1
Student: Ludo Hoenderop
Student Number: 10259732
Faculty: Economics and Business
Program: Corporate Finance
Supervisor: R. Almeida da Matta
The role of CEO overconfidence and cultural differences – Cross-border M&A investigation.
Abstract
This thesis investigates the relationship among the CEO overconfidence of the acquirer and the cultural differences between the acquirer and target during and after takeover activity. A sample of 3,331 U.S cross-border mergers and acquisitions in the period 2000-2016 is used. The
results concludes that the interaction effect of the overconfidence CEO and cultural differences is dimension specific and does not depend on the total cultural differences variable interacted.
Further, shows when CEOs are overconfident and decide to acquirer a foreign target with cultural differences, overconfidence combined with differences in cultural dimensions between
2
Table of Content
1. Introduction……… 3
2. Literature Review……… 5
2.1 Motives for Cross-Border M&A………... 5
2.2. Influence of Cultural Differences in the M&A Market……… 6
2.3. Influence of CEO overconfidence in the M&A Market……… 7
2.4……… 9
3. Hypotheses………10
4. Data and Methodology……….. 11
4.1. Data Collection……… 11
4.2. Event Study………12
4.2.1. Endogeneity………. 13
4.3. CEO Overconfidence……… 14
4.4. Cultural Country Variables……… 14
4.5. Firm- and deal-level characteristics………. 16
4.6. Country Level Variables ……….. 16
4.7 Gravity Model M&A……….. 17
4.8. Tobin’s Q……… 18
5. Results……….. 19
5.1 Descriptive Statistics……… 20
5.2 Results of CARs and BHARs without interaction effect.………. 22
5.3 Results of CARs and BHAR s with interaction effect……… 25
5.4 Results of Gravity Model………... 27
5.5. Results of Tobin’s Q… ……….. 28
6. Robustness……….. 29
6.1 Robustness check on the short-and long-term performance.……….. 30
6.2 Robustness check on Tobin’s Q. ……….. 30
7. Conclusion………... 31 References ……… Appendix………....
3
1. Introduction
U.S. firms spend more than $3.6 trillion on mergers and acquisitions through the last two decades. Global markets are integrated and cross-border M&A is common in the markets. Cross-border mergers could in fact promise greater value than domestic mergers because they include a larger pool of potential merger and acquisition partners, which would allow for greater
potential synergies. In addition, cross-border mergers might generate larger values since they offer greater growth potential in new markets (Ahern, Daminelli and Frascassi, 2015;
Ahammad, Tarba & Glaister,. 2016). However there are still differences between the target and the acquiring nations in terms of country specific (i.e. cultural distance, religion, language) and intangible factors (i.e. CEO confidence). Firms engaging in mergers and acquisitions repeatedly claim that cultural differences are major motive why takeovers have a negative influence on the firm value. A recent survey of more than 800 executives by McGee, Thomas, and Thomson (2015) cites different cultures and difficulty of integrating product lines as partly being guilty for worse ex-post-merger results and a smaller chance of realizing merger synergies. Further, cultural differences between countries are the most common used variable in latest cross-border M&A studies (Eral et al., 2012, Chakrabarti et al., 2009, Han et al., 2016, Lai et al)
Studies have explicitly indicated that cross-cultural difference effect the value on mergers
and negatively and positively M&A. Empirical evidence on the returns of acquirers in the cross-border M&A market has largely focused on the stock performance event studies (Ahern et al 2015, Erel et al 2012, Han et al 2016). In the papers of Lim, Anil & Shenkar (2016); Dutta et al. (2016); Ahern et al. (2015) cultural differences between target and acquirer has a negative effect on the firm performance. Chakrabrati et al (2009) found a positive effect of cultural distance on the long run performance of the cross border mergers. Additional, optimism of a CEO effects the financial intermediation and can affect the corporate financial value and accounting decisions. Optimism, such as CEO overconfidence, plays a critical role in economic decision-making (Puri and Robinson, 2007). In the papers of Malmendier and Tate (2005) and (2008) is argued that CEO overconfidence leads to overinvestment and decreases firms value after the investment. However in their papers they are unable to investigate how cultural distance between the acquirer and target in cross border M&A affects the decision of theoverconfident CEO. Diverging cultures could lead to differences among the companies and could affect the decisions of CEO in possible mergers or acquisitions. In the research of Ferris et al (2013) CEO overconfidence has a negative and significant influence on the decision of CEOs to acquire an unrelated target. Moreover, certain cultural dimensions of Hofstede’s framework, such as individualism and long-term orientation, are even correlated with the overconfidence behavior of a CEO. (Ferris et al., 2013). Though, the cross-border M&A research has not
4 accounted for the effect on firm performance that cultural environments of bidder and target countries have related to the overconfidence of an acquirer CEO.
The research will study the relationship among the CEO overconfidence of the acquirer and
the cultural dimensions differences between the acquirer and target during takeover activity. Further test if CEO and cultural characteristics interacted, affect firm value and stockperformance. Answering these questions is important for international orientated CEOs facing pressure to manage effectively in a globalized financial business environment. The research question of this study is therefore: What are the effectsof CEOs overconfidence concerning acquirers and targets with cultural differences, looking to firm value and stock performance in the U.S. cross-border merger and acquisition market?
This thesis uses a sample of 3,331 U.S cross-border mergers and acquisitions between 2000
and 2016. OLS regressions will be done to find a relationship between CEO overconfidence and the dimensions of cultural differences, looking to the stock performance and the Tobin’s Q. Further a Tobit regression is performed to find the effect on the firm value. The results shows when a CEO is overconfident and decide to acquirer a foreign target with cultural differences (compared to the U.S), overconfidence combined with differences in cultural dimensions between the acquirer and target has a negative effect on the stock performance and firm value. Additional result indicates larger differences in the cultural dimensions and overconfidence of a CEO reduces the deal value on the cross-border merger and acquisitions.In this study, contribute to the growing but still immature literature establishing the
importance of human and cultural characteristics in understanding of corporate decision-making. Moreover, varies from previous cultural differences in the cross-border M&Aresearches by analyzing cultural distance with the cultural difference measure of House et al. (2004), instead of using Hofstede’s (2001) dimensions and country cultures. Hence I contribute on using a new measure of cultural difference in the cross-border market. Lastly, previous research has not yet been focusing on the interaction between overconfidence and cultural distance during cross-border mergers and acquisitions. The purpose of this study is to examine how cultural distance between acquirers (U.S. firms) and targets (cross-border) affects the benefits of acquirers in cross-border mergers and acquisitions. Investigate the impact of cultural differences on the long-run performance of U.S bidder firms that possess intangibles such as CEO overconfidence. Since little is yet known about the increasing effect of a firm’s national cultural background in the relationship among overconfidence and a firm’s performance after a merger or acquisition.
This paper is structured in the following manner. First, relevant literature is discussed and
compared. Second, hypotheses are discussed. Thereafter I elaborate on the sources and treatment of the data. Subsequently, all measures are discussed followed by the summary5 statistics. Then the regressions and their results will be presented. Finally, the robustness checks and the conclusion are presented.
2. Literature Review
2.1. Motives for Cross-border Mergers and Acquisitions
Overall belief is that combined firms will be more profitable than the individual firms. The positive change in firm value can occur for an amount of motives. From a strategic point of view, the most important reason to participate in cross-border M&A is the quick entrance into one or several new foreign markets for their products and services (Stahl and Voigt, 2008; Chakrabarti et al., 2009). This is generally relevant for large listed U.S companies because entering into new foreign markets, particularly those of their competitors, increase the opportunity to improve their market share or seize new market opportunities (Rossi and Volpin, 2004). An investment abroad can be done through Greenfield investments and joint ventures, however a merger or an acquisition is quicker to implement than a Greenfield entry and offers more control over the company as compared to joint ventures (Bauer et al., 2014; Hopkins, 2008; Stahl and Voigt, 2008). In terms of border takeovers, the paper of Lim & Lee (2016) show that a cross-border acquisition deal is more likely to succeed when the degree of relatedness between the target and business is high.
Economic motives appear to be revenue enhancement, succeeding economies of scale, risk spreading, cost reduction, increase bargaining position or responding to market failures abroad (Erel et al, 2012). Another economic reason to involve in cross-border motivations founded by Lim et al. (2016) is market development for cross-border M&A. Whereby market development is described as the growth opportunities in a country abroad. The acquirer firms are mostly related to the low growing developed markets, while the targets in the emerging markets are related to the fast growing opportunities markets.
Coordination and communication amongst the merged entities become more challenging when there are differences between acquiring and target companies culture and nations. Could result in growing cost and effort and for integration. The employees of the merging companies are entrenched in their national cultures, cross-border takeovers could lead to
misunderstandings in decision-making and difficulties during the implementation phase (Bauer et al, 2016). The theory of cultural differences might seem narrow relative to a more classic view of integration risk. Integration risk is frequently connected to employees leaving the firm due to difficult working environments and problems working with different firm cultures. The more ex-ante difficulty of integrating amongst 2 firms, the more the employees of the two firms will have to work together (Hoberg and Philips, 2016). Conquer the problem of ex-ante
6 difficulty and this can results in synergies cost reductions and revenue enhancements. The limitations can be exploited by internalizing the target’s specific intangible assets such as valuable personnel or knowledge skills of the product (Lopez et al. 2010, Reus and Lamont, 2009). Cost reductions can achieved be due to economies of scale or a positive difference in the exchange rate of the target and the acquirer company. The last reason might boost acquisitions, because an acquisition is cheaper when the currency of the acquirer is strong compared to the particular foreign currency (Hopkins, 2008; Stahl and Voigt, 2008)). Revenue enhancements and profits arise when the value of the two firms together is larger than to the sum of the individual’s firm values (Chakrabarti et al., 2009; Morrosini et al, 1998).
2.2. Influence of cultural differences in the M&A market.
Culture is an important description and researchers used different definitions for culture in their papers. Culture is defined as a way a group of societies think, behave, feel and act. Countries have their own cultural identities; people speak different languages, have different religions and different norms and values. Analysis of the cultural differences among the
acquiring firms and the target firms have become the main focus of research (Rossi and Volpin, 2004; Chakrabarti et al, 2009; Ahern et al, 2015). In the research of Hofstede (1980) culture is defined as the collective programming of the mind, separating the people of one group from another. In the paper is discussed that differences in cultural values are a crucial determinant of human organization and behavior. Many researchers suggested that culture is indeed an
element of human behavior and suggested that culture has an effect on economic-decision making (Ahern, Daminelli and Frascassi, 2015; Rossi and Volpin, 2004, House et al., 2004). For example, in a cross-border acquisition the successfully integration of an acquired management team requires careful attention. Firms that fail to recognize the cultural differences can make costly mistakes (Morossini et al. 1998). Further, physical distance can increase the cost or benefits of combining those two. Both cultural differences and geographic distance could
decrease or increase the value of the takeover. In addition, corporate governance considerations can influence the cross-border merger value. (Ahern, Daminelli and Frascassi, 2012; Erel et al 2012). Between the relationship of culture distance and merger and acquisitions value there are two points of views.
Cultural differences between the acquirer and target could be a potential obstacle to achieving integration benefits. Results in cultural differences impose additional cost of the merged or acquired firms, decreasing the synergies and therefore leads to a decrease of the net present value of the deal (Erel et al., 2012). Further, in the paper of Ahern (2015) there are cultural and governance differences between developed and emerging markets and those differences have impact on the firm value. The research observes international mergers and
7 concludes that countries levels of cultural difference in trust and individualism results in lower merger volumes and lower combined abnormal announcement returns. In the research of Antia et al. (2007) cultural differences decreases firm value by imposing a barrier to internalization advantages. The cultural differences have a negative effect in the firm valuation based on Tobin’s Q. Stahl and Voigt (2008) results found cultural differences affect shareholder of acquired firms, since the cultural risk affect the post-performance process. In the research of Conn et al. (2005) cross-border acquisitions results in significantly positive short-term announcement returns. However, the long-term returns are smaller when cultural differences are greater between the U.K. and foreign targets. In addition, Hofstede’s (1980) variables of cultural differences are used to measure and proxy cultural differences. From the research of Conn et al. (2005) is concluded, cultural differences between firms and the acquirers combined with long-term performance are negatively related.
In contrary, in the research of Chakrabarti et. Al (2009) cross-border acquisitions perform better during long-term stock performance if the acquirer and the target come from countries that are culturally more different. Entering cultural different nations could provide competitive advantage by access to unique experiences and capabilities, great learning opportunities and knowledge from the target company not available before (Chakrabarti et al., 2009). Further, in the paper of Han et al. (2016) a measure of trade networks has a positive effect on
announcement returns after controlling for Hofstede’s cultural distance between acquiring and target nations and various firm and deal-specific factor. Steigner and Sutton (2005) investigated how national culture affects the benefits in foreign takeovers. Their evidence discusses that bidders with large intangible assets, such as technological knowledge, create significant profits in culturally distant firms. The current research literature has accepted Hofstede’s cultural dimensions (Ahern et al, 2015; Chakrabarti et al. 2009; Eres et al, 2012; Voight and Stahl, 2008, Reus and Lamont 2009) as a meaningful proxy to measure differences in national culture.
2.3. Influence of CEO overconfidence in the M&A market
The psychology literatures assumes overconfidence as the propensity of CEO’s to think of themselves as ‘above average’ on characteristics like capability, judgment and predictions for successfully outcomes (Alicke & Goverun, 2005). Could simply speak of CEO overconfidence when CEOs are positively biased in the self-assessment of their capabilities and success
probabilities (Breuer et al, 2016). The more a manager climbs on the corporate ladder, the more this manager will experience and create self-serving biases. This ‘above average’ feeling will likely appear by top executives, since executives overall are better skilled compared to other managers. The executives rate him to be superior against other employees. The papers of Malmendier and Tate (2005) and (2008) classify CEOs as overconfident when they insistently
8 expose themselves to risk, specifying that overconfident CEOs have different risk sensitivities. Furthermore, gender effects play a significant role in overconfidence. The research of Barber and Odean (2001) investigate the common stock investment behavior between male and females. The results of this paper presents that a male trade 45% more excessively than female, however suffer worse stock performance.
Overconfidence CEO interacts with a self-serving bias, where CEOs tend to link positive outcomes to their own actions of expertise and negative outcomes to bad luck or external factors (Malmendier and Tate, 2008). CEOs with extreme confidence in their own abilities have a tendency to overestimate the firm’s resources and outcomes under their control. The hubris hypothesis of Roll (1986) arguing that CEOs make relatively few mergers over their careers and are unable to learn from their past mistakes. Roll (1986) argues that acquiring companies perhaps pay too much for the target company due to CEOs suffering from the hubris hypothesis. The CEOs suffering from hubris, value target companies higher than the market values of the target company. The influence of past performance on CEO optimism resulting in an
overpayment. The results of the paper of Doukas and Petzemas (2007) show that overconfident and overoptimistic CEOs complete more mergers and acquisitions. In the research of
Malmendier and Tate (2008) when a CEO is overconfident compared to another CEO, the CEO is more likely to pursue a diversifying merger. In the paper of Malmendier and Tate (2008) is argued that managerial overconfidence leads to overinvestment and destroys firms value, the results shows CEOs overconfidence has a negative and significant influence on the short-term stock performance around merger and acquisition activity.
Overconfidence plays a role by encouraging CEO risk-taking behavior. CEO overconfidence should increase the possibility of making cross border acquisition investments. Because cross-border acquisitions decisions are routinized over time due to learning from previous M&A activity. The overconfident CEO could attribute the success of previous cross-border
acquisitions to his talent and managerial capabilities (Dutta et al., 2014). In most the cases CEO makes the final decision for the firm and they are believed to control the firm. Such a position may make them to believe they can control the outcome (Hirshleifer et al., 2012). This problem frequently occurs with rare large-scale complex corporate decisions (Malmendier and Tate, 2005).
Takeovers have a direct impact on the personal wealth of the CEO. The CEOs appetite for risk (confident or not) may play an important role in the takeover decision (Frijns et al. 2013). CEO’s engage in takeovers, such as cross-border mergers, when this is optimal from the firm’s point of view. However this decision is also affected by the previous personal traits between the
acquirer and target such as the overconfidence of CEOs and cultural differences. Cultural distance between the acquirer and target interacted with CEO overconfidence and his
9 investment decision, could be a potential obstacle to achieve profits after a takeover between culturally different companies.
2.4. Cultural Influences on the overconfident CEO in cross-border M&A decisions.
Culture is part of the CEO daily lives, even though culture is not tangible and hard to determine. Cultural models are based on the idea that the culture of the target company is a critical
determinant of the integration process. Cultural differences play a tremendous role in the M&A cross-border process. The link between the ex-ante merger differences and the post-merger integration outcomes is of great importance (Stahl and Voigt, 2004). Prior researches1 argued that the psychological processes and cultural processes are intertwined and shape among each other. The literature has identified a link to overconfidence CEOs when hubristic CEOs tend to pursue difficult tasks, because CEOs believe themselves too be much better than other
employees (Picone et al., 2014). Takeovers have a direct impact on the personal wealth of the CEO. The CEOs appetite for risk may play an important role in the takeover decision (Frijns et al., 2013). Further, cross-border investment is essentially riskier compared to domestic investment, since there are greater levels of information asymmetry that varies from those of their home country. (Johanson & Vahlne, 2009). There is a possibility that cultural differences could be measured as a risk, has influence on the overconfidence CEO, because managerial overconfidence is shaped partly in their culture. In the research of Ferris et al. (2013) is investigated that certain dimensions of the Hofstede’s cultural dimensions are correlated with the CEO overconfidence. The results show that overconfidence influences the number of acquisitions made and there is a significant cultural influence on the relationship between overconfidence and takeover activity. Breuer and Salzmann (2016) lengthen the findings of Ferris et al. (2013) by regressing the significant dimensions individualism, long-term orientation and uncertainty avoidance on firm’s long-term performance. The negative
relationship between individualism and the merger outcomes challenges the literature. It could be that CEOs in individualistic cultures think more about their own abilities while CEOs in a more collective culture behave more to the ruling norms.
From literature stated above, large cultural differences could lead to greater inconsistency in value between the acquirer and target companies. Challenges in successfully establishing relationships with various local shareholders, thereby obstructing the understanding of the potential benefits from a cross-border M&A (Lopez et al. 2010). To address this problem ensure
1 (I.e. Roll 1986, Alicke & Goverun, 2005; Malmendier and Tate, 2005 & 2008, Hirshleifer, et al., 2012)
10 that host partners will cooperate since business is more efficient with help from local partners, particularly in countries with cultural differences (Lai et al 2016). The results in the paper of Lai et al. (2016) shows that CEOs toward overconfidence increases the propensity of an takeover and their relationship is more pronounced when firms are exposed to greater information asymmetry or environmental uncertainty. This in terms of greater cultural differences, higher host-country risks, and inexperience in the local market. The paper of Reus & Lamont (2009) argues to develop integration abilities to successfully create cross-cultural synergy. CEOs, who highly believe in their own ability to successfully integrate distinct cultures, may choose an acquisition to increase the profits available from investments in countries with highly cultural differences. Research has not yet been focusing on the interaction between overconfidence and cultural distance during cross-border mergers. The purpose of this study is to examine how cultural distance between acquirers and targets affects the benefits of acquirers in cross-border mergers and acquisitions. Investigate the impact of cultural differences on the long-run
performance of U.S bidder firm that possess intangibles such as CEO confidence. Since little is yet known about the increasing effect of a firm’s national cultural background on the variance in the relationship between overconfidence and a firm’s performance after a merger or
acquisition.
3. Hypotheses
In the literature review is argued that CEO overconfidence leads to overinvestment, decreases firm’s value and has a negative and significant influence on the short-term stock performance around merger and acquisition activity. I expect that CEO overconfidence has a negative effect on the firm’s short-term and long-term stock performance of the acquirer. Further, there is concluded that cultural differences have both a positive and negative effect on cross-border acquisitions. I expect a negative effect since there are greater levels of information asymmetry that varies from those of their home country (U.S). According to previous theoretical researches larger cultural distance can delay foreign takeovers by increasing the costs of integration
Hypothesis 1: CEO overconfidence and the cultural dimensions have a negative effect on the short-term stock abnormal returns and the long-term stock abnormal returns.
The greater the differences between the two countries along the cultural dimensions, the lower the combined abnormal returns are in a merger or acquisition in the short and long run.
Expected that hypothesis 1 confirms the importance of national culture and CEO overconfidence during and after an acquisition. Therefore, the effect of overconfidence cannot be tested in isolation, because other variables, such as cultural distance, might moderate the relationship
11 between overconfidence and performance. For the second hypothesis I will test the effect of cultural distance interacted with CEO confidence.
Hypothesis 2: CEO overconfidence interacted with the nine cultural dimensions separately has a negative effect on the short-term stock performance and the long-term stock performance. Further, investigate if cultural differences and CEO confidence has influence on the effect on the deal value of cross-border mergers and acquisitions. I expect that cross-border mergers and acquisitions create value overall, but then again cultural differences and CEO overconfidence decreases a part of the firm value.
Hypothesis 3: Cultural distance and CEO overconfidence has a negative effect on the value created in M&A activity.
In the research of Antia et al. (2007) cultural differences decreases firm value by imposing a barrier to internalization advantages. The cultural differences have a negative effect in the firm valuation based on Tobin’s Q.
Hypothesis 4: CEO overconfidence interacted with the nine cultural dimensions separately has a negative effect on the Tobin’s Q of the acquirer (U.S. firms)
Overall, there is expected that differences in the culture between the acquirer and the target and CEO overconfidence has a negative effect on the short-term and long-term stock performance, the deal value and the Tobin’s Q.
4. Data and Methodology
4.1. Data collection
Sample consists of 3,331 cross-border M&A deals announced and completed during the period 2000 to 2016. The data is obtained from the Securities Data Company (SDC) Platinum M&A
Database. Using the following criteria; (i) the deals are completed, (ii) the acquirer owns 51-
100% of the shares of the target after the transaction, (iii) deals are recognized as cross-border deals by the SDC cross-border flag, (iv) the acquirer is traded publicly, (v) the United States is not the target nation, and (vi) the target have trade volume data. (vii) the transaction values for the cross-border deals must be greater than $10 million, which eliminates small acquisitions without significant economic implications from our data set. Further, transaction and firm-level information was collected from the Thomson One (SDC) database. Such as, friendly versus hostile, payment method, if the cross-border deal is industry related, use of poison pill and if the
12 deal is a tender offer. Comparable to the papers of Malmendier and Tate (2005); Ferris et al (2013); Ahern et al (2015) Lim et al. (2016), Dutta et al. (2016).
Moreover, the sample firms must have return data available from the Center for Research in
Security Prices (CRSP) Database and financial data available from the Compustat database. To
measure CEOs overconfidence, using data of the executives options holding from the Execucomp
database. Cultural differences provided by the GLOBE database as a primary cultural distance
measure (House et al., 2004). Information of the deal level is gained from the SDC Thomson One database. The World Data Bank provides the country level characteristics. Tobin’s q is available in WDRS from database Peters and Taylor’s total Q.
4.2. Event Study
In order to measure the short-term takeover performance look to the effects of the M&A deal on the announcement returns for the acquiring company (U.S.) by means of an event study. The event period will the cover up until the days after the announcement date. The market adjusted return model is used as a benchmark to calculate the return and the abnormal return of the stock prices.
𝑅𝑒𝑡𝑢𝑟𝑛𝑖𝑡= 𝑅𝑚𝑡+ 𝜀𝑖𝑡
The expected return is calculated over the period with event windows of
CARs [−5, +5], CARs [ −3, +3], CARs [−1, +1] and then subtracted from the actual returns. This to calculate the abnormal return of a specific firm in the event of an acquisition or merger.
𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑅𝑒𝑡𝑢𝑟𝑛𝑖𝑡 = 𝑅𝑖𝑡− (𝑅𝑚𝑡+ 𝜀𝑖𝑡)
Where 𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑅𝑒𝑡𝑢𝑟𝑛𝑖𝑡 indicates the abnormal returns for the stock of “𝑖” on day “𝑡”. 𝑅𝑖𝑡 is the return of the stock of “𝑖”” on day “𝑡”. (𝑅𝑚𝑡+ 𝜀𝑖𝑡) is the benchmark return for of stock “𝑡”” on day “𝑡”. From the previous steps the cumulative abnormal returns of a specific firm can be calculated by the following formula. The market return is based on the CRSP market return of the S&P composite index.
𝐶𝐴𝑅𝑖(𝑇1, 𝑇2) = ∑ 𝐴𝑅𝑖,𝑡 1 𝑡=𝑡1
Where 𝐶𝐴𝑅𝑖(𝑇1, 𝑇2) denotes the cumulative abnormal returns over the event windows of 𝑇1, 𝑇2 where 𝑇1= 0 specifies the day of the announcement. ∑1𝑡=𝑡1𝐴𝑅𝑖,𝑡 indicates the sum of the abnormal returns over the event window. The abnormal returns capture the effect of the announcement of the returns received by the company’s stock.
13 method is well used in the literature and the preferred method for long-term return
investigation as it precisely measures investor experience, resulting from the purchase of the respective stock. Examine the 36-month post-merger buy-and-hold returns for the acquirers following Chakrabarti et al. (2009); Steinger and Sutton (2013). Daily returns are compounded in order to calculate the buy-and hold returns for each specific company over the window (12, 24, 36 months). The holding period for stock 𝑖 is 𝑇2, where 𝑇2 is three years or the time until the company is delisted.
𝐵𝐻𝐴𝑅𝑖(1,36)= ∏(1 + 𝑅𝑖𝑡) − 𝑇2 𝑡=1 ∏(1 + 𝑅𝑀𝑡) 𝑇2 𝑡=1
Calculate the return on the acquiring firm’s stock and the market return for the country of the acquirer during this period. For the returns use the closing prices to calculate stock returns. The market return is based on the CRSP market return of the S&P composite index. The difference between the two returns is the BHAR for the acquiring company.
4.2.1 Endogeneity
In the sample, there could be occurrences of endogeneity in the regression. The decision of the company to engage in a cross-border takeover is one itself has fully influence on and could cause self-selection between companies that could potentially engage in merger or acquisition at any given point in time. As an example, Meyers and Malouf (1984) present results from firms that are more likely to participate in a cross-border M&A when their stock is overvalued. A greater market to book ratio or market capitalization could be variables affect for M&A. Statistically would imply that in the regression market capitalization would be correlated with the error term and results in an OLS estimation of market capitalization will be biased and imprecise. In this paper omit the variable market capitalization in the OLS regressions of event studies.
Aimed at another endogeneity problem of survivor bias refers to the exclusion of firms because long-term stock performance does not longer exist 3 years after the announcement data. The paper of Morris (2012) excludes the non-survivors to estimate the abnormal returns for the survivors. The research provides evidence that the exclusion of the non-survivors would not bias the results. Non-survival may not be necessarily an indication of failure. For instance, it is possible that the firms drop out of the sample, not because the firms have failed after the merger or acquisition, but because the firms have been extreme successfully and were acquired by another firm (Morris, 2012). Aimed at my sample, do not exclude firms with stock data available less than 3 years from the regressions, this to avoid survivor bias.
14
4.3. CEO Overconfidence
In short, risk-averse CEOs do not hold their options until expiration; the longer the CEO holds their options, the more confident the CEO is in practice. The overconfidence variable is measured based on the papers of Malmendier and Tate (2005); Malmendier and Tate (2008); Ferris et al (2013) and Campbell and Johnson (2011). The options based measure for
overconfidence is created on the premise that it is optimal for risk-averse, undiversified executives to exercise their own-firm stock option early if the option is in the money. Then define the CEO overconfident as an indicator that equals one if the confidence measure is at least 67%.
First, calculate the average moneyness of the CEOs option portfolio for each year. For each CEO year, calculate the average realizable value per option by dividing the total realizable value of the options by the number of options held by the CEO. Strike price is calculated as the fiscal year end stock price minus the average realizable value. Then calculate the average moneyness of the options as the stock price divided by the estimated strike price minus one. Only interested in options that the CEO can exercise, include only the vested options held by the CEO.
𝑂𝑣𝑒𝑟𝐶𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 𝐶𝐸𝑂 (𝑜𝑝𝑡𝑖𝑜𝑛𝑠) =
𝑟𝑒𝑎𝑙𝑖𝑧𝑎𝑏𝑙𝑒 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑜𝑝𝑡𝑖𝑜𝑛𝑠 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑜𝑝𝑡𝑖𝑜𝑛𝑠 ℎ𝑒𝑙𝑑 𝑏𝑦 𝐶𝐸𝑂
𝑆𝑡𝑜𝑐𝑘 𝑝𝑟𝑖𝑐𝑒 𝑒𝑛𝑑 𝑜𝑓 𝑓𝑖𝑠𝑐𝑎𝑙 𝑦𝑒𝑎𝑟 − (𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑜𝑝𝑡𝑖𝑜𝑛𝑠 ℎ𝑒𝑙𝑑 𝑏𝑦 𝐶𝐸𝑂𝑟𝑒𝑎𝑙𝑖𝑧𝑎𝑏𝑙𝑒 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑜𝑝𝑡𝑖𝑜𝑛𝑠 ) − 1
4.4. Cultural Country Variables
There exist cultural country-level variables that are expected to affect the success of a foreign takeover. One of them is the cultural difference between the acquiring and target country. This variable is particularly important in the case of merger and acquisitions due to the difficulty of integrating already existing foreign management (Kogut and Singh, 1988). It is essential to note that cultural dimensions are society level dimensions. For each takeover we identify the country in which the target firm is headquartered and apply the corresponding cultural values and measure the difference with the acquiring firm (Breuer, 2013). The leading theory behind cultural distance is Hofstede’s six dimensions (Hofstede, 1980; Hofstede, 2001). Quantifying into six dimensions: power distance, individualism versus collectivism, masculinity versus femininity, uncertainty avoidance, long term versus short-term orientation and indulgence versus restraint. Rather than using Hofstede’s dimensions and country culture scores, I use more current data from the GLOBE project with 9 dimensions for cultural differences (House et al., 2004). This variable for cultural differences provided by the GLOBE project as a primary cultural distance measure (House et al., 2004). The reason for this choice is that the GLOBE
15 project has moved beyond Hofstede’s approach, and has designed constructs and scales that are more comprehensive, are cross-culturally developed, are theoretically sound, and have been empirically verified (House et al. 2004; Lim et al., 2016). The GLOBE study is primarily designed to explore the relationship between the cultural values/differences and practices and leadership effectiveness. Variables are relevant for the takeover decision because it captures leadership relevant cultural dimensions of the CEO. The House nine dimensions are the following:
- Assertiveness (dominance in relationships)
- Individualism (focus on individual or collective achievement and interpersonal relationships)
- Institutional Collectivism (desire for institutional based collectivism or nationalism) - Future Orientation (propensity to make future orientated decisions)
- Gender (minimize gender discrimination)
- Performance Orientation (the desire for continued and better performance among companies)
- Human Orientation (sympathetic behavior towards others)
- Power Distance (learning to accept uneven distribution or power in a society) - Uncertainty Avoidance (level of tolerance events and tendency to follow laid down
procedures for uncertain events within the society)
Based on the paper of Antia et al (2007) for each firm 𝑖, nine CD measures are calculated, one for each of House’s nine cultural dimensions. The formula is used to present the difference in cultural in each dimension from the acquirer to the target. Cultural distance is measured by summing over all countries weighted absolute differences between the U.S. and the country dimension score, as follows:
𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠(𝐶𝐷𝑖) = ∑ Wj|𝐷𝑖,𝑗− 𝐷𝑖𝑈𝑆𝐴 | 𝑗
where 𝐷𝑖,𝑗 is the score of one of the cultural dimensions in country 𝑗. The absolute difference of each country’s cultural dimension from that of the U.S is multiplied by the weight Wj. To calculate our primary and alternative measures of total cultural differences between acquirer and target, apply the widely used Kogut and Singh formula (1988). The following formula is build; 𝐶𝐷 𝑇𝑜𝑡𝑎𝑙𝐽= ∑( 9 𝑖=1 √(𝐼𝑖𝑗− 𝐼𝑖𝑢) 2 9 )
16 Where 𝐼𝑖𝑗 is the index of the ith cultural practices dimension and jth country, 𝐼𝑖𝑢 is the index of the ith cultural practices dimension of the U.S. and 𝐶𝐷𝐽 is the cultural distance of the 𝐽𝑡ℎcountry from the US. Prior research stated that firm, deal level and national characteristics affect the international M&A activity of the firms (Chakrabarti et al., 2009, Erel et al., 2012).
4.5. Firm- and deal-level characteristics.
Information of the deal level is gained from the SDC Thomson One database. Deal characteristics have impact on the synergies of the M&A deal. Cash payment of the acquirer is a dummy
variable that takes the value of one if the deal is paid with cash and zero otherwise. Tender offer is the indicator variable that is equal to one if the bid involves a tender offer to the target
shareholder. Further, poison pill takes the value of one if the poison pill affected the bidders cross border merger or acquisition and zero otherwise. The hostile offer variable takes the value of one if the bid is stated as ‘hostile’ by the SDC and zero otherwise. Relatedness is a variable that takes the value of one if the target and the acquirer share the same industry and zero otherwise (Lim et al., 2016; Dutta et al., 2016,). The acquirer firm level characteristics are obtained from the Compustat Database. Target ROA is obtained from Datastream, The return on assets is to control for the for the acquirer and the target’s profitability.
4.6. Country Level Variables
Following to Chakrabarti et al. (2009) there are economic differences between two companies of different nations that have a significant effect on the performance of the M&A. In the thesis is used the differences in GDP, the openness of the target and acquirer to the world, and the corporate governances differences (Anti-directors right) between the target and the acquirer of Djankov & La Porta (2007).
The differences in GDP per capita are revealing for the wealth of one country versus the other country from the World Data Bank. These differences are expected to be important in case of the U.S. takeovers in the developed and emerging economies, therefore relevant to include in the regression.
𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝐺𝐷𝑃=
(𝑃𝑒𝑟 𝑐𝑎𝑝𝑖𝑡𝑎 𝐺𝐷𝑃 𝑜𝑓 𝑈. 𝑆. −𝑃𝑒𝑟 𝑐𝑎𝑝𝑖𝑡𝑎 𝐺𝐷𝑃 𝑡𝑎𝑟𝑔𝑒𝑡 𝑛𝑎𝑡𝑖𝑜𝑛) (𝑃𝑒𝑟 𝑐𝑎𝑝𝑖𝑡𝑎 𝐺𝐷𝑃 𝑜𝑓 𝑈. 𝑆. + 𝑃𝑒𝑟 𝑐𝑎𝑝𝑖𝑡𝑎 𝐺𝐷𝑃 𝑜𝑓 𝑡𝑎𝑟𝑔𝑒𝑡 𝑛𝑎𝑡𝑖𝑜𝑛)
The openness of the targets nation to the global world economy could affect the efficiency with which it can employ its synergies and the easiness with which the acquirer can manage and support the new division of the company abroad. According to Chakrabarti et al. (2009) and Han et al. (2016) the openness of the target nation may have an important bearing on the
17 functioning of the acquired business and is a useful resource for the acquiring nation. Displays how much one country is open to other trade partners or world economies.
𝑇𝑎𝑟𝑔𝑒𝑡 𝑜𝑝𝑒𝑛𝑛𝑒𝑠𝑠 =(𝑇𝑎𝑟𝑔𝑒𝑡 𝑖𝑚𝑝𝑜𝑟𝑡 𝑣𝑜𝑙𝑢𝑚𝑒 + 𝑇𝑎𝑟𝑔𝑒𝑡 𝑒𝑥𝑝𝑜𝑟𝑡 𝑣𝑜𝑙𝑢𝑚𝑒) 𝑇𝑎𝑟𝑔𝑒𝑡 𝐺𝐷𝑃
In the papers of Stulz and Williamson (2003) and Chakrabarti et al. (2009) they make use of an alternative measures for cultural differences origin. The legal origin proxy is based on La Porta et al (1998). This measure should control for culture differences between the two countries. The difference in corporate governance between the acquiring and target nation is the difference in the investor protection. Djankov et al. (2008) constructed a new index of
shareholder protection, based on the director variables of La Porta et al. (1998). The anti-director index ensures the minority shareholders to be protected against self-dealings
transactions benefiting the controlling shareholder. This paper uses the anti-self-dealing index to measure the corporate governance differences. The corporate governance difference indicators control for the corporate governance structure (Breuer et al, 2013).
𝐶𝑜𝑟𝑝𝑜𝑟𝑎𝑡𝑒 𝐺𝑜𝑣𝑒𝑟𝑛𝑒𝑛𝑎𝑛𝑐𝑒 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 =
= 𝐴𝑐𝑞𝑢𝑖𝑟𝑒𝑟 𝐴𝑛𝑡𝑖𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟 𝑟𝑖𝑔ℎ𝑡𝑠 𝑖𝑛𝑑𝑒𝑥 − 𝑇𝑎𝑟𝑔𝑒𝑡 𝐴𝑛𝑡𝑖𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟 𝑟𝑖𝑔ℎ𝑡𝑠 𝑖𝑛𝑑𝑒𝑥. Based on the dependent variables, independent variables and control variables described the following regression will be used to interpret the results. The event windows of CARs[ −5, +5], CARs [ −3, +3], CARs [ −1, +1] CARs are used for the short-term results.
𝐶𝐴𝑅𝑒𝑣𝑒𝑛𝑡𝑤𝑖𝑛𝑑𝑜𝑤= 𝛽0+ 𝛽1 𝐶𝐸𝑂𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽2 𝐿𝑛( 𝐷𝑒𝑎𝑙 𝑉𝑎𝑙𝑢𝑒) + 𝛽3 𝐿𝑛(𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤) + 𝛽4 𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 +
𝛽5 𝐵𝑜𝑎𝑟𝑑𝑠𝑖𝑧𝑒 + 𝛽6 𝐶𝑎𝑠ℎ 𝐷𝑒𝑎𝑙 + 𝛽7 𝑅𝑒𝑙𝑎𝑡𝑒𝑑𝑛𝑒𝑠𝑠 + 𝛽8 𝑂𝑝𝑒𝑛𝑒𝑠𝑠 𝑇𝑎𝑟𝑔𝑒𝑡 + 𝛽9 𝐷𝑖𝑓𝑓 𝐶𝑜𝑟𝑝𝑜𝑟𝑎𝑡𝑒 𝐺𝑜𝑣 +
𝛽10 𝐷𝑖𝑓𝑓 𝑖𝑛 𝐺𝐷𝑃 + 𝛽11 𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 + 𝛽12−20 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 +
𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡
The event windows returns for the buy-and-hold returns of 12 months, 24 months and 36 months is used for the long-term results.
𝐵𝐻𝐴𝑅𝑒𝑣𝑒𝑛𝑡𝑤𝑖𝑛𝑑𝑜𝑤= 𝛽0+ 𝛽1 𝐶𝐸𝑂𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽2 𝐿𝑛( 𝐷𝑒𝑎𝑙 𝑉𝑎𝑙𝑢𝑒) + 𝛽3 𝐿𝑛(𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤) + 𝛽4 𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 +
𝛽5 𝐵𝑜𝑎𝑟𝑑𝑠𝑖𝑧𝑒 + 𝛽6 𝐶𝑎𝑠ℎ 𝐷𝑒𝑎𝑙 + 𝛽7 𝑅𝑒𝑙𝑎𝑡𝑒𝑑𝑛𝑒𝑠𝑠 + 𝛽8 𝑂𝑝𝑒𝑛𝑒𝑠𝑠 𝑇𝑎𝑟𝑔𝑒𝑡 + 𝛽9 𝐷𝑖𝑓𝑓 𝐶𝑜𝑟𝑝𝑜𝑟𝑎𝑡𝑒 𝐺𝑜𝑣 +
𝛽10 𝐷𝑖𝑓𝑓 𝑖𝑛 𝐺𝐷𝑃 + 𝛽11 𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 + 𝛽12−20 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 +
𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡
For the second regression results I will test the moderating effect of nine dimensions cultural differences interacted with CEO confidence, on the short-term and long-term performance.
𝐶𝐴𝑅𝑒𝑣𝑒𝑛𝑡𝑤𝑖𝑛𝑑𝑜𝑤= 𝛽0+ 𝛽1 𝐶𝐸𝑂𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽2 𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓 ∗ 𝐶𝐸𝑂 𝑂𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 +
18 𝛽15 𝐵𝑜𝑎𝑟𝑑𝑠𝑖𝑧𝑒 + 𝛽16 𝐶𝑎𝑠ℎ 𝐷𝑒𝑎𝑙 + 𝛽17 𝑅𝑒𝑙𝑎𝑡𝑒𝑑𝑛𝑒𝑠𝑠 + 𝛽18 𝑂𝑝𝑒𝑛𝑒𝑠𝑠 𝑇𝑎𝑟𝑔𝑒𝑡 + 𝛽19 𝐷𝑖𝑓𝑓 𝐶𝑜𝑟𝑝𝑜𝑟𝑎𝑡𝑒 𝐺𝑜𝑣 + 𝛽20 𝐷𝑖𝑓𝑓 𝐺𝐷𝑃 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 + 𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡 𝐵𝐻𝐴𝑅𝑒𝑣𝑒𝑛𝑡𝑤𝑖𝑛𝑑𝑜𝑤= 𝛽0+ 𝛽1 𝐶𝐸𝑂𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽2 𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓 ∗ 𝐶𝐸𝑂 𝑂𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽3−11 𝐶𝐸𝑂 𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 ∗ 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠 + 𝛽12 𝐿𝑛( 𝐷𝑒𝑎𝑙 𝑉𝑎𝑙𝑢𝑒) + 𝛽13 𝐿𝑛(𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤) + 𝛽14 𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 + 𝛽15 𝐵𝑜𝑎𝑟𝑑𝑠𝑖𝑧𝑒 + 𝛽16 𝐶𝑎𝑠ℎ 𝐷𝑒𝑎𝑙 + 𝛽17 𝑅𝑒𝑙𝑎𝑡𝑒𝑑𝑛𝑒𝑠𝑠 + 𝛽18 𝑂𝑝𝑒𝑛𝑒𝑠𝑠 𝑇𝑎𝑟𝑔𝑒𝑡 + 𝛽19 𝐷𝑖𝑓𝑓 𝐶𝑜𝑟𝑝𝑜𝑟𝑎𝑡𝑒 𝐺𝑜𝑣 + 𝛽20 𝐷𝑖𝑓𝑓 𝐺𝐷𝑃 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 + 𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡 4.7. Gravity Model M&A
With the gravity model of Ahern et al. (2015) analyze the role of cultural differences and CEO overconfidence on wealth creation in the cross-border takeovers. Further, shows if cultural differences and CEO confidence has influence on the effect of volume on cross-border mergers. A gravity model such as Ahern et al. (2015); Frankel & Romer (1998) uses cultural differences to forecast the intensity of cross-country relations. I follow this method but I measure distance as cultural differences, along the nine dimensions of House et al. (2004) and add the variable of CEO overconfidence as an independent variable. More specific, the cultural characteristics and CEO overconfidence could have important effects on the income of the acquisition and merger through their impact on trade. Investigate how culture and CEO overconfidence affects the values of mergers and acquisition from U.S. as acquirer and the rest of the foreign companies as target. Using a Tobin regression model to account for the shortness of observed takeover activity.
𝐿𝑛 (𝐷𝑒𝑎𝑙 𝑉𝑎𝑙𝑢𝑒) = 𝛽0+ 𝛽1𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓 + + 𝛽2−10 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠 + 𝛽11 𝐶𝐸𝑂 𝑂𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 +
𝛽12 𝐷𝑖𝑓𝑓 𝐺𝐷𝑃 + 𝛽13 𝐷𝑖𝑓𝑓 𝐶𝑜𝑟𝑝𝑜𝑟𝑎𝑡𝑒 𝐺𝑜𝑣 + 𝛽14 𝐹𝑖𝑟𝑚𝑠𝑖𝑧𝑒 + 𝛽15𝐿𝑛(𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤) + 𝛽16 𝑅𝑒𝑙𝑎𝑡𝑒𝑑𝑛𝑒𝑠𝑠 +
𝛽17 𝐶𝑎𝑠ℎ𝑑𝑒𝑎𝑙 + 𝛽18 𝑅𝑂𝐴 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 + 𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡
𝐿𝑛 (𝐷𝑒𝑎𝑙 𝑉𝑎𝑙𝑢𝑒) is log the value of all mergers worth at least $10mln when acquirer is from U.S.
and the target is from country j in year t. 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠𝑖𝑗 𝑖s the total difference between two countries for each of our nine cultural value variable. 𝐶𝐸𝑂 𝑂𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 is a dummy
variable that takes one if the CEO is overconfident and zero if the CEO is not overconfident. A gravity model should include country dummy variables and year dummies (Frankel and Romer (1998). Include year-fixed effects to control for worldwide economic shocks, such as crisis, currency crises and changes in the market valuations. Include time-varying country-level variables such as difference in GDP between acquirer and target, difference in corporate governance an openness of the target.
4.7. Tobin’s Q
Tobin’s Q is another measure used to show the firm value for the individual firms and test the firm performance. A new Tobin’s Q measure is used, including intangible capital in measures of
19 investment. In the paper of Peters and Taylor (2014) results indicate that including intangible capital produces a Tobin’s Q proxy that is closer to the true, unobservable Q. Should include intangible capital in proxies for firms’ investment opportunities. This results in a Tobin’s q with a stronger investment-q relation. Specifically, regressions of investment on q produce higher 𝑅2 values, larger slopes, and hence lower implied adjustment costs in both firm-level and
macroeconomic data (Peters and Taylor, 2014). Q is calculated in WDRS from database Peters and Taylor’s total Q. For the firm market value use the Tobin’s Q as a proxy.
𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 𝑡𝑜𝑡𝑎𝑙 = 𝑀𝑘𝑡𝑐𝑎𝑝+𝐷𝑒𝑏𝑡−𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝐴𝑠𝑠𝑒𝑡𝑠
𝐵𝑜𝑜𝑘 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑝𝑟𝑜𝑝𝑒𝑟𝑡𝑦 𝑃𝑙𝑎𝑛𝑡𝑠 𝑎𝑛𝑑 𝐸𝑞𝑢𝑖𝑝𝑚𝑒𝑛𝑡+𝐼𝑛𝑡𝑎𝑛𝑔𝑖𝑏𝑙𝑒 𝐴𝑠𝑠𝑒𝑡𝑠
The study of Malmendier and Tate (2008) shows that CEOs are highly sensitive to cash flows when making investment decisions, therefore control for cash flow. Following the paper of Peters and Taylor (2014) and Antia et al (2007) some factors relative to Tobin’s Q need to be controlled such as the firm size, cash flow, leverage and the total assets. Further control for year fixed effects and country dummies. The individual firms Tobin’s q is regressed on the CEO overconfidence, firm characteristics, cultural differences, year fixed effects and the country dummies.
𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄
𝑖 = 𝛽0+ 𝛽1 𝐶𝐸𝑂𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽2 𝐿𝑛(𝐹𝑖𝑟𝑚 𝑠𝑖𝑧𝑒) + 𝛽3 𝐿𝑛(𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤) + 𝛽4 𝐿𝑛(𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠) +
𝛽5 𝐿𝑛(𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒) + 𝛽6−14 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠 + 𝛽15 𝑇𝑜𝑡𝑎𝑙 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 +
𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡
Further Tobin’s q is regressed on the CEO overconfidence interacted with the cultural differences between acquirer and target, firm characteristics, cultural differences, year fixed effects and the country dummies.
𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 = 𝛽
0+ 𝛽1 𝐶𝐸𝑂𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝛽2 𝐿𝑛(𝐹𝑖𝑟𝑚 𝑠𝑖𝑧𝑒) + 𝛽3 𝐿𝑛(𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤) + 𝛽4 𝐿𝑛(𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠) +
𝛽5 𝐿𝑛(𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒) + 𝛽6−14 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒𝑠 ∗ 𝐶𝐸𝑂 𝑜𝑣𝑒𝑟𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 + 𝑌𝑒𝑎𝑟 𝐹𝑖𝑥𝑒𝑑 𝐸𝑓𝑓𝑒𝑐𝑡𝑠 +
𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑖𝑒𝑠 + 𝜀𝑖𝑗,𝑡 5. Results
In this paragraph, the results from the regressions are stated. First the summary statistics of the cumulative abnormal returns and the buy-and-hold abnormal returns, cultural differences variables, CEO data and deal specific data will be presented. Next, the regression results for the CARs and the BHARs for the acquirer will be presented first without interaction terms and then with interaction terms. Finally, the regression results of the deal value and the Tobin’s q will be presented, Tobin’s q first without interaction terms and then with interaction terms.
20
5.1. Descriptive Statistics
Table 1 in the appendix shows the cumulative abnormal return around the announced takeover for the different event windows earned by the U.S. acquiring company. The market adjusted return model is used as a benchmark to calculate the return and the abnormal return of the stock prices. Since the focus is on the announcement effect there is chosen for 3 different event window based on the announcement date, CARs [-5, +5], CARs [-3, +3] and CARs [-1, +1] (Han et al., 2016, Chakrabarti et al 2009, Antia et al.2007). For the short-term observations, positive cumulative abnormal returns are observed through all the three windows as shown in Table 1. Making it more specific, a t-test is completed of CARs for all three-event windows, which shows the positive and significant results through all the windows. Therefore for the acquiring U.S. firms cross-border M&A is positively and significant estimated when looking to the stock performance around the announcement date, consistent with the findings of Han et al (2016). Table 2 in the appendix present the model which is used to test the influence of the cultural fit and the interaction with the CEO confidence on the long-term performance is based on the methodology of Chakrabarti et al. (2009). Analyzing three different windows of 12 months, 24 months and 36 months following the performance after the takeover. Since not all the data is available for the companies acquired for the entire 36-month post-takeover period, the number of observations declines as the length of the window increases. The buy-and hold averages returns are positive over all the event windows, concluding that for the U.S. acquirers in the long-term there are ‘only winners’ since profits are made on the positive stock performance. Table 3a in the appendix shows the descriptive statistics of the cultural differences per dimension and the total cultural differences of the acquirer (U.S.) and the target. In addition, to analyze the differences between the U.S. and the rest of the countries the scores per dimension of the U.S. are examined as well. The score per dimension is based on the paper of Antia et al (2007) where for each firm, a measurement in cultural difference is calculated. Cultural
differences in cross-border M&A are likely important when companies with possible conflicting values have to coordinate with each other. The total cultural differences are not significant. Although per dimension there are significant differences between the U.S and the foreign targets. The descriptive statistics in table 3a shows a significant negative difference in the
individualism variable and negative significance in the future orientation variable. In addition,
the U.S. culture is individualistic and short term orientated. A possible explanation could be that American companies are more disposed to rapid, short-term and positive results. Further, corresponds to House et al. (2004) that U.S is strongly focused on immediate results and profits.
21 Power distance indicates the concentration of control and inequality within a group. The U.S firm’s scores significant lower than their foreign targets. Power distance could have influence on the organizational culture of the U.S. firms and especially on the corporate governances (House et al., 2004).
For the corporate governance measure, look to the anti-directors right index of table 3d in the appendix. A significant difference is estimated between the foreign targets and the U.S., this could indicate that U.S. firms have stronger hierarchical and authoritative characteristics in their firm culture. Assertiveness (masculinity) is the degree to which individuals are assertive, confrontational, and aggressive in their relationships with others and the difference between the U.S and the foreign target is positive and significant. The assertive pole is related to the masculinity meaning that the CEOs in the U.S companies are more assertive, confrontational and aggressive in their relationships. Uncertainty avoidance is the level to which a society,
organization, or group relies on social norms, rules, and procedures to alleviate the
unpredictability of future events (House et al,, 2004). The U.S. firms have a low preference for uncertainty avoidance, this means U.S. firms are more accepting changes in the company and is capable of taking more risk. From the T-value can be concluded that difference in uncertainty avoidance between the target and the acquirer has a negative and significant influence. An explanation could be that uncertainty avoidance in a country has a negative effect on the M&A performance. The difference of Gender, Human Orientation, Institutional Collectivism and
Performance Orientation between the target and acquirer are not significant.
Table 3b in the appendix, the mean of the CEO age is 53.86 years and the tenure of CEO’s have an average of 8.9 years a stock ownership of 0.02% and vested options of 0.03%, this corresponds to the papers of Malmendier and Tate (2005) & (2008). Looking to the creation of the overconfidence measure in table 3c can be concluded that 18% of the CEOs in the sample are overconfident. However, the measures pay no attention to the return of a CEOs delayed option exercise (Campbell, 2011) and this could affect the CEO overconfidence variable. Further, an overconfident CEO may hold their options for other reasons than being
overconfident. A reason could be inside information to the CEO about the future increase of the firm’s stock price. CEOs want to hold the options longer and earn excessively profits from the stock increase. Therefore holding in the money options could have positive stock return (Malmendier and Tate, 2008). An economic reason could be that CEOs has deep confidence in their correctness and the precision of the upcoming M&A.
When analyzing the Tobin’s Q of descriptive statistics in Table 3d of the appendix, the mean of the Tobin’s Q is 1.37. When Tobin’s q is >1 firms have an incentive to increase their capital stock and buy physical capital. In other words, when q > 1, firms find it profitable to acquire additional capital because value of capital exceeds the cost of acquiring it. This is an incentive
22 for CEO’s to invest, resulting in cross-border M&A activity. Looking at table 3c in the descriptive statistics, the proportion of takeovers involving in cash is around the 64% in cash payment. Since the acquirers use more cash for the takeovers and less stock transactions, cash is the primary medium of payment in cross-border takeovers. The usual argument for method of payment effects is that cash bids do better for bidder shareholders because the market takes this as positive signal of bidder expectations future returns (Conn et al, 2005). There are only 4 hostile offers and 1 poison pill used in the takeovers of the sample. A tender offer is made in only 15% of the cases. Further, among the targets and the acquirers around 17% of the mergers and acquisitions are related in terms of industry. Looking to table 3e in the appendix, the difference in openness measures how much one country is open to other trader partners and world economies. A positive and significant relationship is found between the difference in openness in the foreign targets and the U.S. in the sample. Since a positive and significant relationship between the target openness and the acquirer could explain the improved
performance of the acquiring companies (Han et al. 2016). While cross-border investment are essentially riskier compared, since there are greater levels of information asymmetry , the openness of a target can decrease this problem. The anti-director index is positive and significant if the difference of the anti-directors index between the acquirer and the target is measured. The results present that the minority shareholders of the acquirer better to be protected against self-dealings transactions compared to the targets, during M&A activity.
5.2. Results of CARs and BHARs without interaction effect.
Table 4 and table 5 in the appendix present the results of the OLS regression of the acquirer abnormal returns on the different deal specific variables, country variables and firm data. As can be seen in Table 1 from the t-test the cumulative abnormal returns of the total sample are positive and significant. The multivariate regression measures the stock performance around the announcement date and after. With the event windows of the Cumulative Abnormal
Returns(CAR) before and after the M&A announcement CARs [ -5, +5], CARs [ -3, +3] and CARs[ -1, +1]. For the long run the dependent variable is the buy-and-hold returns from the
announcement date till the length of 1 year, 2 years and 3 years. Looking to the CARs event windows in table 4 and 5 the results shows significant negative and positive results at different levels of 5% and 10% significance, regarding the effect of CEO overconfidence on the firm performance on the short-run. The results in the long-term event windows from the buy-and-hold returns of 1year, 2 year and 3 years presents that the CEO overconfidence has a negative significant influence on the long-run performance, at different levels of 10% significance. The results are consistent with hypothesis 1 where CEO overconfidence has a negative effect on stock performance. The CEO overconfident as an indicator that equals one if the confidence
23 measure is at least 67 and zero otherwise. For an explanation, look to the event windows of column 1 and column 2 of table 4. If a CEO is overconfident it would decrease the CARs event
window[5, +5] with 0.00769% and increase the CARs event windows [-3, +3] with 0.00626%. From table 4 the long event windows of the BHARs of 1 year, an overconfident CEO
coefficient is significant and decreases the BHARs I of 1 year with 0.0233%. Further looking to the BHARs of 3 years, an overconfident CEO coefficient is significant and decreases the BHARs 3 years with 0.0337%. Hirsleifer et al (2012) measured that overconfident CEOs realize
significant success in innovative industries, however it is not specific related to short-term stock performance and cross-border M&A. In the most of the regression the results correspond with the result of the papers discussed in the literature review where overconfidence is observed as a negative variable on firm performance. However there can be established that there is more or less instability in the effect of overconfidence through time. On the days around the takeover decision the overconfidence CEO has negative and positive impact on the firm performance. While on the long-term it has negative effect on the firm’s performance. A possibility is that it takes time for the corporations to integrate the effect of the acquiring CEOs overconfidence. An additional reason could be the difference in length in the event windows used as dependent variable. The coefficients of the control variables Cash Deal, are positive and significant, looking to the long term. This could indicate that a deal payed in cash has a better stock performance. And is consistent with the usual argument for method of payment effects, cash bids do better for bidder shareholders because the market takes this as positive signal of bidder expectations future returns (Conn et al, 2005).
In table 4 and 5 Relatedness is significant and positive as control variable on the different event windows of the buy-and hold abnormal returns. Relatedness between two companies is necessary for positive stock performance after the takeover is and consistent with the findings of Lim & Lee (2016) shows that a cross-border acquisition deal is more likely to succeed when the degree of relatedness between the target and business is high. Tobin’s Q as control variable is only negative significant on the long-term performance, this could indicate that the firm value decreases on the long-term. Looking to the control variable of difference in corporate
governance between the acquirer and the target, the variable has positive and significant influence on the short-and long-term during and after the M&A activity. Consistent with the research of Ahern et al (2015), the anti-director rights index ensures the shareholder to be protected against transaction benefitting the control shareholders.
Table 4 presents the results of the OLS multivariate regression of the acquirer on the
abnormal returns over the short and long-term, now looking to the cultural difference effects. The total cultural difference variable is measured with the formula of Kogut and Singh (1988). In table 4 the total cultural differences has a negative effect on the short- and long-term
24 performance, with a significant level of 10%. Aimed to table 4 column 2 and column 3, if there are total cultural differences between the acquirer and target, the short-term performance decreases with 0.0151% and 0.0982%. Aimed at column 4 and column 5, the total cultural differences between the acquirer and the target decrease the long-term performance of the stock with respectively 0.0611% and 0.1101%. The results are consistent with the findings of (Ahern et al 2015; Eres et al 2012, Morossini et al 1998; Reus and Lamont, 2009) who found that cross-border M&A with greater cultural differences had lower stock performance than targets with smaller cultural differences, only looking to the total cultural differences variable created. Greater cultural distance could lead to more information asymmetry or environmental uncertainty in terms of larger cultural and institutional distances, higher target country risk and inexperience in the market invested (Lai et al, 2016; Johanson & Vahlne, 2009). Interestingly, the measure of cultural differences has the same impact on the short-term abnormal returns and long-term buy-and-hold returns.
Table 5 presents the results of the OLS multivariate regression from the effect of cultural
difference on the returns in cross-border M&A. The results are consistent with hypothesis 1, expected that the difference of the cultural dimensions has a negative effect on stock
performance. The formula of Antia et al. (2007) is used to present the difference in cultural differences in each of the nine dimensions from the acquirer to the target. In this regression cultural differences per dimension, total cultural distance variable, CEO overconfidence, control variables, year fixed effects and the country dummies are included. Only now all the dimensions of the cultural differences are placed in the regression, to look at the effect of each variable separately. The variables of Individualism, Performance Orientation, Uncertainty
Avoidance and Future Orientation and has a significant and negative influence on the stock
performance for most the short-term event windows of the CARs, at different levels of 1%, 5% and 10% significances. Aimed at column 2 in table 5 for Individualism, Performance Orientation and Uncertainty Avoidance, a 1% increase in one of these cultural differences ensures the cumulative abnormal returns will decrease respectively with 0.0568%, 0.0710% , 0.00394% and. The variables of Individualism, Performance Orientation, Uncertainty Avoidance, and Future
Orientation have a significant and negative influence on the stock performance for the long-term
event windows of the BHARs, at different levels of 1%, 5% and 10% significances. Aimed at table 5 column 5 for Individualism, Performance Orientation, Uncertainty Avoidance a 1%
increase in one of these cultural differences ensure the BHARs 2years will decrease respectively with 0.0671%, 0.478% and 0.440%. A possible explanation for the significance of the future orientation in the long run, while in the short-term performance there is no significance, is that it takes time for the corporations to integrate the effect of the future orientated differences between the target and acquirer. The results of the short and long-term performance are
