THE INFLUENCE OF CULTURAL DIFFERENCES: A
STUDY ON CROSS-BORDER MERGERS AND
ACQUISISTION
University of Amsterdam Student: Freek van der Werff 10793003
Economics and Business Finance and Organisation Supervisor Dr. E. Zhivotova 31-01-2018
2
Statement of originality
This document is written by Freek van der Werff who declares to take full responsibility for the content of this paper. I declare that the text and the work presented is original and that no sources other than those mentioned in the references have been used. The Faculty of Economics and
3
Abstract
This study researches the influence of cultural differences on cross-border synergies. The cultural differences are measured according to the Hofstede Cultural Dimensions model. The expectation is that the model will show negative synergies as the level of cultural differences increases. The findings of the study do not support the main hypothesis. The results on other hypotheses are open for discussion.
4
Table of content
1. Introduction……….5
2. Literature review………..7
3. Hofstede Cultural Dimensions……….9
4. Hypotheses………..11
5. Regression……….12
6. Methodology………..13
7. Data and descriptive statistics………14
8. Results……….19
9. Conclusion………22
10. Appendices………..23
5
1. Introduction
We live in a world where globalisation does not seem to stop. The existence of borders can almost be ignored, the other side of the world is a night sleep away. It is offering you the opportunity to enrich yourself with that what lies beyond those physical borders. Nowadays, on a small scale, one can order cheap products made in China via for instance Alibaba.com. Hereby cutting out the
middleman. On a large scale companies can buy other companies from all over the world, hoping to open a niche market or just to come to economics of scale. So it is not a surprise that along with the curve of the growing rate of globalisation, the curve of the growing rate of cross-border Merger and Acquisition is creeping upwards.
However, merging with a company from another country can raise extra difficulties to the
management. The most striking differences in cultural backgrounds. Not only from employees, but also from company values. These differences can have a negative and delaying effect on the process of finishing the merger. Opposite outcomes can also be the case, namely more synergies because of cultural differences. The third possible outcome is no notable difference in the performance. The research question answered in this thesis is: ‘What is de influence of cultural differences on the acquirers performance in a cross-border Merger and Acquisition?’.
The main hypothesis of this study will be that the larger the absolute difference of a cultural
dimension is, the lower the synergy for the acquirer will be. Furthermore there will be hypotheses on some individual cultural dimensions.
The data in this study contains 5115 Mergers and Acquisitions from a time period of 20 years. The M&A are only cross-border deals between 1997 and 2017. The minimum deal value is 1 million dollar and both the acquirer and target firm are listed companies. This was necessary for availability of the required financial data. The synergies are calculated as the Cumulative Abnormal Returns of the acquirer with an event time of 3 days around the announcement date of the M&A.
For the measurement of the cultural differences this study uses the Six Dimensions Model created by the Dutch social psychologist Geert Hofstede in the 1980’s. His cross-cultural framework started out with four dimensions, namely ‘small power distance against large power distance’, ‘individualism against collectivism’, ‘masculinity against femininity’, ‘structured situations against unstructured situations’. Later this was completed with the other two dimensions: ‘short-term orientation against long-term orientation’ and ‘indulgence against restraints’. For each of these dimensions a country is given an index figure for which it is possible to see the countries characteristic in the different dimensions. The Six Dimensions Model is limited to 105 countries. This means that in the dataset of this study only contains these countries to find the Mergers and Acquisitions.
To control for omitted variables the regression model includes control variables for industry, religion, language and legal system. These variables should have a high correlation with the independent variables. When it applies to the two merging companies that the variants in these control variables are the same, there will be likely less problems with the merger ( Ahern, Daminelli, and Fracassi (2012). The control variables are turned into dummy variables. Which means that if for example the
6 two countries from the acquirer and the target speak the same language, the corresponding
coefficient value is included in the regression. If this is not the case the coefficient value is excluded because the value is zero. For the variable Religion the six largest religions in the world are tested namely; Islam, Orthodox, Protestant, Roman Catholic, Buddhism and Jewish. For the Industry control variable the SIC industry codes are used. For each country the primary language is commuted to calculate the Language control variable. To conclude, for the Legal System variable this study separates civil law, common law and religious law or a combination of these. These are the world’s major legal systems (Powell and Mitchell (2007)). The data on Language, Legal system and Religion is collected through the CIA World Factbook. The data for the Industry control variable are from
Zephyr.
The four control variables are included for a couple of reasons. Different language is seen as a big barrier for a smooth cooperation (Harzing, Pudelko and Tenzer (2014)). In their paper they conclude that a common language creates trust. The use of different language is therefor expected to
influence the M&A synergies negatively. Communication plays an important role in corporate connection. When companies have the same legal system it makes it easier to collude. This is especially true for the accountants and legal departments. When company rules and methods need to be changed this will bring additional costs. For a lot of countries religion is still an important part of their cultural characteristics. Different religions are leading to different holidays, ethics and day formats. When two companies merge employees have to adjust to each other. Different religions means different demand, as mentioned above. Which can lead to the increase of costs and mitigate solidarity.
Companies from the same industry are said to understand each other better. Workers who perhaps have to work together in the future have knowledge of the same area.
The purpose of including these control variables is to study the effect of extra common ground between the deal companies. The main study is aiming at the effect of the Six Dimensions Model, these control variables add extra data to study the research question.
This study measures the three days Abnormal Return surrounding the announcement date of the deal. The possible synergies are therefor calculated for the short term. This is according to previous studies (Ahern, Daminelli, and Fracassi (2012). The long term effect of cultural differences is harder to measure, because there is likely to occur misspecified test statistics (Lyon, Barber and Tsai (1999)). The results of this study are not as expected. The main hypothesis cannot be answered.
The study is structured as follows; first there will be review on previous literature, after that the cultural dimensions will be discussed. Following this the hypotheses and the methodology will be formulated. In the middle part the data and descriptive statistics are displayed. The last part of this study are the results and conclusion. At the end there will be the appendix and the references.
7
2. Literature review
In this part existing literature is discussed, researches from the past are explained. On the topic of Merger and Acquisition there has been quite some elaborate research. Also in combination with cross-border M&A. In terms of the effect of cultural differences on M&A the research is minimal. However an article from 2015 gives a good representation for the research in this thesis. In this article Ahern, Daminelli and Fracassi (2015) acknowledge that cultural differences are especially important for cross-border mergers. They provide large-scale evidence to show that national cultural differences have substantial impact on multiple aspects of mergers, including the place where mergers occur and the gains they may take. Problems that might arise when employees have different cultural backgrounds are mistrust, misunderstanding and not having the same goals. This is all reducing coordination which is necessary for good governance. On the other hand they quote that greater cultural difference could increase the likelihood of a successful merger if cultural diversity facilitates innovation and promotes new approaches for problem solving (Page, 2007).
Ahern, Daminelli and Fracassi applied a gravity model to answer their hypothesis. They measure cultural differences with three distinctions. Namely trust, hierarchy and individualism. The study contains a large sample of 52 country’s with mergers between 1991 and 2008. The sample contains 127.950 mergers, of which 30.907 cross-border and 65.796 not from a U.S. acquirer or target. To measure cultural values they use the World Values Survey.
They find strong evidence that the differences in national culture are indeed reducing the volume of cross-border mergers. Hence, the greater the cultural distance between two countries the smaller is the volume of cross border mergers. Larger cultural distance also leads to lower synergy gains Healy, Palepu and Ruback (1992) examine post-acquisition performance for the 50 largest mergers between 1979 and 1984. These merged companies are U.S. public industrial firms. Their research method is to use postmerger accounting data to test directly for changes in the operating
performance.
They conclude that merged firms show significant gains in assets relative to their competitors. The performance improvement is usually the best for merged firms from the same kind of industry. Erel, Liao and Weisbach (2012) analyse a sample of 56.978 cross-border mergers between 1990 and 2007. With their research they conclude that geography clearly matters. The shorter the distance between two countries the more likely there will occur mergers and acquisition between firms from those countries. When two countries are common trade partners this also enhances the number of mergers. Acquiring firms are usually from developed countries and the try to buy target firms with lower accounting standards. In addition, acquirers are more likely to be located in countries with higher corporate tax rates that their targets.
8 They acknowledge the fact that mergers between firms form countries with different identities increase the contracting costs. As examples of different identities they give language, religion and sometimes longstanding feuds. Another way in which cross-border mergers can be effected are corporate governance considerations. Value ca be created for minor shareholders of target firms, because with a merger they get some of the rights of the acquiring firm. An interesting result is that they find that for both domestic and cross-border mergers, acquiring firms outperform their targets. Alexandridis, Petmezas and Travlos (2010) performed a study on gains from Mergers and Acquisitions around the world. They discovered that shareholders experience significant losses or normal returns around the announcement date of a merger. When there is a more competitive market, there will be a premium on biddings. As a consequence the possibility to gain synergies for the acquiring firm is lower. The most competitive markets are those of the United States, United Kingdom and Canada. In this study they also mention that hostile acquisitions result in higher target returns, because of the higher premiums paid in this type of transactions.
Betrand and Zitouna (2008) examined domestic versus cross-border acquisitions. The research mergers and acquisitions in from French firms between 1991 and 2001. The central question in their study is: what would have been the target firms performance if it had not been taken over? They mention that authorities usually foster domestic merger and acquisition and in contrast discourage cross-border take overs.
Overall horizontal mergers and acquisition do not significantly improve the profit of French target firms in the short and long run. However they increase their productivity. When comparing domestic versus cross-border Merger and Acquisition efficiency gains are stronger for cross-border.
Gugler, et al (2003) performed a study on the effect of mergers around the world. At the time this study was the largest cross-national comparison of the effect of mergers. They compared the performance of the merging companies with the performance of nonemerging companies.
Performance measures are profitability and sales. Their findings where that there was an on average increase in profits for mergers, but the sales where reduced. The sample they use is from 1981 to 1998 and contains 69.605 mergers.
A comparison between cross-border and domestic acquisitions is made in the study of Moeller and Schlingemann (2005). They use a sample of 4430 acquisitions between 1985 and 1995. The study looks from the perspective of companies from the United States. It compares US companies who buy cross-border targets and US companies who buy domestic targets. Companies which did cross-border deals had significantly lower changes in operating performance. They also find that global
diversification has a negative influence on stock returns. Just like the study in this thesis Moeller and Schlingemann examined the influence of legal systems on cross-border acquisition. The conclude that acquirer gains are lower for target countries with a more restrictive institutional environment.
9
3. Hofstede Cultural Dimensions
In this paragraph the definition of the six dimensions is explained. Where possible the connection of the dimension and business.
Power Distance
Meant with this dimension is that there is inequality in power between members of a society. A low power distance means being independent. In an organisation power is decentralized and managers rely on the experience of their team members. High power distance stands for a hierarchical society. This means that there is a clear distinction between everybody’s rank and place. A command from a higher rank is easily accepted and justified. The organisation is usually centralised, with a strong and powerful leader, subordinates expect to be told what to do.
Managers from cultures low in power distance tend to more problem solving in their conflict behaviour towards their supervisors than managers from cultures in high power distance (Van Oudenhoven & De Dreu (1997)).
Individualism
The main distinction made here is between a person’s attitude considering a “I” or “We” feeling. Important in this dimension is the degree in which members of society depend on each other. If a country has a low score it means that it can be considered as a collectivistic society. These societies are characterised with a great feeling of family honour and a boundless and close bond to their peers. In an organisational structure management is usually working in groups.
On the other hand when a country has a high index, the society and thus it’s organisational structure is individualistic. It can be seen as a more liberal way of handling things. Members are expected to take care of themselves and are less looked after. Which brings greater responsibility and trust in your own abilities.
Masculinity
A Masculine society is known for its competitive attitude among peers. Winning is important and you are judged by your success. Status is important as well, being able to show your achievements to the outside world is part of that. Management are expected to be decisive. Fighting out conflicts is the way to solve them.
The other side is a feminine society where care and quality of life are more central. In this structure, standing out of the crowd is not necessary preferable. It is considered a greater achievement to have a good quality in life than it is to be the best. Members put in a lot of care in managing a good balance between worktime and leisure time. In an organisation problems are addressed with negotiation and ultimately compromises. Managers are considered to be supportive and involve workers in decision making.
10 Managers from feminine countries tend to be better in the solving of problems with their
colleges(Van Oudenhoven & De Dreu (1997)).
Uncertainty Avoidance
The dimension Uncertainty Avoidance displays the manner in which society deals with the unknown future. People have to consider whether to control the future or just let it happen.
Cultures with a high Uncertainty Avoidance express a need for rules and clarity. People have to feel secure and innovations are not always welcomed.
Societies with low index on Uncertainty Avoidance are considered to be more relaxed. Plans can be altered at short notice and improvisations are common. There is no adversity against taking risk.
Long term Orientation
This dimension describes the difference between pragmatic and normative societies. The main focus is on heaving short lived focus or invest in the future and being able to wait.
Normative societies stick to traditions and dislike societal change. Pragmatic societies encourage thrift and modern developments to prepare for the future.
Indulgence
This dimension is defined as the extent to which people try to control their desires and impulses, based on their upbringing . A relatively weak control over their impulses is called “Indulgence”, whereas a relatively strong control over their urges is called “Restraint”.
People in societies with a high score in Indulgence generally show a willingness to fulfil their impulses and desires, especially with regard to enjoying life and having fun. They possess a positive attitude and have a tendency towards optimism.Societies with a low score in this dimension have a tendency to cynicism and pessimism. Restrained societies do not put much emphasis on leisure time and control the gratification of their desires. People with this orientation have the perception that their actions are restrained by social norms and feel that indulging themselves is somewhat wrong.
11
4. Hypotheses
The main purpose of this study is to research the effect of cultural difference on synergies by cross-border Merger and Acquisition. As told before this study uses the Hofstede six dimensions model to measure for cultural differences. Calculations will be made with these six dimensions together, but also separately. Therefor multiple hypotheses will be mentioned. The research question in this thesis is ‘What is de influence of cultural differences on the acquirers performance in a cross-border Merger and Acquisition?’. Where the synergies are calculated with the Cumulative Abnormal Return around the announcement date between a period of 1997 and 2017.
The first hypotheses will be related to all six dimensions and should answer the research question.
Hypothesis 1: The larger the difference between the six cultural dimensions the lower CAR
Next, for some of the Cultural Dimensions hypotheses are added. Individualism against collectivism, it can be expected from previous literature that this variable will have a negative effect on the synergies. Ahern, Daminelli and Fracassi (2015) used Individualism as independent variable, therefor expected in this study is the same effect.
Hypothesis 2: The larger the absolute difference in the Individualism dimension the lower the CAR
Furthermore in this study a hypothesis about masculinity against femininity is suspected. Mentioned earlier is the fact that managers from feminine countries are better in the solving of problems with their colleges than managers from Masculine countries. This observation is expected to have influence in the model of this study
12
5. Regression
Y
1
=β
0
+β
1
x
1
+β
2
x
2
+β
3
x
3
+β
4
x
4
+β
5
x
5
+β
6
x
6
+β
7
x
7
+ε
Y1= Cumulative Abnormal Returnx1= Absolute difference in index of Power distance dimension x2= Absolute difference in index of Individualism dimension x3= Absolute difference in index of Masculinity dimension
x4= Absolute difference in index of Uncertainty avoidance dimension x5= Absolute difference in index of Long Term orientation dimension x6= Absolute difference in index of Indulgence dimension
x7= control variables ε= error term
13
6. Methodology
To measure the performance of the acquiring company this study uses the Cumulative Abnormal Return. The event window is three days after and three days before the announcement date of the deal (MacKinley (1977)).
To calculate the Cumulative Abnormal Return the Abnormal Return must first be calculated.
AR
it
=R
it
-E(R
it
)
where:
ARit - abnormal return for firm i on day t
Rit - actual return for firm i on day t
E(Rit) – expected return for firm i on day t
The Cumulative Abnormal Return is the sum of ARit
For the Actual Return this study uses the percentage change in the daily returns. Calculated are the daily returns of the three days surrounding the announcement date of the deal. For the Expected Return, the percentage change in the acquirers home country market daily returns are used. Also three days surrounding the deal. For every day the Expected Return is subtracted from the Actual Return. The sum of those differences is processed and this value represents the Cumulative
Abnormal Return. This method is used for every individual deal. The use of the Cumulative Abnormal Return is according to Moeller, et al (2004).
14
7. Data and descriptive statistics
Data
Data for this study is mainly collected from three sources. For the Mergers and Acquisitions this study used the database Zephyr. This database contains comprehensive M&A data with integrated detailed company information.
The time span for the Mergers and Acquisitions is 01/01/1997 and 01/01/2017. This is because the sample should be as large as possible and Zephyr had no data older than 1997. All M&A are cross-border, only limited by countries used in the Hofstede Six Dimensions Model. For both the acquiring and the target company counts that they have to be listed and the deal value had a minimum of 1 million Dollars. They had to be listed because of the availability of financial data. Listed companies are obliged to present those necessary financial figures. All deals in the dataset are completed. Zephyr produced the deal value, company ISIN (International Security Identification Number), country code and deal announcement date. The ISIN and country codes where necessary to calculate the corresponding financial data.
With the ISIN number it was possible to collect the daily returns around the announcement date of the acquiring company. This was done through the database Datastream, for each individual deal. Also the benchmark necessary for the Cumulative Abnormal Return was collected through
Datastream. Used as benchmark are the daily returns around the announcement dates of corresponding countries for the acquiring company.
Eventually the dataset for this study contains 5115 mergers.
For the data on the Six Cultural Dimensions the article of G. Hofstede (1984) was consulted, as well as the website http://www.hofstede-insights.com. This was to collect the necessary indexes to be able to measure the effect of cultural difference. Eventually, the absolute differences between the indices of the acquirer and the target companies are used in the regression model. For calculations with the data Microsoft Excel was consulted and for the statistic measurements Stata was used.
The data on the control variables are collected from the CIA World Factbook. This is true for the variables Religion, Legal System and Language. The data on company variable Industry is from Zephyr.
15
Descriptive Statistics
The original dataset retrieved from Zephyr contained 7025 deals, due to unavailability of financial data, wrong identification codes and generals measurement errors as for example outliers, the number of deals is reduced to 5115. Number of countries with an acquirer is 53 and the number of countries with a target is 68. This means that not all 105 countries specified in the Six Dimensional Model are represented in de dataset.
Graph 1 displays the number of Merger and Acquisition deals per year containing the dataset. Clearly the amount of deals rises per year, especially in the last three years of the dataset. In the year 2008 there is a decline notable, this is probably due to the financial crises.
Graph 1 – Number of M&A deals per year
Table 1 displays the number of acquiring and target companies within a country. The table is limited to a top ten, in the appendix the complete table is displayed. The country codes are also displayed in the appendix. Focussing on the acquiring company, the country that stands out is the United States. It represents roughly one third of all Acquirers. This means that companies of the United States will have a big influence on the research outcome. Another notable conclusion is that eight of the ten countries are considered to be part of the ‘rich western world’. This is according to what can be expected of the wealthy countries and to previous literature. Two relatively small countries
represented are The Netherlands and Singapore. Not to be mistaken is that “CH” is the country code for Switzerland and “CN” is the country code for China.
0 100 200 300 400 500 600 700 800 900 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16
16 Looking at the target company side of the table, it is notable that United States is not prominent in the target company table. On this side is Great Britain the dominant player. As well as in the acquirer side of the table, the largest part of the countries are considered to be from the ‘rich western world’. For the study this is not particularly a bad thing, because also ‘rich western world’ differ a lot in their cultural dimensions, especially in Europe. Another thing that stands out is that Dutch companies are relatively popular for cross-border Merger and Acquisition.
Table 1 - Rank of top acquirer and target countries
This table presents the top ten of the frequently of acquiring and target companies per country.
Rank Acquirer Freq. Percent Target Freq. Percent
1 US 1756 34,3% GB 931 18,2% 2 AU 767 15,0% AU 438 8,6% 3 GB 321 6,3% NL 397 7,8% 4 DE 302 5,9% US 346 6,8% 5 FR 298 5,8% CN 310 6,1% 6 CH 295 5,8% CH 262 5,1% 7 JP 191 3,7% IT 237 4,6% 8 NL 185 3,6% ES 195 3,8% 9 SG 139 2,7% FR 193 3,8% 10 CA 100 2,0% IN 137 2,7%
Table 2 on Cultural Dimensions shows the relation of every individual dimension with the belonging index figures. For all the acquirer and the target companies in the sample the average and standard deviation is calculated. As well as the absolute difference between the deal companies.
Table 2 - Statistics cultural dimensions
This table presents the statistics of the acquirer and target companies on each cultural dimension.
The statistics of the absolute difference used in the regression are also presented.
Acquirer Target
Dimension Average index
Standard Deviation Average index Standard Deviation Average |∆| Standard Deviation Power Distance 44,6 14,7 48,7 18,6 14,0 14,7 Individualism 74,8 22,1 65,5 25,5 20,6 20,7 Masculinity 58,2 15,4 53,6 18,3 16,4 17,1 Uncertainty avoidance 53,9 18,3 54,5 20,7 19,1 15,1 Long Term Orientation 45,1 23,8 52,6 21,0 28,3 17,5 Indulgence 60,4 14,1 54,7 18,4 13,9 15,2
17 Displayed here below are the descriptive statistics on the control variables. Table 3shows per control variable the number of same variants for the acquirer and the target company. For every control variable almost one fourth of the deal companies had the same variant. More than half of the acquirer and target companies had the same legal system.
Table 3 - Amount of identical variants
Number of each M&A deal for which companies have the same dummy control variable variant.
Control variable Number Total Percent
Industry 1223 5115 23,9%
Religion 2129 5115 41,6%
Language 1520 5115 29,7%
Legal System 2689 5115 52,6%
Table 4 shows the data on the specific Religion control variable. Difference is made between the acquirer and target company. Focussing on the acquirer part of the table it is notable that Christian countries dominate the cross-border take-over market. The only country with Hebraic as primary religion is Israel. Which is more dominant that all Orthodox countries, which are represented by multiple countries in Eastern-Europe including Russia.
In the target company part of the table the most notable thing is that companies from Roman Catholic countries are the most dominant, instead of in acquirer table where companies from Protestant countries are the most dominant. Still companies from Christian countries are the most wanted targets. A speculative explanation for the high presence of Dutch companies in the take-over market is that they are used to managing compromises, which is part of the Dutch culture. Also the presence of lots of different religions in The Netherlands make this a possible explanation. Dutch managers are used to work with these differences (Emans, Laskewitz, & Van de Vliert (1994)).
Table 4 - Company amount per religion
This table presents the number and percentage of companies and the corresponding primary religion of their home country.
The table separates the acquirer and target company
Religion Acquirer Percent Target Percent
Islam 93 1,8% 225 4,4% Orthodox 18 0,4% 80 1,6% Protestant 3028 59,2% 2007 39,2% Roman Catholic 1477 28,9% 2070 40,5% Buddism 421 8,2% 487 9,5% Jewish 37 0,7% 41 0,8% Other 41 0,8% 205 4,0% Total 5115 100% 5115 100%
18 Table 5 show the statistics on the control variable Language. By far the most acquirer companies are coming from a English speaking country. Which is in line with the findings in table 1, where countries such as the United States, Australia and Great Britain are highly ranked. Also in line with the
expectation is the relatively high score of Dutch speaking countries. Companies from The
Netherlands as shown above are large players in the cross-border take-over market. In this table they are also accompanied by Belgium, because their primary language is Dutch.
Table 5 - Company amount per language
This table presents the number and percentage of companies and the corresponding primary country language.
The table separates the acquirer and target company.
Language Acquirer Percent Target Percent
English 2981 58,3% 2019 39,5% Spanish 130 2,5% 261 5,1% German 613 12,0% 394 7,7% French 298 5,8% 193 3,8% Arabic 56 1,1% 61 1,2% Portuguese 23 0,4% 138 2,7% Mandarin 221 4,3% 382 7,5% Dutch 214 4,2% 460 9,0% Other 579 11,3% 1207 23,6% Total 5115 100% 5115 100%
The last table is on whether countries from acquirer and target companies use the same legal system For both acquirer and target companies counts that they are likely from a country with either Civil law or Common law.
Table 6 - Company amount per legal system
This table presents the number and percentages of companies and the corresponding legal systems per country.
The table separates the acquirer an target company.
Legal System Acquirer Percent Target Percent
Civil law (1) 1853 36,2% 2698 52,75% Common law (2) 3134 61,3% 2282 44,61% Religious law (3) 7 0,1% 1 0,02% Mix 1&3 28 0,5% 26 0,51% Mix 2&3 67 1,3% 91 1,78% Mix 1,2,3 26 0,5% 17 0,33% Total 5115 100% 5115 100%
19
8. Results
In this part of the thesis are the result of the multiple regression discussed. Every separate hypothesis will be held against the corresponding results.
Table 7 displays eight separate regression analyses on the possible effect on synergies. As told before are the synergies calculated as the Cumulative Abnormal Returns of the acquirer firm around the announcement date. This is the dependent variable. The variables of interest are the six Cultural Dimensions. The absolute difference between the indices of the acquirer an target company on the separate dimensions are the independent variables. The control variables Industry, Religion,
Language and Legal System are expressed as dummy variables. This means that the dummy variable equals one if the variant for the acquirer and target company is the same. The dummy variable equals zero as otherwise. For every regression analysis the error terms are robust. Furthermore, in the table are the number of observation and the R-square presented.
Column 1 to 6 show the regression analysis on each separate cultural dimension, without the control variables. Column 7 shows the multiple regression with each cultural dimension as independent variable, the control variables are excluded. Column 8 shows all Cultural Dimensions and all control variables.
In the regression analysis is shown that the independent variable Uncertainty Avoidance is the only significant variable. It is significant for 10% in column 7 and 8 and 5% significant in column 4. The main hypothesis in this study is that cultural difference has a negative effect on the acquirer
synergies. The regression shows that only the variable Long Term Orientation lowers the Cumulative Abnormal Return. This is not in line with the hypothesis and also not in line with previous literature. Possible explanation is that these cultural dimensions are not a good representation of cultural difference. Another explanation is that the figures are right and the hypotheses is rejected. This means that the more difference there is in culture between an acquirer and an target company, the higher the return is for the acquirer. Taking column 8, the variables with the largest influence on the synergies are from highest to lowest: Uncertainty Avoidance, Masculinity, Long Term Orientation, Individualism, Indulgence and Power Distance.
In table 8 the Cultural Dimension Variables are changed to log variables. This is because the distributions of the absolute differences where far from normally distributed. All of the interest variables have a skewness to the left. This is because of the fact that the largest part of the deal companies had small differences. To solve this problem the variables are changed into log variables. The general idea of table 8 is the same as table 7, the only difference are the changed independent variables. However, the results of the regression differ clearly from the ones in table 7. Column 8 shows that the variable Masculinity is now significant for 10%. Column 2 shows that the variable Individualism is also significant for 10% and column 1 shows that the variable Power Distance is significant for 5%. For the regressions in column 7 and 8 it counts that Indulgence now has a negative effect on the Cumulative Abnormal Return. Also Uncertainty Avoidance has now a negative effect on the Cumulative Abnormal Return. This is in contrast with the figures of table 7 which shows only a negative effect on Long Term Orientation.
20 Hypothesis 2, which states that the larger the absolute difference is in the variable individualism the lower the Cumulative Abnormal Return. This is not true for both the normal absolute difference and the log absolute difference of the variable individualism.
Column 8 also shows that the variables with the largest influence on the synergies are from highest to lowest: Masculinity, Indulgence, Power Distance, Uncertainty Avoidance, Individualism and Long Term Orientation. Hypothesis 3 states that the variable Masculinity has a relatively large influence on the Cumulative Abnormal Return. This is true for both the regression in table 7 and 8. For the
regression in table 7 Masculinity has the second highest influence. As stated before in table 8 Masculinity has the highest influence of al dimensions and it is significant for 10%.
Not as expected is the negative influence of two of the control variables. Namely the dummy variables Industry and Religion show that if the variants of the deal companies within the specific dummy variables are the same, there will be negative effect on the return of the acquirer. The difference in number of observations in both table 7 and table 8 is due to missing values.
21
Table 7 - Regression analysis on cultural difference effect
This table presents the linear regression for the acquirers Cumulative Abnormal Return (CAR). The CAR is given in percentages. The event window is -3 and +3 days surrounding the announcement. Columns 1 to 6 show the regressions of each individual Cultural Dimension. Column 7 shows the regression with all Cultural Dimensions. Column 8 shows the regression with all Cultural Dimensions and the dummy control variables. P-values are between the parentheses and the asterisks denote statistical significance at the 1% (***), 5% (**) or 10% (*) level.
CAR [3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] Independent variable [1] [2] [3] [4] [5] [6] [7] [8] Power Distance|∆| 0,0087 0,00084 0,000445 [0,279] [0,931] [0,965] Individualism |∆| 0,00655 0,00468 0,00421 [0,343] [0,58] [0,642] Masculinity |∆| 0,00403 0,00268 0,00609 [0,46] [0,627] [0,348] Uncertainty Avoidance |∆| 0,0155** 0,0142* 0,0163* [0,036] [0,098] [0,06] Long Term Orientation |∆| -0,00204 -0,00595 -0,00508 [0,77] [0,376] [0,512] Indulgence |∆| 0,00744 0,000377 0,00249 [0,396] [0,969] [0,803] Industry dummy -0,00578 [0,984] Religion dummy -0,5001 [0,164] Language dummy 0,378 [0,287]
Legal System dummy 0,236
[0,52]
Constant 0,157 0,144 0,213 -0,0177 0,352 0,195 0,0404 -0,123
[0,230] [0,319] [0,161] [0,92] [0,103] [0,137] [0,884] [0,779]
N 5115 5115 5115 5115 5078 4997 4997 4997
22
Table 8 - Regression analysis on log cultural difference effect
This table presents the linear regression for the acquirers Cumulative Abnormal Return (CAR). The CAR is given in percentages. The event window is -3 and +3 days surrounding the announcement. Columns 1 to 6 show the regressions of each individual log Cultural Dimension. Column 7 shows the regression with all log Cultural Dimensions. Column 8 shows the regression with all log Cultural Dimensions and the dummy control variables. P-values are between the parentheses and the asterisks denote statistical significance at the 1% (***), 5% (**) or 10% (*) level. CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] CAR [-3,+3] Independent variable [1] [2] [3] [4] [5] [6] [7] [8]
Log Power Distance |∆| 0,164** 0,0735 0,0601
[0,047] [0,589] [0,673] Log Individualism |∆| 0,133* 0,106 0,0256 [0,079] [0,423] [0,895] Log Masculinity |∆| 0,0907 0,222 0,319* [0,303] [0,123] [0,075] Log Uncertainty Avoidance |∆| 0,174 -0,0433 -0,0372 [0,225] [0,814] [0,840]
Log Long Term
Orientation |∆| -0,0126 0,0639 0,0151 [0,917] [0,653] [0,921] Log Indulgence |∆| 0,0256 -0,128 -0,0912 [0,788] [0,304] [0,472] Industry dummy -0,924 [0,776] Religion dummy -0,487 [0,227] Language dummy 0,419 [0,452]
Legal System dummy -0,152
[0,698]
Constant -0,0312 -0,0208 0,0807 -0,167 0,343 0,3501 -0,241 0,0102
[0,852] [0,903] [0,742] [0,682] [0,372] [0,073] [0,679] [0,991]
N 5015 5074 4955 5022 5034 4496 4194 4194
23
9. Conclusions
This study used a large sample of cross-border Mergers and Acquisition to research the effect of cultural differences on the acquirers synergies.
The results turned out to be different than expected. It was not possible to answer the research question with any significance. There was not any evidence to support the main hypothesis. Also the control variables did not show the results in the regression as expected. The results do not
correspond with the results in previous literature.
The third hypothesis , on which stated that the variable Masculinity would have a large influence on the Cumulative Abnormal Return is plausible, only the objective is not significant.
To conclude, this study did not reached the goal it was set out to do. With the obtained results no answer can be given on whether cultural differences effect cross-border M&A. Because of that, more study is necessary, especially with the Six Dimensions Model as independent variables. The model is an elaborate dataset on cultural dimensions of countries. Thus has the potential to be a good measure.
A possibility for the failure of the regression is omitted variables. This is highly likely because there are a lot of factors besides culture that have influence on performance. As the world keeps on globalise further, it will be easier for cross-border mergers to perform without extra cost . Cultural values merge along with the companies and so differences will become smaller.
24
10. Appendices
Appendix 1 - table of Hofstede Cultural Dimensions per country
This table shows all indices per country of Hofstede Cultural Dimensions. The rank of every country is included, as well as the country code.
Power Distance
Individualism Masculinity Uncertainty
avoidance Long Term Orientation Indulgence Country Country code
Index Rank Index Rank Index Rank Index Rank Index Rank Index Rank
Albania AL 90 12,5 20 80,5 80 4 70 40,5 61 21 15 79 Angola AO 83 18 18 87,5 20 93 60 54,5 15 76 83 7 Argentina AR 49 78,5 46 35 56 35 86 18,5 20 71,5 62 27 Australia AU 36 90 90 2 61 25 51 66,5 21 70 71 14 Austria AT 11 103 55 27 79 5 70 40,5 60 23 63 25,5 Bangladesh BD 80 24,5 20 80,5 55 37 60 54,5 47 36,5 20 70 Belgium BE 65 60,5 75 9 54 38 94 7,5 82 7 57 30 Bhutan BT 94 7,5 52 29,5 32 86 28 99 na na Brazil BR 69 51 38 41,5 49 48 76 35,5 44 41 59 28 Bulgaria BG 70 44,5 30 55,5 40 72 85 23,5 69 14,5 16 77 Burkina Faso BF 70 44,5 15 93,5 50 45 55 62 27 60,5 18 72,5 Canada CA 39 88 80 5 52 41 48 80 36 46,5 68 18,5 Cape Verde CV 75 34 20 80,5 15 97 40 89 12 84 83 7 Chile CL 63 65,5 23 75 28 89 86 18,5 31 55 68 18,5 China CN 80 24,5 20 80,5 66 14 30 95,5 87 4 24 68 Colombia CO 67 56 13 99 64 20 80 31,5 13 81,5 83 7 Costa Rica CR 35 92 15 93,5 21 92 86 18,5 na na Croatia HR 73 37 33 49 40 72 80 31,5 58 24,5 33 56,5 Czech Republic CZ 75 34 58 26 57 32 74 38 70 13 29 61 Denmark DK 18 101 74 10 16 96 23 100 35 49 70 15 Dominican Republic DO 65 60,5 30 55,5 65 17 45 84 13 81,5 54 34
25 Ecuador EC 78 30 8 102 63 22,5 67 46 na na Egypt EG 70 44,5 25 69,5 45 56 80 31,5 7 86 4 81 El Salvador SV 66 57,5 19 86 40 72 94 7,5 20 71,5 89 4 Estonia EE 40 86 60 22 30 88 60 54,5 82 7 16 77 Ethiopia ET 70 44,5 20 80,5 65 17 55 62 na na Fiji FJ 78 30 14 97 46 52,5 48 80 na na Finland FI 33 95 63 19 26 91 59 57,5 38 44 57 30 France FR 68 53,5 71 11,5 43 61 86 18,5 63 18 48 40,5 Germany DE 35 92 67 17 66 14 65 48,5 83 5 40 50,5 Ghana GH 80 24,5 15 93,5 40 72 65 48,5 4 87 72 13 Greece GR 60 69 35 46,5 57 32 100 1 45 39,5 50 37 Guatemala GT 95 4,5 6 103 37 82,5 10 102 na na Honduras HN 80 24,5 20 80,5 40 72 50 72 na na Hong Kong HK 68 53,5 25 69,5 57 32 29 97,5 61 21 17 74,5 Hungary HU 46 81 80 5 88 3 82 26,5 58 24,5 31 58 Iceland IS 30 98 60 22 10 99,5 50 72 28 58,5 67 21 India IN 77 32 48 32 56 35 40 89 51 31,5 26 65 Indonesia ID 78 30 14 97 46 52,5 48 80 62 19 38 52,5 Iran IR 58 71,5 41 37 43 61 59 57,5 14 78 40 50,5 Iraq IQ 95 4,5 30 55,5 70 8 85 23,5 25 65 17 74,5 Ireland IE 28 99 70 13,5 68 11,5 35 93,5 24 67,5 65 24 Israel IL 13 102 54 28 47 50,5 81 28 38 44 na Italy IT 50 77 76 8 70 8 75 37 61 21 30 59 Jamaica JM 45 82 39 39,5 68 11,5 13 101 na na Japan JP 54 76 46 35 95 2 92 11 88 3 42 48 Jordan JO 70 44,5 30 55,5 45 56 65 48,5 16 74,5 43 46 Kenya KE 70 44,5 25 69,5 60 27 50 72 na na Kuwait KW 90 12,5 25 69,5 40 72 80 31,5 na na Latvia LV 44 83 70 13,5 9 101 63 52 69 14,5 13 80 Lebanon LB 75 34 40 38 65 17 50 72 14 78 25 66,5 Libya LY 80 24,5 38 41,5 52 41 68 44,5 23 69 34 55
26 Lithuania LT 42 84 60 22 19 94,5 65 48,5 82 7 16 77 Luxembourg LU 40 86 60 22 50 45 70 40,5 64 17 56 32 Malawi MW 70 44,5 30 55,5 40 72 50 72 na na Malaysia MY 100 1,5 26 64 50 45 36 92 41 42 57 30 Malta MT 56 74 59 25 47 50,5 96 4 47 36,5 66 22,5 Mexico MX 81 19,5 30 55,5 69 10 82 26,5 24 67,5 97 3 Morocco MA 70 44,5 46 35 53 39 68 44,5 14 78 25 66,5 Mozambique MZ 85 16,5 15 93,5 38 80,5 44 86,5 11 85 80 9,5 Namibia NA 65 60,5 30 55,5 40 72 45 84 35 49 na Nepal NP 65 60,5 30 55,5 40 72 40 89 na na Netherlands NL 38 89 80 5 14 98 53 65 67 16 68 18,5 New Zealand NZ 22 100 79 7 58 29,5 49 77,5 33 53 75 12 Nigeria NG 80 24,5 30 55,5 60 27 55 62 13 81,5 84 5 Norway NO 31 96,5 69 15 8 102 50 72 35 49 55 33 Pakistan PK 55 75 14 97 50 45 70 40,5 50 33 0 82 Panama PA 95 4,5 11 101 44 59 86 18,5 na na Peru PE 64 63,5 16 90,5 42 64 87 15 25 65 46 42,5 Philippines PH 94 7,5 32 50 64 20 44 86,5 27 60,5 42 48 Poland PL 68 53,5 60 22 64 20 93 9 38 44 29 61 Portugal PT 63 65,5 27 62 31 87 99 2,5 28 58,5 33 56,5 Puerto Rico PR 68 53,5 27 62 56 35 38 91 19 73 99 2 Romania RO 90 12,5 30 55,5 42 64 90 13 52 29,5 20 70 Russia RU 93 9 39 39,5 36 84 95 5,5 81 9 20 70 Saudi Arabia SA 95 4,5 25 69,5 60 27 80 31,5 36 46,5 52 36 Senegal SN 70 44,5 25 69,5 45 56 55 62 25 65 na Serbia RS 86 15 25 69,5 43 61 92 11 52 29,5 28 63,5 Sierra Leone SL 70 44,5 20 80,5 40 72 50 72 na na Singapore SG 74 36 20 80,5 48 49 8 103 72 12 46 42,5 Slovakia SK 100 1,5 52 29,5 100 1 51 66,5 77 10 28 63,5 Slovenia SI 71 38 27 62 19 94,5 88 14 49 34 48 40,5 South Africa ZA 49 78,5 65 18 63 22,5 49 77,5 34 51,5 63 25,5
27 South Korea KR 60 69 18 87,5 39 79 85 23,5 100 1 29 61 Spain ES 57 73 51 31 42 64 86 18,5 48 35 44 45 Sri Lanka LK 80 24,5 35 46,5 10 99,5 45 84 45 39,5 na Suriname SR 85 16,5 47 33 37 82,5 92 11 na na Sweden SE 31 96,5 71 11,5 5 103 29 97,5 53 28 78 11 Switzerland CH 34 94 68 16 70 8 58 59 74 11 66 22,5 Syria SY 80 24,5 35 46,5 52 41 60 54,5 30 56,5 na Taiwan TW 58 71,5 17 89 45 56 69 43 93 2 49 38,5 Tanzania TZ 70 44,5 25 69,5 40 72 50 72 34 51,5 38 52,5 Thailand TH 64 63,5 20 80,5 34 85 64 51 32 54 45 44
Trinidad and Tobago TT 47 80 16 90,5 58 29,5 55 62 13 81,5 80 9,5
Turkey TR 66 57,5 37 43 45 56 85 23,5 46 38 49 38,5
Ukraine UG 92 10 25 69,5 27 90 95 5,5 55 27 18 72,5
United Arab Emirates AE 90 12,5 25 69,5 50 45 80 31,5 na na
United Kingdom GB 35 92 89 3 66 14 35 93,5 51 31,5 69 16 United States US 40 86 91 1 62 24 46 82 26 62,5 68 18,5 Uruguay UY 61 67 36 44 38 80,5 99 2,5 26 62,5 53 35 Venezuela VE 81 19,5 12 100 73 6 76 35,5 16 74,5 100 1 Vietnam VN 70 44,5 20 80,5 40 72 30 95,5 57 26 35 54 Zambia ZM 60 69 35 46,5 40 72 50 72 30 56,5 42 48
28
Appendix 2 - Frequencies of number of companies
This table shows for acquirer and target the number of companies per country. The percentage is also added.
Acquirer Freq. Percent Target Freq. Percent
US 1756 34,3% GB 931 18,2% AU 767 15,0% AU 438 8,6% GB 321 6,3% NL 397 7,8% DE 302 5,9% US 346 6,8% FR 298 5,8% CN 310 6,1% CH 295 5,8% CH 262 5,1% JP 191 3,7% IT 237 4,6% NL 185 3,6% ES 195 3,8% SG 139 2,7% FR 193 3,8% CA 100 2,0% IN 137 2,7% ES 94 1,8% CA 136 2,7% SE 73 1,4% NZ 115 2,2% CN 62 1,2% DE 111 2,2% IT 45 0,9% PT 89 1,7% KR 43 0,8% KR 82 1,6% IL 37 0,7% SE 77 1,5% MY 30 0,6% ID 73 1,4% BE 29 0,6% HK 68 1,3% FI 28 0,5% JP 67 1,3% HK 25 0,5% BE 63 1,2% ZA 22 0,4% BR 49 1,0% KW 20 0,4% FI 49 1,0% TW 20 0,4% MY 49 1,0% MX 18 0,4% RU 45 0,9% AT 16 0,3% SG 44 0,9% IN 16 0,3% IL 41 0,8% LU 15 0,3% NO 37 0,7%
29 AE 14 0,3% TR 36 0,7% CL 14 0,3% PL 35 0,7% BR 12 0,2% JO 31 0,6% PT 11 0,2% TW 28 0,5% DK 10 0,2% CL 23 0,4% HU 10 0,2% AT 21 0,4% RU 10 0,2% IE 21 0,4% LB 9 0,2% GR 20 0,4% PL 9 0,2% LU 20 0,4% TH 9 0,2% PE 20 0,4% GR 8 0,2% TH 19 0,4% NO 8 0,2% ZA 17 0,3% SA 7 0,1% MX 15 0,3% PH 6 0,1% EG 14 0,3% ID 5 0,1% PH 14 0,3% NZ 5 0,1% VN 14 0,3% IE 4 0,1% DK 13 0,3% EG 3 0,1% MA 11 0,2% HR 3 0,1% SK 9 0,2% AR 2 0,04% AR 8 0,2% CZ 2 0,04% HU 8 0,2% MA 2 0,04% IS 8 0,2% TR 2 0,04% HR 7 0,1% JO 1 0,02% LT 7 0,1% PA 1 0,02% MT 7 0,1% PE 1 0,02% RO 7 0,1% PK 6 0,1% CZ 5 0,1% LK 5 0,1% EE 4 0,1%
30 RS 4 0,1% KW 3 0,1% SI 3 0,1% BG 2 0,04% LV 2 0,04% NA 2 0,04% AE 1 0,02% GH 1 0,02% KE 1 0,02% SA 1 0,02% ZM 1 0,02%
31
11. References
Ahern K.R., Daminelli D., Fracassi C. (2015). Lost in translation? The effect of cultural values on mergers around the world, Journal of Financial Economics, 117: 165-189.
Asquith P., Bruner R. F. and Mullins D. W. (1983). The gains to bidding firms from merger, Journal of
Financial Economics, 11(1–4): 121–139.
Alexandridis G., Petmezas D., and Travlos N.G. (2010). Gains from Mergers and Acquisitions Around the World: New Evidence, Financial Management. 1671 – 1695.
Bertrand O. and Zitouna H. (2008). Domestic versus cross-border acquisition: which impact on the target firms’ performance? Applied economics, 40(17): 2221-2238.
Dong, M., Hirshleifer D., Richardson, S. and Teoh S. H. (2003). Does Investor Misvaluation Drive the Takeover Market? The Journal of Finance, 61(2): 725-762.
Dickerson, Gibson A., H. and Tsakalotos E. (1997). The impact of acquisitions on company
performance: evidence from a large panel of UK firms. Oxford Economics Papers, 49: 344– 361.
Emans, B. J. M., Laskewitz, P., & Van de Vliert, E. (1994). The Netherlands. In A.Rahim & A. A.Blum (eds.), Global perspectives on organizational conflict (pp. 53–66), Westport , CT ; Praeger. Erel. I., R. C. Liao. M. S. Weisbach. 2012. Determinants of Cross-Border Mergers and Acquisitions. The
Journal of Finance. 67: 1045-1082.
Franks, J.R., Harris R. and Titman S., (1991). The postmerger share-price performance of acquiring firms. Journal of Financial Economics. 29:9981-9996.
Gugler, K., Mueller D.C., Yurtoglu B.B., and Zulehner C. (2003), The effects of mergers: An international comparison, International Journal of Industrial Organization. 21: 625- 653. Healy, P., Palepu, K., Ruback, R., (1992). Does corporate performance improve after mergers? Journal
of Financial Economics. 31: 135– 175.
Hofstede G. (1984). Cultural Dimensions In Management And Planning. Asia Pacific Journal of
Management. 81-99
Kaplan, S. and Weisbach, M.S. (1992). The Success of Acquisitions: Evidence From Disvestitures. The
Journal of Finance, XLVII(1): 107-138.
Lyon, J. D., Barber, B. M. and Tsai, C.-L. (1999), Improved Methods for Tests of Long-Run Abnormal Stock Returns. The Journal of Finance, 54: 165–201.
MacKinlay, A.C. (1997). Event studies in economics and finance. Journal of economic literature, 35(1), 13-39
32 Martynova M. and Renneboog L. (2008). A century of corporate takeovers: What have we learned
and where do we stand?, Journal of Banking & Finance. 32(10) :2148-2177
Moeller S.B. and Schlingemann F. P. (2005), Global diversification and bidder gains: A comparison between cross-border and domestic acquisitions, Journal of Banking & Finance, 29(3)533– 564.
Moeller S.B., Schlingemann F. P., Stulz R.M. (2004) Firm size and the gains from acquisitions. Journal
of Financial Economics, 73(2): 201-228
Oudenhoven van, J.P.L.M.,& de Dreu, C.K.W. (1997) Managerial conflict management in five European countries: the importance of power distance and uncertainty avoidance, and masculinity. Applied psychology, 10: 97 -120
Powell, E., & Mitchell, S. (2007). The International Court of Justice and the World's Three Legal Systems. The Journal of Politics, 69(2): 397-415.
Schwert G.W. (1996). Markup pricing in mergers and acquisitions. Journal of Financial Economics, 41(2): 153–192.
Stock, J.H., & Watson, M.W. (2003). Introduction to econometrics (Vol. 104). Boston: Addison Wesley.
Tenzer, H., Pudelko, M., Harzing, A. (2014) The impact of language barriers on trust formation in multinational teams. Journal of International Business Studies, 45, 508–535
