1
M&A announcement effect on share price value
European food, beverage and tobacco industry
UNIVERSITY OF AMSTERDAM
FACULTY ECONOMICS AND BUSINESS SCHOOL
BSc Economics & Business
Bachelor Specialisation Finance and Organisation
Author:
L.J.R. van Etten
Student number:
10981721
Thesis supervisor:
Dr. J.J.G. Lemmen
2 This document is written by Luuk van Etten who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
ABSTRACT
The purpose of this thesis is to find a merger and acquisition announcement effect on shareholder value. To find an answer to this topic, an event study and a CAR regression equation will be executed. The results of the [-3,3] window are significant at the 10% level and the results of the [-1,1] window are significant at the 5% level. Moreover, ROE shows significance at the 5% level, %LEVERAGE shows significance at the 1% level and PAYSHARES shows significance at the 10% level. In conclusion, the M&A announcements do affect shareholder value in the European food, beverage and tobacco industry.
Keywords: Event Studies, Announcements, Acquisition, Merger and
Shareholders.
4
TABLE OF CONTENTS
5
LIST OF TABLES
Table 1 Correlations 17
Table 2 Descriptive statistics abnormal returns 18
Table 3 CARs for different event windows 19
Table 4 CAARs for different event windows 19
Table 5 Test Results 22
Table 6 Variables Entered/Removed 23
Table 7 Model Summary 23
Table 8 ANOVA 23
6
LIST OF FIGURES
Figure 1 Mergers and Acquisitions Europe 7
Figure 2 Cumulative abnormal return [-3,3] 20
7
CHAPTER 1 Introduction
1.1 Relevance
Following the articles of previous researchers, mergers and acquisitions come in waves. Out of the available data, concluded can be that the first M&A wave in Europe took place in the late 80s. The construction of the Single Market led to the first wave that Europe experienced. In the late 90s, the second M&A wave emerged. The third and last M&A wave in Europe took place in 2003 and ended with the outbreak of the financial crisis. There are a couple of factors that led to an increase in M&A activity in Europe; e.g. the introduction of the Euro, low interest rates, globalization and technological innovation (Vancea 2013).
After the financial crisis of 2008, company growth increased as a result of increasing economic trust. When a company expects another company to grow it is interesting to take over that particular company. Therefore, mergers and acquisitions are an important factor. The bidding firm expects that taking over the target firm will create value. Following Martynova et al. (2006) most of the takeovers occur in times of economic recovery. According to Dieudonne et al. (2014) the US was in the middle of the seventh M&A wave cycle in 2014. For this reasons there may also be a new wave in Europe after the crisis of 2008. Figure 1 (Source: IMAA) below shows an increase in M&A deal value in Europe after the financial crisis.
Mergers and acquisitions could result in higher stock returns for shareholders. Therefore, it is interesting to investigate the announcement effect of M&As on shareholders. Several conclusions can be derived from previous research on this subject. In general, studies found that M&A investments result in higher stock returns for both acquirers and targets according to e.g. Alexandridis et al. (2010). If we look worldwide, the stock returns of target firms perform better than the stock returns of acquirer firms (Goergen et al. 2003).
0 500 1000 1500 2000 0 5000 10000 15000 20000 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06 20 08 20 10 20 12 20 14 20 16 20 18 … Va lu e of T ra ns ac tion s ( in b il. EUR ) N umb er of T ra ns ac tion s
8 This thesis will look specifically at the European Union because previous research mainly focused on the US and UK. Furthermore, it is easier to understand and find information about the companies and stock prices in the EU than for example Asia. There has been no research focusing on the European food, beverage and tobacco industry so far. This is the main reason why it is interesting to investigate this industry. Moreover, the European food, beverage and tobacco industry is Europe’s biggest manufacturing sector in terms of jobs and added value. Exports of this industry have doubled in the last 10 years according to the European Commission. Therefore, this industry is still growing and thus relevant.
1.2 Research Question and Hypothesis
As described in section 1.1 this thesis will focus on the mergers and acquisitions announcement effect on shareholder value in the European food, beverage and tobacco industry. Summarizing section 1.1 the following research question can be formulated:
To what extent do M&A announcements affect share price value in the European food, beverage and tobacco industry between 2009-2018?
This research question will be investigated according to the following hypothesis:
𝑯𝑯𝟎𝟎: 𝐭𝐭𝐭𝐭𝐭𝐭 𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐭𝐭𝐜𝐜𝐜𝐜𝐭𝐭 𝐜𝐜𝐜𝐜𝐭𝐭𝐚𝐚𝐜𝐜𝐚𝐚𝐭𝐭 𝐜𝐜𝐚𝐚𝐚𝐚𝐚𝐚𝐚𝐚𝐜𝐜𝐜𝐜𝐜𝐜 𝐚𝐚𝐭𝐭𝐭𝐭𝐜𝐜𝐚𝐚𝐚𝐚𝐫𝐫 (𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐫𝐫) 𝐝𝐝𝐚𝐚 𝐚𝐚𝐚𝐚𝐭𝐭 𝐝𝐝𝐜𝐜𝐝𝐝𝐝𝐝𝐭𝐭𝐚𝐚 𝐝𝐝𝐚𝐚𝐚𝐚𝐜𝐜 𝟎𝟎 𝑯𝑯𝟏𝟏: 𝐭𝐭𝐭𝐭𝐭𝐭 𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐭𝐭𝐜𝐜𝐜𝐜𝐭𝐭 𝐜𝐜𝐜𝐜𝐭𝐭𝐚𝐚𝐜𝐜𝐚𝐚𝐭𝐭 𝐜𝐜𝐚𝐚𝐚𝐚𝐚𝐚𝐚𝐚𝐜𝐜𝐜𝐜𝐜𝐜 𝐚𝐚𝐭𝐭𝐭𝐭𝐜𝐜𝐚𝐚𝐚𝐚𝐫𝐫 (𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐫𝐫) 𝐝𝐝𝐚𝐚 𝐝𝐝𝐜𝐜𝐝𝐝𝐝𝐝𝐭𝐭𝐚𝐚 𝐝𝐝𝐚𝐚𝐚𝐚𝐜𝐜 𝟎𝟎
To test this hypothesis a independent T-test will be used. Furthermore, a regression model with relevant variables will be executed and analysed to explain the CAARs. A more detailed explanation will be given in the methodology in chapter 3.
1.3 Research Findings
The findings of the research are as follows: the event study resulted in a significant effect for both the [-3,3] event window and the [-1,1] event window. The [-3,3] window is significant at the 10% level and the [-1,1] window is significant at the 5% level. Therefore, it can be concluded that M&As in the European food, beverage and tobacco industry lead to positive results. Moreover, the regression model showed several significant independent variables. The variables PAYINSHARES, ACQUIRERLEVERAGE and ROE are significant at different
9 levels. PAYINSHARES shows significance at the 10% level, ROE shows significance at the 5% level, ACQUIRERLEVERAGE shows significance at the 1% level and ROE shows significance at the 5% level.
1.4 Research Plan
This thesis is organised as follows: in chapter 2, the literature review is explained. Chapter 3 contains the methodology and the data to answer the research question. Chapter 4 presents the results and finally chapter 5 will present the conclusions and suggestions for further research.
10
CHAPTER 2 Literature Review
2.1 M&A motives
In this section the motives of M&As are described. Mergers and acquisitions continue to be highly popular over time. The numbers are still increasing in countries such as the US, Europe and the UK. For example, entering a new market, achieving synergies, increasing profitability and gaining new scarce resources are motives referring to M&A (Calipha et al. 2010). However, the findings in the existing empirical literature differ sometimes. The importance of various motives changes over time according to Kiymaz and Baker (2008). Following their research, it is useful to group the motives of M&A into categories.
Several studies pointed out that there are two key explanations that synergies are as important as merger motives. These two key explanations are synergies and correction of managerial failure (Martynova and Renneboog 2006). According to Sirower (1997) the definition of a synergy is as follows: an increase in competitiveness and earn cash flows beyond what the companies are likely to accomplish independently. In general, synergies create value for shareholders of combined firms. Multple studies confirm this such as studies from e.g. Kaplan and Weisbach (1992), Andrade, Mitchell and Stafford (2001), Servaes (1991) and Bradley, Desai and Kim (1988). There are two types of synergies created by acquisitions: financial synergies and operating synergies. Financial synergies are related to improvement in the financial metric of the combined companies. Some examples of advantages as a result of financial synergies are:
• Larger customer base, which lead to increased revenues • Streamlined operations, which lead to lower costs
• Technology and talent harmonies
Furthermore, financial synergies may include greater cash flows, tax benefits, lower cost of capital and increased debt capacity. Operating synergies allow firms to achieve higher growth and increase their operating income. Operating synergies arise through the realization of the following:
• Economies of scale and scope • Reduction in agency costs
11 • Greater market shares and lower competition
• Combination of different functional skills by the management team of the bidder
Moreover, firms that are in the same industry are more likely to benefit from operating synergies. However, firms that have different capabilities such as marketing, R&D and finance also show contribution to operating efficiencies. On the other hand, other studies show the role of hostile acquisitions, which correct managerial failure, is an important motive. The management of a company tends to act in their own interest. When shareholders find that the management underperforms, the agency costs will be high. A hostile acquisition occurs when the management is performing poorly and needs to be replaced.
2.2 M&A effects globally
This section will focus on the effects of M&A in different countries is described. There are several studies on M&A effects in different countries globally.
Alexandridis et al. (2010) studied the gains from mergers and acquisitions around the world. They conclude that acquirers beyond the most competitive takeover market do generate gains. The most competitive takeover markets are the U.S., U.K. and Canada (the UUC). These markets create higher gains than acquisitions in other countries. However, the distribution of created value between acquirer and target is depending on the degree of competition in the market. Their empirical findings on public acquisitions show a zero abnormal return for the acquirer.
Campa and Hernando (2004) looked at value created to shareholders by the announcement effect in the European Union (E.U.) in the late 1990s till the beginning of 2000. According to their findings, the cumulative average abnormal return of the target shareholder is 9%, while the cumulative average abnormal return of the acquirer shareholder is 0%. Furthermore, they found that mergers in industries that are heavily regulated or had been under government control create lower value than M&A announcements in unregulated industries. If the merger involves firms from two different countries in regulated industries, this value creation becomes significantly negative. These conclusions are consistent with obstacles such as legal, transaction or cultural barriers. These obstacles lessen the probability of the merger actually being completed and being successful. Therefore, the expected value of the M&As is reduced. Franks and Harris (1989) studied the effects of 1800 U.K. M&As on shareholder value in the mid-1950s till the mid-1980s. The target shareholders gain between 20% and 30% around the announcement date.
12 The acquirer shareholders earn 0%. They conclude that their evidence is similar to comparable studies in the United States. Goergen and Renneboog (2003) analysed the wealth effects of large acquisitions in the E.U. on the short-term. Following their findings, the target firms have a large announcement effect of 9%. The acquirer firm shareholders have an announcement effect of 0.7%. They also found that hostile mergers and acquisitions result in larger price reactions than friendly mergers and acquisitions. The payment method of the acquisitions has a large effect on share price reactions. Moreover, the involvement of an U.K. target or acquirer has more influence on abnormal returns than an European target or acquirer. Finally, they advise the acquirer firms not to take over a target firm that diversifies too much from their core business. Gugler et al. (2003) studied the effects of mergers around the world over the past 15 years. They find that 56.7% of all mergers result in higher than projected profits. After five years, almost the same fraction of mergers resulted in lower than projected profits. According to the results of Gugler et al. (2003), the mergers significantly increase profits on average. However, the mergers reduce the sales of the firms. Across countries, the post-merger patterns of profits and sales look similar. They state that this is one of their most interesting findings. Finally, they did not find a significant difference between domestic and cross-border mergers.
Summarizing all these papers, a few conclusions can be derived.
• The target firms have a large announcement effect (9% or higher)
• The acquirer firms have a zero or close to zero announcement effect (0% or 0.7%) • The announcement effect is lower in regulated countries than in unregulated
countries
• The payment method of M&A has influence on share price reactions • Domestic or cross-border mergers do not differ significantly
13 2.3 Long-term abnormal returns
This section summarizes the M&A literature on long-term abnormal returns. Where short-term abnormal returns are often positive or equal to zero, several studies measure negative long-term abnormal returns (Andrade et al. 2004). Because of this, the net wealth effect becomes negative because of the long-term negative drift overwhelming the positive combined stock price reaction. A difference exists between in the long-term performances in financing the M&As. Acquiring firms that use stock financing earn abnormal returns of -24.2% over the five years after the merger. In contrast, the acquiring firms that make use of cash financing earn an abnormal return of -18.5%. The book-to-market ratio is also an important factor explaining the large difference in long-term abnormal returns. Firms with a high book-to-market ratio are described as ‘’value’’ firms. These firms tend to have higher abnormal returns on average. Firms with a low book-to-market ratio tend to have lower abnormal returns on average, called ‘’glamour’’ firms. The study from Rau and Vermaelen (1998) shows that value acquirers earn abnormal returns of 7.6%, where glamour firms earn abnormal returns of -17.3%.
14
CHAPTER 3 Data and Methodology
3.1 Sample Selection
In this section, the sample selection of the research is described. To find an answer to the research question, an event-study will be used. At first, the data will be obtained from the database Zephyr. Zephyr is an easy database to use and includes M&A deals from all over the world. Using the following restrictions will filter the deals.
• The time period will be from 01/01/2009 till 01/01/2018. • The sector will be food, beverages and tobacco.
• Focus on the European Union.
• Minimum deal value of 0.25 million euro. • The deal status is ‘completed’.
• The deal type is an acquisition or merger. • Both private and public deals will be examined.
• The final stake percentage has to be min. 51% and max. 100%, • The initial stake percentage has to be less than 50%.
We only look at listed acquirers because the targets in this industry are not listed. Therefore, all target firms are private. 79 observations are generated with these restrictions.
3.2 Methodology
In this section the methodology of the research is described. The stock prices of these 79 listed acquirers have to be collected from DataStream. DataStream is a database containing information about stock prices and it is easy to transfer information into Excel. Looking at the stock prices just before the announcement date and just after the announcement date, no other factors have to be taken into account because of the short time period. The event windows will be [-1, 1] and [-3, 3]. The benchmark 𝑅𝑅𝑚𝑚 will be determined following the MSCI Europe Food, Beverage and Tobacco index. This index is composed of large and mid-cap stocks across 15 developed market European countries. The estimation window of the benchmark stock return will be an [-205, -6] interval (Moeller et al. 2005). Within this interval, all the daily returns are calculated. The daily return on the MSCI Europe food, beverage and tobacco index represent the daily market returns. We use the market model to take into account stock price movements. The daily abnormal returns (ARs) have to be determined by
15 the difference between the expected returns and the actual returns. After that, the abnormal returns have to be accumulated (CARs) and the cumulative average abnormal returns (CAARs) have to be determined. Finally, the significance of these outcomes (CAARs) will be tested with a difference T-Test. The significance will show whether M&A announcements do or do not have an effect on acquirer stock prices.
The following formulas will be used according to De Jong (2007).
The market model formula:
𝑅𝑅𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑖𝑖+ 𝛽𝛽𝑖𝑖𝑅𝑅𝑚𝑚𝑖𝑖+ 𝜖𝜖𝑖𝑖𝑖𝑖
Where 𝑅𝑅𝑚𝑚𝑖𝑖 is the return on the market index at time t and 𝑅𝑅𝑖𝑖𝑖𝑖 is the one-period return on asset i at period t. The error term 𝜖𝜖𝑖𝑖𝑖𝑖 is assumed to be temporally uncorrelated normally distributed on asset i at period t. In contrast to the market-adjusted method, which assumes that the 𝛽𝛽 of each stock is equal to one, the market model accounts for differences in the 𝛽𝛽. Therefore, the market model is a better method to define abnormal returns.
The following formula defines the abnormal returns as the residuals or prediction errors:
𝑁𝑁𝑅𝑅𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑖𝑖+ 𝛽𝛽𝑖𝑖𝑅𝑅𝑚𝑚𝑖𝑖
Where 𝑁𝑁𝑅𝑅𝑖𝑖𝑖𝑖 is the normal return on asset i at period t. 𝛼𝛼𝑖𝑖 and 𝛽𝛽𝑖𝑖 are OLS estimates of regression coefficients.
The abnormal return formula:
𝐴𝐴𝑅𝑅𝑖𝑖𝑖𝑖 = 𝑅𝑅𝑖𝑖𝑖𝑖− 𝑁𝑁𝑅𝑅𝑖𝑖𝑖𝑖
Where 𝐴𝐴𝑅𝑅𝑖𝑖𝑖𝑖 is the abnormal return on asset i at period t. 𝑅𝑅𝑖𝑖𝑖𝑖 is defined as the return on asset
i at time t and 𝑁𝑁𝑅𝑅𝑖𝑖𝑖𝑖 is the benchmark or normal return on asset i at period t.
The cumulative abnormal return formula:
𝐶𝐶𝐴𝐴𝑅𝑅𝑖𝑖 = 𝐴𝐴𝑅𝑅𝑖𝑖,𝑖𝑖1+ 𝐴𝐴𝑅𝑅𝑖𝑖,𝑖𝑖2= � 𝐴𝐴𝑅𝑅𝑖𝑖𝑖𝑖
𝑖𝑖2
16 Where the 𝐶𝐶𝐴𝐴𝑅𝑅𝑖𝑖 is the cumulative abnormal return on asset i. The abnormal returns are aggregated from the start of the event period, 𝑡𝑡1, up to period 𝑡𝑡2.
The cumulative abnormal average return formula:
𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅 = 𝑁𝑁 � 𝐶𝐶𝐴𝐴𝑅𝑅1 𝑖𝑖 𝑛𝑛 𝑖𝑖=1
Where the 𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅 is the cumulative abnormal average return, which presents the accumulated 𝐶𝐶𝐴𝐴𝑅𝑅s on asset i divided by the number of observations.
Secondly, the following regression model on the CAR will be used to make this thesis more sufficient.
𝐶𝐶𝐴𝐴𝑅𝑅𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝛽𝛽1 𝑃𝑃𝐴𝐴𝑃𝑃𝑃𝑃𝑁𝑁𝑃𝑃𝑃𝑃𝐴𝐴𝑅𝑅𝑃𝑃 + 𝛽𝛽2𝐶𝐶𝑅𝑅𝐶𝐶𝑃𝑃𝑃𝑃𝐶𝐶𝐶𝐶𝑅𝑅𝐶𝐶𝑃𝑃𝑅𝑅 + 𝛽𝛽3𝑙𝑙𝑙𝑙𝐴𝐴𝐶𝐶𝑙𝑙𝑙𝑙𝑃𝑃𝑅𝑅𝑃𝑃𝑅𝑅𝑃𝑃𝑃𝑃𝑙𝑙𝑃𝑃 + 𝛽𝛽4𝐴𝐴𝐶𝐶𝑙𝑙𝑙𝑙𝑃𝑃𝑅𝑅𝑃𝑃𝑅𝑅𝐴𝐴𝑃𝑃𝐴𝐴𝑃𝑃𝑅𝑅𝐴𝐴𝐴𝐴𝑃𝑃 + 𝛽𝛽5𝑅𝑅𝐶𝐶𝑃𝑃 + 𝜀𝜀𝑖𝑖
The 𝐶𝐶𝐴𝐴𝑅𝑅𝑖𝑖𝑖𝑖 is the cumulative abnormal return of firm i in period t. The independent variables are payment in shares (PAYINSHARES), cross-border deal (CROSSBORDER), size of the acquirer (lnACQUIRERSIZE), leverage of the acquirer (ACQUIRERLEVERAGE) and return on equity of the acquirer (ROE). Payment in shares is equal to 1 if the deal was financed with shares only and 0 otherwise. Cross border is equal to 1 if the there was a cross border takeover and 0 otherwise. The total assets of the acquirer determine the size of the acquirer. The leverage of the acquirer is determined by the percentage of leverage of the total assets. The ROE is the return on equity of the acquirer. 𝜀𝜀𝑖𝑖 is the error term of the regression equation.
17
Table 1. Correlations
CAR(-1,1) PAYINSHARES CROSSBORDER lnACQUIRERSIZE ACQUIRERLEVERAGE ROE
CAR(-1,1) Pearson Correlation 1 .195 .020 .039 -.386** .051
Sig. (2-tailed) .096 .862 .741 .001 .665
N 74 74 74 74 74 74
PAYINSHARES Pearson Correlation .195 1 .056 -.061 -.037 -.025
Sig. (2-tailed) .096 .623 .593 .747 .830
N 74 79 79 79 79 79
CROSSBORDER Pearson Correlation .020 .056 1 -.043 -.058 .054
Sig. (2-tailed) .862 .623 .708 .612 .634
N 74 79 79 79 79 79
lnACQUIRERSIZE Pearson Correlation .039 -.061 -.043 1 -.091 -.001
Sig. (2-tailed) .741 .593 .708 .426 .992
N 74 79 79 79 79 79
ACQUIRERLEVERAGE Pearson Correlation -.386** -.037 -.058 -.091 1 .386**
Sig. (2-tailed) .001 .747 .612 .426 .000
N 74 79 79 79 79 79
ROE Pearson Correlation .051 -.025 .054 -.001 .386** 1
Sig. (2-tailed) .665 .830 .634 .992 .000
N 74 79 79 79 79 79
**. Correlation is significant at the 0.01 level (2-tailed).
The correlations between the variables are shown above. If a correlation is above 0.7, there exists multicollinearity. Therefore, some explanatory variables had to be deleted, which resulted in the table above.
18 3.3 Descriptive Statistics
In this section the descriptive statistics are shown. Table 2 shows the descriptive statistics for the average returns over the different event windows. See the detailed results below.
Table 2. Descriptive statistics abnormal returns
AR [-3] AR [-2] AR [-1] AR [0] AR [+1] AR [+2] AR [+3] N Valid 74 74 74 74 74 74 74 Missing 5 5 5 5 5 5 5 Mean -.0040 .0052 -.0006 -.0013 .0022 .0035 .0006 Std. Error of Mean .0017 .0021 .0017 .0040 .0032 .0020 .0027 Median -.0044 .0041 -.0012 -.0015 .0006 .0015 -.0022 Std. Deviation .0148 .0184 .0147 .0344 .0275 .0172 .0228 Skewness -.139 .998 .190 .923 .179 1.804 3.575 Std. Error of Skewness .279 .279 .279 .279 .279 .279 .279 Minimum -.0443 -.0295 -.0437 -.0854 -.0732 -.0233 -.0410 Maximum .0318 .0784 .0440 .1135 .1008 .0790 .1444 Percentiles 25 -.0095 -.0058 -.0076 -.0150 -.0079 -.0072 -.0077 50 -.0044 .0041 -.0012 -.0015 .0006 .0015 -.0022 75 .0035 .0133 .0072 .0101 .0136 .0099 .0067
The mean shows the mean of the ARs. The ARs before the announcement date show a
negative mean while the ARs after the announcement date show a positive mean. The
standard error of the mean is an estimator of the standard deviation. The median is the middle
value separating the half of a distribution. The standard deviation measures the amount of
variation of the data set. Skewness shows whether a distribution is left or right skewed in
comparison to the normal distribution. Skewness values lower than -1 and above 1 show that
the distribution is not normal. In this case, the values of AR [+2] and AR [+3] are above 1.
Therefore, the AR [+2] and AR [+3] are not normally distributed. The rest of the ARs are
normally distributed.
19
CHAPTER 4 Results
In this section, the results of this thesis are presented. At first, an event study has been done and the results will be described in section 4.1. Secondly, a regression model has been executed. The results will be described and discussed in section 4.2.
4.1 Event Study
The event study has been done with the help of an event study tool of the Erasmus University Rotterdam. This tool is designed to do an event study with DataStream data. Other event study tools such as WRDS are limited to US companies, so the Erasmus Event Study Tool is helpful for this study. Abnormal returns are calculated automatically. Therefore, the cumulative abnormal returns and the cumulative abnormal returns can be determined. The outcomes are shown below in tables 3 and 4.
Table 3. CARs for different event windows
CAR [-3] -0.002876665 CAR [-2] 0.002062174 CAR [-1] 0.001678542 CAR [0] 0.000109364 CAR [1] 0.003636008 CAR [2] 0.005995869 CAR [3] 0.00731419 CARs 0.017919482
Table 4. CAARs for different event windows
CAAR [-3] 0.00256 CAAR [-2] 0.00256 CAAR [-1] 0.00256 CAAR [0] 0.00256 CAAR [1] 0.00256 CAAR [2] 0.00256 CAAR [3] 0.00256
20 Figure 2 shows a graph of the cumulative abnormal returns (CARs) for different event windows within the [-3,3] window, which are consistent with table 2. The cumulative abnormal returns before the announcement are found to be lower than the cumulative abnormal returns after the announcement.
Figure 2. Cumulative abnormal return [-3,3]
Figure 3 shows a graph of the cumulative abnormal returns (CARs) for different event windows within the [-1,1] window, which are consistent with table 2. The cumulative abnormal returns before the announcement are found to be lower than the cumulative abnormal returns after the announcement.
Figure 3. Cumulative abnormal return [-1,1]
-0,004 -0,002 0 0,002 0,004 0,006 0,008
CAR [-3] CAR [-2] CAR [-1] CAR [0] CAR [1] CAR [2] CAR [3]
Cumulative abnormal return [-3,3]
Cumulative abnormal return [-3,3] 0 0,0005 0,001 0,0015 0,002 0,0025 0,003 0,0035 0,004
CAR [-1] CAR [0] CAR [1]
Cumulative abnormal return [-1,1]
Cumulative abnormal return [-1,1]
21
4.1.1 T-Test
An independent T-test is executed to test the significance of the CAAR. The formulas that are used are shown below.
The test statistic for testing the null hypothesis 𝑃𝑃0: 𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅 = 0 is as follows
𝑡𝑡𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 = √𝑁𝑁𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅𝑃𝑃 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶
Where 𝑃𝑃𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 is the standard deviation of the cumulative abnormal returns in the sample
The 𝑃𝑃𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 can be determined by the following formula
𝑃𝑃𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶2 = 𝑁𝑁 − 1 �(𝐶𝐶𝐴𝐴𝑅𝑅1 𝑖𝑖− 𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅)2 𝑁𝑁
𝑖𝑖=1
The test needs to be done for two different windows [-3,3] and [-1,1]. The data for executing these tests are derived from the event study tool.
Testing the null hypothesis 𝑃𝑃0: 𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅 = 0 for the [-3,3] window
𝑡𝑡
𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶[−3,3]= √7
0.0034678 = 1.8082
0.00256
Testing the null hypothesis 𝑃𝑃0: 𝐶𝐶𝐴𝐴𝐴𝐴𝑅𝑅 = 0 for the [-1,1] window22
Table 5. T-test results.
CAAR N
Std.
Deviation T statistic P value [-1,1] 0.00256 3 0.0019925 3.1471** 0.01 < p < 0.02 [-3,3] 0.00256 7 0.0034678 1.8082* 0.05 < p < 0.10 The CAAR [-1,1] is the 3-days cumulative average abnormal return in percent. The CAAR [-3,3] is the 7-days cumulative average abnormal return in percent. The standard deviation gives the amount of variation of data values. T-value gives the probability that the CAAR is different from zero. The p-value is the estimated probability of rejecting the null-hypothesis. The level of significance is higher when the p-value is small.
*Statistical significance at the 10% level. **Statistical significance at the 5% level. *** Statistical significance at the 1% level.
From the results of the T-tests, it can be concluded that the outcomes are significant. For the [-3,3] window the t value is 1.8082, which can be concluded as significant at the 10% level. This means that the cumulative average abnormal return (CAAR) for the [-3,3] window is significantly different from 0 and thus the null hypothesis is rejected at the 10% level. For the [-1,1] window the t value is 3.1471, which can be concluded as significant at the 5% level. This means that the cumulative average abnormal return (CAAR) for the [-1,1] window is significantly different from 0 and thus the null hypothesis is rejected at the 5% level. In conclusion, this means that M&A in the European food, beverage and tobacco industry lead to positive results.
23 4.2 Regression Model
For the regression model, the variables return on equity (ROE), payment in shares (PAYINSHARES), total assets of the acquirer (lnACQUIRERSIZE), cross-border deal (CROSSBORDER), percentage leverage of the acquirer (ACQUIRERLEVERAGE) and earnings before interest and tax (lnACQUIREREBIT) have been entered, which are all independent variables. The dependent variable is the cumulative abnormal return (CAR). The CAR(-1,1) is best to use given the fact that level of significance. See table 6 below for detailed results.
Table 6. Variables Entered/Removed
Model Variables Entered Variables Removed Method
1 ROE lnACQUIRERSIZE PAYINSHARES CROSSBORDER ACQUIRERLEVERAGE Enter
a. Dependent Variable: CAR(-1,1) b. All requested variables entered.
The model summary shows the R, R Square, Adjusted R Square and the Standard Error of the Estimate. The R shows the correlation between the predicted values and the observed values of the dependent variable. The R square, the coefficient of determination, represents the proportion of the variance in the dependent variable that is explained by the model (independent variables). The adjusted R square represents the proportion of the variance in the dependent variable that is explained by only those independent variables that really affect the dependent variable. The standard error of the estimate measures the accuracy of the predictions. The most important factor to look at is the R square. The R square is 0.233, which means that 23.3% of the dependent variable is explained by the independent variables. The percentage is not that high. However, it is acceptable because the abnormal returns are explained. See table 7 below for detailed results.
24
Table 7. Model Summary
Model R R Square Adjusted R Square
Std. Error of the Estimate
1 .482a .233 .176 .0488
a. Predictors: (Constant), ROE, lnACQUIRERSIZE, PAYINSHARES, CROSSBORDER, ACQUIRERLEVERAGE
The ANOVA table shows us a test on the model. The F-test is used to determine whether at least one independent variable has influence on the dependent variable. The F-value is 4.121 and the p-value is 0.003. From this values, it can be concluded that at least one independent variable shows a linear relation with the dependent variable. See table 8 below for detailed results.
Table 8. ANOVA
Model
Sum of
Squares df Mean Square F Sig.
1 Regression .049 5 .010 4.121 .003b
Residual .162 68 .002
Total .211 73
a. Dependent Variable: CAR(-1,1)
25 The coefficients table shows the t-values and p-values of the coefficient separately. The coefficient of CROSSBORDER is not significant, which means that cross-border deals do not have influence on the CAR(-1,1). The coefficient of PAYINSHARES is positively significant at the 10% level, which means that the method of payment does affect the CAR. If a company chooses to pay the deal in shares, the potential synergy gains are expected to move beyond the premium. The coefficient for lnACQUIRERSIZE is not significant, which tells that the total assets of the acquirer do not affect the CAR. The coefficient of ACQUIRERLEVERAGE is negatively significant at the 1% level, which means that the percentage of leverage of the acquirer does affect the CAR strongly. A company with a high percentage of leverage gives a sign of financial weakness, economic distrust and no potential growth. Therefore, it is not interesting for shareholders to invest in this particular company, which will result in a lower CAR. The coefficient of ROE is positively significant at the 5% level, which tells that the return on equity does affect the CAR. When a company has a high return on equity, it is likely that the CAR is also expected to be high. See table 9 below for detailed results.
𝐶𝐶𝐴𝐴𝑅𝑅𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝛽𝛽1 𝑃𝑃𝐴𝐴𝑃𝑃𝑃𝑃𝑁𝑁𝑃𝑃𝑃𝑃𝐴𝐴𝑅𝑅𝑃𝑃 + 𝛽𝛽2𝐶𝐶𝑅𝑅𝐶𝐶𝑃𝑃𝑃𝑃𝐶𝐶𝐶𝐶𝑅𝑅𝐶𝐶𝑃𝑃𝑅𝑅 + 𝛽𝛽3𝑙𝑙𝑙𝑙𝐴𝐴𝐶𝐶𝑙𝑙𝑙𝑙𝑃𝑃𝑅𝑅𝑃𝑃𝑅𝑅𝑃𝑃𝑃𝑃𝑙𝑙𝑃𝑃 + 𝛽𝛽4𝐴𝐴𝐶𝐶𝑙𝑙𝑙𝑙𝑃𝑃𝑅𝑅𝑃𝑃𝑅𝑅𝐴𝐴𝑃𝑃𝐴𝐴𝑃𝑃𝑅𝑅𝐴𝐴𝐴𝐴𝑃𝑃 + 𝛽𝛽5𝑅𝑅𝐶𝐶𝑃𝑃 + 𝜀𝜀𝑖𝑖
Table 9. Coefficients of the regression model
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) .056 .063 .888 .378 PAYINSHARES .030 .017 .181 1.704 .093* CROSSBORDER -.002 .017 -.013 -.118 .906 lnACQUIRERSIZE .001 .003 .017 .159 .874 ACQUIRERLEVERAGE -.144 .035 -.477 -4.097 .000*** ROE .018 .009 .246 2.111 .038**
a. Dependent Variable: CAR(-1,1) b. * Statistical significant at the 10% level
** Statistical significant at the 5% level *** Statistical significant at the 1% level
26
CHAPTER 5 Conclusions
This study investigated the effect of M&A announcements on share price value in the European food, beverage and tobacco industry between 2009 and 2018. Previous research focused mainly on the UK and US. Therefore, it was interesting to look at the E.U.. Furthermore, the food, beverage and tobacco industry is one of the largest manufacturing industries. Therefore, this industry is relevant.
A sample of 79 observations was observed with several restrictions through Zephyr. Furthermore, DataStream was used to find the stock prices of the different companies for the different time periods. The benchmark return of the market was determined by the MSCI European food, beverage and tobacco industry index.
To find an answer to the research question, an event study and a regression model have been performed. An event study tool of the Erasmus University Rotterdam has been used to determine the abnormal returns. Thereafter, the cumulative abnormal returns and the cumulative average abnormal returns were calculated. The results of the event study are significant for both event windows [-3,3] and [-1,1]. However, the [-1,1] window showed significance at the 5% level where the [-3,3] window showed significance at the 10% level. Therefore, it was best to use the CAR(-1,1) in the regression model. The results of the regression model show that ROE, ACQUIRERLEVERAGE and PAYINSHARES are significant at different levels. ROE shows significance at the 5% level, ACQUIRERLEVERAGE shows significance at the 1% level and PAYINSHARES shows significance at the 10% level.
In conclusion, the results show that the M&A announcements do significantly affect shareholder value in the European food, beverage and tobacco industry between 2009 and 2018. There are several limitations in this thesis given the limit of time. Further research could focus at a larger time period and/or lower deal value, which will result in more observations. Moreover, more observations could lead to private and public target firms, which could then be analysed separately. A larger event window e.g. [-10,10] could give the graphs of CAR even more credibility. Furthermore, the regression model could be adjusted. The R square of this research is 0.233, which means that 23.3% in the variation of the dependent variable is explained by the independent variables. Therefore, it could be interesting to look if there are other independent variables, which could take the adjusted R square to a higher level e.g. crisis dummy or public/private dummy.
27
REFERENCES
Alexandridis, G., Petmezas, D., Travlos, N.G. (2010). Gains from Mergers and Acquisitions Around the World: New Evidence. Financial Management, 39(4), 1671-1695.
Andrade, G., Mitchell, M., & Stafford, E. (2001). New Evidence and Perspectives on Mergers. The Journal of Economic Perspectives, 15(2), 103-120.
Andrade, G., Stafford, E. (2004). Investigating the economic role of mergers. Journal of
Corporate Finance, 10(1), 1-36.
Bjorvatn, K. (2004). Economic integration and the profitability of cross-border mergers and acquisitions. European Economic Review, 48(6), 1211-1226.
Brown, S.J., Warner, J.B. (1980). Measuring security price performance. Journal of Financial
Economics, 8(3), 205-258.
Campa, J. M., & Hernando, I. (2004). Shareholder value creation in European M&As. European financial management, 10(1), 47-81.
Cybo-Ottone, A., Murgia, M. (2000) Mergers and shareholder wealth in European banking.
Journal of Banking & Finance, 24(6), 831-859.
De Jong, F. (January 2007). Event Studies Methodology. Tilburg University.
Dieudonne, S., Cretin, F., Bouacha, S. (2014) M&A Activity: Where are we in the Cycle?
28 Franks, J.R., Harris, R.S. (1989). Shareholder wealth effects of corporate takeovers: The
U.K. experience 1955-1985. Journal of Financial Economics, 23(2), 225-249.
Goergen, M., Renneboog, L. (2003). Value creation in large European mergers and acquisitions. In Advances in Mergers and Acquisitions (pp. 97–114). Emerald Group Publishing Limited.
Gugler, K., Mueller, D.C., Yurtoglu, B.B., Zulehner, C. (2003). The effects of mergers: an international comparison. International Journal of Industrial Organization, 21(5), 625-653.
Healy, P. M., Palepu, K., Ruback, R.S. (1992). Does Corporate Performance Improve after Mergers? Journal of Financial Economics, 31(2), 135-175.
Kaplan, S.N., Weisbach, M.S. (1992). The Succes of Acquisitions: Evidence from Divestitures. Journal of Finance, 47(1),107-138.
Kiymaz, H., Baker, H.K. (2008). Short-Term Performance, Industry Effects, and Motives: Evidence from Large M&As. Quarterly Journal of Finance and Accounting, 47(2), 17-44.
Martynova, M., Oosting, S., Renneboog, L. (2006). The Long-Term Operating Performance of European Mergers and Acquisitions.
Martynova, M., Renneboog, L. (2008). A Century of Corporate Takeovers: What Have We Learned and Where Do We Stand? Journal of Banking and Finance, 32(10), 2148-2177.
29 Martynova, M., Renneboog, L. (2006) Mergers and Acquisitions in Europe. Advances in
corporate finance and asset pricing, 13-75.
Mariana, V. (2013). Mergers and Acquisitions Waves from The European Union Perspective.
Annals of Faculty of Economics, 1(2), 272-283
Moeller, S.B., Schlingemann, F.P., Stulz, R.M. (2005). Wealth Destruction on a Massive Scale? A Study of Acquiring-Firm Returns in the Recent Merger Wave. The Journal
of Finance, 60(2), 757-782.
Rau, P.R., Vermaelen T. (1998). Glamour, Value and The Post-Acquisition Performance of Acquiring Firms. Journal of Financial Economics, 49(2), 223-253.
Servaes, H. (1991). Tobin’s Q and the Gains from Takeovers. Journal of Finance, 46(1), 409-419.
Shleifer, A., Vishny, R.W. (2001). Stock Market Driven Acquisitions. Journal of Financial
