Shareholder wealth effects of cross-border and domestic deals within Europe

(1)

Shareholder wealth effects of cross-border and domestic deals within Europe

Author: R.J.Bruggeman

Student number: s1347764

e-mail address: r.j.bruggeman@gmail.com

Telephone number: 06-54758264

(2)

Shareholder wealth effects of cross-border and domestic deals within Europe

R.J. Bruggeman University of Groningen 2

Table of content

1. INTRODUCTION ... 4

2. THEORETICAL FRAMEWORK AND EMPIRICAL EVIDENCE ... 5

2.1 THEORETICAL FRAMEWORK... 5 2.1.1GEOGRAPHICAL DIVERSIFICATION... 6 2.1.2METHOD OF PAYMENT... 8 2.1.3 PRODUCT DIVERSIFICATION... 9 2.1.4 INVESTOR PROTECTION... 10 2.2 EMPIRICAL EVIDENCE... 11

3. DATA AND METHODOLOGY ... 16

3.1 DATA... 16

3.1.2 SUMMARY STATISTICS... 17

3.2METHODOLOGY... 18

3.2.1 STRUCTURE OF AN EVENT STUDY... 18

3.2.2 MODEL SPECIFICATIONS... 20

3.2.3 FACTORS INFLUENCING CUMULATIVE ABNORMAL RETURNS... 21

3.2.4STATISTICAL TESTS AND ROBUSTNESS TEST... 22

4. RESULTS... 25

4.1.GEOGRAPHICAL DIVERSIFICATION... 25

4.2METHOD OF PAYMENT... 27

4.3 PRODUCT DIVERSIFICATION... 28

4.4 INVESTOR PROTECTION... 30

4.6 REGRESSION ANALYSIS USING ALL VARIABLES... 31

5. CONCLUSIONS... 32

LITERATURE LIST... 35

(3)

Abstract

(4)

1. Introduction

In recent years the number of acquisitions made by companies in foreign countries has increased substantially (Shimizu et al., 2004). Product and capital markets continue to become more integrated, new markets are emerging, and globalization has become an important strategic issue for corporations (Moeller and Schlingemann, 2005) On a global scale, cross-border acquisitions worldwide during 1986-2000 accounted for 26% of the value of total acquisitions (Conn et al., 2005). While the majority of M&As involves two firms within the same country, over 40% of the M&As that were completed between 1999 en 2000 involved firms headquartered in two different countries (Shimizu et al., 2004). Historically, a large number of deals were concentrated in the U.S. and the UK. However, in the 1990s continental Europe enters the mergers and acquisitions foray (Aw and Chatterjee, 2004) The industrial structure across Europe is characterized by having relatively small firms with their activity heavily concentrated within their national borders, especially when compared to the industrial structure of the United States, an economic union approximately equal in size to the European Union (Campa and Hernando, 2004). The concentration of activity taking place in Europe is still very driven by national boundaries. Within-border mergers and acquisitions account for 50% of all transactions involving a European firm (Campa and Hernando, 2004). Apparently, there are still a number of barriers why cross-border mergers and acquisitions did not emerge with the establishment of a single currency and the promotion of integrating national markets into a single European market. While the occurrence of cross-border M&As has grown dramatically in the last few years, academic research on this type of strategic action has not kept pace with the changes (Shimizu et al., 2004).

The focus of this paper is to analyze whether bidder and target announcement returns differ significantly between cross-border and within-border mergers and acquisitions within Europe. Furthermore, I also analyze whether the method of payment, the level of diversification and the way countries protect shareholder rights influence the announcement returns for bidders as well as targets.

(5)

returns if they are taken over by a company located in another country. Furthermore, for both within-border bidders and within-border targets, offers consisting of cash trigger significant higher announcement returns than do offers consisting of shares. By performing an ordinary least regression analysis, offers consisting of shares destroy value for within-border bidder shareholders. The remaining results are insignificant.

The paper is structured in the following way. Section 2 discusses the relevant literature and provide empirical evidence. Section 3 describes the procedure how the data is collected and the descriptive statistics of the dataset. Furthermore, the methodology used in this is study is explained. The fourth section presents the results and section 5 gives the conclusions and suggestions for further research.

2. Theoretical framework and empirical evidence

2.1 Theoretical framework

The main question one should ask himself is why merger and acquisitions exist? In a merger, two companies are combined to achieve certain strategic objectives. If a company believes that by merging with another company, the future prospects of the combined entity are bigger than the two stand-alone prospects, then it is a logical consequence to merge. When there is a merger announcement, investors change their expectations about the future prospects and that is reflected in share prices. Are the future prospects better than the stand-alone value, then the share price will rise and visa versa.

There are several perspectives why companies undertake a merger or acquisition (Sudarsanam, 2003). The first one is the economic perspective on mergers and acquisitions. Firms undertake mergers or acquisitions in order to gain competitive advantage over their rivals through cost reduction or increased market power (Sudarsanam, 2003). Think about economy of scale and economy of scope. The main focus lies on cost reduction.

(6)

The next perspective is the finance theory. The previous two described perspectives perceive mergers and acquisitions as ways of achieving competitive advantage. The finance theory considers merger and acquisition decision of firms within the framework of the conflicts of interest among various financial claim holders of those firms. Managers of a firm run a company, but a company is owned by the shareholders. Shareholders are only interested in the wealth maximization of their investment, while managers may also pursue their own interests. This is known as the agency model. Managers may undertake mergers and acquisitions that do not satisfy the wealth maximization criterion.

The fourth and last perspective is the managerial perspective on mergers and acquisitions. Managers, in making acquisition decisions and in reacting to takeover bids, will be motivated by both pecuniary and non-pecuniary considerations. Managers reject a cash bid to avoid control loss and make a cash bid to preserve control. Furthermore, in order to align the manager with the interests of the shareholders, they receive shares (stock options). Because managers hold an undiversified portfolio of shares, they want to diversify the company in order to decrease the volatility of the firm’s earnings and their own investment. Another factor of the managerial perspective is hubris. Acquiring managers overestimate their capacity to create value out of acquisitions and often overpay for them.

All these perspectives have different hypotheses and different consequences for bidder and target returns. In the next subsection I explain which hypotheses and the effects on bidder and target returns could be relevant for the merger and acquisition process within Europe.

2.1.1 Geographical diversification

(7)

foundations on which cross-border M&A research was based (Shimizu et al., 2004). More recent research focuses on organizational learning perspective and the resource based view. In general, there are three different streams of literature concerning cross-border M&As. The first stream sees cross-cross-border M&As as a mode of entry in a foreign market, the second as a dynamic learning process and the last stream as a value-creating strategy (Shimizu et al., 2004). In this research, cross-border M&As is seen as a value-creating strategy. The first question that needs to be answered is whether acquiring firms create value for their shareholders by taking over targets located in another country. The second question is whether companies which are taken over by a company located in another country earn significant abnormal returns. Aw and Chatterjee (2004) state that cross-border bidder returns are higher if the acquirer can apply one country’s superior management techniques. Management techniques are different around the world and if a bidding firm knows that it can improve the performance of the target by replacing the management, then it will consequently result in higher abnormal returns. Firms enter foreign markets in order to exploit the firm’s specific resources. By exploiting these firm specific resources value is created. By taking over companies in other countries and transferring the knowledge (intangible assets), synergy is created (Moeller and

(8)

will be lower (Moeller and Schlingemann, 2005). Because of the more competitive cross-border market, it is more likely that targets taken over by bidders originated from another country earn higher returns than targets taken over by a company located in the same country. In the context of M&A activity in Europe during the period 1999-2006, I expect that cross-border bidder returns differ from within-border bidder returns. This is due to the fact that European companies perceive the integrated market as their “home” market. Companies must get a strong foothold on the European market in order to survive in the competitive environment; therefore I also expect that cross-border target returns differ significantly from domestic target returns.

2.1.2 Method of payment

Another characteristic that can influence the abnormal returns is method of payment. It describes how the deal is financed. The different methods of payments are cash, all-shares and a combination of cash and all-shares. The first article which theoretically

describes how methods of payment influence the transacting process is written by Hansen (1987). The transaction process is treated as a two-agent bargaining game under

imperfect information. If the bidder is uncertain about the target’s value, a bid consisting of shares forces target shareholders to share the risk that the bidder may have overpaid (Hansen, 1987). But managers are unlikely to offer shares, when they think that their shares are undervalued. Managers are only interested in offering shares when their shares are overvalued by the market (Bruner, 2002). For target shareholders, payment with cash will also have a positive influence on abnormal returns. If the offer is made in cash, the target shareholders are immediately liable to pay their taxes. If the method of payment is shares, then taxes are deferred. So the tax argument expects that cash offers have higher returns than an offer consisting of shares to compensate target shareholders for the immediate payment of taxes (Huang and Walkling, 1987).

(9)

documented in Huberman (2001). This forces the bidder to pay with cash. Therefore, the tax argument and the signaling hypothesis are no longer relevant. Therefore, I expect that there exists no difference between the method of payment for cross-border bidder and target returns.

To summarize the hypotheses, I expect that cash offers earn higher abnormal returns for within-border bidder and target returns than offers consisting of shares. For cross-border deals, I expect that there exists no difference between cash and shares offers for bidders as well as targets.

2.1.3 Product diversification

(10)

pay more for the target than they are worth to the bidding firm shareholders. Based on this finding I expect that diversifying deals leads to lower bidder returns than non-diversifying deals and that targets earn higher announcement returns if they are taken over by a company operating in an unrelated industry. I expect this diversification effect to hold for within-border as well as cross-border deals.

Another question that needs to be answered is whether there exists a difference between within-border diversifying deals and border diversifying deals. A cross-border diversifying deal has also to deal with the costs of coordinating corporate policies and the difficulties in monitoring managerial decision making in globally diversified firms (Denis et al., 2002). These costs are likely to be bigger for cross-border diversifying deals. Therefore, I expect that within-border diversifying deals earn higher returns than cross-border diversifying deals. For targets, I expect that there exist no difference between within-border diversifying deals and cross-border diversifying deals. The effects mentioned above are only relevant for bidder returns and has no influence on target returns.

2.1.4 Investor protection

The last characteristic is investor protection. This variable describes whether the bidder originates from a common law country and the target company is located in a civil law country. In the last few years, there has been done a lot of research on investor protection (Rossi and Volpin, 2004; La Porta et al., 2000; La Porta et al., 1997).

(11)

et al., 2000). The reason why the common law and civil law differs from one another is rooted in the history of the countries.

When investor rights are well enforced, they will pay more for their investments and this will make it more attractive for entrepreneurs to issue shares. Furthermore, La Porta et al., 2002) found that firms located in countries with better investor protection have higher Tobin’s Q than do firms located in countries with inferior protection. A higher Tobin’s Q means that there are growth opportunities present. La Porta et al. (2000) argue that an increasingly more important mechanism for a firm to increase value is to be acquired by a firm that has a relatively better regime of corporate governance. In a country with low shareholder protection, there are large benefits of private control and the market for corporate control does not operate freely (Rossi and Volpin, 2004). Therefore, I expect that bidder returns and target returns are higher when the bidder originates from a common law country and the target from a civil law country.

2.2 Empirical evidence

The empirical and theoretical literature on cross-border M&As is not as large as on within-border M&As. Few studies compare domestic M&As with cross-border M&As. Morck and Yeung, (1992) analyzed 322 cross-border acquisitions in the period 1978-1988 by US firms. They found significant positive abnormal returns for bidders. But the model they used in order to test the hypothesis is arbitrary. Instead of using the market model, they just subtracted the market return from the daily return. Eun et al. (1996) tested in their paper the combined wealth gains for foreign acquisitions of US firms. They found that the combined gain was significant positive ($68mln on average), but the returns for the targets were very high (37.02%) and for the bidders they were negative. Moeller and Schlingemann (2005) researched whether there existed a difference between cross-border and within-border acquisitions. Their sample consisted of 4430 acquisitions between 1985 en 1995 in the US. The results indicated that cross-border acquisitions earned significantly lower abnormal returns than within-border acquirers, but still positive.

(12)

cross-Shareholder wealth effects of cross-border and domestic deals within Europe

border bidder returns. They used the Netherlands, because the Netherlands have a very open economy. Corhay and Rad (2000) made a distinction between acquisitions done by Dutch firms for three different regions. European acquisitions, acquisitions made in the U.S. and East-European acquisitions. European acquisitions yielded the highest returns. Lowinski et al. (2004) researched whether within-border mergers and acquisitions in Switzerland differs from cross-border mergers and acquisitions made by Swiss firms. They concluded that there exists no difference between cross-border and within-border mergers and acquisitions. Conn et al. (2005) researched whether there exists a difference between cross-border and within-border deals. They came to the same results as Moeller and Schlingemann (2005). Besides that, their results also showed that the long run returns are also lower for cross-border deals than for domestic deals. Conn et al. (2005) and Moeller and Schlingemann (2005) also used almost the same time span. Danbolt (2004) focused in her research only on target returns of UK firms and analyzed a possible cross-border effect. Her results showed that cross-cross-border targets gain significantly more than domestic deals during the event window.

(13)

(14)

Table 1: Empirical studies focusing on short-term announcement effects surrounding a bid

Author (Year) Country Period Model Sample size Event

window CAR Bidder within- border CAR Target within- border CAR bidder cross-border CAR target cross-border Goergen and

Renneboog (2004) European bidder and target

1993-2000 CAPM 56 (49) cross-border bidders (targets) and 86 (85) within-border bidders (targets) [-1,0] -0.45% 10.22% 2.38% 11.25% Moeller and Schlingemann (2005)

U.S. bidder and

foreign target 1985-1995 Market model 383 cross-border and 4047 within-border

[-1,1] 1.173% - 0.307% -

Conn, Cosh, Guest

and Hughes (2005) UK bidder and foreign target 1984-1998 Market model 1140 cross-border and 3204 within-border

[-1,1] 0.68% - 0.33% -

Danbolt (2004) UK targets and

foreign bidders 1986-1991 Market model, CAPM 96 cross-border and 389 within-border [0,1] in months - 20.10% - 22.35% Lowinski, Schiereck and Thomas (2004) Swiss bidders and foreign targets

1990-2001 Market model 91 cross-border and 23 within-border

[-1,1] -0.21% - 0.63% -

Campa and

Hernando (2004) European union bidder and target returns

1998-2000 CAPM 80 cross-border and 182 within-border

(15)

Author (Year) Country Period Model Sample size Event

window CAR Bidder within- border CAR Target within- border CAR bidder cross-border CAR target cross-border Corhay, Rad

(2000) Dutch bidder and European target

1990-1996 Market model 84 cross-border [-5,5] - - 1.44% - Morck and Yeung

(1992) U.S. bidders and foreign targets 1978-1988 Daily return-market return 322 foreign acquisitions by U.S. bidders 0 - - 0.29% Eun, Kolodny,

Scheraga (1996) U.S. (foreign) bidders and foreign (U.S.) targets

1979-1990 Mean adjusted 117 foreign acquirers and 213 U.S. targets

(16)

3. Data and methodology

3.1 Data

In this section, I describe the construction and the characteristics of the dataset used to test the hypotheses formulated above. The starting point for this study is to collect all possible information on cross border deals that were involved in M&A activity during the 1999-2006 period. These data comprise my direct test sample. In addition, I also need a control sample in order to benchmark the performance of returns on M&A transactions carried out domestically.

I start out with a list of merger and acquisitions deals in the period between 01-01-1999 and 31-12-2006. This list is extracted from the Zephyr database. I require that both target and bidder originate from the EU, and that the bidder or the target is listed on a European stock exchange. Furthermore, I require that the method of payment is cash or shares or a combination of both. The acquired stake should be minimal 50% with a maximum of 100%. Additionally, the deal value should be known, with a minimum of 1 million euros. Besides that, I also extract the Industry Classification Benchmark and the country where the bidder/target is located from the Zephyr database. Following Martynova and Renneboog (2006) financial institutions are left out. This selection procedure yields a total of 1309 deals. Out of these 1309 deals, 153 are cross-border deals and 1156 are within-border deals. In order to get a control sample, I also extract 153 border acquisitions. Based on the deal value of the cross-border deal, the within-border deal is chosen where the deal value is closest to the cross-within-border deal. The next step is to filter out deals where the acquirer or target carried out multiple deals within 12 months. If a bidder takes over another company a year before, the estimation window could be biased and consequently result in biased abnormal returns. I did not filter out other value relevant information, such as profit warnings. This study focuses on the announcement effects of mergers and acquisitions, and only those effects are filtered out.

(17)

also extracted from DataStream. Lack of share price and or market value information reduces the sample to 306 announcements.

3.1.2 Summary statistics

Table 2 and 3 presents the summary statistics of my sample. In total, my sample consists of 149 bidders and 157 targets. Out of these 306 announcements, 164 are cross-border and 142 are within-border. 60% of the bidders pay with cash, while 56% of the targets are paid with cash. 23% of the bidders pay with shares and 25% of the targets get paid with shares. The remainder is paid with a combination of cash and shares. These results coincide with the results of Goergen and Renneboog (2004). They also find that 23% of the bids are financed by shares. Furthermore, 26% of the bidders make a deal that is diversifying and 30% of the targets is taken over by a company operating in an unrelated industry. 18% of the cross-border bidders who originate from a common law country take over a company located in a civil law country and 18% of the targets located in a civil law country are taken over by a company located in a common law country.

Table 2: Sample composition; distinguishing between cross-border and domestic deals, method of payment and diversification.

Total

sample All-cash bid All-shares bid Mixed payment bid Diversifying Investor protection Cross-border bidders 76 55 14 7 20 14 Within-border bidders 73 35 20 18 19 - Bidders 149 90 34 25 39 14 Cross-border targets 88 61 17 10 26 16 Within-border targets 69 28 23 18 21 - Targets 157 89 40 28 47 16

(18)

of the targets). Goergen and Renneboog (2004) have almost the same sample, 63% of all within-border bids take place in the UK.

Table 3: sample composition; presents where the bidder and target originate from.

Country Cross-border bidder Cross-border target Within-border bidder Within-border target Austria 2.63 % 1.14 % - - Belgium 2.63 % 3.41 % 2.74 % 2.90 % Denmark 5.26 % 3.41 % - - Finland 7.89 % 2.27 % - - France 17.11 % 15.91 % 13.70 % 15.94 % Germany 13.16 % 4.55 % 8.22 % 11.59 % Ireland 3.95 % 4.55 % - 2.90 % Italy 6.58 % 1.14 % 6.85 % 10.14 % Luxembourg 1.32 % - - - Netherlands 6.58 % 22.73 % - 2.90 % Poland - 1.14 % - - Portugal - 1.14 % - - Spain 7.89 % 3.41 % - 1.45 % Sweden 5.26 % 6.82 % 2.74 % 7.25 % UK 19.74 % 28.41 % 65.75 % 44.93 % Total 100 % (n=76) 100 % (n=88) 100 % (n=73) 100 % (n=69) 3.2 Methodology

The methodology used in this study is the event study. An event study is built on the efficient market hypothesis. When a merger or acquisition is announced, investors change their expectations about the future prospects of the company. A merger or acquisition announcement lead to the fact that share prices deviate from their normal return when the market reacts to the unexpected release of value-relevant information. This section describes the structure of an event study and the way how the hypotheses are tested.

3.2.1 Structure of an event study The event window

(19)

abnormal return is the actual ex post return of the security over the event window minus the normal return of the firm over the event window. The normal return is defined as the expected return without conditioning on the event taking place (MacKinlay, 1997). As can be seen in the figure, the abnormal returns increase around the day of the announcement. In order to define the event window, I measure the 95% confidence interval1. The event window is set at the days where the abnormal returns are out of the 95% confidence interval. For bidders, the cumulative abnormal returns are calculated for the period [-2,3]; for targets the cumulative abnormal returns are calculated for the period [-5,1].

Figure 1: Abnormal returns for all bidders and all targets.

Estimation window

(20)

common choice is using the period prior to the event window for the estimation window. But, the closer the end of the event window is set at the day of the announcement, the more likely it is that the model parameters are influenced by the announcement effects. So, in this study the estimation window is defined as [-260,-11].

Normal performance model

There are two common choices for modeling the normal return, the constant mean return model and the market model (MacKinlay, 1997). The constant mean return assumes that the mean return of a given security is constant through time. The market model assumes a stable linear relation between the market index return and the security return. I chose to use the market model, because in a security price, there is always a risk factor that influences the share price. By using the market model, the risk factor common to all securities is eliminated. The constant mean return does not take the common risk factor in account. The next subsection explains which formulas I used to test the hypotheses formulated in section 2.

3.2.2 Model specifications

The market model is a statistical model which relates the return of any given security to the return of the market portfolio. For any security i the market model is:

it mt i i it R R =α +β +ε

t = -260,…, -11 (1) ) 0 ( it = E ε

var(ε_it)=σ2εi

Where R_it and R_mt are the period-t returns on security i and the market portfolio. In this research, the market portfolio is MSCI Europe. This benchmark is chosen, because this study focuses on merger announcements within Europe.

(21)

mt i i it it R R AR = −αˆ −βˆ (2)

After calculating the abnormal returns, the abnormal returns must be aggregated through time and across securities in order to draw overall inferences for the event of interest (MacKinlay, 1997). Formula (3) calculates the cumulative abnormal return (CAR) for company i in the sample and formula (4) aggregates the CAR’s calculated in formula (3).

= = 2 1 ) ( 1 2 τ τ τ τ τ it i AR CAR (3) and ( , ) 1 ( 1, 2) 1 2 1 τ τ τ τ = = N i i CAR N CAR (4)

where τ1 represents the first day of the event window and τ2represents the last day of the

event window.

3.2.3 Factors influencing cumulative abnormal returns

In the literature are some factors mentioned which influence CAR’s. In the following subsection I explain which control variables I use and how they are measured. The correlation matrix between the control variables is given in the appendix.

Method of payment

In order to test whether method of payment influence the CAR’s of the companies in my sample. Method of payment consists of cash, shares and mixed payment. Method of payment is divided in three different control variables. If a deal is paid with cash the value is 1, 0 otherwise. If a deal is paid with shares, the value is 1, 0 otherwise.

Product diversification

(22)

Investor protection

The indicator equals 1 if the bidding firm is incorporated in a country of English legal origin (common law) and the target originates from a country of civil law and 0 otherwise. The common law countries are Ireland and the UK. It is computed based on the LaPorta et al. (1997) classification. This control variable is only used for cross-border deals.

Way of testing the hypotheses

The hypotheses are tested in two different ways. The first step is to split the sample up with respect to the control variables and test whether the means of the sub samples differ significant from each other. The other way is by performing an ordinary least squares regression (OLS). All the control variables are taken together and it measures the impact of the different control variables on the CAR’s. The regression used in this study is as follows: ε β β β β α+ + + + +

= cash shares diversification investorprotection

CAR 1 2 3 4 (5)

For within-border deals, the control variable investor protection is left out. This is done, because for domestic deals investor protection is not relevant.

The results of this regression are tested for heteroskedasticity using the White test. This test refers to unequal variance in the regression errors. The null hypothesis underlying the test assumes that the errors are both homoskedastic and independent of the regressors. If the results obtained from the White test are insignificant, it is correct to use the values derived from the OLS regression.

3.2.4 Statistical tests and robustness test

(23)

0 )) , ( var( ) , ( : 2 1 2 1 2 1 0 = τ τ τ τ CAR CAR H (6) 0 )) , ( var( ) , ( : 2 1 2 1 2 1 _≠ τ τ τ τ CAR CAR Ha (7)

Furthermore, I also test whether different sub samples differ significant from each other. This is done through comparing the means of the sub samples. I use a two-tailed t-test to compare the means. The null hypothesis is tested using

0 ) )) , , ( var( )) , , ( var( ( ) , , ( ) , , ( : 2 1 2 1 2 1 2 1 2 1 0 = + − i i i i i i n CAR n CAR CAR CAR H κ τ τ κ τ τ κ τ τ κ τ τ (8) 0 ) )) , , ( var( )) , , ( var( ( ) , , ( ) , , ( : 2 1 2 1 2 1 2 1 2 1 _≠ + − i i i i i i a n CAR n CAR CAR CAR H κ τ τ κ τ τ κ τ τ κ τ τ ₍₉₎

where κiis a sub sample and n is the number of observations in the sub sample. i

(24)

By implementing the Corrado rank test, the first step is to transform the abnormal returns into their respective rank.

) ( it it rank AR K = bidder t = -260,…,3 target t = -260,…,1 k N i s i D K N d Z ) _σ 2 1 ( 1 1 0 = + − = (10)

The standard deviation is calculated using the entire sample period:

2 )) 2 1 ( 1 ( 1 1 3 260 + − = = = − = s N i it t t s k D K N D σ , where (11)

d= number of days in the event window

=

0

i

K average rank of the event window

=

s

D Number of days in the sample

=

N Number of events

(25)

4. Results

This section describes the empirical results. At first, I look at the descriptive statistics of all bidders and all targets in my sample. Table 4 describes the statistics of all bidders and all targets. 46% of the bidder returns result in positive cumulative abnormal returns. Campa and Hernando (2004) find the same percentage of positive cumulative abnormal returns in their sample. The mean is 0.3% and positive. The mean target return is 11.7% and 75% of the target cumulative abnormal returns are positive. Campa and Hernando (2004) find that only 60% of the targets earned positive CAR’s. The probabilities of the Jarque-Bera test are very close to zero, indicating that the data is not normally distributed. As a robustness test, I also give the results for the Corrado rank test. If the results coincide, the t-statistic is not influenced by outliers or a skewed distribution. In the following subsections, I focus on the differences between the different sub samples. I start with geographical diversification, followed by method of payment, product diversification and investor protection.

Table 4: Descriptive statistics of the CAR’s in my sample

All bidders All targets

Mean 0.003 0.117 Median -0.003 0.085 Maximum 0.330 1.063 Minimum -0.411 -0.364 Std. dev. 0.085 0.178 Jarque-Bera 178.57 250.57 Prob. JB 0.00 0.00 % positive 46% 75% 4.1. Geographical diversification

(26)

cross-Shareholder wealth effects of cross-border and domestic deals within Europe

border bidders earn lower CAR’s than within-border bidders. Based on the results provided in table 5a I do not find support for the hypothesis that cross-border bidder returns and within-border bidder returns differ significantly from each other.

In table 5b are the announcement effects for targets presented. All the CAR’s are significantly different from zero and positive. By performing the Corrado rank test as a robustness test, the target returns remain significant. The CAR’s in my sample are in line with the findings of Goergen and Renneboog (2004). Campa and Hernando (2004) find lower CAR’s for the targets. Both papers find that cross-border targets earn higher CAR’s than within-border targets, although the difference is not significant. Danbolt (2004) focus in her study on abnormal announcement returns for targets. She finds a significant cross-border effect during the month surrounding the bid. It is difficult to compare the results of Danbolt (2004) with my results, because this study focuses on short-term announcement effects. Contrary to the results of Goergen and Renneboog (2004) and Campa and Hernando (2004), there exists a cross-border target effect in my sample. Therefore, the results show support for the hypothesis that cross-border targets earn significantly higher CAR’s than within-border targets.

Table 5a: CAR’s of bidders and t-test in order to test for differences.

All bidders Within-border

bidders (1) Cross-border bidders (2) Difference between (1) and (2)

CAR 0.003 -0.001 0.008

T-value 0.478 -0.133 0.883 0.670

Corrado rank

test -0.213 -0.520 0.212

Table 5b: CAR’s of targets and t-test in order to test for differences.

All targets Within-border

targets (1) Cross-border targets (2) Difference between (1) and (2) CAR 0.117 0.082 0.145 T-value 8.229*** _4.902*** _6.768*** _2.207** Corrado rank test 8.377 *** _4.432*** _7.264***

(27)

4.2 Method of payment

In order to answer the question whether the method of payment has an influence on cumulative abnormal returns, I split the sample into four sub samples, because I expect a difference between cross-border and within-border method of payment effects. The results for bidders are presented in table 6a and the results for the targets in table 6b. The results of the Corrado rank test show that cross-border bidder returns deviate significantly from zero when the offer is made in shares. This is not supported by the value. The t-value should not be used when the data is not normally distributed. The probability of the Jarque-Bera test is 0.772. In this sub sample the data is normally distributed, so it is correct to compare the means using a t-test. The results indicate that a within-border cash bid generates significant higher CAR’s than a within-border bid consisting of shares. These findings correspond with the findings of Conn et al. (2005). They also find that all cash bids generate higher CAR’s than non cash bids. Unfortunately, they did not perform a statistical test whether the differences between a cash bid and a non cash bid are significant. The difference between a border bid consisting of shares and a cross-border cash bid is not significant. This is consistent with the hypothesis presented in section 2. Cash bids may be forced by the reluctance on the part of cross-border target shareholders to accept foreign equity (Conn et al., 2005; Moeller and Schlingemann, 2005). This is confirmed by table 2, because 73% of the cross border bidders are paid with cash, while only 48% of the within-border bidders are paid with cash. Other factors influence the cross-border method of payment.

(28)

one outlier is removed. The mean CAR with the outlier is 24.2% and an offer consisting of shares generates significantly higher CAR’s than do offers consisting of cash (p=0.04). Following Brooks (2002), if an outlier has a significant influence on the results, then it must be removed. Therefore, I find support for the hypotheses that there exists a payment effect in within-border deals and that the payment effect diminishes in cross-border deals for target- as well as bidder shareholders.

Table 6a: CAR’s of bidder firms by method of payment and the significance of difference between methods of payment

All bidders cash All bidders shares Within-border cash (1) Within-border shares (2) Cross-border cash (3) Cross-border shares (4) Difference between (1) and (2) Difference between (3) and (4) CAR 0.008 -0.017 0.011 -0.038 0.006 0.014 T-value 1.032 -0.912 0.956 -1.553 0.560 0.531 2.039** _0.356 Corrado rank test -0.259 0.375 0.476 -1.108 -0.711 1.908 *

Table 6b: CAR’s of target firms by method of payment and the significance of difference between methods of payment

All targets cash All targets shares Within-border cash (1) Within-border shares (2) Cross-border cash (3) Cross-border shares (4) Difference between (1) and (2) Difference between (3) and (4) CAR 0.123 0.096 0.122 0.031 0.124 0.191 T-value 7.237*** _3.313*** _4.845*** _1.089 _5.592*** _3.782*** _2.425** _1.338 Corrado rank test 7.680 *** _3.221*** _4.459*** _1.531 _6.256*** _3.159***

***,**,* indicates significance at the 1% , 5% and the 10% level.

4.3 Product diversification

(29)

non-diversifying deals. But again, the difference is not significant. Moeller and Schlingemann (2005) show in their results that there are lower announcement returns when a company diversifies geographically as well as industrially. Returns for cross-border diversifying deals are lower than for within-border diversifying deals, but the results are insignificant. Therefore, I reject the hypotheses that diversifying deals generates lower bidder announcement returns than non-diversifying deals. I also reject the hypothesis that cross-border diversifying deals result in significant lower bidder returns than within-cross-border diversifying deals. This is consistent with the findings of Conn et al. (2005). Their results also show that relatedness has no significant influence on announcement returns, although a non-diversifying deal has a positive impact on announcement returns.

Table 7b present results for target announcement returns with respect to diversification. The Corrado rank test shows that within-border diversifying deals generate a less significant result for target returns. The announcement returns deviate significantly from zero, but only at the 10% level. The descriptive statistics3_{show that the}

sub sample is normally distributed so it is correct to use the t-value. The diversification effect is strongest in within-border deals. Target returns are higher in within-border diversifying deals than in within-border non-diversifying deals. But the difference is not significant. Cross-border diversifying target returns are almost the same as cross-border non-diversifying target returns. The difference between within-border target diversifying returns and cross-border target diversifying returns is not significant. Target shareholders value the cross-border effect, but not the diversification effect.

Table 7a: CAR’s of bidder firms by diversification and the significance of difference between diversification

(30)

Table 7b: CAR’s of target firms by diversification and the significance of difference between diversification

All targets diversif ying All targets non-diversifyi ng Within-border diversifyi ng (1) Within-border non-diversifying (2) Cross-border diversifyi ng (3) Cross-border non-diversifying (4) Difference between (1) and (2) Difference between (3) and (4) Difference between (1) and (3) CAR 0.131 0.111 0.117 0.067 0.144 0.145 T-value 4.371*** _7.025*** _4.162*** _3.265*** _2.877*** _6.510*** _1.367 _0.034 _0.442 Corrado rank test 3.502 *** _7.719*** _1.943* _4.029*** _2.963*** _6.736***

***,**,* indicates significance at the 1% , 5% and the 10% level

4.4 Investor protection

In order to answer the question whether investor protection has an influence on the cross-border announcement returns, the cross-cross-border sample is split up into two different sub samples. The results for the bidders are in table 8a and the results for the targets in table 8b. Table 8a shows that the announcement returns are higher for bidders who take over a company located in a civil law country. This is contrary to my expectations. The difference is also insignificant.

The results of the target returns show that targets located in a civil law country and who are taken over by a company located in a common law country generate higher CAR’s than targets taken over by companies located in civil law countries. This is consistent with the expectations of Rossi and Volpin (2004). The premium paid for targets located in civil law countries is higher when the bidder is located in a common law country. But the results are insignificant. Therefore, I reject the hypotheses that bidder and target announcement returns are significantly higher when the bidder is located in a common law country and the target is located in a civil law country.

Table 8a: CAR’s of bidder firms by investor protection and the significance of difference

(31)

Table 8b: CAR’s of target firms by investor protection and the significance of difference

All

cross-border targets Investor protection high (1) Investor protection low (2) Difference between (1) and (2) CAR 0.145 0.171 0.139 T-value 6.768*** _2.650** _6.293*** _0.569 Corrado rank test 7.264 *** _3.018*** _6.608***

4.6 Regression analysis using all variables

This section presents the results from the OLS regression. The CAR is the dependent variable in each of the four regression models. As a robustness check, I also perform the regression with the rank as the dependent variable (results in appendix). Following Danbolt (2004), coefficients reported in italics indicate that the OLS and rank regression provide conflicting results as to the sign or whether or not the coefficient is statistically significant. The White test is used to test whether the standard errors are homoskedastic. In each of the four regressions there are no signs that I have to reject the hypothesis that the error terms are not homoskedastic. The main result in table 9a is that shares have a significant negative influence on the CAR. Investors expect significantly less value creation from the firm’s decision to pay a within-border deal with shares. The signs for cash and diversification for within-border bidders are not the ones I expected (expectations confirmed by rank regression), but the results are insignificant. For cross-border bidders none of the controlling variables have a significant influence on the CAR’s. Other factors, not investigated by me, may influence the CAR’s of cross-border bidders.

(32)

high investor protection has a negative impact on target shareholders announcement returns. But the results are not significant.

To summarize the findings of the OLS regression. The only control variable that has a significant influence on the dependent variable is shares. If a bidder undertakes a within-border deal and pay with shares, then shareholder value is destroyed for the bidder company shareholders.

Table 9a: CAR₌α₊β1cash₊β2shares₊β3diversification₊β4investorprotection₊ε Within-border bidders Cross-border bidders

Constant 0.013 0.015 Cash -0.003 -0.001 Diversification 0.004 -0.028 Shares -0.05* _0.008 Investor protection - -0.001 Adjusted R2 _0.021 _-0.027 White (F-statistic) 0.616 0.338

Table 9b: CAR=α+β1cash+β2shares+β3diversification+β4investorprotection+ε Within-border targets Cross-border targets

Constant 0.062 0.129 Cash 0.047 0.007 Diversification 0.051 -0.035 Shares -0.045 0.072 Investor protection - -0.020 Adjusted R2 _0.068 _-0.012 White (F-statistic) 0.050 0.818

5. Conclusions

(33)

Table 10: Summary of hypotheses and the corresponding results.

Hypotheses Results

Geographical diversification

Cross-border bidder returns differ significantly from within-border bidder

returns Rejected

Cross-border target returns differ significantly from within-border target

returns Not rejected

Method of payment

Within-border bidder returns where the offer consists of cash differ significantly from within-border bidder returns where the offer is made in shares

Not rejected Within-border target returns where the offer consists of cash differ

significantly from within-border target returns where the offer is made in shares

Not rejected Cross-border bidder returns where the offer consists of cash do not differ

significantly from cross-border bidder returns where the offer is made in shares

Not rejected Cross-border target returns where the offer consists of cash do not differ

significantly from cross-border target returns where the offer is made in shares

Not rejected

Product diversification

Within-border bidder returns where the deal is diversifying differ significantly from within-border bidder returns where the deal is non-diversifying

Rejected Within-border target returns where the deal is diversifying differ

significantly from within-border target returns where the deal is non-diversifying

Rejected Cross-border bidder returns where the deal is diversifying differ

significantly from cross-border bidder returns where the deal is non-diversifying

Rejected Cross-border target returns where the deal is diversifying differ significantly

from Cross-border target returns where the deal is non-diversifying Rejected Within-border bidder returns where the deal is diversifying differ

significantly from cross-border bidder returns where the deal is diversifying Rejected Within-border target returns where the deal is diversifying do not differ

significantly from cross-border target returns where the deal is diversifying Not rejected

Investor protection

Cross-border bidder returns where the bidder is located in a common law country and the target in a civil law country differ significantly from cross-border bidder returns where the bidder is not located in a common law country and the target in a civil law country

Rejected

Cross-border target returns where the bidder is located in a common law country and the target in a civil law country differ significantly from cross-border target returns where the bidder is not located in a common law country and the target in a civil law country

(34)

The results show that bidder returns are not influenced whether the bid is cross-border or not. On the other hand, cross-border target returns are significantly higher than within-border target returns. For within-within-border deals, the method of payment is of significant influence. For both bidders and targets, offers consisting of cash trigger significant higher announcement returns than do offers consisting of shares. The OLS regression shows that offers consisting of shares destroy value for within-border bidder shareholders. All the other results shows no significant difference between the different control variables.

(35)

Literature list

Aw, M.S.B., Chatterjee, R.A., 2004, The performance of UK firms acquiring large cross-border and domestic takeover targets. Applied Financial Economics 14. pp. 337-349.

Brooks, C., 2002, Introductory Econometrics for Finance. Cambridge University Press. Cambridge, MA.

Brown, S.J., Warner, J.B., 1985, Using daily stock returns, the case of event studies, Journal of Financial Economics, Vol. 14, pp. 3-31.

Bruner, R.F, 2002, Does M&A pay ? A survey of evidence for the decision maker, Journal of Applied Finance, Spring/summer, pp.48-68.

Campa, J.M., Hernando, I., 2004, Shareholder value creation in European M&As, European Financial Management, Vol. 10, pp. 47-81.

Chang, S.C., Wang, C.F., 2007, The effect of product diversification strategies on the relationship between international diversification and firm performance, Journal of World Business, vol. 42, pp. 61-79.

Conn, R.L., Cosh, A., Guest, P., and Hughes, A., 2005, The impact on UK acquirers of domestic, cross-border, public and private acquisitions, Journal of Business Finance and Accounting, Vol.32, No. 5 & 6, pp.815-70.

Corhay, A., Rad, A.T., 2000, International acquisitions and shareholder wealth : Evidence from the Netherlands, International Review of Financial Analysis, vol. 9:2, pp. 163-174.

(36)

Cowan, A.R., 1992, Nonparametric event study tests, Review of Quantitative Finance and Accounting, Vol. 2, pp. 343-358.

Denis, D.J., Denis, D.K., Yost, K., 2002, Global diversification, industrial diversification, and firm value, The Journal of Finance, Vol. LVII, pp. 1951-1979.

Danbolt, J., 1995, An analysis of gains and losses to shareholders of foreign bidding companies engaged in cross border acquisitions into the United Kingdom, 1986-1991, European Journal of Finance, Vol.1, No. 3, pp.279-309.

Danbolt, J, 2004, Target company cross-border effects in acquisitions into the UK, European Financial Management, Vol.10, No.1, pp. 83-108.

Eun, C.S., Kolodny, R. and Scheraga, C., 1996, Cross-border acquisitions and shareholder wealth: Tests of the synergy and internalization hypotheses, Journal of Banking and Finance, vol. 20, pp. 1559-1582.

Goergen, M., Renneboog, L., 2004, Shareholder wealth effects of European domestic and cross-border takeover bids, European Financial Management, Vol. 10, pp. 9-47.

Hansen, R.G., 1987, A theory for the choice of exchange medium in mergers and acquisitions, Journal of business, vol. 60, pp. 75-95.

Hitt, M.A., Hoskisson, R.E, Kim, H., 1997, International diversification effects of innovation and firm performance on product diversified firms, Academy of Management, Vol. 40, pp.767-798.

(37)

Huberman, G., 2001, familiarity breeds investment, Review of financial studies, vol. 14.3, pp. 659-680.

Jensen, M.C., 1986, Agency costs of free cash flow, corporate finance, and takeovers, The American Economic Review, Vol. 76 , pp. 323-329.

Kang, J., 1993, The international market for corporate control: mergers and acquisitions of U.S. firms by Japanese firms, Journal of Financial Economics, Vol. 11, pp.345-71.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R., 1997, Law and Finance, Journal of Political Economy 106, pp. 1113–1155.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R., 2000, Investor protection and corporate governance, Journal of Financial Economics, vol. 58, pp. 3-27.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R., 2002, Investor protection and corporate valuation, The Journal of Finance, Vol. LVII, pp. 1147-1170.

Lewellen, W.G., 1971, A pure financial rationale for the conglomerate merger, The Journal of Finance, Vol. 26, pp. 521-539.

Lowinski, F., Schiereck, D., Thomas, T.W., 2004, The effect of cross-border acquisitions on shareholder wealth- evidence from Switzerland, Review of Quantitative Finance and Accounting, Vol. 22, pp. 315-330.

MacKinlay, A. C. 1997. Event Studies in Economics and Finance, Journal of Economic Literature, vol. 35, pp. 13-39.

(38)

Martynova, M., Renneboog L., 2006, The performance of the European market for corporate control: evidence from the 5th takeover wave, working paper 2006-118.

Moeller, S., Schlingemann F., and Stulz R., 2004, Firm size and the gains from acquisitions, Journal of Financial Economics 73, 201-228.

Moeller, S.B., and Schlingemann, F,P.,2005, Global diversification and bidder gains: a comparison between cross-border and domestic acquisitions’, Journal of Banking and Finance, Vol. 29, pp. 533-564.

Morck, R., Shleifer, A., Vishny, R., 1990, Do managerial objectives drive bad acquisitions, The Journal of Finance, Vol. 45, pp. 31-48.

Morck, R., Yeung, B., 1992, Internalization: an event study, Journal of International Economics, vol. 33, pp. 41-56.

Rajan, R., Servaes H., and Zingales L., 2000, The cost of diversity: The diversification discount and inefficient investment, Journal of Finance 55, 2537-2564.

Roll, R., 1986, The Hubris Hypothesis of Corporate Takeovers, Journal of Business 59, pp. 197-216.

Rossi, S., Volpin, P.,2004, Cross-country determinants of mergers and acquisitions, Journal of Financial Economics, Vol. 74, pp. 277-304.

(39)

Stein, J.C., 1997, Internal capital markets and the competition for corporate resources, The Journal of Finance, Vol. 52, pp. 111-133.

Sudarsanam, S., 2003, Creating value from mergers and acquisitions, Pearson education Limited. Harlow.

(40)

Appendices

Appendix 1: measurement of the 95% confidence interval

The confidence interval is a range of values. The sample mean is at the center of this

range and the range is mean ± confidence. By taking a confidence interval of 95%, the

values are calculated using the following formula. ) ( 96 . 1 n mean± σ

The abnormal returns for all bidders and all targets are aggregated for the event window [-20,20]. The event window is set where the abnormal returns lies outwards the confidence interval.

Appendix 2a: Correlation matrix bidders

Cash Shares Diversification Investor

protection Cash 1 Shares -0.76 1 Diversification 0.10 0.02 1 Investor protection -0.01 0.04 0.18 1

Appendix 2b: Correlation matrix targets

Cash Shares Diversification Investor

protection Cash 1 Shares -0.74 1 Diversification -0.16 -0.00 1 Investor protection 0.12 -0.08 0.02 1

Appendix 3: descriptive statistics of cross-border bidders paying with shares

(41)

Appendix 4: descriptive statistics of within-border targets taken over by a company operating in an unrelated industry Cross-border bidder Mean 0.117 Median 0.137 Maximum 0.374 Minimum -0.164 Std. dev. 0.128 Jarque-Bera 0.183 Prob. JB 0.913

Appendix 5a: Rank₌α₊β1cash₊β2shares₊β3diversification₊β4investorprotection₊ε

Within-border bidders Cross-border bidders

Constant -2.378 -1.543 Cash 6.210 -0.112 Diversification -5.805 -10.895 Shares -3.589 18.353 Investor protection - 10.133 Adjusted R2 _-0.02 _0.03 White (F-statistic) 0.290 1.158

Appendix 5b: Rank₌α₊β₁cash₊β₂shares₊β₃diversification₊β₄investorprotection₊ε

Within-border targets Cross-border targets

(42)