationalization processes in today’s world. More and more firms decide to spread their products and services across borders. Acquiring a foreign firm is one possible strategy that firms pursue in order to achieve these internationalization goals. In fact, these cross-border acquisitions are today’s most important foreign direct investment vehicles (Zander & Zander, 2010) and higher investments are made into these acquisitions every single year (UNCTAD, 2016a). Still the growth in this business field is not accompanied by a growth in the corresponding literature.

So far, much literature that has been conducted clarifies which factors are important for the acquiring firm in determining its cross-border investment commitment. Here, both target country and target firm specific factors seem to matter. On a country level analysis, the proximity to customers, the mitigation of trade barriers, a highly driven workforce, and mounting local markets (Overesch & Wamser, 2010) are determined to be important selection criteria. Access to valuable resources (Barney, 1991; Hajzler, 2014) and unique competencies (Hitt, Dacin, Levitas, Arregle, & Borza, 2000) are proven to be imperative firm selection criteria.

9

institutional distance between home and host country affect investment commitments of cross-border acquisition deals. Investment commitment can be explained by the willingness of firms to invest large amounts of money into a project (which is in this thesis: a foreign firm).

Next to the external push and pull factors mentioned above, there is a set of internal aspects that may influence the outcomes of the internationalization process. The current literature fails to incorporate some internal actors which may influence internationalization decisions that might be worth looking at. Therefore, this study adds a new and under-researched actor to the academic field, namely political ideologies of individual executives. More specifically, liberal ideologies on the one hand and conservative ideologies on the other. So far, some scholars were able to identify the effect of political ideologies of executives on areas such as CSR (Chin, Hambrick, & Trevino, 2013) and tax sheltering (Francis, Hasan, Sun, & Wu, 2016), but no research was conducted on investment commitment in cross-border acquisitions.

Hence, this study seeks what effect political ideologies of CEOs have on cross-border acquisition investment commitment in unfamiliar environments. More specifically, only firms

and executives originated from the United States are incorporated in this thesis. This decision has been made since the political preferences in the United States consist in most cases out of solely two clearly distinctive ideologies, namely either republican (often referred to Elephants) or democratic (often referred to Donkeys).

The answers of the above mentioned research question contributes further to the literature to gain understanding what actors influence investment commitment in cross-border acquisitions. From a managerial perspective, the answers may help multinational enterprises to recognize that the appointment of an individual executive with a particular political ideology, can shift the outcomes of cross-border acquisition investments.

10

2. THEORY

In today’s world, internationalization goals are an indispensable force in many firm’s corporate strategies. These aims do often lead to, the widely researched field of, foreign direct investment (FDI). According to Conconi, Sapir and Zanardi (2016), FDI is an advanced internationalization process that goes beyond solely importing and exporting. In the latter case, there is neither physical foothold nor control in the foreign environment or organization. This is in contradiction with FDI, where a firm obtains always some degree of control inside the host country’s operations (Kariithi, 2007).

On its turn, FDI can be categorized in multiple entry modes depending on their degree of control inside the foreign firm, varying from low levels of control (e.g. licensing) towards high levels of control (e.g. greenfield investments and cross-border acquisitions). Each entry mode is most suitable for a specific strategy and can therefore in many cases not be generalized by academics (Hollensen, Boyd, Marie, & Ulrich, 2011). With that argument, this study focusses itself on one subdivision of FDI, namely cross-border acquisitions (CBAs). This particular selection has been made as CBAs are today’s most important foreign direct investment vehicles, while comprehensive research on this field is lacking (Zander & Zander, 2010).

CBAs occur in any industry. Studies have shown that CBA deals in all industries are relatively comparable amongst each other. To some extent the total value of CBAs amongst every industry (relative to the total number of industry) does not differ significantly (UNCTAD, 2016b). In other words, the amount of investment per CBA seems relatively equal in every industry. Furthermore, sectors follow similar trends throughout time. Rao and Reddy (2015) found that CBAs in most sectors react parallelly when an event, as an economic crisis, occurs. Hence, this field of research is highly suitable for a study encompassing numerous industries.

11

literature denotes that around 70 till 90 percent of all acquisition deals fail (Martin, 2016; Zhou, Xie, & Wang, 2016). These significant percentages do partially arise from a simple source. Namely the need and greed from the acquiring CEO and/or board. They focus too much on gaining from the CBA instead of investing in the relation between acquiring and target firm (Grant, 2014).

2.2 Host country specific factors 2.2.1 Uppsala model

As an explanatory introduction of the independent variables, this thesis brings the Uppsala model to the table, which helps to explain the underlying arguments and the eventual hypothesis. According to Johanson and Vahlne (1977), firms internationalize on a deliberate pace. The model states that foreign market investments only increase if the host market knowledge has been increased on forehand. In other words, host countries with a distinctive market environment will at first receive smaller amounts of investments from an internationalizing firm, compared with countries with a similar market environment as the home country. In later stages, the internationalizing firm increases its investments in the dissimilar host environment, when the market knowledge has been enlarged. This incremental way of internationalization is extremely suitable for corporate risk management.

CBAs, which represent a predominant share of foreign direct investment, may also follow the path of incremental internalization. According to the Uppsala model and in order to reduce risks, the acquiring firms may adopt various strategies to decrease their investments in host environments that stand both cultural- and institutional wise far away from the home environment. To illustrate, an example where a US based acquirer, internationalizes towards both Canada and a China: According to the Uppsala model, the initial investment in the environment with a low psychic distance (here Canada), is assumed to be higher than the investment in the environment with a high psychic distance (here China). Due to the acquaintance with Canada, the US acquirer faces less uncertainty and is therefore more willing to make a large initial investment, while the internationalization towards China incorporates incremental steps and smaller investments at a time.

12

be one example of how the Uppsala model influences the height of foreign investment (Forsgren & Kinch, 1970).

A second practice to reduce investment commitment in a dissimilar host country environment is the acquirement of minority control shares in a foreign firm (Demirbag, Glaister, & Tatoglu, 2007; Xu, Pan, & Beamish, 2014). This strategy might be favourable in situations with high risks, i.e. unknown and dissimilar market environments. In this case, a firm can easily observe the practices of the target firm, in order to get acquainted with the host environment, before fully acquiring the target firm in a later phase. This practice mitigates both investment costs and risk related costs (and hence the potential costs of withdrawment).

With the above mentioned theory in mind, the paper now further encounters the cultural and institutional distance effects on internationalization strategies. More specifically how cultural and institutional distances influence foreign investment commitments.

2.2.2 Cultural distance

Cultural distance is one of the most researched and measured variables in today’s IB academic field. This conception is even more emphasised by Cho and Padmanabhan (2005), who argued that no IB study can be validated unless an explicit variable controlling for cultural distance is included. Therefore, this IB thesis utilizes cultural distance as an independent variable.

According to leading scholars, cultural distance can be defined as the extent to which the shared values, norms and beliefs in one country diverge from those in another (Hofstede, 2001; Kogut & Singh, 1988). These distances can also be found within corporate settings. In general literature states that, the greater the cultural differences between countries, the greater the variances between the firms’ managerial and organizational practices of these nations (Larimo, 2003).

13

entry modes. They exclude investment choices and commitment within one entry mode (here: Cross border acquisitions).

Only few scholars performed in-depth analyses regarding investment commitment within one entry mode. In a study regarding international cooperative ventures the authors assert that when there is widening cultural distance between the environments of two countries, lacking host country knowledge discourages large resource commitments (Lee et al., 2008). Further evidence suggests that cultural distance as measured by Hofstede’s cultural dimension of uncertainty avoidance, between firms from the United States and host countries has a significant effect on the share of sought equity by the acquiring party (Chari & Chang, 2009).

Furthermore, and as mentioned before, the Uppsala model argues that when the psychic distance increases, investment commitment decreases. Cultural distance is depicted as one subcategory of psychic distance (Johanson & Vahlne, 1977).

Hence, when combining all arguments above this thesis proposes that:

Hypothesis 1: A greater cultural distance between two countries results in a lower relative CBA investment commitment by the acquiring party.

Note that this thesis solely incorporates relative investments in its hypotheses. By discursive persuasion, it seems clear that large multinational enterprises with large amounts of total assets, revenues and financial resources have in general much more to spend than its smaller counterparts. Hence, it is statistically relevant to only address the relative investment of a firm.

2.2.3 Institutional distance

14

Similar to the cultural distance literature, the effect of institutional distance is mostly measured as an effect on entry mode choice and country selection. Theories argue that firms with a global strategy invest in host countries where institutional distances are smallest (Xu & Shenkar, 2002). Also countries with underdeveloped institutional environments found to obtain less FDI, including cross-border acquisitions, than its developed counterparts (Xu & Shenkar, 2002). In a study, analysing the corruption component of the institutional environment, Cuervo-Cazurra (2006) found that countries with high levels of corruption receive most FDI from firms that originated from countries with similarly high levels of corruption and vice versa. This finding further implies that firms tend to seek for comparable institutional environments while internationalizing. Furthermore, when the regulative and normative distances enlarge between two countries, a multinational enterprise prefers less ownership and low equity control over the foreign investment, respectively (Xu & Shenkar, 2002). Hence, the firm exerts less investment commitment towards the host firm.

Since this thesis aims to investigate CBAs made by firms and executives of the United States of America, additional literature of the American institutional environment is needed. Addressing the database developed by Kaufmann et al. (2011), it becomes visible that the United States enjoys high institutional standards. Therefore, it can be assumed, according to the arguments above, that American firms tend to have less investment commitment to firms that are situated in incomparable institutional environments, which in this case means countries with underdeveloped institutions. In other words, since the US has a developed institutional environment and because the paper only incorporates US based firms, large institutional distances can in this paper only occur in weak institutionalized host countries. This argument is supported by the world governance indicators database which states that the development of the institutional environment of the United Sates belongs to the global top 15% (Kaufmann, Kraay, & Mastruzzi, 2013).

Also, when referring back to the Uppsala model, the psychic distance pillar of the framework incorporates institutional components (e.g. political stability, education levels, business practices, etc.). And as mentioned before, the Uppsala model concludes that when the psychic distance increases, investment commitment in the foreign market and/or firm decreases (Johanson & Vahlne, 1977).

Hence, when combining all arguments above this thesis proposes that:

15

2.3 Political ideology

The other aim of this thesis is to test what the effect of political ideologies from CEOs is on their investment commitment of CBAs. Before incorporating political ideologies in this paper, it is relevant to first further investigate the role of CEOs inside a firm, addressing both personal behavioural- and power distribution aspects of his/her function.

2.3.1 Upper echelons theory (UET)

The central framework used in this paper is the logic of the upper echelons theory (UET). The fundamental argument of the UET is that every executive perceives certain circumstances dissimilarly than a collegial executive would have observed it (Hambrick, 2007). This event exists due to the fact that idiosyncratic preferences will always remain intact while dealing with human beings. These, so called, “individualized lenses” are originated though a set of personal characteristics, including factors, as experiences, traits, values and norms (Hambrick & Mason, 1984). Concluding and according to this theory, firm strategies are likely to vary amongst different CEO’s.

This is in contrast with neoclassical economy theories and reasoning from economists as Milton Friedman. From their point of view, managers and executives are supposed to solely act in the best interests of the shareholders. Personal preferences of any kind are therefore off the table and all decisions must be made on rational grounds only. However, in the current literature, most academics follow the path of the UET (Cannella & Holcomb, 2005). Therefore, the remainder of this thesis elaborates also on this theory.

16

2.3.2 Liberalism vs. Conservatism

This study distinguishes two types of political ideologies that executives can pursue, namely liberalism on one side and conservatism on the other. This discrepancy is exceptionally visible in the United States of America where the large majority of the population adheres either one of the above mentioned currents. In practice, this liberal and conservative ideology leads in most cases to a cling towards the democratic or republican party, respectively (Diaz-Veizades, Widaman, Little, & Gibbs, 1995; Dunlap & Allen, 1976). In last decades, this generalization has even been proven to be more significant than it was during the 20th century (Hetherington, 2009). In order to add some number to these statements Hutton, Jiang, and Kumar (2014) demonstrated that only 3% of the members of the republican party identified themselves as liberal, the remaining 97% were either moderately or strongly conservative. Hence, numerous studies follow this American generalization (see Francia, Green, Herrnson, Powell, & Wilcox, 2005 for reference). This thesis also pursues this common generalization.

In a study performed by Adams, Licht, and Sagiv (2011), the authors found that if CEOs either hold self- or other- regarding values, this would affect the extent to which he adopted a shareholder- versus a stakeholder oriented governance mode, respectively. Consistently, academia value conservative CEOs as shareholder-oriented and liberal CEOs as stakeholder-oriented (Tetlock, 2000). For example, liberal CEOs do care more and spend more financial resources on CSR related issues, which is in line with the stakeholder theory (Chin et al., 2013). Another study found that conservative CEOs receive higher payments, especially during economic prosperous times, than liberal ones (Gupta & Wowak, 2016), which demonstrates a high degree of self-regarding values that ultimately leads to a more shareholder oriented governance mode (Adams et al., 2011).

17

acceptance is essential in the internationalization process towards a cultural distant environment (Contractor, Lahiri, Elango, & Kundu, 2014).

Hence, when combining all arguments above, this thesis proposes that:

Hypothesis 3:

The negative relation between cultural distance and relative investment commitment per CBA will be strengthened when the acquiring firm is led by a conservative leaning CEO.

Even though most scholars are congenial about the above mentioned effects of cultural distance, when addressing the effect of political ideologies on CBAs in institutional distant environments, literature seems to be more ambiguous. As discussed previously, conservatives are more uncertainty avoidant while operating in a distant and unrelated environment (Contractor et al., 2014). Therefore, it seems logical that, due to its unfamiliarity with the host institutions, conservative CEOs would pursue less investment commitment in institutional distant environments.

However, when dealing with institutional environments, other forces seem to play a role. As addressed by Kostova (1997: p.180), regulatory institutions are “all laws and rules in a particular national environment which promote certain types of behaviors and restrict others”. Hence, a strong institutional environment is characterized by low corruption figures, independent regulatory organizations and enforcement of laws and policies (Kaufmann et al., 2011). In other words, institutions add boundaries to the market environment.

Being more specific by only encountering CBAs made by American firms, the arguments above come down to the following situation. The World Governance Indicators define the quality of the institutional environment in the US as very high. This means that when American firms internationalize, a great institutional distance can only occur when firms aim for countries with low quality institutions. In a broad sense, in this thesis, a high institutional distance is equal to low quality institutional countries, and vice versa.

18

Hence political ideologies can both have a negative and a positive effect on the relation of institutional distance on the investment commitment of a CBA. However, due to the abundance of evidence leaning towards the positive side, the paper presents the following exploratory hypothesis:

Hypothesis 4:

The negative relation between institutional distance and relative investment commitment per CBA will be weakened when the acquiring firm is led by a conservative leaning CEO.

*H4 - *H3 + CEO’s political degree of Conservatism (vs. Liberalism) H1 - Cultural distance of target country Relative investment commitment per Cross-border acquisition (CBA) Institutional distance of target country Host country Specific factors H2 - Firm size Host country experience Internationalization experience Vendor usage Control variables

19

3. METHODOLOGY AND

DESCRIPTIVE STATISTICS

In order to test the hypotheses of this study, a sophisticated analysis is required, supported by a sufficiently large data set. The complete data set is gathered from the database Zephyr (Bureau van Dijk). The database keeps track of millions of mergers and acquisitions made by firms. This particular study sought only for completed cross-border acquisitions made by American firms in the period of 2013 until March 2017, from which the deal value was

known. The total number of records of this search strategy displayed by Zephyr were 279 CBAs.

The final requisite set in this study is the exclusion of deals made by firms in the financial industry. These firms tend to make unproportionable amounts of investments per acquisition that are not in line with other industries (UNCTAD, 2016b). For strategic and accounting purposes they engage in significantly more CBAs than firms in other industries (Mudambi & Nicosia, 1998). This will cause issues when including the CEO’s political ideology, in which one CEO would be overly present in the dataset. Therefore, this thesis excludes CBAs made by firms active in the financial sector, as they potentially may bias results. Coming from a total number of 279 deals, 64 of them remained to be incomplete (mostly due to a lack of financial company data) and therefore were impractical for further analyzation. Hence, this thesis addresses a total amount of 215 cross-border acquisition deals, spread over 184 different CEOs. This fact shows that there have been CEOs that participated in more than one CBA. In order to avoid letting one CEO and one firm bias the results significantly, the study delisted all records of a firm when the CEO participated in more than four (> 2% of the total data set) CBA deals, in the set time frame.

Dependent variable – Investment commitment

20

assets and revenue have on general much more to spend than its smaller counterparts. Therefore, it is statistically relevant to only address the relative investment of a firm. The amount of capital the acquiring firm is committed to pay for a target firm as an percentage equivalent of the acquiring firm’s total assets. In other words, the total investment of one deal (I) will eventually be divided by the value of the total assets of the firm (A), within maximum one year before the CBA was finalized, in order to arrive at the relative investment commitment (IC):

(1) IC = I / A

The total assets of a firm have been derived from data in Orbis. This database displays extensive financial company data, including total assets. For any cases where the value of the total assets of a firm a year before the acquisition was inadequate, estimated or completely missing, additional annual reports and 10-K SEC reports were examined.

Independent variables Cultural distance

The cultural distance variable is measured through the Hofstede database. Due to its ground-breaking development, the Hofstede database is still used in an abundance of academic papers. Even though the framework originated a long time ago, it has not lost its value. A recent study showed that the numbers in the Hofstede model might have changed slightly over time, but all in the same direction which makes the model still viable (Beugelsdijk, Maseland, & van Hoorn, 2015). The paper also considered the use of other cultural databases such as the cultural dimensions of Schwartz. However the explanatory power of the Schwartz and Hofstede model is comparable and can both be used by IB scholars in order to arrive at reliable results (Drogendijk & Slangen, 2006).

21

(2) Cd = (Δ PDIus-targ) + (Δ IDVus-targ) + (Δ MASus-targ) + (Δ UAIus-targ) + (Δ LTOus-targ) + (Δ INDus-targ)

Institutional distance

The second independent variable incorporates the institutional distance between the home country (here United states) and the target country. The variable is measured via the World Governance Indicators (WGI) database created by Kaufmann, Kraay and Mastruzzi. Similar to the cultural dimensions of Hofstede, the WGI also measures the institutional quality of a country on six different dimensions (voice and accountability (VA), political stability (PS), government effectiveness (GE), regulatory quality (RQ), rule of law (RL) and control of corruption (CC)).

In the database, each country is bounded to a score varying from -2,5 to +2,5 on any dimension. A zero score represents moderate quality of the institutional dimension. A negative and a positive score denote a weak and strong quality of the institutional dimension, respectively. The difference (Δ) between the American and target country’s institutional score on every dimension will be summed in order to arrive at the institutional distance variable. A Cronbach’s Alpha reliability test (a=0,817) approved the usage of all six institutional scores into one factor. A minimum score of 0,7 was used as threshold (Peterson, 1994).

(3) Id = (Δ VAus-targ) + (Δ PSus-targ) + (Δ GEus-targ) + (Δ RQus-targ) + (Δ RLus-targ) + (Δ CCus-targ)

Moderator variable - political ideology

All firms that occur in the database of Zephyr are also archived in the Orbis database. Both databases are services owned by the overarching data sourcing company, Bureau van Dijk. With the BvD number in Zephyr, researchers are able to track the firm in Orbis. In Orbis it is possible to identify the CEO corresponding to the BvD number. The executive first, middle and last name were gathered, before starting with defining their political ideology. However, an important concern needed to be accounted for. Namely that the date of completion of the CBA, should fell later than the appointment date of the current CEO. For any CBAs in which this was not the case, the prior CEO who completed the deal was incorporated in the study instead.

22

Commission, records all contributions of minimal 200USD to political action committees, to national and local parties, to campaign committees for federal office and to individual candidates. For every CEO, the study reached for the opensecrets.org database to obtain political donation data. The Federal Election Commission database displays some personal information. In first, the full name of the donor (first, middle and last name) is exhibited. Furthermore, the contributor’s residence area and his / her occupation is public. In most occasions the CEO’s name was not the only name occurring in the system. Therefore each name needed to be verified before addressing them to one CEO. Not only the full names, but also short names, background, previous occupations and addresses were carefully checked. Bloomberg and LinkedIn acted in this study as validation systems. Bloomberg is a publicly available database exhibiting extensive executive’s information. LinkedIn provides similar information, but is, in contradiction with Bloomberg, not shaped by a third party, but by the executive him / her-self.

Hence, the paper measured each CEO’s political conservatism by investigating the degree to which the executive supported the Republican Party, as opposed to the Democratic Party. All donations to democratic parties and candidates by a CEO were added up and ultimately detracted from the total height of donations to the republican parties. In other words, a positive amount represents a republican favor, while a negative amount of total donations of one executive represents a democratic preference. For example if a CEO contributes with three donations of $500 each to democratic candidates and one donation of $200 to a republican candidate, his or her political ideology, with a strength of -$1,300, is liberal. However, no assumption can be made that a CEO contributing $15,000 to a republican party is ten times more conservative-minded than a CEO who donates $1,500. This thesis controls for this issue by scaling the donations into six groups, in order to determine the degree of conservatism, displayed in the table below.

Group Donation boundaries Label

0 Under < > $-25,000 Highly liberal ,2 $-24,999 < > $-7,500 Moderately liberal ,4 $-7,499 < > $-1 Slightly liberal ,6 $1 < > $7,499 Slightly conservative ,8 $7,500 < > $24,999 Moderately conservative

23

The histogram below illustrates all CEOs political ideology records. After examination it becomes visible that the group is negatively skewed with a value of -0,269 and has a mean of 0,513. These numbers imply that in general CEOs are leaning more towards a conservativism ideology, which is a fair representation of the proportions in the real business world and consistent with finding within prior research (Bonica, 2014; Chin et al., 2013; Gupta & Wowak, 2016).

Control variables

Prior experience internationalization

A prior existence of an owned subsidiary abroad does familiarize the firm with the process of internationalization (Eriksson, Johanson, Majkgard, & Sharma, 1997). As a result, the unfamiliarity regarding internationalization in general decreases, which eventually may lead to a higher investment commitment towards international investments (Johanson & Vahlne, 2003). Therefore this study controls for the prior experience of internationalization of MNE’s. The number of active foreign subsidiaries of the MNE is taken as a proxy for internationalization experience.

Prior experience host country

24

the experience in the country in which the CBA took place. Being already active in the host country before engaging in the CBA, acquaints the MNE with some degree of host country cultural- and institutional- environmental knowledge. In other words, the liability of foreignness mitigates (Zaheer, 1995). As a result the investment commitment increases (Johanson & Vahlne, 2003), The number of active subsidiaries of the MNE in the CBA’s host country is taken as a proxy for host country experience.

Firm size

Even though this thesis accounts for the distribution in wealth of large firms by solely calculating the relative CBA investment, firm size still needs to be addressed as a control variable. Due to the liability of size, larger firms do often have more resources to manage bad times and failures (Colombo & Delmastro, 2000). Hence, larger firms may invest a higher investment / total asset- ratio in CBAs than smaller firms would do (Moeller, Schlingemann, & Stulz, 2004). The total number of employees of the firm is used as a proxy for firm size. These figures are derived from Orbis and annual reports.

Vendor usage

The final control variable zooms in on a specific deal characteristic. Before engaging in a CBA, the acquiring firm can decide to assign a third party (so called, vendor). These specialized vending parties can help to overcome liabilities of newness faced by the acquiring party. As a result, firms may be more willing to ensure a larger investment as they are more secure about the CBA (Weber, Current, & Benton, 1991). Zephyr displays for each deal whether a vendor was appointed to streamline the acquisition. This led to a dummy variable in which the possible outcomes either denoted a 0 for no vendor used or a 1 for a vendor used.

For a more comprehensive overview of all variables one can consult appendix 1. This variable overview is provided as purpose for detailed clarification, enhanced transparency and possible reproduction of the study.

Descriptive statistics

25

SPSS provides. The first two mentioned methods are mostly implemented by scholars, as the latter is unnuanced and may lead to invalid results (Peugh & Enders, 2004).

List-wise deletion removes the entire record when only one value in the equation is missing. The CEO’s political ideology variable is crucial for the independent/dependent relationship analyzation and would therefore follow a list-wise deletion approach, which let in some models to 122 records instead of the total 215. However, list-wise deletion was in some cases unnecessary. Pair-wise deletion, which only removes a part of the equation, was therefore used for dealing with missing values in control variables that were not crucial for the regression.

Table 2 - descriptive statistics

Highlighting the most important descriptive statistics one can observe that on average a firm invests an amount that is equal to 6,36% of their total assets in a CBA. But with a relatively large standard deviation these percentages can fluctuate heavily from 0,001% till up to almost 44%. Further, and as indicated before, there is a small prevalence of conservatism CEOs on liberal CEOs, 51% on 49% respectively.

Variable Measurement or proxy N Minimum (lowest possible value) Maximum (highest possible value) Mean Std. deviation Relative

investment CBA deal value / _{total assets} 215 0,001 (0>) 43,79 (∞) 6,36 9,80 Cultural distance Hofstede difference US and host country 214 19 (0) 268 (600) 113,58 68,67 Institutional distance WGI difference US and host country 215 0,83 (0) 12,58 (30) 3,13 2,54 Conservatism index CEO

26

3.2 Research methods

The conceptual model seeks to explain the relation between independent, dependent and moderator variables. In this case, an OLS linear regression analysis is required. Before the test was executed, multiple assumptions and requisites had to be met.

In first, a regression analysis requires a sufficiently large data set. Guidelines for similar researches (with 3 or less predictor variables) demand for a minimum of around 100 records in order to perform a regression analysis of decent quality (Combs, 2010; Green, 1991; Nunnally, 1978). This thesis incorporates 215 records (and 122 in some models), and therefore passing this criterion resplendently.

Secondly, all outliers needed to be accounted for. This paper uses a method named winsorizing to encounter potential outliers. The data is winsorized at a point where the Z-values extend the commonly accepted threshold of 2,5 (Field, 2009). In other words, all records that exceeded a value of two-and-a-half times the standard deviation added on or detracted from the mean, were considered as outliers. Instead of deleting these records, winsorizing argues that the value of all outliers gets lowered to the highest (or lowest) non-outlier record. In total a set of eight data points of the CBA investment commitment have been lowered to the threshold. Winsorizing has also been applied on other variables that were prone to outliers.

The final assumption addresses the multicollinearity issue. In table below all dependent, independent and control variables are tested on multicollinearity in a Pearson correlation matrix.

***Correlation is significant at the 0.01 level (2-tailed) ** Correlation is significant at the 0.05 level (2-tailed)

Table 3 - Pearson correlation matrix

Investment Commit. Cultural distance Institutional distance prior experience international. prior experience host country Firm size Vendor usage DEP: Investment commitment

cross-border acquisition 1

Cultural distance -,189*** 1

Institutional distance -,215*** ,438*** 1

prior experience international. -,312*** ,086 ,135 1

prior experience host country -,220*** -,070 ,075 ,572*** 1

Firm size -,258*** ,090 ,083 ,426*** ,154** 1

Vendor usage ,015 ,038 -,163** ,099 ,062 -,029 1

27

Correlation between the two independent variables is 0,45 and statistically significant. This is fairly below the threshold of 0,7, which indicates signs of multicollinearity. Also all control variables included, not any value exceeds demarcation point of 0,7 (Nachtsheim, Neter, Li, & Kutner, 2004). Therefore the study concludes that no multicollinearity exists. The correlation with the highest measured value is found between the control variables of ‘prior internationalization experience’ and ‘prior host country experience’ (=0,572**). This fairly high correlation can be explained due to the fact that when firms already had prior experience in the host country of the CBA, they naturally also had internationalization experience.

One can further experience a negative value of prior internationalization experience, prior host country experience and firm size on the dependent variable. Even though these are solely loose correlations, there are already some inferences possible. Because, according to the explanation in the control variable description, the thesis expected a positive correlation on these three items instead. The result section will elaborate on this issue more thoroughly.

After testing and serving all assumptions the OLS regression was ran. At first the model was tested without moderating variables. The tested equations to find evidence for hypothesis one and two (that explain investment commitment (IC)) are denoted below.

(4) IC = β0 + β1Cultural distance score + β2Prior internationalization experience + β3Prior host country experience + β4Firm size + β5Vendor usage + Ɛ

(5) IC = β0 + β1Institutional distance score + β2Prior internationalization experience + β3Prior host country experience + β4Firm size + β5Vendor usage

+ Ɛ

28

(6) IC = β0 + β1(ZscoreCultural distance * ZscoreCEO’s conservatism) + β2Cultural distance score + β3CEO’s conservatism + β4Prior

internationalization experience + β5Prior host country experience + β6Firm

size + β7Vendor usage + Ɛ

(7) IC = β0 + β1(ZscoreInstitutional distance * ZscoreCEO’s conservatism) + β2Institutional distance score + β3CEO’s conservatism + β4Prior

internationalization experience + β5Prior host country experience + β6Firm

29

4. RESULTS

The main objective of this section is to analyze whether the previously debated set of hypotheses can be confirmed. Table 4 displays the core results of the OLS regression tests ran on all data. In total seven models have been established in order to provide answers to the four hypotheses.

Table 4 - regression analysis on cross border acquisition investment commitment

***Significant at the 0.01 level ** Significant at the 0.05 level * Significant at the 0.10 level

¹ Missing political ideology data led to a decrease in sample size in this model.

Regression on dependent variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

Mod. CEO Conservatism

x Institutional distance -0,018 (,994) CEO Conservatism x Cultural distance -0,037 (1,059) Indep. Institutional distance -0,662** (,262) -0,493 (,385) -0,666* (,347) Cultural distance -0,024** (,010) -0,015 (,014) -0,023* (,013) CEO Conservatism 6,266* (3,714) 6,067* (3,681) 5,631 (3,737) 6,414* (3,684) Contr.

prior experience international.

-0,042** (,019) -0,055** (,026) -0,036* (,019) -0,037* (,019) -0,047* (,026) -0,047* (,026) -0,049* (,026)

prior experience host country

-0,250 (,250) -0,139 (,334) -0,338 (,334) -0,250 (,247) -0,196 (,334) -0,241 (,338) -0,147 (,336)

Firm size (in Th.)

30

4.1 Hypotheses testing

Model 1 acts as base model of the thesis. This model solely incorporates all control variables, tested on the dependent variable. The most outstanding result in this model is the negative relation between internationalization experience and CBA investment commitment. Hence for every foreign subsidiary a MNE owns, the firm invests a lower amount of money in a CBA. To be precise, an amount equivalent to 0,042% of their total assets less. The coefficients and significance levels of the internationalization experience variable are relatively stable throughout all other models. Further, also firm size is partially significant throughout the models. This is all contradictory to the proposed effect of the control variables. Seemingly, experience and firm size make MNEs more careful in their foreign investment commitments.

Model 2 adds the CEO conservatism variable into the model, which is later also used to calculate the interaction term. A significant coefficient is denoted at the CEO’s conservatism index. The findings suggest that highly conservative CEOs, on average engage in higher investments per CBA (≈6,3% more relative to its total assets), than severe liberal executives.

Model 3 tests hypothesis one. Hypothesis one proposed that “A greater cultural distance

between two countries results in a lower relative CBA investment commitment by the acquiring party”. The regression test denotes a coefficient of -0,024 (p <0,05) amongst the variables. This

implies that an increase in the cultural distance between the home and host country does certainly lead to a lower investment commitment. Hence one can conclude that there is enough evidence to confirm hypothesis one.

Model 4 tests hypothesis two. Hypothesis two proposed that “A greater institutional

distance between two countries results in a lower relative CBA investment commitment by the acquiring party.” The regression test denotes a coefficient of -0,662 (p <0,05) amongst both

variables. This implies that an increase in the institutional distance between the home and host country does certainly lead to a lower investment. Hence one can conclude that there is enough evidence to confirm hypothesis two.

31

amongst both variables we divide the institutional coefficient through 20 (=600/30). This changes the coefficient to β’ ≈ -0,031 (= -0,662/20). Both independent variables do operate at the same measurement scale now which makes a comparison possible. Therefore, one can conclude that the institutional distance (β’ ≈ -0,031) has a greater influence on CBA’s investment commitment than cultural distance (β = -0,024) (-,024/ -,031 = 29% greater).

Combining all independent variables in a multiple regression test (as done in model 5) slightly increases the (Adjusted) R². However, in this case, the significant levels of both independent variables (cultural and institutional distance) converted into inappropriately high values (p>0,1). Consequently, they are not useable for further inferences. By referring to table 3 (Pearson’s correlation matrix), this outcome is not surprising as the independent variables already correlate significantly amongst each other with a relative high value of ,438**. Even though the sign of both independent variable’s coefficient remains negative, as proposed, no conclusions can be drawn due to insignificant p-values.

Model 6 tests hypothesis three. Hypothesis three proposed that “The negative relation

between cultural distance and relative investment commitment per CBA will be strengthened when the acquiring firm is led by a conservative leaning CEO”. Including the moderator into

the model does not exacerbates the negative effect tested in hypothesis one, which is denoted by the moderating coefficient of -0,037. A significant negative coefficient determines that the proposed negative independent-dependent relation (of -0,023) gets weakened when the CEO’s ideology is conservative leaning. In other words, conservative leaning CEOs do engage in higher CBA investment commitments in culturally distant countries, than liberal ones. This effect is contradicting to the expected outcomes. Nevertheless, for a valid interpretation of the results, the interaction effect needs to be statistically significant. The significance level is lacking in this outcome. Hence this thesis concludes that no inferences can be made from these results. Therefore, hypothesis three will be rejected.

Finally, model 7 tests hypothesis four. Hypothesis four proposed that “The negative

relation between institutional distance and relative investment commitment per CBA will be weakened when the acquiring firm is led by a conservative leaning CEO”. Including the

32

measured in model 5, this moderator shows a nonsignificant value (p>0,1), which is crucial for further inferences. Therefore, also hypothesis four will be rejected.

The Adjusted R²varies slightly amongst all seven models (from ,10 as minimum to ,13 as maximum). These values can be explained as follows: for example 12,1% of all the factors that affect CBA investment commitment are present in model 7. Even though this value seems relatively low, it is an acceptable percentage in social / behaviour studies, in which often countless factors do influence the dependent variable (Frost, 2013).

Summarizing the main results and answering the hypotheses of the study, table 5 is composed. It provides an overview of the conclusions to all hypotheses.

Table 5 – Overview hypotheses testing

¹ Non-significant interaction effects

HYPOTHESIS DESCRIPTION RESULTS

H1 Cultural distance has a negative effect on CBA’s investment commitment Confirmed H2 Institutional distance has a negative effect on CBA’s investment commitment Confirmed H3 The negative relation between cultural distance and CBA’s investment

commitment is stronger when CEOs are leaning more towards conservatism Rejected¹

H4 The negative relation between institutional distance and CBA’s investment

33

4.2 Robustness Checks

Multiple robustness checks have been conducted to examine whether the regression models behave differently when interchanged with relatable variables. According to Lu & White, (2014) the tests help to find and highlight imperfections initiated from the original regression analysis. As a result of these tests, a conclusion can be drawn whether the model can also function adequately in the presence of invalid inputs (ANSI, 1991). In total a number of twelve robustness checks have been ran. These checks have been conducted on two independent variables (i.e. cultural and institutional distance). Main argument is the relative high correlation amongst them as shown in the Pearson correlation matrix in table 3. By interchanging these variables with comparable constants, the robustness check test whether the models remain valid. The first six checks acted as substitutions of the cultural distance –independent- variable. The original independent variable conjugates all Hofstede’s cultural dimensions. For the robustness check, every dimension is tested separately. The exact results of the renewed regression tests can be found in appendix 2A.

The latter six checks tested the correctness of the second independent variable, namely the institutional distance. The institutional distance in this thesis is measured by the six World Governance Indicators. For the robustness check, every indicator is tested separately. The exact results of the renewed regression tests can be found in appendix 2B.

-35

5. DISCUSSION AND

CONCLUSION

5.1 Discussion and limitations

Out of a study of 215 deals and 122 CEOs this thesis is able to give a comprehensive insight of the effects on political preferences on CBA investments. More concrete, by analysing the empirical results explained in the previous section, this section provides clarifications of the hypothesis testing.

The research incorporates two distinctive parts. The firsts segment checks whether a relation exists between the unfamiliarity of environments and CBA investment commitment. The fact that both hypothesis one and two are significant and negative, does confirm that the more cultural- and institutional unfamiliar a host country is, the lower the investment commitment of a firm tend to be. These findings are in line with many other scholars who find that a widening cultural distance between two countries discourage large resource commitments (Chari & Chang, 2009; Lee et al., 2008). This same effect is found by scholars who focused on institutional distance among countries (Cuervo-Cazurra, 2006; Xu & Shenkar, 2002).

The results also showed that the negative institutional distance effect is stronger than the cultural distance effect, implying that MNEs do respond more severe to institutional discrepancies. Corresponding literature does likewise support this outcome (Kostova, 1997; Shenkar, 2001; Xu & Shenkar, 2002). They argue that institutional distance is a more accurate predictor of a MNE’s behaviour than cultural measurements.

36

the characteristics of an entire society. As a result the explanatory power of these models may not cover the loading completely. To be more specific, the Hofstede’s cultural dimensions (as used in this paper) is found to be a framework that not completely covers a one-on-one interpretation of individual firm behaviour (McSweeney, 2013; Wolfgang, 2016). Translating this issue towards this thesis, one can argue whether the usage of societal models (here: World Governance Indicators (WGI) and Hofstede’s cultural dimensions) does satisfactory describe the corporate culture and corporate institutions of an individual firm, or not. In other words, this thesis implies that all corporate cultures and institutions are equal within every US-based firm. An assumption that is argued in the literature (Grant, 2013). Nevertheless, the use of societal models in order to explain individual behaviour is still a very common application in the cultural and institutional literature and due to the absence of more refined and sophisticated frameworks, the use of societal models can be seen as the best alternative (Brewer & Venaik, 2014; Fellows & Liu, 2013). Future research in this field may attempt to catalogue corporate cultures and institutions instead of national ones in order to derive more accurate findings.

Furthermore, because the paper only incorporates US based firms and since the US has a developed institutional environment, large institutional distances can only occur in weak institutionalized host countries. Therefore caution should be taken into account in the interpretation. The results solely imply that a high institutional distance leads to a decrease in CBA investment commitment, when the acquiring firm is American. Hence, the thesis does not imply that the effect is somewhat similar when the acquiring firm is non-American. Thus, the effect of institutional distance in non-American countries may be significant different, than the effect that has been found in this study.

37

thesis do not follow the hypothesized pattern. However, due to the significant negative relationship found in this study, one can still argue that the Uppsala model remains valid. This is in line with recent findings of comparative studies. In a recent study drawn from the financial sector, the authors stress the relevance of knowledge gained by experience (Lindh, Thelinius, & Hadjikhani, 2016). Similar effects have been found some decades before by Eriksson et al., (1997), highlighting the constant relevance of incremental internationalization throughout time. In a different comparative study about subsidiary’s decision making autonomy, De Jong, Van Dut, Jindra, & Marek (2015) furthermore proved that uncertainty caused by large country’s psychic distances indeed and still play a larger role in MNE’s behaviour than potential trust benefits. Concluding, the divisions of opinions (here: incremental vs born-global strategy) are common in the IB literature, due to the fact that theories follow a social science approach. Therefore, in opposing theorem we cannot state that one theory is ultimately right and the other wrong. However, this paper can state that the results of this research do emphasise the continuous importance of cross country differences.

That said, the author acknowledges that because of globalization, businesses may be considered as a part of networks and that spatial- and psychic distances may fade over time. In other words, the internationalization structure changes from ‘liability of foreignness’ to ‘liability of outsider ship’ (Johanson & Vahlne, 2009). Future research can elaborate on this by incorporating network determinants (e.g. learning and cooperation capabilities) instead of solely psychic determinants (e.g. cultural and institutional distances), in order to predict MNEs behaviour.

38

recent evidence of multiple studies that found the ongoing explanatory power of the Upper Echelons Theory (Plöckinger, Aschauer, Hiebl, & Rohatschek, 2016; Wang, Holmes, Oh, & Zhu, 2016; Yamak, Nielsen, & Escribá-Esteve, 2014).

Even more, this thesis makes a generalization that is worth mentioning for a proper interpretation. Namely the postulation that executives who support republican and democratic parties are on forehand considered as conservative and liberal, respectively. Even though this assumption is followed by multiple leading studies (as in: Chin et al., 2013; Gupta & Wowak, 2016; Tetlock, 2000), the decision can be debated. Data shows that not only that the degree of liberalism and conservatism amongst democratic and republicans varies within those groups, but that in some cases democrats may even be more leaning towards conservative ideologies and vice versa (Francia et al., 2005). The thesis acknowledges that these generalization may have led to a few false records in the dataset. However, these errors are expected to be minor and non-frequent.

Finally, a last explanation that may have caused insignificant moderating results, is the limited amount of CEOs incorporated in the analyses. Due to a lack in transparency and in the amount of data regarding executive’s political ideologies, the paper was only able to extract 122 CEOs. This amount, even though acceptable, is lower than most related studies in the same field. A greater number of CEOs may have helped to arrive at more satisfactory results.

5.2 Future research opportunity

39

trust (Gulati, 1995), which eventually may lead to the intensifying operations within the familiar environment.

5.3 Conclusion

The aim of this thesis was to test what effect political ideologies of CEOs have on

cross-border acquisition investment commitment in unfamiliar environments. Discursive persuasion

40

BIBLIOGRAPHY

Adams, R. B., Licht, A. N., & Sagiv, L. 2011. Shareholders and stakeholders: How do directors decide? Strategic Management Journal, 32(12): 1331–1355.

Andersen, O. 1993. On the Internationalization Process of Firms: A Critical Analysis.

Journal of International Business Studies, 24(2): 209–231.

ANSI. 1991. Standard Glossary of Software Engineering Terminology. The Institute of

Electrical and Electronics Engineers Inc.

Barney, J. 1991. Firm Resources and Sustained Competitive Advantage. Journal of

Management, 17(1): 99–120.

Bebchuk, L. A., & Cohen, A. 2005. The costs of entrenched boards. Journal of Financial

Economics, 78(2): 409–433.

Benediktsson, M. O. 2010. The Deviant Organization and the Bad Apple CEO: Ideology and Accountability in Media Coverage of Corporate Scandals. Social Forces, 88(5): 2189– 2216.

Beugelsdijk, S., Maseland, R., & van Hoorn, A. 2015. Are Scores on Hofstede’s Dimensions of National Culture Stable over Time? A Cohort Analysis. Global Strategy Journal, 5(3): 223–240.

Bilkey, W. J., & Tesar, G. 1997. The export behaviour of smaller-sized Wisconsin manufacturing firms. Journal of International Business Studies, 8(1): 93–8. Bonica, A. 2014. Mapping the Ideological Marketplace. American Journal of Political

Science, 58(2): 367–387.

Brewer, P., & Venaik, S. 2014. The ecological fallacy in national culture research.

Organization Studies, 35(6): 1–24.

Briscoe, F., Chin, M. K., & Hambrick, D. C. 2014. CEO ideology as an element of the corporate opportunity structure for social activists. Academy of Management Journal, 57(6): 1786–1809.

Cannella, A. A., & Holcomb, T. R. 2005. A Multi-Level Analysis of the Upper-Echelons Model. Research in Multi-Level Issues.

Cavusgil, S. T. 1980. On the internationalisation process of firms. European Research, 8(6): 273–281.

Chari, M., & Chang, K. 2009. Determinants of the share of equity sought in cross-border acquisitions. Journal of International Business Studies, 40(8): 1277–1297.

Chin, M. K., Hambrick, D. C., & Trevino, L. K. 2013. Political Ideologies of CEOs: The Influence of Executives’ Values on Corporate Social Responsibility. Administrative

41

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

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

Christensen, D. M., Dhaliwal, D. S., Boivie, S., & Graffin, S. D. 2015. Top management conservatism and corporate risk strategies: Evidence from managers’ personal political orientation and corporate tax avoidance. Strategic Management Journal, 36(12): 1918– 1938.

Colombo, M. G., & Delmastro, M. 2000. A note on the relation between size, ownership status and plant’s closure: sunk costs vs. strategic size liability. Economics Letters, 69(3): 421–427.

Combs, J. 2010. Big samples and small effects: let’s not trade relevance and rigor for power.

Academy of Management Journal.

Conconi, P., Sapir, A., & Zanardi, M. 2016. The internationalization process of firms: From exports to FDI. Journal of International Economics, 99: 16–30.

Conover, P. J., & Feldman, S. 1981. The Origins and Meaning of Liberal/Conservative Self-Identifications. American Journal of Political Science, 25(4): 617–645.

Contractor, F., Lahiri, S., Elango, B., & Kundu, S. 2014. Institutional, cultural and industry related determinants of ownership choices in emerging market FDI acquisitions.

International Business Review, 23(5): 931–941.

Cuervo-Cazurra, A. 2006. Who cares about corruption? Journal of International Business

Studies, 37(6): 807–822.

de Jong, G., van Dut, V., Jindra, B., & Marek, P. 2015. Does country context distance determine subsidiary decision-making autonomy? Theory and evidence from European transition economies. International Business Review, 24(5): 874–889.

Demirbag, M., Glaister, K. W., & Tatoglu, E. 2007. Institutional and transaction cost

influences on MNEs’ ownership strategies of their affiliates: Evidence from an emerging market. Journal of World Business, 42(4): 418–434.

Detomasi, D. A. 2008. The political roots of corporate social responsibility. Journal of

Business Ethics.

Diaz-Veizades, J., Widaman, K. F., Little, T. D., & Gibbs, K. W. 1995. The Measurement and Structure of Human Rights Attitudes. The Journal of Social Psychology, 135(3): 313– 328.

DiMaggio, P., & Powell, W. 1983. The Iron Cage Revisited. American Sociological Review. Drogendijk, R., & Slangen, A. 2006. Hofstede, Schwartz, or managerial perceptions? The

effects of different cultural distance measures on establishment mode choices by multinational enterprises. International Business Review, 15(4): 361–380. Dunlap, R. E., & Allen, M. P. 1976. Partisan Differences on Environmental Issues: A

Congressional Roll-Call Analysis. The Western Political Quarterly, 29(3): 384–397. Eisenhardt, K. M., & Bourgeois, L. J. 1988. Politics of Strategic Decision Making in

High-Velocity Environments : Toward a Midrange Theory. Academy of Management

42

Eriksson, K., Johanson, J., Majkgard, A., & Sharma, D. D. 1997. Experiential knowledge and cost in the internationalization process. Journal of International Business Studies, 28(2): 337–360.

Ess, C., & Sudweeks, F. 2001. Culture, technology, communication : towards an intercultural global village. Suny series in computer-mediated communication.

https://doi.org/303.48/33, NetLibrary Inc. C N - HM851 .C85 2001eb.

Fellows, R., & Liu, A. M. M. 2013. Use and misuse of the concept of culture. Construction

Management and Economics, 31(5): 401–422.

Field, A. 2009. Discovering Statistics Using SPSS. Sage Publication, vol. 58. https://doi.org/10.1234/12345678.

Finkelstein, S. 1992. Power in Top Management Teams : Dimensions , Measurement , and Validation. Academy of Management Journal, 35(3): 505–538.

Forsgren, M., & Kinch, N. 1970. Företagens anpassning till förändringar i omgivande system (The adaptation of the firm to changes in surrounding systems). Uppsala Department of

Business Studies.

Francia, P. L., Green, J. C., Herrnson, P. S., Powell, L. W., & Wilcox, C. 2005. Limousine liberals and corporate conservatives: The financial constituencies of the democratic and republican parties. Social Science Quarterly.

Francis, B. B., Hasan, I., Sun, X., & Wu, Q. 2016. CEO political preference and corporate tax sheltering. Journal of Corporate Finance, 38: 37–53.

Frost, J. 2013. Regression Analysis: How Do I Interpret R-squared and Assess the Goodness-of-Fit? The minitab blog.

Grant, A. 2013. Givers take all: The hidden dimension of corporate culture. McKinsey

Quarterly, (2): 52–65.

Grant, A. 2014. Give and Take (1st ed.). Penguin Books.

Green, S. B. 1991. How Many Subjects Does It Take To Do A Regression Analysis?

Multivariate Behavioral Research, 26(3): 499–510.

Gulati, R. 1995. Does Familiarity Breed Trust? The Implications of Repeated Ties for Contractual Choice in Alliances. Academy of Management Journal, 38(1): 85–112. Gupta, A., & Wowak, A. J. 2016. The Elephant (or Donkey) in the Boardroom: How Board

Political Ideology Affects CEO Pay. Academy of Management. https://doi.org/10.1177/0001839216668173.

Hajzler, C. 2014. Resource-based FDI and expropriation in developing economies. Journal of

International Economics, 92(1): 124–146.

Hambrick, D. C. 2007. Upper echelons theory: An update. Academy of Management Review. Hambrick, D. C., & Mason, P. A. 1984. Upper Echelons: The Organization as a Reflection of

Its Top Managers. Academy of Management Review, 9(2): 193–206.

Hetherington, M. J. 2009. Putting Polarization in Perspective. Source: British Journal of

Political ScienceJ.Pol.S, 39(39): 413–448.