The influence of cultural distance on company performance in
the process of German acquisitions
By Gert Jan Nagel*
Aquamarijnstraat 865
9743 PX Groningen
(06) 12142383
g.j.nagel@student.rug.nl
Student number 1410636
University of Groningen
Faculty of Economics & Business
Department of International Business
Supervisor: Dr. A.B Kibriscikli-Ozcandarli
Co-Assessor: Dr. C. Dörrenbächer
April 4
th2008
“Culture is more often a source of conflict than of synergy. Cultural differences are a nuisance at best and often a disaster.” (Hofstede, 1997:10)
ABSTRACT
Although several studies justified the negative influence of cultural differences in the process of mergers and acquisitions, this study indicates that cultural distance on Hofstede’s (2001) dimensions power distance, individualism, uncertainty avoidance and masculinity has no significant influence on the post-acquisition performance of German acquiring companies. A multiple regression analysis shows that power distance, uncertainty avoidance- and masculinity differences even have a positive relationship with the post-acquisition performance. This might be due to what O’Grady and Lane (1995) describe as the ‘physic distance paradox’. It emphasizes that operations like acquisitions in psychically close countries are not necessarily easy to manage.
Keywords: merger, acquisition, cultural distance, post-acquisition performance, power distance, individualism, uncertainty avoidance, masculinity.
TABLE OF CONTENT
1. INTRODUCTION...4
2. LITERATURE REVIEW...6
2.1 Mergers and acquisitions...6
2.2 National – and organizational culture...7
2.3 The theory of Hofstede and the German scores... 8
2.3.1 Power distance...9
2.3.2 Individualism ...10
2.3.3 Uncertainty avoidance...10
2.3.4 Masculinity ...10
2.3.5 Characteristics of the ‘German way’ of doing business...11
2.4 Influences of cultural distance on the post-acquisition performance……….... 12
2.5 Conceptual framework………..…. 15
3. RESEARCH METHODOLOGY……….. 19
3.1 Research setting and sample………...… 19
3.2 Data collection………...…. 20
3.3 Measure of variables………...………… 21
4. ANALYSIS AND RESULTS ...24
4.1 Descriptive Statistics……….... 24
4.2 Multiple Regression requirements...29
4.3 Correlation analysis……….. 31
4.4 Multiple Regression Analysis………... 32
5. CONCLUSION AND DISCUSSION ...37
REFERENCES...44
1. INTRODUCTION
In today’s world of business, mergers and acquisitions are very common all around the world. It is interesting, that despite the popularity of mergers and acquisitions, research indicates that between 55-70 per cent of mergers and acquisitions fail to meet the anticipated purposes (Carleton, 1997). Very often a lack of cultural fit between the merging companies is mentioned as an important factor of the failure (Nahavandi and Malekzadeh, 1988, Weber and Sweiger, 1992). In the case of domestic mergers and acquisitions, cultural differences will play a less important role than in case of international mergers and acquisitions (Weber and Camerer, 2003). International mergers and acquisitions bring together not only two different organizational cultures, but also two different national cultures (Schneider, 1988). Both companies are embedded in different national cultures, and therefore have different beliefs, values and assumptions. They also operate within different business systems and share different perceptions of information and time (Claes and Gerritsen, 2002). The increasing cost of integration, as a result of this national cultural distance, might influence the cross-border acquisition performance (Singh and Montgomery, 1987). Therefore, mergers and acquisitions between corporations from different national cultures might include many extra difficulties.
The aim of this research is to make the reader aware of the importance of cultural aspects in case of acquisitions and to investigate whether differences in national culture between companies will significantly influence the post-acquisition performance. Therefore, the following main research question will be answered: “What is the influence of cultural differences on the post-acquisition performance of the acquirer”?
This research is part of the broader stream of research about mergers and acquisitions. Nevertheless, as can be written in the research question, the focus is on acquisitions only.
Furthermore, the research has been done in a German context, which means that all acquiring companies are German. Lots of M&A literature is based on American companies and only few come up with empirical evidence. Therefore, a choice has been made for a European perspective rather than an American. Germany is the largest economy in Europe and after the United States and Japan the third biggest in the world. For these reasons, German acquiring companies are the focus of this research.
four dimensions of national culture: power distance, individualism, uncertainty avoidance and masculinity. This measure for cultural distance has been used because of the extensive evidence of the validity and reliability of the underlying Hofstede’s (2001) national cultural scores (Shane, 1993) Jemison and Sitkin (1986) argue that higher levels of cultural distance between firms have been associated with higher degrees of conflict during the post-acquisition integration period. As a result, it might influence the overall performance of the companies which engaged in this acquisition. Therefore, a relationship between cultural distance and the performance of these companies in case of an acquisition can be assumed. This assumed relationship will be tested by a multiple regression analysis in order to answer the main research question.
2. LITERATURE REVIEW
This part of the research will make the reader familiar with terms like mergers, acquisitions, national culture and organizational culture. Secondly, the theory of Hofstede (2001) will be described and applied to the German context. Thirdly, it will consist of the most relevant literature concerning the influences of cultural distance between companies in case of mergers and acquisitions. Furthermore, a theoretical framework will be presented in order to create a suitable foundation for this research. 2.1 Mergers and acquisitions
The main idea behind mergers and acquisitions is that one plus one makes three. Two companies together may be more valuable than two separate companies (25-08-2007, www.investopedia.com). At least that is the reasoning behind it. Often, terms like ‘merger’ and ‘acquisition’ are used as though they are synonymous. Actually, it has to be made clear that both terms are not the same. According to Sudarsanam (1995:1) “two corporations come together to combine and share their resources to achieve common objectives”, in case of a merger. The shareholders of both companies often remain as joint owners of the combined entity. An acquisition resembles more of an arm’s-length deal, with one firm purchasing (acquirer) the assets or shares of another company (target). In a merger, a new entity may be formed subsuming the merging firms. On the other hand, the acquired firm becomes the subsidiary of the acquirer in case of an acquisition (Sudarsanam, 1995). Both mergers and acquisitions are part of a broader company aim, namely to grow and expand its activities. Other goals of M&A are acquiring new knowledge and technology or the creation of synergies through economies of scale and – scope. Mergers and acquisitions can take place in one country but also across borders. These are so called international mergers and acquisitions.
Existing literature presents a variety of mergers. Based on the relationship between the merging companies, there are five types (25-08-2007, www.investopedia.com):
1. Vertical mergers, where a customer and company or a supplier and company are merging. 2. Horizontal mergers, where two companies are merging that are competitors. For instance, they
produce the same products or operate in the same market.
3. Market-extension mergers, when two corporations are merging that sell the same products but operate in different markets.
4. Product-extension mergers, when two companies are selling different but related products in the same market are merging.
Unlike mergers, most acquisitions involve one company purchasing another. There is no exchange of stock or consolidation as a new company. Acquisitions may sometimes be friendly when all parties feel satisfied with the deal. They may also be hostile (Gertsen, Søderberg and Torp, 1998). In case of an acquisition or takeover, a company is able to buy another company with stock, cash or a combination of both. It is possible that a certain company acquires all the assets of another company but this mostly happens in smaller deals.
As already mentioned, the main force behind M&A is that ‘one plus one makes three’. Synergy is the force that makes it possible for one plus one to become three. According to Brock (2005:271) “synergy occurs when the profitability of the combined entity is more than the profit that would have been generated by the two firms independently”. Other authors like Cartwright and Cooper (1992) and Sirower (1997) describe it as the 1 + 1 > 2 effect. Brock (2005) makes four distinctions between synergies:
1. Financial synergies which are for instance including reduced overhead costs and tax benefits. 2. Production synergies which can be achieved through economies of scale, increased
purchasing power or elimination of inefficient product lines.
3. Technological synergies are realised by transferring specific technology or technical processes (Kitching, 1967).
4. Organizational synergies are generated by eliminating duplicated functions or realizing employee creativity through improved personnel administration.
In this study, synergies will be measured by the improvement of the acquiring company’s Return on Total Assets. The ROA can be calculated by dividing the company’s profits by the amount of total assets. A clarification for this choice will be given in the methodology part of this paper. In view of the fact that company profits can be influenced by several factors, this ROA can be a result of all types of above described synergies, depending on the goals of both companies, prior to the merger or acquisition.
To sum up, the main forces for companies to engage in mergers and acquisitions are to generate synergies by creating economies of scale, economies of scope, acquiring new technologies and improving market share by for instance activity expanding (Sudarsanam, 1995).
2.2 National - and organizational culture
general, mostly called national culture. Often, the terms national – and organizational culture are used as if they mean the same. It has to be made clear that these two concepts are related to each other but are not the same. The concept of organizational culture became popular in the late 1970s. Organizational researchers like Pettigrew (1979), Ouchi (1981) and Peters and Waterman (1982) began paying serious attention to it. According to Pettigrew (1979:18) organizational culture refers to “the collection of beliefs, values and assumptions held by the members of an organization”. This organizational culture or corporate culture is influenced by a broader concept of culture, namely national culture. Schneider (1988) points out that in large multinationals, known for their strong corporate cultures, national differences are still very strong represented. Hofstede (1991:5) provide a comprehensive definition of culture as “the collective programming of the mind that distinguishes the members of one category of people from those of another”. Hofstede (1980) defines national culture as the collective programming of the human mind. Culture is composed of certain values, which shape people’s behaviour. Adler (1997) argues that national culture influences our values, which in turn affects our attitudes, and as a result our behaviour. This assumption has been empirically justified by Homer and Kahle (1988).
As already mentioned, culture is hard to define and therefore difficult to measure. An approach which focuses on corporate culture alone leads to the development of uninational theories which are of little use beyond borders (Weber et al. 1996). Organizational culture is embedded in a broader national culture. Furthermore, there is extensive evidence of the validity and reliability of the national cultural scores from Hofstede used in this study (Shane, 1993).
Therefore, a choice has been made for culture as being ‘national’ using Hofstede’s (2001) national cultural scores.
2.3 The theory of Hofstede and the German scores
Because the theory of Geert Hofstede is an important foundation of this research, a short description will be given. Furthermore, this research focuses on the performance of German acquiring companies, and as a result, their culture is very important for the success of the acquisition. Each target company, with its particular national culture, will be confronted with the culture of the acquiring German firm. Therefore, the German culture will shortly be described according to the dimensions of Hofstede (2001). Furthermore, in international acquisitions, different business systems are confronted. These different systems may collide and influence the company performance. Therefore, some characteristics of the German business system will be pointed out.
countries. Hofstede developed a model (figure 1) that identifies four primary dimensions to support in distinguishing cultures; power distance, individualism, uncertainty avoidance and masculinity. The arrows indicate that the scores on each dimension create the specific national culture which differs from other countries. The scale of the scores is of interval level. Therefore, the distance between the scores does have meaning. For example, the distance from 10-20 is the same as from 30-40.
Later on, Hofstede added a fifth dimension after conducting an additional international study with a survey instrument developed with Chinese employees and managers. This fifth dimension, called long-term orientation was based on Confucian dynamism and was applied to 23 countries. In this research, only the first four dimensions are used in order to measure culture. Because the fifth dimension was applied to only 23 countries it is not part of this research. The four dimensions are described below and applied to Germany.
Figure 1: Hofstede’s Model of Cultural Differences (Hofstede, 1991:31)
2.3.1 Power distance
2.3.2 Individualism
According to Hofstede (2001) individualism focuses on the degree the society reinforces individual or collective achievement and relationships. It “represents the relation between an individual and her/his fellow individuals” (Swaidan and Hayes, 2005:10). The opposite of individualism is collectivism. Here, we find societies in which people from birth onwards are integrated into strong, cohesive groups. In the case of Germany a score of 67 has been assigned. In contrast, some Anglo-Saxon countries like Great Britain, Canada and the United States of America have higher scores in this variable. The USA has a score of 91 which indicates that individuality and individual rights are paramount within the society. Individuals in these societies may tend to form a large number of looser relationships.
2.3.3 Uncertainty avoidance
Hofstede (1983) describes uncertainty avoidance as the degree a certain society is tolerant towards uncertainty and ambiguity. Furthermore, it indicates to what extent a culture programs its members to feel either uncomfortable or comfortable in unstructured situations. Germany scores relatively high (65) on this dimension in comparison with other economically well developed countries like Great Britain (35) and the United States of America (46). On the other hand, Japan (92) scores even more higher on this dimension which means that Japanese people try to minimize the possibility of novel, surprising and unusual situations. It can be imagined that M&A, as a result of this cultural distance on uncertainty avoidance between German and Japanese companies, are more difficult and it might influence the post-mergers or acquisition performance negatively. The same difficulties can arise when German firms engage in M&A with American or British corporations.
2.3.4 Masculinity
Masculinity refers to the role between genders. The opposite of masculinity is femininity.
A high masculinity ranking indicates the country experiences a high degree of gender differentiation. In these cultures, males dominate a significant portion of the society and power structure, with females being controlled by male domination. A low masculinity ranking indicates the country has a low level of differentiation and discrimination between genders. In these cultures, females are treated equally to males in all aspects of the society. Aspects of more masculine oriented societies are assertiveness and competitiveness. When we apply the dimension masculinity to the German context, a score of 66 on this dimension can be found.
for each citizen. Furthermore, high scores on uncertainty avoidance and individualism illustrates that Germans do not really like surprises or sudden changes and are relatively individualistic.
Figure 2: The German scores on Hofstede’s cultural dimensions (21-10-2007, www.geert-hofstede.com)
2.3.5 Characteristics of the ‘German way’ of doing business.
Not only cultural distance but also ‘the way of doing business’ can have its influence on the performance of companies in case of M&A. Every company has its own rituals which may differ significantly from others. To ensure the achievement of business success with a German corporation you have to be informed about the cultural differences and expectations in Germany and the situations in which they are important.
When we take a look at the German decision-making process it turned out to be extremely time consuming and much slower compared to for instance the USA. This can be troublesome to US executives when German and USA companies engage in M&A (Shibata et al., 1991). These differences in decision-making are partly a result of the scores on the cultural dimensions. For instance, German power distance (35) is lower than the USA (40). This might indicate that power inequalities in US companies are more present and people who make decisions have relatively more power. As a result, the decision-making process might be quicker than in German companies. According to uncertainty avoidance, Germany (65) scores higher than the USA (46). In Germany, each possible consequence is taken into account before the decision is made. Therefore, the decision-making process is more time consuming than in US companies where they are more tolerant towards future uncertainties.
rituals is the fact that Germans keep a larger personal space around them compared to for instance North Americans do. When German people are engaged in a conversation they keep a considerable space between them, as where North Americans are closer and even touch each other while they are talking. Age is also very important in Germany. In business meetings, age takes precedence over youth. When entering a room, the eldest person will go first.
These aspects are important factors which should be taken into account in doing business with German companies, for example in the context of international mergers and acquisitions where different cultures are confronted.
2.4 Influences of cultural distance on the post-acquisition performance
As will be shown, many authors have written about the influence of cultural distance between corporations in the case of M&A but not all authors agree on this topic. Although most believe that differences in national culture affect performance in acquiring firms, some studies find cultural differences are problematic, others find they add value to the parent firm. Other studies find variable or no effects (Brock, 2005). As already mentioned, many agree that national cultural distance influences the performance, but study of the relationship between cultural differences and acquisition performance have produced conflicting findings (Weber, Shenkar, and Raveh, 1996). Hofstede (1980) and Cartwright and Price (2003) argue that cultural differences create organizational challenges that hinders integration and increase acquisition costs. Gomez-Mejia and Palich (1997) also state that cultural distance between the acquired and acquiring firm requires increased integration and therefore higher costs. Jemison and Sitkin (1986) share these ideas and state that higher levels of cultural distance have been associated with greater conflict in an acquisition over day-to-day decisions. Furthermore, these differences between the acquired corporation and the parent company can lead to cultural clash among employees (Brock, Barry and Thomas, 2000). Cartwright and Cooper (1993) argue that different types of culture create different psychological environments for the acquisition or merged company. These differences in practices have negative influence on performance. “The degree of cultural fit that exists between combining organizations is likely to be directly correlated to the success of the combination” (Cartwright and Cooper, 1993:60).
Summarizing, these findings suggest that higher cultural differences are associated with higher costs. These costs are not just money but also costs in the sense of less tangible aspects. For instance, when companies are not well integrated, employees may feel less committed. As a result they may be less motivated and less productive. It therefore reduces the post-acquisition performance.
clarified by Morosini, Shane and Singh (1998) who found a significant positive relationship between cultural distance and post-acquisition performance. Their study included 52 companies and field interviews with managers. In a more general sense, O’Grady and Lane (1996) indicate that operations, like mergers and acquisitions, in psychically close countries are not necessarily easy to manage. This is because assumptions of similarity can prevent executives from learning about critical differences. O’Grady and Lane did research on thirty-two Canadian retail companies. Evidence shows that only seven (22%) were functioning successfully in the United States. Empirical evidence confirms greater cultural differences between Canada and the U.S than assumed previously.
In sum, different from the focus on the higher costs of integration, these authors emphasize the learning benefits provided by cultural differences.
As already indicated, a third stream of research on this topic yielded mixed results and is not sure about whether cultural distance influences performance. Calori and Lubatkin (1996) found that the post-acquisition performance is sometimes influenced by national culture, but not always in the same direction. Gomez-Mejia and Palich (1997) conducted a study of performance in Fortune 500 firms between 1985 and 1994 and found that neither cultural difference nor culturally unrelated differences were associated with firm performance in cases of mergers and acquisitions.
All above mentioned conflicting results in this specific field of interest is another convincing reason for this study to further explore the relationship between cultural differences and company performance in the context of acquisitions. The research question that will be answered is therefore: What is the influence of cultural differences on the post-acquisition performance of the acquirer?
Literature shows that firms in countries which are significantly distant along Hofstede’s (1980) power distance and individualism dimensions, present specific differences in their decision-making practices and in their power and control structures. This might create conflict when companies engage in M&A (Bourgoin, 1989; Kreacic and Marsh, 1986).
potential for poor integration, lower managerial commitment and less resource sharing is created (Brock, 2005). This mismatch is more likely to be present in international M&A because different national cultures get together. Cultural aspects like routines and repertoires related to innovation and inventiveness, as well as the degree of entrepreneurship, have been found to vary significantly across countries along Hofstede's (1980) individualism-collectivism variables (Shane, 1993). These differences in the level of power distance and individualism might result in conflicts which eventually can have a negative effect on the post-merger or acquisition performance.
As a result of these findings, it can be imagined that this created potential has a negative influence on the performance of both the parent – and the acquired company after the acquisition has been completed. Based on the findings in the above described literature, the following two hypotheses are formulated:
H1: A higher difference in ‘power distance’ between companies will lead to a lower post- acquisition performance
H2: A higher difference in ‘individualism between companies will lead to a lower post- acquisition performance
If the level of uncertainty avoidance in a company or within a nation is high, people are less willing to take risks and are feeling uncomfortable in unstructured situations. The level of uncertainty avoidance has strong influence on the decision-making process of managers. Managers who dislike unstructured situations and are reluctant to take risks will be less willing to make big investments and turn out to be less innovative (Chuck et al., 2006). This might create serious discrepancy between the parent – and acquired company in the case of M&A. When for instance, a German corporation with a score of 65 on uncertainty avoidance acquires a Japanese company (score of 92) two different decision-making processes come together which results in a so called ‘cultural clash’. As mentioned before, Jemison and Sitkin (1986) share these ideas and state that higher levels of cultural distance have been associated with greater conflict in an acquisition over day-to-day decisions. Not only managers, but also the employees of both companies have their own perception about uncertainty avoidance. Japanese employees are more resistant to changes in structure and the adaptation of different practices (Brock, 2005). This might also be an obstacle to the realisation of synergies and therefore the post-acquisition performance. Therefore, a third hypothesis has been formulated:
The fourth dimension in the theory of Hofstede (1980) is masculinity. Cultures with high scores on the masculinity dimension are mostly experiencing a high degree of gender differentiation. These cultures are characterised by males dominating a great portion of the society and power structure. Usually, females are being controlled by male domination. A low masculinity ranking indicates the fact that females are treated equally to males in all aspects of the society.
Jemison and Sitkin (1986) found that higher levels of cultural distance on masculinity between firms have been associated with higher degrees of conflict during the post-merger and acquisition integration period. Just like the other cultural dimensions, the level of masculinity has its implications for the decision-making process and distribution of power within a corporation. As a result, companies that are more distant on this dimension may face troubles integrating both decision-making structures and power structures. Conflicts may arise when employees, used to male typical forms of behaviour like assertiveness and aggressiveness, are now controlled by female supervisors or are subjected to more female types of behaviour like carefulness. This potential for conflicts might have its influence on the performance of the company after the acquisition has been accomplished. Therefore, the following hypothesis has been formulated:
H4: A higher difference in ‘masculinity’ between companies will lead to a lower post- acquisition performance
2.5 Conceptual framework
Figure 3 can be interpreted as a conceptual framework for this research. It shows the relationships between all variables. Both the acquiring- and target companies have their own national cultures which influences the values for all four dimensions of Hofstede (1980). This is represented by the arrows 1 and 2. The cultural distance between both companies is the result of the companies’ cultural scores. Arrow 4 and 5 indicate this. According to the earlier described literature, when these two companies engage in an acquisition this cultural distance might create a certain level of conflict and therefore it is likely that the post-acquisition performance will be influenced (arrow 3).
Figure 3: Conceptual framework
As shown in figure 3, two country control variables are included. The average growth of the Gross Domestic Product per capita (GDP per capita) of the target company’s country is the first country control variable. This to control, to a certain extent, for the overall economic situation of the home countries of the companies countries involved. The post-acquisition performance might be influenced by the overall economic situation of the countries of both companies. Since all acquiring companies are from Germany it would have had no influence if the GDP growth of Germany had been chosen. Therefore, a choice has been made for the GDP growth of the target company’s country in order to control for this overall economic country performance. A choice has been made for GDP per capita in order to control for country size. Large countries often have higher GDP and by dividing this GDP by the total number of inhabitants, GDP of all countries is comparable. The second country control variable is the distinction between Coordinated Market Economies (CME), Liberal Market Economies (LME) and Mixed Market Economies (MME). The ‘Varieties of Capitalism” approach advocated by Hall and Soskice (2001), stresses the notion that the way firms resolve many of the coordination problems they are confronted with, varies across countries (Akkermans et al. 2007). Both CMEs and
National culture Company X National culture Company Y Acquiring company X’s score on: - Power distance (P) - Individualism (I) - Uncertainty avoidance (UA) - Masculinity (M)
Target company Y’s score on: - Power distance (P) - Individualism (I) - Uncertainty avoidance (UA) - Masculinity (M)
Country control variables: - GDP per capita growth - Type of capitalism
o CME
o LME
o MME
Acquiring company control variables:
- Level of international experience
LMEs can be seen as two archetypes representing the extremes of a continuum. According to Hall and Soskice (2001:8), in LMEs:
“… firms coordinate their activities primarily via hierarchies and competitive market arrangements.
(…) Market relationships are characterized by the arm’s-length exchange of goods or services in a context of competition and formal contracting. In response to the price signals generated by such markets, the actors adjust their willingness to supply and demand goods or services (…)
In CMEs, on the other hand:
“… firms depend more heavily on non-market relationships to coordinate their endeavours with other
actors and to construct their core competencies. These non-market modes of coordination generally entail more extensive relational or incomplete contracting, network monitoring based on the exchange of private information inside networks, and more reliance on collaborative, as opposed to competitive, relationships to built the competencies of the firm. (…) the equilibria on which firms coordinate in coordinated market economies are more often the result of strategic interaction among firms and other actors” (Hall & Soskice, 2001:8)
Typical examples of LMEs are the U.S. and the UK. Typical examples of CMEs are Germany and Japan. However, Hall and Soskice acknowledge that several countries do not fit within this continuum of CME-LME. Therefore, they created a third category that is called Mediterranean (or Mixed) Capitalism. Countries with this classification are distinguished by their recent histories of extensive state intervention and large agrarian sectors. Typical examples are France, Italy, Spain, Portugal, Turkey and Greece (Hall and Soskice, 2001:21).
3. RESEARCH METHODOLOGY
3.1 Research setting and sample
This research is part of the broader stream of merger and acquisition literature and the influence of culture in this context. It can be considered as a deductive research because a conceptual framework and theoretical structure has been developed prior to its testing through empirical observation (Gill and Johnson, 2002). In this research, quantitative data is gathered and statistically tested.
Initially, the data for this study comes from 509 German companies who did, according to the Zephyr database, an acquisition in the year 2000. These 509 companies was the starting point of this research but because data availability problems, the actual number of acquisitions that remained part of the research turned out to be 303. According to Brewerton and Millward (2001) this amount exceeds the required number of observations (107) when doing multiple regression analysis with eight or more independent variables. It therefore meets the requirements of achieving adequate statistical power in order to infer accurate conclusions from statistical testing (Brewerton and Millward, 2001). A choice for the year 2000 has been made because at this time an enormous wave of mergers and acquisitions was taking place. The choice for Germany has already been explained in the introduction.
All 303 acquisitions are shown in table 1. In these acquisitions the acquiring companies are all German. The acquisitions can be divided in national acquisitions and international acquisitions where a total amount of 150 are German-German acquisitions. The other 153 are acquisitions where a German company acquires a foreign target company. These will be referred to as German-foreign acquisitions or international acquisitions. The fact that almost half of the acquisitions are national acquisitions is interesting. It provides an opportunity to compare the post-acquisition performance of both national – international acquisitions since cultural distance is less important in national acquisitions than in international ones.
Table 1: Distribution of acquisitions
When taking a closer look at the data, the foreign companies come from different countries with each its particular culture. Besides the target companies from Germany itself, 21 other countries are involved in this research (table 2). The United States of America (43) and Great Britain (28) are both the largest components within this group.
Total number of German-foreign acquisitions 153 Total number of German-German acquisitions 150
Germany 150 Canada 3
USA 43 Australia 3
Great Britain 28 Finland 3
Switzerland 13 Israel 2 France 12 Turkey 2 Italy 8 Argentina 2 Netherlands 7 Belgium 2 Spain 7 India 1 Austria 5 Mexico 1
Denmark 4 South Africa 1
Sweden 4 Japan 1
Total number of target companies’ countries 303
Table 2: Distribution of countries of the target companies
3.2 Data collection
The data for this study has been obtained from several resources. According to the Zephyr database 509 German companies did an acquisition in the year 2000. The Zephyr database is published by Bureau van Dijk and contains information on M&A activity, IPO’s, joint ventures and private equity deals.
The performance measure, ROA is the dependent variable. The percentage change of the ROA of the acquiring company between 1999 and 2002 will be measured, which is the ROA one year before and two years after the acquisition. The acquisitions are all taking place in 2000. A choice has been made for two years because a wide body of literature suggests that the first two years after the acquisition are critical to its overall performance (Jemison and Sitkin, 1986; Balloun and Gridley, 1990). Jemison and Sitkin (1986) also argue that by the end of the two-year period, the process of combining the firms usually has been completed and the underlying results of integration can be measured effectively. A choice has been made for measuring performance through return on total assets in order to control for company size. Some researchers use other financial figures to measure company performance, for instance profits. When comparing the profits of companies, these are often biased because of the fact that bigger companies may have relatively higher profits. Other researchers use stock prices as a company performance indicator, but these are strongly related and depending on the overall performance of the stock market and overall performance of the economy.
Financial and contains information on equities, indices, macro economic data and financial data on companies. After this second step, the remained missing values were tried to derive from the company websites, through annual reports. As a fourth and final step an e-mail has been sent (appendix 1) to the companies with a question if they want to participate in this study and provide the required missing figures. As expected, this email had a very low response rate. Only eight out of 134 companies reacted on the email and three stated that they were not willing to participate in my research. This is a positive response rate of 5 out of 134 which is 3.7%. In appendix 2 some examples of replied emails can be found.
The scores for the four variables concerning the cultural dimensions are based on the outcomes of the theory of Hofstede (2001). The company control variables are mainly obtained from the Amadeus database. The missing company data was also obtained from DataStream, company websites and by e-mail. Furthermore, the country control variable GDP growth per capita is derived from the website of the International Monetary Fund (14-11-2007, www.IMF.nl). The second country control variable is the distinction between CMEs, LMEs and MMEs. The countries of origin from the target companies are classified as being a CME, LME or MME and this is based on the “Varieties of Capitalism” school of Hall and Soskice (2001).
3.3 Measure of variables
In this study an answer will be given on the main research question through testing the following four hypotheses empirically:
H1: A higher difference in ‘power distance’ between companies will lead to a lower post- acquisition performance
H2: A higher difference in ‘individualism between companies will lead to a lower post- acquisition performance
H3: A higher difference in ‘uncertainty avoidance’ between companies will lead to a lower post-acquisition performance
H4: A higher difference in ‘masculinity’ between companies will lead to a lower post- acquisition performance
These hypotheses are tested by a multiple regression analysis. The regression equation is formulated as follows:
Where:
Y = percentage change return on assets of the acquiring company between 1999 and 2002 ROA = profits before tax / total assets
β0= intercept
X1= difference in power distance
(score power distance acquiring company – score power distance target company) X2= difference in individualism
(score individualism acquiring company – score individualism target company) X3= difference in uncertainty avoidance
(score uncertainty avoidance acquiring company – score uncertainty avoidance target company)
X4= difference in masculinity
(score masculinity acquiring company – score masculinity target company)
Variable 5, 6 and 7 are three dummy variables concerning the level of international experience of the acquiring company. Here, three levels of international experience are distinguished: X5 = general international experience (0 = yes and 1 = no)
X6 = local experience (0 = yes and 1 = no)
X7 = experience with international acquisitions (0 = yes and 1 = no)
X8= size of the acquiring company (measured by the number of employees at the time of the acquisition in 2000)
X9 = GDP growth per capita of the target company’s country in US dollars X10= Type of capitalism
4. ANALYSIS AND RESULTS
This part of the paper will consist of a basic description of the data that is used in this research. Furthermore, the requirements for doing correlation- and multiple regression analysis will be pointed out. Finally, a correlation – and multiple regression analysis is used to answer the hypotheses on page 21. These results are the foundation of an answer to the research question of this study.
4.1 Descriptive statistics
As shown in the methodology part, eleven variables are of importance in this study. Table 3 will describe the key features of all variables except for the dummy variables and variable X10which only have scores of 0, 1 and 2. Here, N is the total number of observations, PD is Δ power distance, I is Δ individualism, UA is Δ uncertainty avoidance and M is Δ masculinity between the acquiring company and target company. Size TC is the size of the acquiring company.
Variable ROA PD I UA M Size AC GDP growth
N 303 303 303 303 303 303 303 Mean -2.33 4.86 8.22 10.03 6.81 21492 -0.04 Mode -0.59 0 0 0 0 68 -0.08 Median -0.68 0 1 5 0 398 -0.08 Minimum -67.44 0 0 0 0 2 -0.32 Maximum 11.56 46 49 42 66 440200 0.29 Std Deviation 7.29 9.34 10.4 11.63 13.96 70685.23 0.07 Kurtosis 46.90 4,04 -0.56 -0.64 6.21 21.1 2.09 Skewness -6.14 2.22 0.85 0.76 2.61 4.48 0.45
Table 3: Descriptive statistics of the entire sample
the deviation from the mean and can be calculated by extracting the square root of the variance. From table 3 it can be concluded that the standard deviations are not small. This indicates that they are not tightly grouped around the mean and the data includes many outliers. The kurtosis represents the relative flatness or peak height of each variable. A positive kurtosis indicates that the distribution of this variable contains a relatively high peak and a negative kurtosis indicates that the distribution of this variable is relatively flat. For ROA, the kurtosis is 46.90 which indicate that there is a very high peak. The skewness of each variable refers to the amount the distribution is shifted to the left or to the right. A negative skew means that more than 50% of the scores are towards the right on the histogram. A positive skew means the opposite.
In sum, the entire data of this study represents many outliers especially within the variables ROA and size acquiring company.
Variable ROA PD I UA M Size AC GDP growth
N 153 153 153 153 3153 153 153 Mean -2.65 9.63 16.28 19.78 13.09 33146 0.008 Mode 0.11 5 24 19 4 68 -0.08 Median -0.83 5 22 19 4 398 -0.08 Minimum -67.44 0 1 5 0 9 -0.32 Maximum 11.56 46 49 42 66 440200 0.29 Std Deviation 8.32 11.28 9.10 8.69 16.83 86016 0.07 Kurtosis 47.74 0.35 -0.27 -0.21 1.67 12.73 5.76 Skewness -6.46 1.24 -0.13 .29 1.64 3.53 -1.13
Table 4: Descriptive statistics of the international acquisitions
Table 4 shows that the worst performing company, with a ROA decrease of 6744%, is part of the group of international acquisitions. According to the literature this is normal because cultural distance is higher in international acquisitions. Nevertheless, it shows that the best performing company is also part of this group. This is contradictory to what one would assume. Furthermore, table 4 indicates that the minimum difference in power distance and difference in masculinity is zero. This is the result of a German-Great Britain acquisition. Germany and Great Britain have similar, 35 and 66 respectively, scores on both power distance and masculinity.
hypotheses of this research, it can be assumed that when companies engage in national acquisitions, the non-existence of cultural distance might positively influences the post-acquisition performance. Table 3 shows a ROA percentage change of all acquiring companies (303) of 2.33 which means -233%. On the other hand, table 4 shows an average percentage change of German-foreign acquisitions of -2.65 (-265%). Table 5 shows that the mean percentage change of the acquiring companies in case of German-German acquisitions is -1.99 (-199%). This simple analysis shows that German acquiring companies that engage in national acquisitions are performing better, than German acquiring companies in international acquisitions, according to their ROA growth. In the light of this research, this makes sense because cultural differences are less present within national acquisitions and therefore do not have to be dealt with. It has to be mentioned that acquiring companies in both national – and international acquisitions are performing worse in comparison with one year before the acquisition. This is indicated by the negative growth of the ROA. This might be the result of several other reasons, both within the company itself and outside influences like the overall performance of the industries and economies involved. Although, from this study it is not allowed to draw conclusions about this because there is no other insight company information, that might have its influence on the company performance. What we can say, is that the variable that represents the overall economic performance of the countries involved (GDP growth) has a mean of -0.04. This indicates that on average, the economies are also performing worse. This might be a possible influence of the worsened performance of the companies. Nevertheless, this -0.04 is a very small percentage in comparison with the enormous average worsening of firm performance.
Mean Δ% ROA German-German acquisitions -199% Mean Δ% ROA German-foreign acquisitions -265% Mean Δ% ROA in all cases -233%
Table 5: Average Δ% of ROA
In order to draw a more solid conclusion about whether companies in national (German-German) acquisitions are performing significantly better than the ones in international (German-foreign) acquisitions an independent-sample T-test is conducted. The T-test is usually used to see whether there are differences between groups (Brewerton & Millward, 2001). Here, an independent-sample T-test is used because the samples are unmatched. The requirement of having ratio or interval measurement level is met because ROA are ratio level. A confidence interval of 95% is used.
In this independent T-test the hypotheses are formulated as follows: H0: the means of both groups are the same
The results of the T-test are shown in table 6, which consists of both the group statistics (6a) and the independent sample test (6b). The group statistics indicate that 150 national – and 153 international acquisitions are part of the research. According to the group statistics, the means of national – and international acquisitions are -1.99 (-199%) and -2.65 (-265%) respectively. These numbers are all in line with the results of tables 3, 4 and 5.
6a. Group Statistics
Type of acquisition N Mean Std. Deviation Std. Error Mean Percentage change National acquisitions
ROA 1999-2002 International acquisitions 150 153 -1.9995 -2.6540 6.08815 8.32449 .49710 .67288 6b. Independent Sample T-Test
Table 6: Results of the T-test
The statistical program SPSS uses the Levene’s Test for Equality of Variances to analyse whether variances of two groups are the same. The Levene’s Test results in the acceptance of equal variances. Therefore, the T-test will be calculated according the equal-variances method. Table 5 shows that the p-value is 0.436. Because the test is one-tailed the p-value is 0.436/2 = 0.218. Therefore, we reject hypothesis 1 because 0.218 > 0.05. The average ROA of national – and international acquisitions are not significantly different.
To summarize, it can be stated that the post-acquisition performance of the acquirer in national acquisitions is not significantly better than in case of international acquisitions. Alternatively, the post-acquisition performance of the acquirer in international post-acquisitions is not significantly worse than in national acquisitions.
Levene’s Test for Equality
of Variances t-test for Equality of Means
F Sig. T Df Sig.
2-tailed
Mean
Difference Std. ErrorDifference
The dummy variables are ‘level of international experience’ and ‘type of capitalism’. Firstly, the level of international experience has three categories of experience. Table 7 summarizes the characteristics of each category.
Level of international experience
General international
experience Local experience Experience with international acquisitions
Yes No Yes no yes no
Number of companies 302 1 286 17 262 41
Table 7: Level of international experience
As table 7 indicates, only 1 acquiring company has no former international experience before they acquired another company. It was a German company called Globeground GmbH that acquired an American division of the company called Aero Groundservices. The post-acquisition performance of this company, measured by the ROA growth is -143%. Compared to the mean average growth ROA of all acquisitions (-233%) Globeground GmbH performed better two years after the acquisition took place. This is interesting because they had no opportunity to learn from former international experience, compared to the other acquiring companies in this study. Nevertheless, they performed better in terms of ROA. A possible explanation for this might be the fact that the data of this study contains several outliers that have influenced the means of ROA growth. Furthermore, table 7 shows that only 17 acquiring companies had no local experience before they engaged in the acquisition. Finally, 41 acquiring companies did an acquisition without having any former experience with international acquisitions. When taking a closer look at the post-acquisition performance of the latter group, the data shows that the 262 acquiring companies with experience with international acquisitions (-201%) perform better than the 41 companies without this type of experience (-436%). This result seems clear because the companies have gained former experience and by making use of this learning advantage they perform better. However, this conclusion can not be drawn with much certainty. It has to be made clear that these three groups are not mutually exclusive. Within the group of 262 acquiring companies with experience with international acquisitions also national acquisitions are present. Here, former gained international experience is of less importance. Although, the fact that they do have experience with acquisitions in general, might be an advantage.
argumentation that companies with similar types of capitalism and business systems perform better after they engaged in an acquisition. The average ROA change of 77 CME-LME acquisitions is -3.83 which is worse than the average ROA change of -2.33. This is also in line with the argumentation that these differences between types of capitalism might have a negative influence on the post-acquisition performance. According to the literature (Taylor, 2006) the acquiring companies in CME-MME acquisitions are assumed to perform somewhere in the middle because the MME classification is somewhere in the middle on the CME-LME continuum. This turned out to be true for the remaining 34 CME-MME acquisitions. A -2.00 change in ROA is present in this group. Table 8 gives a clear overview of these figures.
Table 8: ROA changes according to type of capitalism
The USA, Great Britain, Australia and Canada are classified as being LMEs. Furthermore, France, Italy, Spain and Turkey are MMEs (Hall & Soskice, 2001). The other countries are not directly mentioned in the work of Hall and Soskice (2001) but other literature (Taylor, 2006) indicates that Israel and India are both MMEs (Taylor, 2006). Because there is not sufficient information about the type of capitalism of Argentina, Mexico and South Africa, these countries are not classified as being a CME, LME or MME and are therefore not part of the statistics above.
4.2 Multiple regression requirements
Before conducting a multiple regression analysis the data in this study has to meet several requirements or restrictions (Brewerton & Millward, 2001:149):
1. the dependent and independent variables should be of interval or ratio-level (except for the dummy variables)
2. linear relationship between the variables
3. homoscedasticity (residuals and variables should be normally distributed) 4. large sample
Except for the variables about the level of experience and type of capitalism, all variables are classified as being ratio or interval measurement level. This indicates that requirement 1 has been met. The second requirement entails the fact that there has to be, at least assumed, a linear relationship between the variables. Based on the described literature, a linear relationship between the variables is assumed.
Type of acquisition Average ROA change
CME-CME -1.76
CME-LME -3.83
As indicated by requirement 3, the variables should be normally distributed. This will be tested by the Bowman-Shelton test (Newbold et al., 2003). The hypotheses concerning this test are:
H0: the variable is normally distributed H1: the variable is not normally distributed
The test statistic represented by the Bowman-Shelton test is B and can be calculated by the following formula (Newbold et al., 2003):
B = n * ( skewness 2 + (kurtosis - 3) 2) 6 24
Where n is the number of observations.
Based on the skewness and kurtosis from table 3, this will be tested for each variable. For instance, applying this formula to the dependent variable ROA, the following calculation has to be made:
B = 303 * ( -6.14 2 + (46.90 - 3) 2) = 303* (-6.28 + 80.3) = 22417.07 6 24
With the help of table 9 (Newbold et al., 2003:565), which shows the number of observations with the associated appropriate values of B in case of several sample sizes, it is possible to draw a conclusion on whether this variable is normally distributed.
This study contains of 303 observations and a significance level of 5%, so B should be smaller than 4.6. In the case of ROA B is equal to 22417.07, which is above 4.6 and therefore H0 is rejected. To conclude, the variable ROA is not normally distributed.
Sample size N 10 % point 5 % point
200 3,48 4,43 250 3,54 4,51 300 3,68 4,6 400 3,76 4,74 500 3,91 4,82 800 4,32 5,46 >800 4,61 5,99
In case of the variable PD, the formula is the following:
B = 303* ( 2.22 2 + (4.04 - 3) 2) = 303* ( 0.82 + 0.045 ) = 262.12 6 24
B is 262.12 > 4.6, so H0 is rejected and this indicates that variable PD is not normally distributed. The same procedure has been followed for the other variables and the results are shown in table 10:
Variable ROA PD I UA M Size AC GDP growth
B-value 22417.07* 262.12* 196.36* 196.45* 474.10* 5149.81* 20.68* Table 10: Variables and B-values
* = not normally distributed
From table 10 it follows that all variables are not normally distributed. Requirement 3 is therefore not met. To overcome this problem, several sensitivity analyses have been done in order to exclude the outliers who are mainly responsible for the fact that the data is not normally distributed. The upper and lower 10 of the outliers had been excluded but this had no result. Furthermore, the outliers above and below three times the standard deviation had been excluded. This had no result either. This limitation is considered as an important limitation in this research. In the discussion part more attention will be paid to this.
As already mentioned in the methodology part, the sample of this study (303) is large enough to achieve adequate statistical power in order to infer accurate conclusions from statistical testing (Brewerton and Millward, 2001). This means that requirement 4 has been met.
4.3 Correlation analysis
Percentage change ROA 1999-2002 Difference in power distance Difference in
individualism Difference in uncertainty avoidance Difference in power distance Correlation .041 Significance* .242 Df 291 Difference in individualism Correlation -.058 -.218 Significance* .161 .000 Df 291 291 Difference in uncertainty avoidance Correlation .028 .142 .515 Significance* .315 .007 .000 Df 291 291 291 Difference in masculinity Correlation .052 .448 .062 .495 Significance* .189 .000 .145 .000 Df 291 291 291 291
* = 1-tailed Table 11: Results of partial correlation analysis
The following hypotheses are relevant in case of a correlation analysis: H0: there is a correlation coefficient of zero
H1: there is a correlation coefficient different than zero
Table 11 shows that none of the cultural dimensions are significantly correlated with the post-acquisition performance. This appears from the significance levels of .242, .161, .315 and .189 which are all higher than .05 (α = 5%). Therefore, hypothesis 1 is rejected in all cases. As a result, it can be concluded that all independent variables are not significantly correlated with the dependent variable. A correlation of 1.00 would indicate that there is a perfect correlation between the variables. The difference in power distance, difference in individualism, difference in uncertainty avoidance and difference in masculinity show a correlation of .041, -.058, .028 and .052 respectively. This indicates that there indeed is a correlation with the performance indicator. However, the low numbers indicate a very weak relationship.
It is interesting to see that only one variable, namely difference in individualism, has a negative correlation with the percentage change ROA. As the four hypotheses on page 20 of this study indicate, all cultural dimensions are assumed to be negative correlated with the percentage change ROA. However, the correlation analysis shows that only difference in individualism is negatively correlated. Furthermore, table 11 shows that all cultural dimensions, except for masculinity and individualism, are significantly correlated to each other. This might be a result of the fact that 150 out of 303 acquisitions are national ones. In these national acquisitions cultural distance is not present. As a result, 150 out of 303 acquisitions have a cultural distance of zero on all four cultural dimensions.
The multiple regression analysis is done in order to test the hypotheses stated on page 21 and to answer the research question of this study. In general, a regression analysis is used to estimate a linear relationship between a dependent variable and one or more independent variables (Brewerton & Millward, 2001). Here, the dependent variable is Y. It is the percentage change ROA of the acquiring company between 1999 and 2002. The independent variables are X1up till X4 and the other variables are control and/or dummy variables.
Table 12 shows the results of the multiple regression analysis. It consists of three different parts. Firstly, 12a shows the overall results of the regression analysis. This model summary shows an R² of .052. The R² is the measure of how well the independent variables predict the dependent variable (Allison, 1999). This means that all ten independent variables only predict 5% of the variable Y. This small R² means not that the model is not valid or even worthless. It also does not mean that there is no relationship. It just indicates that there is no linear relationship, which is one of the requirements for using multiple regression analysis (Huizingh, 2005). “Although it is certainly true that a higher R² is better, there is no reason to reject a model if the R² is small. It is quite possible that a model with a high R² could be a terrible model, for instance when wrong variables are included” (Allison, 1999:31).
12a. Model Summary
12b. ANOVA
12c. Coefficient
Model Understandardized
Coefficients StandardizedCoefficients
t Sig.*
B Std. Error Beta
1 (Constant -1.953 .961 -2.033 .022
Difference in power distance .001 .078 .001 -.235 .497
Difference in individualism -.113 .090 -.153 -1.248 .107 Difference in uncertainty avoidance .055 .072 .087 .765 .223
Difference in masculinity .014 .040 .026 .346 .365
General international experience 8.35 7.69 .066 1.085 .139
Local experience -2.254 2.12 -.065 -1.063 .145
Experience with int. acquisitions -3.031 1.321 -.142 -2.294 .023
Size acquiring company 8.28E-006 .000 .080 1.345 .011
GDP growth TC -8.388 9.596 -.076 -.874 .196
Type of capitalism -.316 1.221 -.029 -.259 .398
* = 1-tailed Table 12: Results of multiple regression analysis Model R R Square Adjusted R
Square
St. Error of the Estimate
1 .227 .052 .019 7.27573
Model Sum of
Squares df Mean Square F Sig. 1 Regression 829.959 10 82.996 1.568 .116
Secondly, table 12b consists of the results of the Analysis of Variance (ANOVA). Here, the following two hypotheses are used:
H0: all regression coefficients, except for the intercept, are zero H1: at least one regression coefficient is not zero
Table 12b shows a p-value of 0.116, which is higher than 0.05. Therefore, hypothesis H1 is rejected and hypothesis H0, which states that all regression coefficients are zero, is accepted.
The third part of table 12 consists of a description of the estimated equation of the regression. The first column, under B, shows the regression coefficients. The regression equation is therefore:
Y = -1.953 + .001 X1- .113 X2 + .055 X3 + .014 X4 + 8.35 X5- 2.254 X6 - 3.031 X7 + 8.28E-006 X8 -8.388 X9-.316 X10 + ri
When taking a closer look at the regression equation and the related regression coefficients, only B2 is negative. According to the hypotheses on page 21, all relationships between the ROA growth and the cultural dimension were assumed to be negative. Table 12c shows that only hypothesis two has a negative relation with the dependent variable Y which is:
H2: A higher difference in ‘individualism between companies will lead to a lower post- acquisition performance
post-acquisition performance. Another possible explanation for these contradictory findings may be due to the fact that the data is perhaps biased because half of the acquisitions are national (German-German) ones or that the other studies used different measurement instrument of performance or cultural distance.
One should be aware of the fact that not all variables are of the same measurement level. Therefore, the regression coefficients do not indicate their relative value. In order to compare the regression coefficients in a better way, the coefficients are calculated under the heading Standardized Coefficients - Beta. These figures indicate the relative value of the different variables.
The T-test is used to test the four most important hypotheses of this study. The results of this test can be found in table 12c. As mentioned several times, the hypotheses are:
H1: A higher difference in ‘power distance’ between companies will lead to a lower post- acquisition performance
H2: A higher difference in ‘individualism between companies will lead to a lower post- acquisition performance
H3: A higher difference in ‘uncertainty avoidance’ between companies will lead to a lower post-acquisition performance
H4: A higher difference in ‘masculinity’ between companies will lead to a lower post- acquisition performance
The first hypothesis is tested in the following way:
H0: the regression coefficient of X1(difference in power distance) is zero
H1: the regression coefficient of X1(difference in power distance) is less than zero
Table 12c shows a p-value of .497. As hypothesis H1 makes clear, a one-sided T-test is used. It has to be mentioned, that the program SPSS only tests 2-tailed. Therefore, all p-values in table 12c are divided by two. Based on a p-value of .497, which is higher than .05 (α = .05), hypothesis H0 is accepted. Therefore, the first hypothesis that states that a higher difference in power distance between companies will lead to a lower post-acquisition performance does not significantly hold.
The same procedure is applied to hypothesis 2:
H0: the regression coefficient of X2 (difference in individualism) is zero
Table 12c shows a p-value of .107. Based on a p-value of .107, which is higher than .05 (α = .05), hypothesis H0 is accepted. Therefore, the second hypothesis that states that a higher difference in individualism between companies will lead to a lower post-acquisition performance does not significantly hold.
This same procedure is followed for hypotheses three and four. Here, table 12c shows p-values of .223 and .365 respectively. Hypotheses three and four are rejected because both .223 and .365 are higher than .05. Furthermore, these hypotheses are, together with the first hypothesis, positive. As a result, the alternative hypothesis would not have been accepted at all. Even if the p-values were significant. As a result, it can be concluded that hypothesis three, which states that a higher difference in uncertainty avoidance will lead to a lower post-acquisition performance, is not significantly true. Finally, hypothesis four, which states that a higher difference in masculinity will lead to a higher post-acquisition performance, is also not significantly true.
Another remarkable finding is the fact that experience wit international acquisitions turned out to be negatively related to post-acquisition performance. Furthermore, it is interesting to see that this relationship is almost significant. This negative relationship is due to the fact that ‘experience with international acquisitions’ is marked with a zero (0) and ‘no experience with international acquisitions is marked with a one (1). This is in line to what is expected according to the literature that says that companies with former experience in international acquisitions can use this learning advantage in future acquisition activities.
Table 13 is a summary of the multiple regression coefficients and p-values of all four hypotheses. As can be seen, all four hypotheses in this study are rejected. This indicates that cultural differences on all four dimensions of Hofstede (2001) have no significant influence on the performance of German acquiring companies after an acquisition.
Hypothesis Regression coefficient Multiple regression
P-value Accepted/Rejected
H1 power distance .001 .497 Rejected
H2 individualism -.113 .107 Rejected
H3 uncertainty avoidance .055 .223 Rejected
H4 masculinity .014 .365 Rejected
Table 13: Overall multiple regression results of the hypotheses
