Long-term performance of Western European Industrial Acquirers:
Results of Germany, France and the United Kingdom
University of Groningen
Author: Christiaan Huitema Studentnr: 1510207
TABLE OF CONTENTS
1 INTRODUCTION... 3
2 LITERATURE ... 4
2.1 OVERVIEW OF MERGER WAVES... 4
2.2 INCENTIVES TO ACQUIRE... 5
2.3 RETURN OF ACQUIRERS... 7
2.4 CROSS-BORDER ACTIVITY AND DOMESTIC ACQUISITIONS... 7
2.5 EXPERIENCE IN ACQUISITIONS... 8
2.6 PAYMENT METHOD... 8
2.7 TARGET SECTOR... 9
2.8 RELATIVE SIZE OF THE ACQUIRER... 10
2.9 SHORT-TERM POST-ANNOUNCEMENT PERFORMANCE... 10
3 SAMPLE, DATA AND METHODOLOGY ... 12
3.1 INITIAL SAMPLE... 12
3.2 FINAL SAMPLE... 12
3.3 METHODOLOGY... 13
3.4 UNIVARIATE TESTS... 15
3.5 MULTIPLE CROSS-SECTION REGRESSION ANALYSIS... 15
3.5.1 ADDITIONAL SECOND CROSS-SECTION REGRESSION... 16
3.6 DESCRIPTIVE STATISTICS... 17
3.6.1 DEPENDENT VARIABLES... 18
3.6.2 INDEPENDENT AND INTERESTING VARIABLES... 19
3.6.3 COMPILATION OF SAMPLE... 19
3.7 INDUSTRIAL BENCHMARK... 20
4 RESULTS ... 21
4.1 CAARS OF ACQUIRERS... 21
4.2 UNIVARIATE RESULTS... 22
4.2.1 DOMESTIC RETURNS VERSUS CROSS BORDER RETURNS... 23
4.2.2 FREQUENT ACQUIRERS VERSUS LESS FREQUENT ACQUIRERS... 23
4.2.3 CASH VERSUS OTHER METHODS OF PAYMENT... 24
4.2.4 INDUSTRIAL TARGETS VERSUS NON-INDUSTRIAL TARGETS... 24
4.2.5 LARGE ACQUIRERS VERSUS SMALL ACQUIRERS... 25
4.2.6 POSITIVE VERSUS NEGATIVE SHORT-TERM POST-ANNOUNCEMENT RETURNS... 25
4.3 MULTIPLE CROSS-SECTION REGRESSION ANALYSIS... 26
4.4 SECOND CROSS-SECTION REGRESSION RESULTS... 28
5 CONCLUSION ... 29
1 INTRODUCTION
Several studies in the past have taken a look to the short-term performance of acquisitive companies. These studies found in general that the post-merger performance of acquisitive companies was negative between the announcement date and the merger date. The most recent merger wave took place in the 1995-2002 period (see first table). This merger wave had the highest number of acquisitions of all merger waves with a top number of mergers in the year 2000. The 2000-2001 period will be used for this thesis because of the high merger activity and the fact that it was the most recent merger wave.
In the financial literature there are fewer studies of mergers and acquisitions on the long-term performance of the acquirer than on the short-term performance. Therefore this study is focused on the long-term performance of the acquiring firms. The performance of the firms will be examined up to five years after the announcement date. Most of the literature and research papers that can be found about mergers and acquisitions are based on samples of American companies. Merger activity in European countries is less examined by researchers. Therefore, it is interesting to know how European countries, like the United Kingdom, France and Germany are performing in the take-over market. In this study the long-term performance and the characteristics of the deals made, will be compared. The companies of the sample will all be manufacturing companies. This thesis also examines cross border activity of acquiring companies in comparison with domestic acquirers. The overall goal of the thesis is to find out if acquiring is a successful strategy for a company in the long run. A second purpose of this thesis is to examine if the industrial European companies of the sample outperform the benchmark significantly and positively. This thesis is useful for pension funds, institutional investors and other investors with a long-term investment focus. The most striking conclusions of this thesis are that industrial European acquirers do indeed outperform the benchmark significantly. Furthermore, the target sector variable as well as the domestic variable are the determinants which are most accountable for this positive outperformance.
2 LITERATURE
In this chapter I will give an overview of the literature on the performance of acquiring companies. First in section 2.1, an overview of the mergers waves in the past will be given as well as the driving forces behind such waves. I will continue with the incentives to acquire companies in section 2.2. Section 2.3 deals with the return of acquirers and sections 2.4 - 2.9 discuss the determinants of the return of the acquirers and the related hypotheses. The discussed determinants in chronological order are: cross-border activity and domestic acquisitions, experience in acquisitions, payment method, target sector, relative size of the acquirer and the short-term post-merger performance. Finally, table 2 summarizes the determinants used by the authors of the discussed literature.
2.1 Overview of Merger Waves
Table 1: Merger waves history
Waves Period Change in Environment and
Technology
1st Merger wave 1897-1904 Industrial revolution, steam
engine, horizontal mergers
2nd Merger wave 1920-1929 Vertical and conglomerate
mergers, network opportunities 3rd Merger wave 1965-1975 Diversification of products,
acquiring firms of other markets
4th Merger wave 1984-1988 Combination of technology and
production activity
5th Merger wave 1995-2002 Globalization and deregulation, cross-border M&A
Source: Kleinert and Klodt (2002)
As can be seen in table 1, the fifth merger wave took place in the 1995-2002 period. The key items of that wave are: globalization, deregulation and cross border M&A’s. Globalization can be described as changes in societies and the world economy, which are the result of increased trade and cultural exchange. In economic contexts, it is often understood to refer almost exclusively to the effects of trade, particularly trade liberalization or free trade.1 Deregulation is a reduction of government regulation of private industry.2 Due to these changes before and in the fifth merger wave, the markets of companies expanded and the opportunities to penetrate foreign markets increased for acquiring companies. As a result of these changes there were more cross-border acquisitions and mergers in the fifth merger wave, compared with the preceding merger waves which were more locally focused and less cross-border oriented.
2.2 Incentives to acquire
Although most of the studies in the financial literature conclude that the acquirer will not benefit from acquisitions, companies continue to merge. However, in the management literature evidence can be found that acquirers can benefit from mergers. According to the industrial management literature, firms can gain from mergers because of synergies. Synergies arise if two firms merge and the whole company works more efficiently and profitable than the individual companies. Lubatkin (1983) mentioned three sources of synergies, namely:
• Technical economies: Lowering the average cost of the firm due to experience and know how
in common technology.
• Pecuniary economies: Increased market power purely through size increased and therefore
power to demand lower prices from suppliers.
• Diversification economies: Improving performance in relation to the risk of the firm. More
diversification in revenue, the use of different sales areas, will make the combined firm less risky and this will, for instance, lower the interest rates of banking facilities.
Late in the sixties, Mueller (1969) already found that the payment of the management was related to the size and the growth of the acquiring company and not to the financial performance of the company. This might also induce managers to take wrong decisions on mergers. A more recent study of Seth et al. (2000) examined three motives for acquisition. These motives are the synergy hypothesis, the managerialism hypothesis and the hubris hypothesis. The synergy hypothesis matches with the synergies discussed by Lubatkin (1983); the value of the combined firm exceeds the value of the firms separately. It is assumed that the managers of the firm are motivated to increase the economic value of the firm in favor of the shareholders and that they have the ability to select targets which improve the value of the combined firm (Seth et al. 2000). The managerialism hypothesis suggests that managers make acquisitions to maximize their own wealth at the expense of the shareholders of the firm. It is assumed that managers of the firm are aware of overpaying for acquisitions but continue to do so, because their managerial compen-sation is based on the size of the firm and not on performance in terms of profits (Seth et al. 2000 and Mueller 1969). In comparison with the managerialism hypothesis, the hubris hypothesis (Roll, 1986) suggests that the manager is not aware of overpaying for acquisitions. The managers of the firm make mistakes in evaluating the targets, but assume that their valuations of the targets are correct (Seth et al. 2000). These valuation mistakes lead to overpayment and eventually losses for the shareholders of the firm.
2.3 Return of Acquirers
A recent study provided by Moeller and Schlingemann (2005) examines the announcement returns of American companies which acquire cross-border firms. These results are compared with the announcement returns of American companies which have taken over domestic firms. Their study concludes that the announcement returns of cross-border firms are significantly lower than the announcement returns of domestic firms. The difference in returns of domestic and cross-border returns equals 1%. Also Chatterjee and Aw (2004), found for United Kingdom (UK) companies that the post-takeover returns of cross-border acquisitions in the United States were significantly lower than the returns of acquisitions in the domestic UK market. Chatterjee and Aw (2004) used different time frames up to two years namely t+6, t+12, t+18 and t+24 (all in months) to measure the performance of acquiring UK firms. In contrast, Ben-amar and André (2006) found that the average acquiring firm announcement period abnormal return for the Canadian market is positive. Cullen & Baker (1984); Digman (1986); Lenz (1981) and Lubatkin and Shrieves (1986) argued that it will take quite some time before the acquisition will affect the performance of the acquiring firm. Fowler and Schmidt (1988) therefore used a nine year time-frame to measure the long-term performance. For the Canadian market, Eckbo and Thornburn (2000), found evidence in line with the results of Chatterjee and Aw (2004). The performance of Canadian acquirers is significantly higher in the domestic market than in the foreign markets. In this thesis it is expected that acquirers in the industrial sector can significantly outperform the sector as a whole represented by the Dow Jones Western Europe Industrial Index in the long-term. The 22 studies of long-term performance discussed in Agrawal and Jaffe (2000) showed more negative than positive stock price performance of the acquirers in the long run. Therefore, the hypotheses to be tested are defined as follows:
H1,0 : Long-term abnormal performance of all acquirers is significantly positive.
H1,1 : Long-term abnormal performance of all acquirers is zero.
2.4 Cross-border activity and domestic acquisitions
hand, a loss of control due to distance in cross-border deals might be the reason for lower returns in comparison with the domestic acquisitions (Ravenscraft and Scherer, 1987). Eckbo and Thornburn (2000) found that the abnormal returns of domestic acquisitions in Canada outperform the foreign acquisitions. Based on the findings of Scherer and Ross (1990) and Ravenscraft and Scherer (1987), this thesis examines the returns of cross-border and domestic take-over activity. The administrative problems of acquisition mentioned by Lubatkin (1983) and the problems of distance (Ravenscraft and Scherer, 1987) are expected to be higher for cross-border acquisitions than for domestic acquisitions. Therefore, the following hypotheses are defined:
H2,0 : The abnormal long-term returns of acquirers taking over targets domestically are higher
than abnormal long-term returns of acquirers taking over cross-border targets.
H2,1 : The abnormal long-term returns of acquirers taking over targets domestically are not higher
than abnormal long-term returns of acquirers taking over cross-border targets.
2.5 Experience in Acquisitions
The result of the study performed by Cools and Van der Laar (2006) shows that acquirers have equal or better long-term returns than companies which grow with organic or growth strategies. The high growth rate of acquiring companies is the main factor for good stock market performance. Cools and Van der Laar (2006) conclude that acquisitions drive performance and not the other way around, which seems to be an answer to the question of Lubatkin (1983): ‘Does more merger activity leads to better performance of the firm?’ According to Cools and Van der Laar (2006) more acquisitions will lead to better performance of the acquiring firm. This thesis will test if frequent acquirers outperform less frequent acquirers in the long run. The following hypotheses will be tested:
H3,0 : More frequent acquirers will have higher long-term abnormal returns than less frequent
acquirers.
H3,1 : More frequent acquirers will not have higher long-term abnormal returns than less frequent
acquirers.
2.6 Payment Method
suggest that the method of payment reflects inside information of the acquirer’s own stand-alone value or the value of the target’s resources under control of the acquirer. Shareholders react more positive to cash in case of acquisitions than on stock payments. Loughran and Vijh (1997) studied the term returns to acquirers and found that acquirers paying with cash earned positive long-term abnormal returns and acquirers paying with stock earned negative long-long-term abnormal returns. Ben-amar and André (2006) found that the cash payment have a positive impact on value creation. The assumption is made that paying with stock reduces the dividends paid in the future to the current shareholders of the firm due to dilution. It may, more over, be a sign of overpriced stock. According to Hansen (1987) cash financing is only done when the acquirer is undervalued and equity financing is likely to occur when the acquirer is overvalued. Martynova and Renneboog (2006) state that cash-offers signal that the acquiring firm wants to pay off the target shareholders in order to not pay future value increases of the merged firm. Stock-offers on the other hand, signal that the acquiring firm wants to share the risk of the merged firm with the target shareholders.
In this thesis the returns of acquirers paying with cash, are expected to be significantly positive compared to acquirers with other methods of payment like shares, debt, loan notes or differently. The following hypotheses will be tested:
H4,0 : Acquirers paying acquisitions in cash are performing better in the long run than acquirers
paying in shares, debt, loans or differently.
H4,1 : Acquirers paying acquisitions in cash are performing not better in the long run than
acquirers paying in shares, debt, loans or differently.
2.7 Target sector
in conglomerate mergers than in conglomerate ones. They defined a merger as non-conglomerate if the acquirer and its target are in the same industry, measured by the 4 digit SIC-code. Vertically integrating activities of targets are therefore less positively received by the shareholders of the firms than horizontal integration. The following hypothesis will be tested: H5,0 : The long-term performance of acquirers with targets in the same industry is better than the
long-term performance of acquirers with targets in non-related sectors.
H5,1 : The long-term performance of acquirers with targets in the same industry is not different
from the long-term performance of acquirers with targets in non-related sectors.
2.8 Relative size of the acquirer
According to the financial literature, the relative size of the acquirer is often examined as a control variable of the acquirer’ return. Moeller, Schlingemann and Stulz (2004) find that there are no dollar synergy gains for acquisitions made by large firms. Eckbo and Thorburn (2000) found that the returns of acquirers decrease when the relative size of the acquirer tends to increase. Asquith et al. (1983) report that the size of the target relative to the size of the acquirer is positively correlated with the Cumulative Abnormal Returns (CAR) of the acquirer. In contrast, Loughran and Vijh (1997) found that abnormal returns become smaller and eventually
negative as the relative size of the target to acquirer increases.
To examine if the size of the acquirer in relation to the size of the target influences the returns of the acquirer in the long run, the following hypotheses will be tested:
H6,0 : Relative large acquirers have lower abnormal long-term returns than small acquirers.
H6,1 : Relative large acquirers do not have lower abnormal long-term returns than small acquirers.
2.9 Short-term post-announcement performance
positive short-term abnormal returns will also have positive long-term returns. The following hypotheses are therefore univariately tested:
H7,0 : There is a significant positive relation between the short-term abnormal returns and the
long-term abnormal returns.
H7,1 : There is not a significant positive relation between the short-term abnormal returns and the
long-term abnormal returns.
An overview of the literature used in this thesis is given in table 2. It can be seen that the size of the acquirer and the payment method as well as the cross-border or domestic determinants are most frequently used by researchers to determine the returns of the acquirers. All studies discussed are based on the long-term performance, except for Faccio, McConnell and Stolin (2006), Moeller, Schlingemann and Stulz (2004) and Ben-Amar and André (2006) which have a short-term performance focus.
Table 2: Overview of the determinants of performance used in the literature Cross-border or domestic Experience of Acquirer Payment
Method Sector of target
Relative size of the
acquirer Agrawal, Jaffe,
Mandelker (1992) x
Ben-Amar and André
(2006) x x
Capron (1999) x x
Chatterjee and Aw
(2004) x
Cools and Van der Laar
(2006) x x x
Eckbo and Thorburn
(2000) x x x x x
Faccio, McConnell and
Stolin (2006) x x x
Loughran and Vijh
(1997) x x
Moeller, Schlingemann
and Stulz (2004) x x x x
Martynova and
3 SAMPLE, DATA AND METHODOLOGY
In section 3.1 the initial sample is presented and in section 3.2 the final sample is shown. The methodology of the tests is explained in section 3.3 and the univariate tests are discussed in section 3.4. The multiple cross-sectional regression analysis is shown in section 3.5 and in subsection 3.5.1 the variables of the additional second (control) regression analysis are discussed. The descriptive statistics of all relevant variables are described in section 3.6. In the subsections 3.6.1, 3.6.2 and 3.6.3 the characteristics of the dependent variables, the independent and interesting variables and the compilation of the sample are presented, respectively. In the final section of this chapter, section 3.7, the benchmark will be discussed.
3.1 Initial sample
The initial sample3 is derived from the Zephyr database. Zephyr contains deal information of mergers and acquisitions. The acquirers in the sample are listed on the London Stock Exchange, the Euronext Paris or the Frankfurt Stock Exchange. The acquirers are English, French or German companies only. The acquirers are all industrial companies in category 2000 according to the Industry Classification Benchmark of Zephyr. All deals made by the acquirers are completed deals and the deal type is acquisition. Only deals with a known value are selected in the sample. The period selected was 1/1/2000-31/12/2001 because of the fact that I wanted to study a five years performance of acquirers too and because it is the end of the last merger wave. These selection criteria resulted in 728 deals.
3.2 Final sample
values in the turnover of the target and also in the turnover and the assets of the acquirer resulted in a sample of 192 deals. The next step selected the first deal4 made by the acquirers in the sample in the 1/1/2000 - 31/12/2001 period. This deal is chosen as the representative deal for the performance of the acquirer. Each deal represents one acquirer in the sample. This step reduced the number of unique acquirers to 100 deals. One of the returns of the acquirers, retrieved from DataStream, generated an error for ROK Property Solutions. This company is deleted from the sample. Deutsche Post is also deleted from the sample. DataStream showed that Deutsche Post was first listed on 20 November 2000. The acquisition of Deutsche Post was made before being listed; as a result no stock data could be retrieved from DataStream for this deal. These further adjustments have led to the final sample of 98 deals.
3.3 Methodology
Brown and Warner (1980) compared 5 different methodologies to measure the price performance of securities. The bottom line of their study is that there is no evidence that using complicated methodologies are better than using simpler models. Moreover, they presented evidence that more complicated methodologies make the researcher worse off. Although there are several methodologies (for instance the market model) which estimate the of the acquirer by regression in the estimation period5, it is likely that the of the acquirer will be different before and after the acquisition due to changes in the structure of the firm, especially if we examine the long-term performance of the acquirer up to 2 and up to 5-years. For these reasons, the Market Adjusted Returns (MAR) model will be used in this thesis.
A recent paper of Chatterjee and Aw (2004) used t+24 months for the long-term performance of UK firms. A substantial part of the sample of this thesis consists of UK firms and t+24 (months) will here for be used to measure the long-term performance. This time frame is also chosen for comparing the results with other findings of other researchers. In the literature overview of post-acquisition performance studies in Agrawal and Jaffe (2000), 60 months after the announcement date is also frequently used by the researchers. I will therefore also use a t+60 (months)
4 Most of the acquirers made more than one acquisitions in the sample period. Only the first deal made by each
performance indicator. This is already a long time period and using a longer time-frame might reduce the value of the results. A nine-year timeframe (as used by Fowler and Schmidt, 1988), for example, may include in my opinion generally too many external factors and events that can affect the performance of the companies in the sample. The Market Adjusted Model gives the following equation:
where is the Abnormal Return (AR) on acquirer iat time t. represents the return of acquirer i at time t. is the return on the benchmark (b), at time t. Adding the returns of equation leads to the following equation:
-
where represents the Cumulative Abnormal Return for each acquirer (i) for event window (T). The chosen long-term performance event windows are: T+24 (months) and T+60 (months). Cumulative Average Abnormal Return (CAAR) for all acquirers can than be calculated by equation :
where is the result of the sum of the CAR at time T for all acquirers divided by the amount of acquirers (N) in the sample.
By using the one sample T-test, it is tested if the outcomes of CAART for the event windows
(T+24) and (T+60) are significantly different from zero at the 1% level. The equation to be tested is shown below:
0
3.4 Univariate tests
The independent variables, related to the literature in section 2, are ratio and ordinal (dichotomous) variables. To test the impact of a single independent variable on the dependent variable, univariate tests are made. The singular impact of the variables on the dependent variables can also be shown by calculating a correlation table. The correlation coefficient of the normal variables shows the steepness and direction of the relation. The dichotomous (dummy) variables in this thesis consist of only 2 groups, the reference group and the omitted group (others). In the univariate tests the difference between the mean of the reference group and the mean of the omitted group will be examined on significance with the T-test.
The univariate cross-section regression analysis shows how much the independent variables in the model can explain of the variance of the dependent variable. It is assumed that there is a linear relation between the determinants and the dependent variable. The adjusted R2 is used instead of the R2 as an indicator of the explained percentage of the variance of the dependent variable in the model.
3.5 Multiple cross-section regression analysis
This analysis examines the impact of the 6 characteristics discussed in the literature of this thesis on the two-year and five-year CARs of the acquirers. Subsection 3.5.1 gives an additional second regression analysis. The following equation will be tested where i stands for one individual acquirer:
=
Where:
CAR = is the dependent variable. It represents the Cumulative Abnormal Returns (CARs) of 24
months (first regression) or 60 months (second regression).
LOCALD = is the dummy which controls for cross-border activity and domestic acquisitions. The
made by acquirers are numbered 1; the resulting deals are considered cross border and defined as 0 in this dummy.
is the ACQuirer Frequency (ACQF) variable. This variable indicates the experience of the acquirer.6 The number of extra acquisitions made by each acquirer after the first acquisition,
but still before the end of the sample-date (31/12/2001), is taken for this variable.
a dummy for the payment method of the acquisition. The dummy can be defined as: a number of 1 when the acquisition has been funded with cash only and a value of 0 is given when it has not been funded with cash.
is the Target Sector Dummy (TSD). It resembles a dummy for horizontal acquisitions. If
the target of the acquiring firm is in the same industry as the acquirer, according to the Standard Industry Classification (SIC) Benchmark codes in Zephyr, a number of 1 is given to deals with the same first digit. A value of 0 is given to deals where the first digit of the target code is not equal to the first digit of the SIC-code of the acquiring firm.
is the ACQuirer Turnover divided by the Target Turnover (ACQTTT). This variable controls for relative size. 7 It is defined as the turnover of the acquirer divided by the turnover of
the target.
= is the ABNormal Return of 1 week (ABNR1). The short-term abnormal returns of
one week after the announcement date are used as a representative of shareholders’approval towards the acquisition. A short-term positive or negative reaction of shareholders might be indicative of the long-term abnormal returns of the acquirer. This variable also represents the short-term post-acquisition performance of the acquirers.
is the error term.
3.5.1 Additional second cross-section regression
To check the ACQF, ACQTTT and ABNR1 variables, a second cross-section regression analysis is presented. The mentioned variables ACQF, ACQTTT and ABNR1 in the first regression will be replaced by the following dummies:
6 For example: An acquirer which have taken over 8 targets in the 1/1/2000-31/12/2001 period, gets an acquirer
frequency value of 7 (8-1). The first acquisition is not included in this variable, as this is taken to be the major acquisition. An acquirer with only one single acquisition in the period will therefore have a value of 0 for the acquirer frequency variable.
is the ACQF dummy. Less frequent acquirers are companies which have taken over only one target in the sample-period. More frequent acquirers are defined as companies which have taken over two or more companies. The less frequent acquirers group is given a dummy value of 0, the more frequent acquirers group gets a dummy value of 1.
is the ACQTTT dummy. If the turnover of the acquirer is at least 10 times larger8 as the turnover of the target, it is considered as a large acquirer (relative) and the dummy variable gets a value of 1. If the turnover ratio is smaller than 10, the acquirer is considered a small acquirer (relative) and a value of 0 is given.
= is the ABNR1 dummy. The 1-week post-announcement abnormal returns are divided in a group with positive 1-week CAARs and a group with negative 1-week CAARs. The deals in the first group are given a value of 1 and the deals in the second group get a value of 0.
The ACQFD dummy is inserted to show the differences in long-term returns between more frequent acquirers and less frequent acquirers. This dummy is also used in the univariate tests to test the impact on the long-term performance of the acquirers. ACQTTTD is inserted because of outliers in the underlying turnovers of the acquirers and the targets, which are used to determine ACQTTT. Finally, the ABNR1Ddummy is used in the univariate tests as a control variable of ABNR1 and will therefore be included in the second cross-section regression analysis.
3.6 Descriptive statistics
The descriptive statistics of the dependent, independent and other relevant variables are shown and the interesting statistics are highlighted in table 3. Furthermore, an overview of the compilation of the sample will be given and the characteristics of the observations will be discussed.
8 This division of relative large and small acquirers is based on the method used by Very et al. (1997). In their
Table 3: Descriptive statistics of all variables
(N=98) Mean Median Minimum Maximum Std. Deviation
DEPENDENT VARIABLES ABNR2 17,77% 7,87% -64,62% 219,34% 61,57% ABNR5 31,67% 6,66% -101,06% 626,96% 119,34% INDEPENDENT VARIABLES LOCALD 0,63 1 0 1 0,485 ACQF 0,94 0 0 9 1,565 ACQFD 0,42 1 0 1 0,496 CASHD 0,57 1 0 1 0,497 TSD 0,55 1 0 1 0,500 ACQTTT 53515 11 0 5234848 528790 ACQTTTD 0,51 1 0 1 0,502 ABNR1 1,78% 1,77% -21,19% 45,67% 8,92% ABNR1D 0,57 1 0 1 0,497 INTERESTING VARIABLES DEALV (x 1mln €) 397,6 23,0 0,4 9313,4 1396 TT (x 1mln €) 404,3 28,6 0,001 7561 1195,3 ACQT (x 1mln €) 2618,2 342,2 0,02 42619,2 5614,2
Benchmark 2-years CR 5-years CR
DJWII* -18,79% -16,64%
*Benchmark return measured from 4-1-2000 till 4-1-2002 for 2-years CR and from 4-1-2000 till 4-1-2005 for 5-years CR. For an explanation of the independent variables see section 3.5. Interesting variables are discussed in 3.6.2, including the benchmark.
3.6.1 Dependent variables
3.6.2 Independent and interesting variables
Except for the ACQF, ACQTTT and ABNR1 variables, all other independent variables are dummy variables. Dummy variables, in this thesis, have a minimum of 0 and a maximum of 1 and these values are given to select groups. For example, the CASHD variable shows that the sample consists of 57% deals paid with cash and that also 57% of the sample had a positive 1-week post-announcement abnormal return (ABNR1D). The mean frequency of additional acquisitions (ACQF) is 0,94, which indicates that almost every acquirer, on average, has taken over one additional target in the sample period. ACQTTT has a mean of 53515, which means that the average acquirer is 53515 times as large as the target, based on turnover. This is caused by 1 single outlier, because the target reported a turnover of only 1,32 (x1000). To control for such outliers, the ACQTTTD is added to the thesis. The mean and median of the ABNR1 (1,78% and 1,77% respectively) are almost identical. However, regarding the high standard deviation of ABNR1 (8,92%) this occurrence indicates that the 1-week abnormal returns are not concentrated around the mean or median.
An interesting variable is the deal value (DEALV). The mean value of an acquisition in the sample is almost 400 million Euros and the largest acquisition is approximately 9,3 billion Euros. Target Turnover (TT) and ACQuirer Turnover (ACQT) are shown because these values are used to determine the ACQTTT variable as well as the ACQTTTD dummy.
The cumulative return of the benchmark for the 2-years as well as for the 5-years period is given. For instance, an investor who invested money in a basket, which resembles the exact weights of all the securities in the benchmark index, would have lost 18,79% of his money in the 2-years period. The loss for the 5-years period would have been 16,64%, indicating that the three years following the 2-years period were on average slightly positive.
3.6.3 Compilation of sample
Table 4: Number of deals made by acquirers domestic and cross-border
N=98 Number of Observations
Acquiring countries DOMESTIC EUROPE USA TOTAL
United Kingdom 47 9 15 71
Germany 4 3 0 7
France 11 5 4 20
The United Kingdom covers the main part of the number of deals in this sample, namely 71. This is in line with the sample-distribution of previous studies of European acquirers. Faccio, McConnell and Stolin (2006) also found in their study that United Kingdom deals were dominating. Initially it was the intention to include a subgroup Germany-USA, but the absence of any deal in this category made this impossible.
The United Kingdom, France and Germany are not equally distributed in this sample. Because of the under representativeness of France and Germany, together 27 deals, there will be made no further separation between the acquiring countries’ long-term, domestic and cross-border results, which we intended to study originally too.9
3.7 Industrial benchmark
The Cumulative Average Abnormal Returns (CAARs) are based on a benchmark index, which is here the Dow Jones Western Europe Industrials Index10 (ticker symbol E1IDU). This benchmark is a good representative of the returns of the whole industrial sector of Western Europe. The index contains top industrial companies of Europe. The Thomson Financial DataStream database is selected to retrieve the weekly closing ratings from 1/1/2000 till 31/12/2006. These closing rates are converted into euros.11
9 In appendix 2 an overview of the long-term CAARs for each group, mentioned in table 4, is given for review (these
returns are not tested for significance).
10http://www.esignalcentral.com/support/symbol/esigind_d1.asp viewed at 6 Jun 2007.
4 RESULTS
This section of the paper shows the outcomes of the univariate tests and the multiple cross-section regression analysis. In subcross-section 4.1 the results of the CAARs of the acquirers will be discussed and linked to the main hypotheses. The results of the univariate tests are shown in subsection 4.2. In subsection 4.2.1 - 4.2.6, the independent variables are discussed separately. Finally, the results of the multiple cross-section regression analysis are presented in subsection 4.3 and the determinants are discussed in 4.3.1 - 4.3.6. Results of the second regression are discussed in subsection 4.4.
4.1 CAARs of Acquirers
In this subsection the results of the event study will be presented and the hypotheses based on the return of acquirers (H1,0 and H1,1) are discussed. In table 5 an overview of the mean returns and
significance of ABNR2 and ABNR5 is shown.
Table 5: One-sample T-test of long-term CAARs
Test value = 0 ONE-SAMPLE T-TEST
Variables Mean return Df t-value Sig. (%)
ABNR2 17,77% 97 2,858 0,5***
ABNR5 31,67% 97 2,627 1,0***
***Significance at the 1% level.
According to the one-sample T-test in table 5, the 2-years and 5-years CAARs (ABNR2 and ABNR5, respectively) differ from the hypothesized test value of 0. Based on these results it can be said that the industrial acquirers in Western Europe have significantly higher CAARs for the two-year and five-year period after the announcement-date compared with industrial companies of the Dow Jones Western Europe Industrials Index. Hypothesis H1,0 should therefore not be
4.2 Univariate results
The results of the Independent Sample T-test are shown in this subsection. In 4.2.1 - 4.2.6 the results of the dummy variables will be discussed individually as well as with their related hypotheses.
In table 6 the CAARs of the independent dummy variables are shown for the 2-years and 5-years period separately. The dummy variables are shown vertically in the left column. The first number of observations (N) for each dummy variable resembles the first group (dummy = 1) to be tested and the second number resembles the second group (dummy = 0) with the remaining observations of the sample (total number of both groups is 98). The mean is the CAAR for each group. Mean Differ is the difference between the CAAR of the first and the second group. Sign.(%) shows if the mean difference is significant. Finally, the Hypothesis accepted gives an overview of the accepted hypothesis based on the results of this univariate independent sample T-test. A correlation table is included in Appendix 1, for the correlation coefficients of the normal variables this can be looked at for reviews.
Table 6: Results of Independent sample T-test
2-YEARS Period 5-YEARS Period
Dummy
Variables N Mean SD Differ. Mean Sign. (%) Hypothesis accepted Mean SD Differ. Mean Sign. (%) Hypothesis accepted UNIVARIATE RESULTS LOCALD 62 0,2538 0,68 0,21 7,5* a H2,0 0,4686 1,37 0,41 5,7* a H2,0 36 0,0467 0,45 0,0550 0,76 ACQFD 41 0,0952 0,62 -0,14 26,3 H3,1 0,1685 0,92 -0,25 30,0 H3,1 57 0,2372 0,61 0,4233 1,35 CASHD 56 0,2386 0,61 0,14 26,0 H4,1 0,4051 1,19 0,21 40,0 H4,1 42 0,0964 0,62 0,1988 1,20 TSD 54 0,2902 0,70 0,25 3,7** a H5,0 0,5588 1,33 0,54 2,1** a H5,0 44 0,0396 0,46 0,0196 0,93 ACQTTTD 50 0,1399 0,45 -0,08 54,2a H6,1 0,3205 1,20 0,01 97,4 H6,1 48 0,2171 0,75 0,3127 1,19 ABNR1D 56 0,2932 0,65 0,27 3,1** H7,0 0,4927 1,28 0,41 9,2* H7,0 42 0,0238 0,54 0,0820 1,04
*Significance at the 10% level, **Significance at the 5% level and ***Significance at the 1% level.
4.2.1 Domestic returns versus Cross border returns
In table 6, the CAAR of the domestic deals of the three acquiring countries together is 25,38% for the 2-years period (first group of LOCALD) and only 4,67% (CAAR) for the cross-border deals (second group of LOCALD). The mean difference of LOCALD (21%) is significant at the 10% level.
The mean return of the 5-years period is 46,86% for domestic deals and 5,50% for cross-border deals. For the 5-years period the mean difference of 41% is also significant at the 10% level. It can be seen that the mean difference of LOCALD for the 5-years period is more significant compared to the mean difference for the 2-years period (5,7% and 7,5% respectively).
The Levene’s T-test for equality of Variances12 found that the variances of LOCALD may not be assumed equal because the significance of the Levene’s test, which tests the equality of the variances, is 3%. Therefore, the significance values of the Independent Sample T-test with unequal variances assumed are used to examine the mean difference of LOCALD.
Based on these univariate results and a significance-level of 10%, it can be said that the long-term CAARs of domestic deals are significantly different compared to the long-long-term CAARs of the cross-border deals made by the industrial acquirers. Therefore, H2,0 should not be rejected but
the alternative hypothesis is.
However, based on the 5% significance-level, the mean differences of the 2-years and 5-years periods are not significantly different. In this case, the H2,0 should be rejected and H2,1 will be
accepted.
4.2.2 Frequent acquirers versus less frequent acquirers
According to Cools and Van der Laar (2006) acquisitions drive performance. However, the univariate independent sample T-test shows that the more frequent acquirers group has a less positive return than the less frequent acquirers group (9,52% and 23,72% respectively) for the 2-years period. The mean difference of the 2 groups is not significant at the 10% level. The mean difference of the 5-years period is even worse. The less frequent acquirers group has a 25% higher CAAR compared with the more frequent acquirers group. However, the result is even more insignificant (30% > 26,3%) compared with the 2-years period significance.
There is no evidence found that there is a significant difference between the returns of more or less frequent acquirers in the 2-years and 5-years periods. Therefore, the H3,0 hypothesis should
be rejected and its alternative (H3,1) should be accepted. These findings contradict the findings of
Cools and Van der Laar (2006). More frequent acquirers have no higher long-term returns than less frequent acquirers.
4.2.3 Cash versus other methods of payment
As can be seen in table 6, 56 of the 98 acquisitions have been paid with cash. The mean of the first group, cash payments only, is 23,86% for the 2-years period. The mean of the second group, other payments than cash, is only 9,65%. The mean difference of 14% of the 2 groups is insignificant at the 10% level.
The 5-years period CAAR is similar to the results of the 2-years period and also very insignificant. However, the CAAR of the cash payments is 40,51% which is more than two times as positive as the CAAR of the group with other payments as cash (19,88%).
Because of the lack of significance found for the 2-years period as well as the 5-years period, the H4,0 hypothesis should be rejected and the H4,1 hypothesis should be accepted. There is no
evidence found that acquirers, paying acquisitions with cash, are performing better in the long run than acquirers paying in shares, debt or different.
4.2.4 Industrial targets versus non-industrial targets
Matching the acquirer’s industry with the industry of the target with the same ISIN category (in this thesis the ISIN 2000 sector) resulted in 54 acquisitions (first group) of the acquirers in the same industry and 44 acquisitions (second group) in different categories. The CAAR of the acquirer in the same industry for the 2-years period is 29,02%. The CAAR of the second group has a CAAR of only 3,97%. The mean difference of these groups is 25%, which is significant at the 5% level.
The 5-years CAAR for the first group is higher than the CAAR of the 2-years period. Moreover, the 5-years CAAR of the second group are even worse compared to the 5-years CAAR. The mean difference of the 5-years period is also more significant than the 2-years period (2,1% < 3,7%). Both CAARs are significant at the 5% level, therefore the H5,0 hypothesis should not be
acquirers taking over targets in the same industry category perform better in the long run than industrial acquirers taking over targets in non-related sectors.
4.2.5 Large acquirers versus small acquirers
The first group of ACQTTTD represents the acquirers with a turnover which is more than 10 times as large as the turnover of the target. These acquirers are the (relative) large acquirers. The second group consists of acquirers with a turnover smaller than 10 times and is called the (relative) small acquirers.
The mean difference of the 2 groups for the 2-years period is negative (-8%), indicating that the CAAR of large acquirers is worse than the CAAR of the small acquirers. However, the mean difference is not significant at the 10% level (54,2%).
For the 5-years period the CAARs for both group are almost similar, the mean difference is only 1%. This indicates that the relative size of the acquirer has little impact on the long-term performance of acquirers. The significance of the mean difference level is 97,4%, which is far from significant.
Based on these univariate findings for the 2-years and 5-years periods, the H6,1 hypothesis should
be accepted and the H6,0 hypothesis should be rejected. It can be said that relative large acquirers
do not have significant lower abnormal long-term returns than smaller acquirers.
4.2.6 Positive versus negative short-term post-announcement returns
It is expected that the positive short-term post-announcement abnormal returns, with an event window of t+1 week (ABNR1), leads to a higher CAARs for the 2-years and 5-years periods. The first group of ABNR1 consists of all deals with a positive short-term post-announcement abnormal return. The CAAR for the 2-years period is 29,32% which is substantially higher than the CAAR for the group with negative post-announcement abnormal returns (2,38%). The mean difference is positive (27%) and significant at the 5% level (3,7%).
For the 5-years periods the mean difference increased to 41%. However, this difference is not significant at the 5% level but at the 10% level (9,2%). This result suggests that the impact of the short-term post-announcement abnormal returns becomes less significant over time.
The H7,0 hypothesis should be accepted and H7,1 should be rejected for the 2-years period. It can
the 5-years period this relation is weaker. Based on a confidence-level of 90%, the H7,0 should be
accepted and its alternative rejected.
4.3 Multiple cross-section regression analysis
This subsection shows the regression results of the 6 independent variables (LOCALD, ACQF, CASHD, TSD, ACQTTT and ABNR1) as a determinant for the 2 long-term dependent variables (ABNR2 and ABNR5). Differences between the first and the second (control) regression analyses are discussed in 4.4. In table 7, B and Sig. represent the regression coefficients of the independent variables and their significance, respectively. The adjusted R2, is the alternative statistic of the R2 statistic (which measures goodness of fit of a model) and adjusts for the extra number of varia-bles included in the model. Adjusted R2 only increases if the extra variable improves the model more than would be expected by chance.13
Table 7: Results of the cross-section regression analyses
N=98 FIRST REGRESSION N=98 SECOND REGRESSION
ABNR2 ABNR5 ABNR2 ABNR5
Independent
Variables B Sig. B Sig. Independent Variables B Sig. B Sig. constant -0,132 0,358 -0,242 0,375 constant -0,195 0,272 -0,450 0,194 LOCALD 0,222 0,083* 0,477 0,052* LOCALD 0,200 0,118 0,419 0,094*
ACQF -0,025 0,529 -0,081 0,280 ACQFD -0,095 0,445 -0,188 0,441 CASHD 0,073 0,560 0,048 0,840 CASHD 0,090 0,466 0,104 0,666 TSD 0,227 0,069* 0,465 0,050** TSD 0,195 0,116 0,464 0,056*
ACQTTT 1,07E-007 0,363 4,45E-007 0,049** ACQTTTD -0,029 0,814 0,115 0,633 ABNR1 1,106 0,112 1,468 0,267 ABNR1D 0,247 0,048** 0,361 0,137 Adjusted R2
0,054 0,087 Adjusted R2 0,062 0,052
*Significance at the 10% level, **Significance at the 5% level and ***Significance at the 1% level.
variable is significant at the 10% level in the univariate test and remains also significant at the 10% level for the cross-section regression analysis for the long-term periods.
The regression coefficient of the ACQF variable is relative small and negative for the ABNR2 and ABNR5, indicating that the frequency of acquirers has a negative impact on the return of acquirers in the long run. The ACQF as a determinant of long run CAARs is highly insignificant for both the ABNR2 and the ABNR5 variable (52,9% and 28%). These results are in line with the univariate results mentioned in the Pearson Correlation (PC) table in Appendix 1. Based on these findings, it can be concluded that there is no relationship between the frequency of the acquirer and the long-term CAARs of the acquirers.
The payment method of acquisitions, represented by the CASHD, has a positive regression coefficient of 0,073 for ABNR2 and 0,048 for ABNR5. A higher amount of cash payments in a sample might lead to higher CAARs in the long run. However, the cross-section regression results (as well as the univariate test results) are insignificant at the 10% level (56% and 84%). Therefore, it can be said that cash payments have almost no impact on the CAARs of the acquirer of the firm in the long run.
The determinant for horizontal acquisitions (TSD) is positive and significant at the 10% level for ABNR2 (6,9%) and even more significant for ABNR5 (5%). The regression coefficient becomes more positive for ABNR5 in comparison with ABNR2, 0,465 and 0,227 respectively. These findings imply that the sector of the target, chosen by the acquirer, is important for the CAAR of the acquirer. Univariate results of the independent variable (TSD) report significance at the 5% level. There is a strong and positive relation between related (matching) sectors in acquisitions and the returns of the acquirers in the long run.
Finally, the importance of the sixth determinant of the long run performance of the acquirer will be discussed. The regression coefficients of ABNR1 (scale) are positive (1,106 and 1,468 respectively). These coefficients indicate that an increase of 1% of acquirers with a positive short-term post-announcement abnormal return might lead to higher CAARs for the long-term period (ABNR2 and ABNR5). However, both coefficients are, based on a confidence-level of 90%, insignificant in the first regression. The univariate results, mentioned in the PC-table in Appendix 1, are similar regarding the significance and the signs of the coefficients. Because of no significance found it can be said that there is no relation between the short-term announcement CAAR of the acquirer and the long-term CAARs of the acquirer. The determinants of the model declare only 5,4% of the variance of ABNR2. This percentage increases for ABNR5 to 8,7%. Based on the relative low adjusted R2 of the first cross-section regressions (for ABNR2 and ABNR5), it can be said that there are also other explanatory factors of the long-term CAAR of the acquirer which are not examined in this thesis.
4.4 Second cross-section regression results
This regression is made to control for robustness of the variables. The ACQF, ACQTTT and ABNR1 will be replaced by the ACQFD, ACQTTD and ABNR1D dummy variables respectively. These variables are used in the univariate tests.
Replacing the variables in the second regression did not lead to substantial changes in significance compared to the first regression, except for ABNR1D. In the second regression ABNR1D is significant for the 2-years CAAR at the 5% level (4,8%), assuming that there is a relation between the short-term post-announcement abnormal return and the 2-years CAAR. This finding is different from the first regression, indicating that differences in measurement can influence significance. However, these results are in line with the univariate results.
5 CONCLUSION
The main findings of this thesis are that Western European Industrial acquirers have significantly higher long-term returns, measured in CAARs, than industrial companies in the Dow Jones Western Europe Industrial Index if their first acquisition is made in the 1/1/2000-31/12/2001 period. The 2-years and 5-years CAAR are respectively, 17,77% and 31,67%, which are both significant at the 1% level. These results are in line with researchers who report positive long-term returns (Cools and van der Laar (2006) and Loughran and Vijh (1997)).
The long-term CAARs for domestic deals made by acquirers and the sector of the target are significant in the univariate test and remain significant in the cross-sectional regression analysis. These characteristics do influence the performance of the acquirer in the long-term periods positively.
The indicator for the short-term post-announcement performance is positively related, as expected, to ABNR2 and ABNR5 and is significant in univariate tests, however in the cross-sectional regression its significance diminished. Only for the 2-years period, the short-term post-announcement period remains significantly positive (second regression). It seems that the short-term post-announcement performance can be used as a warning instrument for long-short-term oriented investors when they select their investments.
The experience of the acquirer has no positive sign as expected and is insignificant for the long-term periods. These findings indicate that in the long run a more frequent industrial acquirer cannot outperform a less frequent industrial acquirer. The method of payment has the expected positive sign, but is not significant. Therefore, it cannot be said that the experience of the acquirer as well as the payment method are relevant as a determinant for the long-term performance of the acquirer in this paper. An insignificant method of payment is in contrast with other researchers who found that cash had a significant positive impact on the returns of an acquirer in the long run (Loughran and Vijh, 1997). The difference in results might be caused by the sample period and methodology used.
Based on the most striking findings of this thesis, I conclude that the acquisitive strategy is profitable for industrial acquirers in the long run and even more profitable if the acquisitions are made in the same sector and in the same country as the acquirer. Enlarging market share by taking over other companies is most profitable if the acquirer has sufficient knowledge about the market of its targets. It is obvious to see that acquirers have more knowledge about their domestic market than cross-border markets, which is confirmed by this thesis.
This thesis has several limitations. Firstly, it must be reckoned that the sample is only based on industrial acquirers and that the sample has only 98 observations. This number of observations is relative small and consists only of complete acquisitions. Furthermore, there is no separation made between different types of acquisitions. Therefore the stated conclusions cannot be used in general.
REFERENCES
Agrawal, A. and J.F. Jaffe, 1999. The Post-merger Performance Puzzle. University of Alabama and University of Pennsylvania, mimeo.
Agrawal, A., J.F. Jaffe and G.N. Mandelker, 1992. The Post-merger Performance of Acquiring Firms: A Re-examination of an Anomaly. Journal of Finance, Vol. 47, No. 4, 1605-1621.
Asquith, P., R.F. Bruner, and D.W. Mullins, Jr., 1983. The Gains to Bidding Firms from Merger.
Journal of Financial Economics, 11, 121-139.
Ben-amar, W. and P. André, 2006. Separation of Ownership from Control and Acquiring Firm Performance: The Case of Family Ownership in Canada. Journal of Business Finance &
Accounting, 33(3) & (4), 517–543.
Chatterjee, R., Aw and M., 2004. The Performance of UK firms Acquiring large Cross-border and Domestic takeover Targets. Judge Institute of Management Studies Research Paper
WP07/00, Cambridge, UK.
Cools, K. and Van der Laar, M., 2006. The Performance of Acquisitive Companies in the US. In
Advances in Corporate Finance and Asset Pricing, Elsevier, Amsterdam.
Cullen, J.B. and Baker, D.D., 1984. Administration Size and Organisation Size: An Examination of the Lag Structure. Academy of Management Journal, 27, 644-653.
Digman, L.A., 1986. Strategic Management. Concepts, Decisions, and Cases. Tex. Business
Publications, Plano.
Eckbo, B.E. and Thorburn, K.S., 2000. Gains to Bidder Firms revisited: Domestic and Foreign Acquisitions in Canada. Journal of Financial and Quantitative Analysis 35 (1), 1-25.
Faccio, M., J. McConnell and D. Stolin, 2006. Returns to Acquirers of Listed and Unlisted Targets. Journal of Financial and Quantitative Analysis 41, 197-220.
Fowler, K.L. and Schmidt, D.R., 1988. Tendor Offers, Acquisitions, and Subsequent Performance. Academy of Management Journal 31, 962-974.
Hansen, R.G., 1987. A theory for the choice of Exchange Medium in Mergers and Acquisitions,
Journal of Business 60, 75-95.
Halpern, P., 1983. Corporate Acquisitions: A Theory of Special Cases? A Review applied to Acquisitions. Journal of Finance 38, 297-317.
Hong, H. and M. Shum, 2002. Increasing Competition and the Winner's Curse: Evidence from Procurement. The Review of Economic Studies, Vol. 69, No. 4. (Oct., 2002), 871-898.
Kleinert, J. and Klodt, H., 2002. Causes and Consequences of Merger Waves. Kiel working paper No. 1092, Kiel.
Lenz, R.T., 1981. Determinants of Organizational Performance. An Interdisciplinary Review.
Strategic Management Journal 2, 131-154.
Loughran, T. and A.M. Vijh, 1997. Do Long-Term Shareholders Benefit From Corporate Acquisitions? The Journal of Finance, Vol. 52, No. 5, 1765-1790.
Lubatkin, M.H., 1983. Mergers and the Performance of the Acquiring Firms. Academy of
Management Review 8, 218-225.
Lubatkin, M.H. and Shrieves, R.E., 1986. Towards reconciliations of market performance measures to strategic management research. Academy of Management Review 11, 497-512. MacKinlay, A.C., 1997. Event Studies in Economics and Finance. Journal of Economic
Literature 35, 13-39.
Martynova, M. and Renneboog, L., 2006. Mergers and Acquisitions in Europe, Discussion Paper,
Center for Economic Research, Tilburg University, Tilburg.
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 29, 533-564.
Mueller, D.C., 1969. A Theory of Conglomerate Mergers. Quarterly Journal of Economics 83, 644-660.
Scherer, F.M. and Ross, D., 1990. Industrial Market Structure and Economic Performance. Third edition. Boston: Houghton-Mifflin.
Seth, A., K. P. Song and R.R. Pettit, 2000. Synergy, Managerialism or Hubris? An Empirical Examination of Motives for Foreign Acquisitions of U.S. Firms. Journal of International
Business Studies, Vol. 31, No. 3. (3rd Qtr., 2000), 387-440.
Ravenscraft, D. and Scherer, F.M., 1987. Life after takeover. Journal of Industrial Economics 36(2), 147-156.
Weston, J.F. and Chung, K.S., 1983. Some Aspects of Merger Theory. Journal of the Midwest
APPENDIX 1
Pearson Correlation (PC) table
ABNR2 ABNR5 LOCAL ACQF ACQFD CASHD TSD ACQTTT ACQTTTD ABNR1 ABNR1D ABNR2 PC 1 ,719*** ,163 -,079 -,114 ,115 ,203** ,103 -,063 ,156 ,218** Sig. ,000 ,109 ,438 ,263 ,260 ,045 ,315 ,538 ,125 ,031 N 98 98 98 98 98 98 98 98 98 98 98 ABNR5 PC ,719*** 1 ,168* -,127 -,106 ,086 ,226** ,203** ,003 ,103 ,171* Sig. ,000 ,098 ,214 ,300 ,400 ,025 ,045 ,974 ,311 ,092 N 98 98 98 98 98 98 98 98 98 98 98 LOCAL PC ,163 ,168* 1 ,024 -,083 ,024 ,036 -,133 -,154 -,039 -,061 Sig. ,109 ,098 ,811 ,415 ,811 ,728 ,191 ,130 ,701 ,550 N 98 98 98 98 98 98 98 98 98 98 98 ACQF PC -,079 -,127 ,024 1 ,711*** -,127 -,075 -,061 ,171* ,042 ,098 Sig ,438 ,214 ,811 ,000 ,214 ,463 ,550 ,092 ,679 ,335 N 98 98 98 98 98 98 98 98 98 98 98 ACQFD PC -,114 -,106 -,083 ,711** 1 -,102 -,066 -,086 ,128 -,031 -,018 Sig. ,263 ,300 ,415 ,000 ,320 ,517 ,400 ,211 ,760 ,861 N 98 98 98 98 98 98 98 98 98 98 98 CASHD PC ,115 ,086 ,024 -,127 -,102 1 ,089 ,088 ,018 ,118 ,083 Sig. ,260 ,400 ,811 ,214 ,320 ,384 ,389 ,863 ,246 ,415 N 98 98 98 98 98 98 98 98 98 98 98 TSD PC ,203** ,226** ,036 -,075 -,066 ,089 1 ,092 -,064 -,033 ,130 Sig. ,045 ,025 ,728 ,463 ,517 ,384 ,370 ,533 ,749 ,201 N 98 98 98 98 98 98 98 98 98 98 98 ACQTTT PC ,103 ,203** -,133 -,061 -,086 ,088 ,092 1 ,100 ,050 ,088 Sig. ,315 ,045 ,191 ,550 ,400 ,389 ,370 ,329 ,625 ,389 N 98 98 98 98 98 98 98 98 98 98 98 ACQTTTD PC -,063 ,003 -,154 ,171 ,128 ,018 -,064 ,100 1 -,054 ,018 Sig. ,538 ,974 ,130 ,092 ,211 ,863 ,533 ,329 ,596 ,863 N 98 98 98 98 98 98 98 98 98 98 98 ABNR1 PC ,156 ,103 -,039 ,042 -,031 ,118 -,033 ,050 -,054 1 ,710*** Sig. ,125 ,311 ,701 ,679 ,760 ,246 ,749 ,625 ,596 ,000 N 98 98 98 98 98 98 98 98 98 98 98 ABNR1D PC ,218** ,171* -,061 ,098 -,018 ,083 ,130 ,088 ,018 ,710*** 1 Sig. ,031 ,092 ,550 ,335 ,861 ,415 ,201 ,389 ,863 ,000 N 98 98 98 98 98 98 98 98 98 98 98
APPENDIX 2
Number of Observations Acquiring
countries Number of deals acquisition Type of 2-years CAAR Sign. 5-years CAAR Sign.
United Kingdom 47 Domestic 21,6% 0,035** 37,28% 0,044**
9 Europe -1,76% 0,894 11,41% 0,567 15 US 13,7% 0,330 15,78% 0,384 Germany 4 Domestic 74,35% 0,235 157,15% 0,395 3 Europe -20,41% 0,161 -32,89% 0,245 France 11 Domestic 23,63% 0,199 47,69% 0,122 5 Europe 19,53% 0,502 21,44% 0,759 4 US -14,5% 0,143 -37,48% 0,193
