The Wealth Effect of Acquiring Firm M&A Announcements

(1)

The Wealth Effect of Acquiring Firm M&A

Announcements

An Event Study of Fast Moving Consumer Goods Industry Acquirers from Developed Markets

University of Groningen Faculty of Economics and Business

MSc International Business & Management – specialization in International Financial Management

Author: David Sipkens

(2)

Abstract

In this paper the effect of cross-border merger and acquisition announcements on acquiring firm’s shareholder wealth in the fast moving consumer goods (FMCG) industry is being studied. The sample is comprised of 124 M&As which were announced in period between January 2003 and August 2008. This empirical research is conducted by making use of event study methodology. Results are based on cumulative abnormal returns over a 41-day event window.

This study has led to the conclusion that cross-border M&As in the FMCG industry lead to significant positive abnormal returns for acquiring firm’s shareholders. Hence, cross-border M&As can be seen as a value creating strategy for international growth. The studied research variables; market integration, firm diversification and payment method were found to have no significant influence on abnormal returns. This study also tests the effect of two control variables on the research variables. The control variable firm size does not influence the research variables. However, the research variables market integration and firm diversification have shown to be significantly influenced by the fact whether the target firm is publicly quoted or privately held.

Keywords: cross-border mergers and acquisitions, fast moving consumer goods industry,

(3)

1. Introduction

Cross-border mergers and acquisitions (M&As)1 are a popular strategy for firms to expand internationally. Mergers and acquisitions occur in cyclical waves (Goergen and Renneboog, 2004). During the late 1990’s there has been a rapid increase in the number of Cross-border M&As, which are part of the “fifth M&A wave” that took place between 1993 and 2000. Again, in the recent period starting in 2003 a M&A wave has emerged.

Different reasons could lay the foundation for a firm to decide to expand internationally through M&As. Some of these reasons could be economies of scale, diversification, increased market power, and tax benefits. However, following the neo-classical theory of the firm, firms should be motivated by competitive market forces to maximize shareholder wealth. Taking this into account, firms should only engage in M&A activities if it increases shareholder wealth for the acquiring firm (Manne, 1965). This is however a point of discussion in the literature, because researchers far from agree on the effect that cross-border M&As have on the wealth of acquiring firm’s shareholders.

In this study, the empirical evidence will be presented on the shareholder wealth effect of cross-border M&As during the current M&A wave. More specifically, cross-border M&As within the fast-moving-consumer-goods (FMCG) industry will be studied. The FMCG industry consists of manufacturers of non-durable consumer goods with an expected lifespan of one year or shorter. Examples of products produced within this industry are food, beverages, tobacco, toiletries, and cosmetics. As will be explained later in more detail, studying FMCG industry’s cross-border M&As is interesting, because no such research has been conducted earlier for this industry. Besides, since the last decade a strong concentration within de industry due to cross-border M&As is noticeable.

(5)

The focus of this paper will be on the shareholder wealth effect of the acquiring firms, because this effect remains unclear in previous literature. In doing so, one of my goals will be to see if any differences exist in the acquiring firm shareholder wealth effect between cross-border M&As to developed markets and to emerging markets. Furthermore, I will study some aspects of the M&A which could influence the shareholder wealth effect and should therefore be included in this research, such as the diversity of the acquired M&A targets, and the way in which the M&As are paid for. This will give a better understanding in the end of how shareholder wealth is being affected.

The main research question of this paper will be:

“How does the announcement of cross-border M&As in the FMCG industry affect the acquiring firm’s shareholder wealth”

(6)

2. Research framework

This paper intents to investigate acquiring firm’s shareholder wealth effects of cross-border M&As within the FMCG industry through an event study of stock price reactions to M&A announcements. In addition, this paper tries to contribute to the discussion of causes for value creation or destruction of cross-border M&As.

Many of the previously written papers on M&As analyze shareholder wealth effects of acquisition transactions from before the 1990s or from the fifth wave which started around 1995. Contrary to these papers, this paper will focus on M&As which are part of the most recent M&A wave. Figure 1 shows yearly worldwide M&A deal volumes since 1980. This graph clearly shows the start of the latest M&A wave in 2003.

Figure 1: Wordwide M&A deal volumes

Source: www.worldpress.com (adjusted version)

(7)

transactions. Doukas and Travlos (1988) and La Porta et. al. (2000) also find that cross-border M&As are value enhancing for acquirers. Other studies that also incorporate non-US acquiring firms in their sample do not support these findings. For instance Brealey et. al. (1998) have studied a more recent sample of world-wide cross-border M&As. As a result of their research they find significant gains for acquired firms but only negligible abnormal returns for acquirers. Moeller and Schlingemann (2005) and Datta and Puia (1995) even find that cross-border M&As tend to destroy value for shareholders of acquiring firms. Moreover, a study of KPMG found that only 17% of cross-border M&As created shareholder value, while approximately 53% destroyed it (Economist, 1999). As can be seen from the examples of previous conducted research, there is no consensus on the effects of cross-border M&As on the firm value of acquiring firms. Overall, the results indicate that cross-border M&As are not highly successful.

In contrast to the studies described above which analyzed the fourth or fifth merger waves, Moeller (2006) recently studied a sample of 1400 companies which were involved in M&A transactions of the current (deals which are closed since 2003) M&A wave. They found that deals in the current M&A cycle are creating shareholder value in the short term. By contrast, as a result to their previously conducted analyses concerning the merger cycles of the late 1980s and the late 1990s they found that shareholder value was destroyed. The analyzed success of the latest cycle is global; but, Asian deals appeared to be more successful than European ones. Moeller (2006) argues that the difference between recent M&As and those from earlier waves is that companies are now seeking advice from outside and inside the company on how to do deals better.

(8)

being revealed by the change in stock price due to information contained in the acquisition announcement.

2.1 Development of hypotheses

In the following part of the research framework I will go deeper into the development of several hypotheses. For every hypothesis I will start by giving theoretical arguments based on underlying literature which support the hypothesis.

2.1.1 Industry returns

All earlier mentioned studies have one thing in common, which is the focus on multiple industries in studying shareholder wealth effects. However, without making distinctions between certain industries it is possible to overlook relevant issues. Some industries could benefit more form cross-border M&As as a strategy to expand internationally than others. Hence, this research distinguishes from these studies by focusing on the FMCG industry only.

(9)

positive for M&As in the FMCG industry. Based upon the above discussion the following hypothesis has been formulated:

Hypothesis 1:

Cumulative abnormal returns around the announcement of a cross-border M&A in the FMCG industry are positive.

2.1.2 Market integration

Cross-border M&As often involve different markets. In the specific case of this study two different markets can be separated. While M&As with a developed market2 bidder are being studied, the targets are sited in both developed and emerging markets3. The term emerging markets is often defined vaguely and therefore needs to be explained more extensively. The World Bank uses the definition of countries with low to middle per capita income, where companies can seek for worthwhile medium to long term investment opportunities, because of their potentially dynamic and fast growing economy. The Morgan Stanley Emerging markets index is being used in this study, in order to get an exact list of these markets. This list includes countries in Asia, Latin-America, Eastern-Europe and some African countries.

An essential difference between developed and emerging target markets is the degree of integration. In his research from twenty-four national markets Korajczyk (1996) indicates that segmentation tends to be much larger for emerging markets than for developed markets. Danbolt (1995) and McCann (2001) conclude that the degree of capital market integration may have an impact on the emergence of abnormal returns to acquiring firm’s shareholders. However, they leave the way in which it affects abnormal returns unknown. From a theoretical point of view, Bjorvatn (2004) argues that due the to “business stealing effect”, mergers in integrated markets are not likely to be profitable. By his opinion it is therefore a reasonable conjecture that closer integration of markets would reduce the attractiveness of cross-border mergers and acquisitions.

(10)

An important period concerning market integration in emerging markets worth mentioning is the period between the late 1980s and the early 1990s. This period saw a liberalization of financial markets in emerging market countries; thereby, these countries allowed foreign firms to acquire domestic firms. Although normally liberalization is synonymous with integration, these countries were at best only partly integrated into world capital markets (Francis et.al., 2008). Therefore, despite liberalization tendencies, these markets are still much more segmented than developed markets nowadays. Based upon the above described market integration effects, a reasonable assumption will be that M&A deals between a bidder and a target from less integrated markets lead to higher abnormal returns than M&A deals between two parties from integrated markets. Based on this the following hypothesis has been formulated:

Hypothesis 2:

Higher cumulative abnormal returns are expected from M&As to emerging markets than from M&As to developed markets in the FMCG industry.

2.1.3 Firm diversification

Managers of companies involved in M&As often argue that they strive to create synergy by acquiring another company. Synergy could be defined as the idea that the value and performance of two companies combined, through for instance a M&A, will be greater than the sum of the two individual companies. In this research, synergy will be taken into account by studying the effects of whether the target companies are from a related or an unrelated industry.

(11)

products. Another reason is that a diversification strategy can result in growth. However, M&As with targets from other industries are expected to generate less synergy. Less synergy is expected because synergy estimations are based on value creation due to the fact that similarities exist between the characteristics of the acquiring firm and the target firm (Homberg et. al., 2008). In addition to synergy effects, M&As between unrelated firms are also expected to be riskier, because acquirers have a lack of experience in skills and techniques in the unrelated target industry. Therefore, is assumed that related M&As with targets from within the FMCG industry are valued higher by shareholders and will consequently generate a higher cumulative abnormal return around the announcement of such M&As. Based on this the following hypothesis has been formulated:

Hypothesis 3:

Higher cumulative abnormal returns are expected for cross-border M&A announcements of firms which are related than for announcements of unrelated M&As.

2.1.4 Method of payment

(12)

expected for M&As paid in cash. Based on this the following hypothesis has been formulated:

Hypothesis 4:

Higher cumulative abnormal returns are expected for cross-border M&A announcements of M&As which are paid in cash than those paid in shares.

By accepting or not rejecting the above stated hypotheses I will attempt to give an answer to the main research question. By studying different factors which are of great importance in cross-border M&As, I will be able to conclude how these different factors like market integration, firm diversification, and payment method influence the shareholder wealth effect of acquiring firms. The conclusions which I will draw regarding the four hypotheses will together give an answer to the way in which the shareholder wealth effect is being influenced due to announcements of cross-border M&As in the FMCG industry.

2.2 Control variables

Having developed three variables on which will be tested on in this research, some control variables will be used in this paper as well in order to see whether the research variables described in the hypotheses above are heavily influence by certain factors. In other words, these control variables can explain differences in why some companies do receive positive shareholder wealth effects on a certain variable and others receive a zero or negative return.

2.2.1 Firm size

(13)

large bidding firms. This effect is could be explained by the fact that large firms offer larger premiums compared to small firms. Furthermore, there is a theory that managerial hubris plays a more significant role in large firm decisions. Firm size will be measured in this research by total enterprise value on the balance sheet of the year preceding the M&A. Enterprise value is calculated as the firm’s market capitalization plus minority interest, preferred shares and dept, minus total cash and cash equivalents. Enterprise value is chosen as a measure for firm size, because it is considered more accurate than looking at market capitalization only.

2.2.2 Public or private targets

The second control variable used in this research is the fact whether targets are publicly quoted or privately held. In this research both public and private targets are included in the sample. Conn et. al. (2005) conclude in their paper on over 4,000 acquisitions of UK public acquirers, that a difference exist between cross-border M&As with public and privately held firms. Positive abnormal returns were found in case of privately held firms, while shareholder wealth effects of publicly held targets resulted in zero announcement returns. Therefore, the difference between public and private targets will be used as a control variable in this research.

Based on the hypotheses described above in combination with the control variables the following conceptual model has been composed:

Figure 2: Conceptual model

Market integration Developed/Emerging market Payment method Paid in cash or in shares Diversification Related or unrelated industry target

Stock price around the announcement of FMCG industry

M&As

Acquiring firm size

= positive or negative effect

(14)

3. Data

Having discussed relevant literature and formulated hypotheses, this paper will now proceed by giving insight in the data collection process for this research. Two databases are used to gather the data included in this research, which are the Zephyr database and Datastream. In the upcoming part, the process of collecting data by making use of these databases will be elaborated on more extensively.

3.1 Zephyr database

The start of this research was made by gathering a sample of M&A cases which fit the criteria for this specific paper. In order to find these cases, the ZEPHYR4 database was used. The database provides possibilities to specify a sample according to specific needs. In order to get to the final sample of 189 M&A cases which were drawn from the Zephyr database, a list of restrictions had to be composed. This list consists of the following restrictions; firstly, all M&As had to be cross-border M&As. Secondly, only developed market acquirers included in the Morgan Stanley Developed Market Index were allowed in the sample and the targets had to be from either emerging markets or developed markets and also needed to be included in one of the Morgan Stanley Developed/Emerging Market indices. Thirdly, the acquirers had to be FMCG manufacturers. Obtaining a sample which solely consists of FMCG industry companies was difficult, because a lot of the fast moving consumer goods (e.g. chemical household products like soap) are produced in sectors which also include many non-FMCG producers. Nonetheless, a division based on industry sectors had to be made. Therefore, a small definition of the FMCG industry was used in this research, which includes the food, beverage and tobacco manufacturing industry (UK SIC 15 and 16). The next restriction was that all acquirers had to be quoted companies, because acquiring firm’s shareholder wealth effects are studied in this paper. Another restriction was that the M&As were paid

4

(15)

for in either shares, cash, or a mix of both. Finally, the last restriction was that all M&A announcements should have taken place in 2003 or later, because only M&As of the most recent wave are being considered in this research. Based on these restrictions a sample has been formed of 189 cross-border M&As within the FMCG industry.

Because the Zephyr database contains additional information on companies involved in M&As as well, the database was also used to find information on whether the target firms operated in related or unrelated industries. Also information about firm size, and the fact whether targets are public or private could be gathered from Zephyr.

After having found a preliminary sample of 189 cases, all mergers and acquisitions of the same company which occurred within 40 working days from each other were excluded from the sample. Excluding these M&A cases was necessary in order to make sure that no confounding effects are being found as a result of an overlap in event windows.

3.2 Datastream

After the initial sample had been drawn from the Zephyr database and was adjusted for overlaps, the corresponding stock prices around the announcement dates for each company had to be found. The stock price information has been derived from the Datastream5 database.

Through Datastream stock prices for a large part of the sample could be found. For some companies the stock prices could not be found by Datastream. Therefore, local stock markets were consulted to find information on these missing companies. Furthermore, stock information on some companies could not be found by Datastream, because these firms had been acquired themselves after being involved as an acquirer in an earlier M&A. The final research sample of which all available stock prices have been found consists of 124 M&A cases. Besides event window stock prices, also information

(16)

concerning stock indices which are used for the estimation window has been deducted from Datastream.

3.3 Sample characteristics

In this paragraph the characteristics of the final sample which consists of 124 cross-border M&A cases in the FMCG industry will be discussed shortly. Table 1 shows the distribution of M&As over the different years of the latest M&A wave.

Year 2003 2004 2005 2006 2007 2008 Total Number of M&As 17 31 30 21 19 6 124 Table 1: Distribution of M&A cases over different years

Table 2 shows which portion of the sample has a certain variable characteristic. The sample portions are rather equally divided over the first variable, which indicates whether the M&A targets are developed market or emerging market companies. Since this is the main variable of this research, this was the first search characteristic during the data collection process. Furthermore, the majority of the 124 cases consists of companies involved in M&As with a target company active in a related industry. Moreover, 93% of the M&As involve in this sample was paid in cash and just a small minority was paid in shares. Therefore, the earlier described preference of paying in cash is supported. The control variables are also shown in table 2. The sample consists for roughly two third of large acquiring companies. 19% of the targets were listed on a stock index. The other 81% were private companies.

Control variables

Total deals Divided into

(17)

4. Methodology

After having described how the data for this research has been collected, I will now elaborate more on the methodology used in this research. In the next paragraphs the fundamentals of the event study model will be explained, followed by the explanation of specific choices which had to be made for the particular event study model of this research.

4.1 Efficient Market Hypothesis

Before going deeper into the specific event study methodology it is important for a paper like this to elaborate some more on the fundamentals of an event study. Therefore, attention will be paid to the Efficient Market Hypothesis (EMH). The EMH is an important concept which has become widely accepted in the economic literature since the late 1950s. The hypothesis expects a market to be efficient with respect to a certain information set, if it is impossible to derive economic gains by trading based on this information set (Jensen, 1978). Moreover, the Efficient Market Hypothesis is based on the assumption that stocks are in equilibrium at all times, i.e. all relevant available information is incorporated and reflected by the capital market’s total value. Therefore, it is assumed that investors cannot constantly “beat the market”. When new information is available to the market, the EMH assumes that overreaction to this information occurs at the same amount as underreaction, causing the sum effect to reflect the stock price at its intrinsic value (Fama, 1998).

Several versions of the EMH have been discussed in the literature which has led to the construction of three broad forms of the Efficient Market Hypothesis. These are the following:

(18)

(2) The Semi-Strong Form of the Efficient Market Hypothesis suggests that all publicly available information is incorporated into the current stock price. Therefore, also in this form of the EMH no abnormal returns can be obtained by either fundamental or historical analysis. Only investors having information about a stock which is not publicly available can earn an abnormal return.

(3) The Strong Form of the Efficient Market Hypothesis suggests that all information is incorporated and reflected by current stock prices. This includes both publicly available information and private or insider information. According to this form no returns exceeding normal returns can be made.

Event study theory assumes that markets have a Semi-Strong Form Efficiency. Because the Semi-Strong Form Efficiency assumes all new information available in the market to be directly reflected in the share price, this form of the EMH is important for event studies, which measure how newly available information affects stock prices. Moreover, the main assumption of an event study is that in case any useful or surprising information reaches the market an abnormal return will arise.

4.2 The Event Study model

The methodological approach that is used in this research follows the work of MacKinlay (1997). His article forms a broad base for many event studies and clearly explains the different steps one has to undergo carrying out an event study. Also the study of Warner and Brown (1985) which specifically focuses on event studies with daily stock returns has been consulted.

(19)

general procedure for conducting an event study. These steps will be followed more or less in the next sub-paragraphs.

4.2.1 Defining the event window

The final sample of this research consists of 124 events. In figure 3, a timeline for an event study has been given, on which this research will be based. The announcement day of each of the 124 M&As is similar to the event day and will be defined as day zero in this study. It is now necessary to specify a length of observation interval. The studied M&A announcements in this paper take place in one day and therefore the interval is set to one day as well. Consequently, daily stock returns will be used in this research.

Figure 3: Time line for an event study

Having specified the observation interval, the next step is to decide on the length of the total observation. The total observation has to be split up in two parts, respectively the estimation period and the event window. Between the estimation and event window no overlap may occur, because in that case the event returns could influence the normal returns (MacKinlay, 1997). The estimation period in this research lasts from -200 days (τ = T0 + 1) to -20 days (τ = T1), and is denoted as L1. A period of 180 days is about the

average for event studies using daily stock returns (Peterson, 1989). This period will be sufficient in order to make a good prediction.

The event window for this study will be 41 days and will last from day -20 (τ = T1 + 1) to

+20 day (τ = T1), and is denoted as L2. A 41 day event window is also recently used by

Lowinski et. al. (2004) who conducted a study on shareholder wealth effects of cross-border M&As in Switzerland. Some event studies also use shorter event windows. However, effects around the announcement of M&As might arise over a longer period, caused by for instance bidding wars or rejections of an initial offer. Besides, in some cases it might be possible that some pre-announcement information has leaked which also

Estimation window

0

T0 T1 T2

(20)

causes effects to occur over a longer period. Therefore, a somewhat longer event window has been chosen in order to pick up these effects. In figure 4, the specific time line for this event study is illustrated.

Figure 4: Specific event study time line

4.2.2 Abnormal returns

To evaluate the impact of an event, abnormal returns have to be measured. The abnormal return is the actual return of a security over the period of the event window minus the normal return of the company over the same period. The normal return can be defined as the expected return without conditioning on the event taking place (MacKinlay, 1997). For company i on event date τ the abnormal return is

ARiτ = Riτ- E(Riτ|Xτ) [1]

Where ARiτ is the abnormal return, Riτ is the actual return, and E(Riτ|Xτ ) is the normal

return for time period τ. MacKinlay (1997) provides two choices for modeling the normal return, which will be discussed in the next paragraph.

4.2.3 Market model

An analysis of abnormal returns starts by calculating the normal returns. MacKinlay (1997) provides two different approaches in order to calculate normal returns, which are the Constant Mean Return Model, and the Market Model. In this paper, the market model will be used to measure normal performance. The market model is a statistical model which relates the return of any given security to the related market portfolio. The linear specification of the model follows from the assumption that asset returns are jointly normally distributed. The market model is preferred over the constant mean return model,

L1: Estimation L2: Event

(21)

because it removes the proportion of the return that is related to variation in the market’s return. Hence, the variance of the abnormal return is decreased. Consequently, this can lead to increased ability to detect M&A announcement effects.

The Dow Jones Titans Food and Beverage Index will be used as market index in this research. The securities underlying the index are assumed to be jointly normally distributed, independently and identically distributed through time. The market model for any company i is

Rit = αi +βiRmt +εit [2]

Here, Rit is the period-t return on security i, and Rmt is the period-t return on the market portfolio. εit is the zero mean disturbance term. αi and βi are the parameters of the market model. The parameters are estimated through line estimation from the regression Rit on Rmt. Clearly, these are estimated over the estimation window period (180 days). The market model assumes that if the event would not occur, the relationship between the returns of company i and the market index remains unchanged. Furthermore, the disturbance term’s (εit) expected value is zero. In this way, through the use of the market model, the expected normal returns for the event window are calculated. The next step is to calculate the abnormal returns for each day of the event window period.

4.2.4 Average and cumulative abnormal returns

Once the normal returns have been calculated using the market model, it is now possible to measure abnormal returns. The sample abnormal return ARiτ follows from the following formula τ τ τ i αi βi m i R R AR ∧ ∧ − − = . [3] Here i ∧ α and _i ∧

(22)

securities and through time (Campbell et. al., 1997) First, aggregation through time will be considered for an individual security. Therefore, cumulative abnormal return (CAR) will be measured for each company in the sample. The CAR is the sum of all abnormal returns. The cumulative abnormal return of company i in the sample is formulated by the following formula

∑

= = 2 1 ) 2 , 1 ( τ τ τ τ τ τ τ ARi CAR with [4] 2 2 ) 1 1 2 ( ) 2 , 1 ( i i τ τ τ τ σε σ = − + . [5]

Once the CARs have been calculated for all companies in the sample it is now possible to aggregate through time and across securities. Therefore, the average cumulative abnormal return will be measured. The average abnormal return for all companies in the sample is formulated by the following formula

) 2 , 1 ( 1 ) 2 , 1 ( 1 τ τ τ τ

∑

= = N i i CAR N CAR with [6] ) 2 , 1 ( 1 )) 2 , 1 ( var( 1 2 2 σ τ τ τ τ

∑

= = N i i N CAR . [7]

4.2 Statistical tests

Once the average cumulative abnormal returns have been measured it is now necessary to see whether they significantly differ from zero. The concept of perfect capital markets assumes that all securities are positioned on the security market line (SML), because all securities are correctly priced. So, in perfect capital markets no excess returns occur. This explains why in paragraph 4.2.3 the disturbance term’s (εit) expected value is zero. The following hypotheses can be formulated in order to test the significance of average CARs

(23)

These hypotheses can be tested by making use of a one-sample Kolmogorov-Smirnov test. The Kolmogorov-Smirnov is a non-parametric test and is used in this research because the CARs are not normally distributed. In order to use parametric tests (e.g. t-tests) the criteria of normal distribution had to be fulfilled, so testing with parametric test would not be valid in this research. In order to answer hypothesis one, which involves the whole sample, the one-sample Kolmogorov-Smirnov test will be used. However, for testing the research variables two-sample tests need to be used.

The Mann-Whitney test will be used to test if significant differences exist between the research variables included in hypotheses two, three, and four. This test is often used instead of a two-sample t-test, if the criteria for performing a t-test can not be fulfilled.

Besides the fact that the Mann-Whitney test can be used for data which is not normally distributed it has another advantage for this specific research. This advantage is that non-parametric tests like the Mann-Whitney test are more robust for differences in sample size. Since the sample groups of the second and third research variable (diversification and payment method) are not of similar size, testing with a parametric test would have been less valid.

(24)

5. Empirical results

In this chapter the results of the event study performed in this research will be discussed. First, the more general primary results will be discussed, which will be followed by the statistical results of testing the hypotheses this research.

5.1 Primary results

Before going into the statistical results of this research I will discuss the primary results. Figure 5 shows the average 41-day CAR around the announcement of all the M&As of FMCG acquirers included in the sample. The average CAR for the total sample is positive, and has a 41-day CAR of 5,5%. The 41-day CAR was positive for 65% of the sample (80 companies) and negative for the other 35% of the sample. Furthermore, the figure shows that the steepest 3-day CAR increase over the 41-day event window occurs directly around the announcement date (day 0). However, the increase in CAR around day zero is not extreme compared to the increase in CAR over the pre-announcement period. The average CAR increases with almost the same regularity over the pre-announcement period, and the increase is weakening in the post-pre-announcement period.

Average CAR sample (N=124)

0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 Days Return

Avarage CAR sample

(25)

In figure 6 the average CARs for the different years in the sample are shown. A few aspects are worth mentioning. First, the average CARs of M&As which took place in the year 2008 are moving up and down the 0% line. With an average 41-day CAR of 0,002% there is on average no effect on shareholder wealth as a result of FMCG industry M&As in this year. This effect can be explained by the credit crisis which started to affect share prices in 2008. The years 2004 and 2007 also stand out in figure 5, as they show average CARs which are higher than the other years, and have 41-day CARs of 8,1% and 11,9% respectively.

Average CARs for different years

-6.00% -4.00% -2.00% 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 Days Return 2003 2004 2005 2006 2007 2008

Figure 6: Average CARs for every year included in the sample

5.2 Statistical results

(26)

variables. The summarized test results will be presented together with the hypotheses, whereas the complete results can be found in the appendices.

5.2.1 Industry returns

This paragraph will discuss the results concerning the total sample of 124 FMCG industry cross-border M&As. The following hypothesis had to be tested in order to draw inferences about the whole sample:

Hypothesis 1: Cumulative abnormal returns around the announcement of a cross-border

M&A in the FMCG industry are positive.

The summarized results of the performed one-sample Kolmogorov-Smirnov test are listed in table 3. The results show a mean of 5,5%, which is equal to the average 41-day CAR which was already discussed in paragraph 5.1. Since the resulting p-value from this test is 0,000, the null hypothesis

can be rejected with a confidence interval of 95%. Hence, this leads to the conclusion that cross-border M&As in the FMCG industry with a developed market acquirer lead to significant positive cumulative abnormal returns. In other words, M&A transactions in the FMCG industry do create wealth on the short-run for their shareholders. This result is consistent with the work of Moeller (2006) who also found a significant positive shareholder wealth effect for M&As which occurred during the latest M&A wave. As described earlier the link between industry concentration and M&A activity is found to be the strongest for the food and drink industry (Chapman, 2003). This link can also be explained by the significant positive abnormal returns for M&As within this industry. However, this result differs from earlier studies which mainly argue that cross-border M&As result in negligible or sometimes negative abnormal returns for acquirers (Brealey et. al., 1998; Datta and Puia, 1995). The difference with the previous described papers might be explained by the different period in which the M&As took place, and by the fact that this research only includes M&As of FMCG industry firms. The conclusion can be

(27)

drawn that nowadays cross-border M&As form a value creating international growth strategy for companies in the FMCG sector, since they on average affect the shareholder wealth positively.

CAR

N 124

Normal Parameters(a,b) Mean ,05483010

Std. Deviation _,206755770

Most Extreme Differences Absolute ,251

Positive ,251

_Negative _-,160

Kolmogorov-Smirnov Z 2,789

Asymp. Sig. (2-tailed) ,000

Table 3: Results entire sample

5.2.2 Developed vs. Emerging markets

In the next paragraphs the results concerning the research variables will be discussed. The first research variable which will be discussed is market integration. In order to see whether FMCG industry cross-border M&As with targets from less integrated capital markets result in higher abnormal returns the following hypothesis had to be tested:

Hypothesis 2: Higher cumulative abnormal returns are expected from M&As to

emerging markets than from M&As to developed markets in the FMCG industry.

(28)

emerging market target are the highest, this result can not be generalized. The differences in means within this sample should be considered to have occurred by chance.The claim of Bjorvatn (2004), who argues that M&As with developed market targets will lead to lower abnormal returns, cannot be confirmed based on this result, although the sample seems to be in line with his findings. The conclusion should be drawn that shareholders do not value a FMCG company cross-border M&A differently whether the M&A includes a target from an integrated or more segmented financial market. However, measuring the effect of market integration on a larger sample will probably give a higher chance of finding statistically significant results, given the fact that average CARs of emerging market targets are almost twice as high as CARs of developed market targets.

Target market N Mean Rank

CAR Emerging 53 62,83

Developed 71 62,25

Total 124 Asymp. Sig. (2-tailed) ,930

Table 4: Summarized results market integration

Average CARs Developed/Emerging Market Targets

0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 Days Return

Average CAR D-E Average CAR D-D

Figure 7: Average CARs developed vs. emerging market targets

5.2.3 Related vs. Unrelated targets

(29)

Hypothesis 3: Higher cumulative abnormal returns are expected for cross-border M&A

announcements of firms which are related than for announcements of unrelated M&As.

Table 5 lists the result of the Mann-Whitney test which is performed in order to test the significance of this hypothesis. The p-value of 0.966 shows that no significant support has been found for the differences stated in hypothesis 3. Consequently, the hypothesis should be rejected. This means that CARs of M&As with related targets must be assumed equal to CARs of M&As with unrelated targets. The lower expected synergy levels which Homber et. al. (2008) have found to be related to acquiring unrelated targets, do not influence the way in which shareholders value M&As in the FMCG industry. An explanation for the absence of this difference could be that shareholders may also see an advantage in a diversified business portfolio of FMCG firms. The conclusion which can be drawn based on this result is that shareholders do not value announcements of cross-border M&As with related targets significantly higher than those with unrelated targets.

Diversification N Mean Rank

CAR Related target 112 62,54

Unrelated target 12 62,08

Table 5: Summarized results diversification

5.2.4 Payment in cash vs. shares

The third research variable which will be discussed is the influence of the payment method on the existence of abnormal returns in cross-border M&As in the FMCG industry. The following hypothesis is tested:

Hypothesis 4: Higher cumulative abnormal returns are expected for cross-border M&A

(30)

The summarized results of the test are listed in table 6. The p-value of 0,096 indicates that with a 95% confidence interval no significant difference exist between the cumulative abnormal return means of M&As paid in cash and those paid in shares. Consequently, because no clear difference can be determined through this analysis, CARs of M&As paid either in cash or with shares have to be assumed equal. The results of Bruner (2001), who found that shared paid M&As lead to lower shareholder wealth, cannot be supported. However, the result does show a higher mean for M&As paid in shares within this sample. Higher means for M&As paid in shares must therefore be assumed to have occurred by chance.

Payment method N Mean Rank

CAR Cash 115 60,99

Shares 9 81,78

Table 6: Summarized results payment method

5.2.5 Control variables

This paragraph will discuss the influence of the control variables on the sample. The two control variables which were tested are acquiring firm size and the fact whether targets were publicly quoted or privately held firms. The influence of the control variables will first be discussed for the whole sample, after which the influence on the independent research variables will be discussed.

(31)

Control Size N Mean Rank

CAR Large acquirers 86 62,02

Small acquirers 38 63,58

Table 7: Summarized results control variable size on whole sample

In table 7 the results are listed for the effect of firm size on the whole sample. The results indicate that no significant difference in means between small or large acquirers can be found. This means that the returns of the sample as a whole are not being affected by the size of the acquiring firm.

To test whether the size of acquiring firms did have an impact on a single research variable, for instance to see if M&As with emerging market targets have significantly higher abnormal returns if they were performed by small firms instead of large firms, Mann-Whitney test have been performed for every single research variable. The results of these tests can be found in appendix E. Looking at the results it appears that none of tests has a significant outcome. Therefore, the conclusion can be drawn that the size of acquiring firms does not have a significant influence on the research variables which have been tested in this paper.

The second control variable which was tested on the whole sample referred to whether the acquired firms were publicly quoted or privately held firms. The expected difference was that acquiring unquoted targets would lead to higher abnormal returns than acquiring quoted targets. The results for testing this control variable on the whole sample are listed in table 8.

Control Quoted N Mean Rank

CAR Quoted targets 24 76,29

Unquoted targets 100 59,19

(32)

The results show a significant difference between the means of quoted and unquoted targets. However, the effect is different than expected. With a confidence interval of 95% the conclusion can be drawn that M&As with a quoted target result in significantly higher cumulative abnormal returns than M&As with privately held targets. These results are not in line with the in paragraph 2.2.2 discussed results of Conn et. al. (2005), who only found positive abnormal returns for privately held firms. In the current research M&As with quoted and with unquoted targets both result in significant positive abnormal returns, however, those of quoted firms were significantly higher.

This result might be explained by the fact that it is easier for shareholders of acquiring companies to gain insight in the financial records of publicly quoted targets than in financial records for privately held targets. Consequently, acquiring firm’s shareholders might value M&As with targets in which they can gain insight higher than M&As with targets they know much less about.

Also this control variable was tested for every single research variable. Of the six tests which were performed, two resulted in a significant outcome. The summarized results of the significant tests are listed in table 9 and 10. First, it appeared that in case an emerging market target was acquired, the targets which were quoted had a significantly higher cumulative abnormal return than those which were unquoted. However, this was not the case for developed market targets. Therefore, the conclusion can be drawn with a confidence interval of 95% that the fact whether a target is quoted or not has a significant impact on the shareholder returns of M&As with emerging market targets in the FMCG industry. It seems that shareholders value cross-border M&As with emerging market targets significantly higher if these targets are publicly quoted. This result might as well be seen as an unrealized potential which could be used in future managerial decisions concerning potential cross-border M&A targets from emerging markets.

Control Quoted vs. Emerging

Market targets N Mean Rank CAR Quoted Emerging market _targets ₁₁ _35,73

Unquoted Emerging market

targets 42 24,71

Total 53 Asymp. Sig. (2-tailed) ,035

(33)

Secondly, it has been found that if a related target was acquired the CARs were significantly higher if this target was also quoted. Therefore, the conclusion can be drawn with a 95% confidence interval that when a related target is acquired the resulting abnormal returns will be significantly influenced by the fact whether the target is quoted or unquoted.

Control Quoted vs. Related

targets N Mean Rank CAR Quoted Related targets 21 70,19

Unquoted Related targets 91 53,34

(34)

6. Conclusion and future research

In this section the conclusions which are draw from the results described above will be given. In addition, suggestions for future research will be given in the second paragraph of this section.

6.1 Conclusion

In this study the effect of cross-border M&A announcements on acquiring firm’s shareholder wealth in the FMCG industry has been studied. The sample was comprised of 124 M&As which were announced in period between 2003-2008. The research was conducted using event study methodology. The shareholder wealth effect was measured using 41-day cumulative abnormal returns.

This study has shown that cross-border M&A announcements within the FMCG industry, which took place during the current M&A wave, result in significant positive abnormal returns. With this result it can be shown that shareholders of FMCG firms value M&As high as a strategy for international growth. The conclusion can be drawn that cross-border M&As are besides a popular growth strategy also a value creating growth strategy for FMCG firms. The managerial implication which coheres with this result is that managers of companies active in the FMCG industry should consider M&As as a growth strategy if they want to expand across borders, because through this strategy managers can maximize shareholder wealth.

(35)

market targets, even though average CARs for emerging market targets are higher in this sample.

The second research variable which has been tested was firm diversification. The results of the test indicate that no significant differences exist between the means of related and unrelated targets. Because expected differences could not be proved, equal means have to be assumed. The conclusion must be drawn that possible synergy advantages and less risk which are expected to go hand in hand with acquiring related targets do not influence the way in which shareholders value M&As in the FMCG industry.

Finally, the difference in CARs between cash payments and share payments has been tested. With just 9% of the sample consisting of shared paid M&As, a clear preference for paying in cash could be determined. However, no significant difference between the means could be found. The conclusion which should be drawn based on this study is that shareholders value FMCG industry M&As equally, whether they are paid for in cash or with shares.

(36)

6.2 Future research

After having summarized the results of this study in the previous paragraph, I remain with future challenges in this field of research. At first, in this study no distinction has been made between the levels of internationalization of the acquirers. It could be argued that shareholders of firms which are already highly internationalized react different to a cross-border M&As than those of firms which have only operated on domestic markets before. However, almost all the firms included in this sample are large global companies and therefore this aspect is not expected to influence the results strongly. Future research regarding M&As in the FMCG industry could include an analysis of this effect to see the how it influences the results.

Secondly, in this study the second and third research variables (firm diversification and payment method) were not distributed equally among the sample. Both consisted of rather dissimilar groups. Even though the used Mann-Whitney test is considered to be rather robust to size differences, having groups of a more similar size would be preferable. Hence, future researchers should attempt to gather a larger sample, which might help to increase the size of the smaller groups.

Thirdly, this study could be extended in time by including M&As which were announced in 2009. It will than be possible to see if cross-border M&As are also a value creating growth strategy in times of a financial crisis. This study could as well be extended to a more in-depth case study. In this way it will also be possible to say something about effects which occurred after the M&A deal is closed. For instance, something could be said about realized synergies. However, measuring these effects will be complicated because there has to be controlled for many external factors influencing the effect.

(37)

References

Bjorvatn, K. (2004), “Economic integration and the profitability of cross-border mergers and acquisitions”, European Economic Review, 48, pp. 1211-1226.

Brealey, R.A., I.A. Cooper, and E. Kaplanis, (1998), “International sources of risk: evidence from crossborder mergers”, Working paper, London Business School.

Bruner, R., (2001), “Does M&A Pay? A Survey of Evidence for the Decision-Maker”, Working Paper; University of Virginia Darden Graduate School of Business.

Campbell, John Y., Andrew W. Lo and A. Craig MacKinlay (1997): The Econometrics of Financial Markets. Princeton University Press; Princeton, New Jersey.

Chapman, K. (2003), “Cross-border mergers/acquisitions: a review and research agenda”, Journal of Economic Geography, 3, pp. 309-334.

Conn, R., A. Cosh, P. Guest, and A. Hughes, (2005), “The Impact on UK Acquirers of Domestic, Cross-border, Public and Private Acquisitions”, Journal of Business Finance & Accounting, 32, pp. 815-870.

Danbolt, J. (1995), “An Analysis of Gains and Losses to the Shareholders of Foreign Bidding Companies Engaged in Cross-Border Acquisitions into the United Kingdom, 1986–1991.” The European Journal of Finance 1, pp.279–309.

Datta, D., and G. Puia, (1995) “Cross-border acquisitions: an examination of the influence of related and cultural fit on shareholder value creation in US firms”, Management International Review, 35, pp. 337-359.

(38)

Doukas, J., N.G. Travlos, (1988). “The effect of corporate multinationalism on shareholders’ wealth: evidence from international acquisitions”, The Journal of Finance, 43, pp. 1161–1175.

The Economist (1999). Faites vos jeux, December 4, 63.

Francis, B., I. Hasan, and X. Sun (2008), “Financial market integration and the value of global diversification: Evidence for US Acquirers in cross-border merger and acquisitions”, Journal of Banking & Finance, 32, pp. 1522-1540.

Fama, E. (1998) "Market efficiency, long-term returns, and behavioral finance." Journal of Financial Economics, Vol. 49, pp. 283-306.

Goergen, M. and L. Renneboog (2004), “Shareholder wealth effects of European Domestic and Cross-Border Takeover Bids”, European Financial Management, 10, pp. 9-45.

Groothedde, B., C. Ruijgrok, and L. Tavasszy (2005), “Towards collaborative, intermodal hub networks; A case study in the fast moving consumer goods market”, Transportation research part E, 41, pp. 567 – 583.

Henry, D. and F. Jespersen (2004), “Mergers, Why most big deals don’t pay off”, Business Week, October 14.

Homberg, F, R. Rost, and M. Osterloh (2008), “Do Synergies Exist in Related Acquisitions? A Meta-Analysis of Acquisition Studies”, Working paper; Institute for Organization and Administrative Science University of Zurich.

(39)

Korajczyk, R. (1996) “A Measure of Stock Market Integration for Developed and Emerging Markets.” World Bank Economic Review, 10, pp.267-289.

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

Lowinski, F., D. Schiereck, and T. Thomas, (2004), “The Effect of Cross-Border Acquisitions on Shareholder Wealth – Evidence from Switzerland”, Review of Quantitative Finance and Accounting, 22, pp. 315-330.

MacKinlay, C. (1997) "Event Studies in Economics and Finance." Journal of Economic Literature, Vol. 35 No. 1, pp. 13-39.

Manne, Henry G. (1965), “Mergers and the Market for Corporate Control.” Journal of Political Economy, Vol.75, pp. 110-112.

McCann, M. (2001), “Cross-border Acquisitions: The UK experience.” Applied Economics 33, pp. 457–461.

Moeller, S., (2006), “Studying M&A targets”, Financial Times, October 5.

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

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

(40)

Appendix

Appendix A: Morgan Stanley Market Indices

As of June 2006, the Morgan Stanley Emerging Markets Index included:

• Argentina • Brazil • Chile • China • Colombia • Czech Republic • Egypt • Hungary • India • Indonesia • Iran • Israel • Jordan • Malaysia • Mexico • Morocco • Pakistan • Peru • Philippines • Poland • Russia • South Africa • South Korea • Taiwan • Thailand • Tunisia • Turkey • Vietnam

As of May 2008, the Morgan Stanley Developed Markets Index included:

(41)

Appendix B: List of all M&A deals included in this event study

Acquirer Country Acquirer Target Country Target Date

Constellation Brands Inc. United States BRL Hardy Ltd Australia 16-1-2003 Raisio Yhtymä Oyj Finland Diffchamb AB Sweden 13-2-2003 Zagro Asia Ltd Singapore Fezagro Co., Ltd Thailand 28-2-2003 Nestlé SA Switzerland Nestle (M) Behad Malaysia 19-3-2003 Kerry Group plc Ireland Guernsey Bel Inc United States 2-4-2003 Asia Pacific Breweries Ltd Singapore Hatay Brewery Ltd Vietnam 2-5-2003 Centrale Suiker Maatschappij NV Netherlands Unilever Group's Hungarian baking-ingredients

business Hungary 20-5-2003 Yakult Honsha Co., Ltd Japan Yakult (Singapore) Pte Ltd Singapore 26-5-2003 Ajinomoto Co., Inc. Japan Ajinomoto Genetika Research Institute Russian Federation 7-7-2003 Nestlé SA Switzerland Nestlé India Ltd India 8-7-2003 Provimi SA France Rolimpex SA Poland 19-7-2003 Interbrew SA Belgium Conistrade (M) Sdn Bhd Malaysia 5-9-2003 Saputo Inc. Canada Molfino Hermanos SA Argentina 3-10-2003 Ajinomoto Co., Inc. Japan Orsan SA France 16-10-2003 Ebro Puleva SA Spain Riceland-Magyarorszag Kft Hungary 6-11-2003 Heineken NV Netherlands BBAG Oesterreichische Brau-Beteiligungs AG Austria 19-11-2003 Lion Nathan Ltd Australia Changzhou Hua Xia Brewing Co., Ltd China 17-12-2003 Devro plc United Kingdom Cutisin AS Czech Republic 6-1-2004 Carlsberg A/S Denmark Holsten-Brauerei AG Germany 20-1-2004 Heineken NV Netherlands Brau-Union AG Austria 4-2-2004 Danisco A/S Denmark Rhodia SA's food additives division France 5-2-2004 Tate & Lyle plc United Kingdom McNeil Nutritionals' Alabama-based sucralose

manufacturing plant United States 19-2-2004 Interbrew SA Belgium Braco SA Brazil 3-3-2004 Associated British Foods plc United Kingdom Unilever Group's Mexican food oils and fats brands Mexico 9-3-2004 Anglo-Eastern Plantations plc United Kingdom Bina Pitri Jaya PT Indonesia 18-3-2004 Kerry Group plc Ireland Oregon Chai Inc United States 23-3-2004 Naturex SA France Hauser inc.'s rosemary extracts activities United States 8-4-2004 Interbrew SA Belgium Zhejiang Shiliang Brewery Co., Ltd China 21-6-2004 DCC plc Ireland Bottle Green Ltd United Kingdom 7-7-2004 Associated British Foods plc United Kingdom Tone Brothers Inc. United States 22-7-2004 Ebro Puleva SA Spain Riviana Foods Inc. United States 23-7-2004 Ridley Inc. Canada Sweetlix LLC United States 23-7-2004 Asia Pacific Breweries Ltd Singapore DB Breweries Ltd New Zealand 26-7-2004 Coca-Cola Company, The United States Joya brand Mexico 28-7-2004 Heineken NV Netherlands Tsentral'naya Evropeiskaya Pivovarennaya

Kompania Russian Federation 10-8-2004 InBev SA Belgium Sun Interbrew OAO Russian Federation 12-8-2004 HQ Sustainable Maritime

(42)

SABMiller plc United Kingdom Amalgamated Beverage Industries Ltd South Africa 22-9-2004 Heineken NV Netherlands VINAP Russian Federation 1-10-2004 Koninklijke Ahold NV Netherlands ICA Ahold AB Sweden 25-10-2004 McCormick & Company Inc. United States CM van Sillevoldt BV Netherlands 2-11-2004 Cadbury Schweppes plc United Kingdom Orangina Schweppes SAS France 2-11-2004 Schweitzer Mauduit International

Inc. United States

KCPI's tobacco-related paper manufacturing

operations Philippines 15-11-2004 Glanbia plc Ireland Kortus Food Ingredients Services GmbH Germany 7-12-2004 PepsiCo Inc. United States Snack Ventures Europe Belgium 13-12-2004 Koninklijke DSM NV Netherlands Roche (Shanghai) Vitamins Ltd. China 27-1-2005 Danisco A/S Denmark Genencor International Inc. United States 27-1-2005 Groupe Danone SA France Casancrem brand Argentina 10-2-2005 InBev SA Belgium Companhia de Bebidas das Americas Brazil 14-2-2005 Ebro Puleva SA Spain Panzani SAS France 23-2-2005 SABMiller plc United Kingdom Birra Peroni SpA Italy 23-2-2005 Coca-Cola Hellenic Bottling

Company SA Greece Multon ZAO Russian Federation 31-3-2005 InBev SA Belgium Sun Interbrew OAO Russian Federation 12-4-2005 NBTY Inc. United States SISU Inc. Canada 25-4-2005 Heineken NV Netherlands Patra Brewery Russian Federation 6-5-2005 Saputo Inc. Canada Schneider Cheese Inc. United States 16-5-2005 LifeBrandz Ltd Singapore ThaiNutri Co., Ltd Thailand 1-6-2005 Diageo plc United Kingdom Old Bushmills Distillery Company Ltd, The Ireland 6-6-2005 SunOpta Inc. Canada Cleugh's Frozen Foods Inc. United States 20-6-2005 HJ Heinz Company United States HP Foods Ltd United Kingdom 21-6-2005 Heineken NV Netherlands Kombinat Pivovaryonoi I Bezalkogolnoi

Promyshlenosti Imeni Stepana Razina Russian Federation 6-7-2005 DCC plc Ireland Brett Fuels Ltd United Kingdom 6-7-2005 SunOpta Inc. Canada Pacific Fruit Processors Inc. United States 13-7-2005 SABMiller plc United Kingdom Grupo Empresarial Bavaria SA Colombia 19-7-2005 Spectrum Brands Inc. United States VARTA Consumer Batteries GmbH & Co. KGaA Germany 19-7-2005 Kerry Group plc Ireland Noon Products Ltd United Kingdom 7-8-2005 Det Østasiatiske Kompagni A/S Denmark Dumex Malaysia Sdn Bhd Malaysia 17-8-2005 Carlsberg A/S Denmark Carlsberg Brewery Hong Kong Ltd Hong Kong 1-9-2005 InBev SA Belgium KK Brewery China 2-9-2005 USANA Health Sciences Inc. United States Personal care products manufacturing facility China 20-9-2005 Petra Foods Ltd Singapore Sime Darby Marketing Sdn Bhd Malaysia 29-9-2005 SABMiller plc United Kingdom Grupo Empresarial Bavaria SA Colombia 28-10-2005 Nestlé SA Switzerland Dreyer's Grand Ice Cream Inc. United States 1-12-2005 Charlie's Group Ltd New Zealand Phoenix Organics Ltd United Kingdom 2-12-2005 Tate & Lyle plc United Kingdom Continental Custom Ingredients Inc. United States 5-12-2005 Ajinomoto Co., Inc. Japan Hong Kong Amoy Food Group Hong Kong 12-1-2006 InBev SA Belgium Fujian Sedrin Brewery Co., Ltd. China 23-1-2006 Petra Foods Ltd Singapore Nestle Philippines Inc.'s chocolate manufacturing _{and distribution assets} Philippines 23-1-2006 New Dragon Asia Corporation United States Chengdu based state-owned noodle manufacturing

(43)

Koninklijke Wessanen NV Netherlands Bio Slym Srl Italy 21-3-2006 Constellation Brands Inc. United States Vincor International Inc. Canada 3-4-2006 Cadbury Schweppes plc United Kingdom All American Bottling Corporation United States 25-4-2006 Asia Pacific Breweries Ltd Singapore Aurangabad Breweries Ltd India 2-5-2006 Cosentino Signature Wines plc United Kingdom Lorenza-Lake Winery United States 12-5-2006 Rieber & Søn ASA Norway Frödinge Mejeri AB Sweden 19-5-2006 Agrana Beteiligungs AG Austria Xianyang Andre Juice Co., Ltd China 12-6-2006 Carlsberg A/S Denmark Holsten-Brauerei AG Germany 26-6-2006 SABMiller plc United Kingdom McKenzie River Corporation's Sparks and Steel _{Reserve malt beverage brands} United States 3-7-2006 Coca-Cola Hellenic Bottling

Company SA Greece Fonti Del Vulture Srl Italy 5-7-2006 DCC plc Ireland Carlton Fuels Ltd United Kingdom 7-7-2006 SABMiller plc United Kingdom Foster’s India Ltd India 4-8-2006 Coca-Cola Hellenic Bottling

Company SA Greece Yoppi Kft Hungary 22-8-2006 SunOpta Inc. Canada Hess Food Group LLC United States 7-11-2006 Dairy Crest Group plc United Kingdom St Hubert SAS France/Italy 9-11-2006 Wilmar International Ltd Singapore PPB Oil Palms Bhd Malaysia 14-12-2006 Naturex SA France Hammer Pharma SpA Italy 23-1-2007 Lännen Tehtaat Oyj Finland Maritim Food AS Norway 6-2-2007 Saputo Inc. Canada Dansco Dairy Products Ltd United Kingdom 23-3-2007 J & J Snack Foods Corporation United States CoolBrands International Inc.'s Whole Fruit and

Fruit-a-Freeze assets Canada 2-4-2007 Hershey Company, The United States Godrej Beverages and Foods Ltd India 3-4-2007 Nestlé SA Switzerland Gerber Products Company United States 12-4-2007 SunOpta Inc. Canada Baja California Congelados SA de CV's certain

assets Mexico 4-5-2007 Nutreco Holding NV Netherlands Maple Leaf Animal Nutrition Canada 21-5-2007 Zhongpin Inc. United States Deyang East China Food Co., Ltd's total assets China 29-6-2007 Coca-Cola Hellenic Bottling

(44)

Appendix C: Kolmogorov-Smirnov Test total sample 41-day CAR

Test 1

Descriptive Statistics

Percentiles

N Mean Std. Deviation Minimum Maximum 25th 50th (Median) 75th

CAR ₁₂₄ _,05483010 _,206755770 _-,255050 _1,312916

-,03192368 ,02782405 ,08905637

One-Sample Kolmogorov-Smirnov Test

CAR N ₁₂₄ Mean ,05483010 Normal Parameters(a,b) Std. Deviation _,206755770 Absolute _,251 Positive ,251

Most Extreme Differences

Negative _-,160

Kolmogorov-Smirnov Z _2,789

Asymp. Sig. (2-tailed) ,000

(45)

Appendix D: Mann-Whitney tests for all variables

Test 2

Ranks

TargetMarket N Mean Rank Sum of Ranks

1,00 53 62,83 3330,00 2,00 71 62,25 4420,00 CAR Total 124 Test Statistics(a) CAR Mann-Whitney U 1864,000 Wilcoxon W 4420,000 Z -,088

a Grouping Variable: TargetMarket

Test 3

Ranks

Diversification N Mean Rank Sum of Ranks

1,00 112 62,54 7005,00 2,00 12 62,08 745,00 CAR Total 124 Test Statistics(a) CAR Mann-Whitney U 667,000 Wilcoxon W 745,000 Z -,042

Asymp. Sig. (2-tailed) _,966

(46)

Test 4

Ranks

PaymentMethod N Mean Rank Sum of Ranks

1,00 ₁₁₅ _60,99 _7014,00 2,00 ₉ _81,78 _736,00 CAR Total ₁₂₄ Test Statistics(a) CAR Mann-Whitney U 344,000 Wilcoxon W 7014,000 Z -1,671

a Grouping Variable: PaymentMethod

Test 5

Ranks

ControlSize N Mean Rank Sum of Ranks

1,00 86 62,02 5334,00 2,00 38 63,58 2416,00 CAR Total 124 Test Statistics(a) CAR Mann-Whitney U _1593,000 Wilcoxon W _5334,000 Z _-,222

Asymp. Sig. (2-tailed) _,824

The Wealth Effect of Acquiring Firm M&A Announcements