Shareholder wealth effects of mergers and acquisitions in Western Europe Amsterdam Business School Name Xander Jager Student number 10660755 Program Economics & Business Specialization Economics and Finance Number of ECTS 12 Supervisor Mr. R.J. (Robin) Döttling Target completion 31/1/2018 Abstract This thesis investigates the short-term wealth effects of acquiring firm’s shareholders in Western Europe from mergers and acquisitions announcements, over the period 2010 and 2017. I examine this with the event study methodology described by MacKinlay (1997). I provide detailed information about what mergers and acquisitions are, motives of M&A, merger waves, the method of payment and bid characteristics. Also, I analyze the expected shareholder wealth effect through testing shareholders’ Cumulative Average Abnormal Return (CAAR). Denmark, France and the Netherlands experience a statistically significant negative CAAR in different event windows around the announcement date. Furthermore, the results in this thesis suggest a statistically significant negative CAAR of the total Western European sample over a 10-day, 5-day and 1-day window centered around the event day. Additionally, I find that different methods of payment have a negative impact on the share prices of acquirer firms. Keywords: mergers and acquisitions; merger waves; method of payment; short-term wealth effects JEL classification: G32, G34
27
Embed
Shareholder wealth effects of mergers and acquisitions in ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Shareholder wealth effects of mergers and acquisitions in
Western Europe
AmsterdamBusinessSchool
Name Xander Jager
Student number 10660755
Program Economics & Business
Specialization Economics and Finance
Number of ECTS 12
Supervisor Mr.R.J.(Robin)Döttling
Target completion 31/1/2018
Abstract
This thesis investigates the short-term wealth effects of acquiring firm’s shareholders in Western
Europe from mergers and acquisitions announcements, over the period 2010 and 2017. I examine this
with the event study methodology described by MacKinlay (1997). I provide detailed information
about what mergers and acquisitions are, motives of M&A, merger waves, the method of payment and
bid characteristics. Also, I analyze the expected shareholder wealth effect through testing
shareholders’ Cumulative Average Abnormal Return (CAAR). Denmark, France and the Netherlands
experience a statistically significant negative CAAR in different event windows around the
announcement date. Furthermore, the results in this thesis suggest a statistically significant negative
CAAR of the total Western European sample over a 10-day, 5-day and 1-day window centered around
the event day. Additionally, I find that different methods of payment have a negative impact on the
share prices of acquirer firms.
Keywords: mergers and acquisitions; merger waves; method of payment; short-term wealth effects
JEL classification: G32, G34
2
Statement of originality
This document is written by Student Xander Jager 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
2. LITERATURE REVIEW ............................................................................................................................. 6
2.1 WHAT IS M&A? ............................................................................................................................................. 6 2.2 MOTIVES OF MERGERS AND ACQUISITIONS ................................................................................................... 6 2.3 MERGER WAVES ............................................................................................................................................. 8 2.5 BID CHARACTERISTICS ................................................................................................................................. 10
3. DATA AND METHODOLOGY ................................................................................................................ 11
3.1 DATA ............................................................................................................................................................ 11 3.2 METHODOLOGY ............................................................................................................................................ 11
3.2.1 Event Study ........................................................................................................................................... 11
4.1 EVENT STUDY RESULTS ................................................................................................................................ 15 4.2 IMPACT OF THE METHOD OF PAYMENT ......................................................................................................... 18
1. Introduction Mergers and acquisitions (M&A) are a popular way of corporate development. In 2017, companies
announced over 50.000 transactions with a total value of more than 2.9 trillion EUR. The number of
deals in 2014 increased by 12.63% compared to 2013 and the total value of the transactions increased
by 107,16%. The acquisition of SABMiller by Anheuser-Busch InBev SA/NV, BG Group by Royal
Dutch Shell and Covidien by Medtronic are examples of sizable acquisitions from Western Europe in
the last few years with a deal value of 91.93, 64.4 and 31.56 billion euros, respectively (IMAA, 2017).
Figure 1: Number and Value of M&A in Western Europe. Source: IMAA
M&A profitability generated a small peak of research in the last 40 years. There is disagreement
among researchers about the announcement wealth effects of bidding firms. Walker (2000), Mitchell
and Stafford (2000), Akbulut and Matsusaka (2010) and Healy et al. (1992) find small negative
returns for bidding companies. On the other hand, researchers reveal small or zero positive abnormal
returns (Martynova and Renneboog, 2006; Eckbo and Thorburn, 2002; Macquira et al., 1998;
Schwert, 2000; Loderer and Martin, 1990). Bidders and targets shareholders combined earn positive
adjusted returns so that one should suggest that M&A does pay. However, much research illustrates
around a zero return to bidding companies. Synergies, efficiencies and value-creating growth seem
hard to realize, so managers should be careful with participating in M&A activity (Bruner, 2002).
The most research of mergers and acquisitions focus on the United States or M&A activity in one
country. The literature on domestic mergers and acquisitions is consistent. The target firm’s
shareholders receive a premium of 20-40% on average relative to the stock price before the
announcement (Goergen and Renneboog, 2004). Mulherin and Boone (2002) report abnormal
announcement returns of 21% in the USA in the 1990s. Besides, average target abnormal returns in
5
the USA are 29% for 1963-1986, 24% for 1972-1987 and 27% for 1971-1982 (Jarrel and Poulsen,
1988; Servaes, 1991; Kaplan and Weisbach, 1992).
This thesis adds to this literature by examining the short-run effect of an M&A announcement in
Western Europe on stock prices. I provide further evidence on the creation of short-term shareholder
wealth effects expected by M&A activity. The research question is:
What is the short-term effect of an M&A announcement on abnormal returns for acquirer Western
European shareholders from 2010 till 2017?
I investigate the short-term returns for Western Europe mergers and acquisitions with the event study
methodology described by Mackinlay (1997). The event study has many applications. In finance
research, event studies practiced to a diversity of firm-specific and economy-wide events. To measure
the short-term shareholder value creation of M&A announcements, I use the methodology of the
market model. I apply the event study to a sample of 81 M&A announcements of Western European
acquiring companies from 2010 till 2017. Also, I investigate the impact of the method of payment on
stock prices.
I find a statistically negative CAAR for Denmark, France and the Netherlands in different event
windows around the M&A announcement date. The acquirer shareholders in Denmark, France and the
Netherlands lose money when a merger or acquisition occurs. Further, I reveal a statistically negative
CAAR of acquirer firm’s shareholder from Western Europe.
Also, all-cash bids, all-equity bids and the combination of cash, debt and stock bids in the
sample generate a negative price correction.
Finally, I claim that managerial hubris and the agency motive restrict the bidding firm’s
shareholders to experience significant positive abnormal returns in the seventh merger wave.
I organized this thesis as follows: section 2 shows the literature about mergers and acquisitions.
Section 3 describes the research methodology, variables and data sources. Section 4 provides the
results of the event study and the impact of the method of payment. Section 5 summarizes the main
conclusions from this study.
6
2. Literature review
2.1 What is M&A?
Mergers and acquisitions (M&A) are often referred as the market for corporate control. There is
typically a buyer, the “acquirer” or “bidder”, and a seller, the “target”, when one firm acquires
another.
The target company can merge with the acquirer company. Also, a corporation, firm or group
of individuals can acquire the target firm. These are the two primary mechanisms by which control
and ownership of a public corporation can change. The acquiring firm absorbs the target firm and
goes further on as one company with an acquisition. In the case of a merger, the acquirer and target
acknowledge going further as one new firm rather than two different companies separately owned.
The transaction with a merger or an acquisition is the same, because the acquirer must buy the shares
of the target firm. When an acquisition occurs, it is not necessary that the acquirer purchases the full
hundred percent shares of the target. There are also examples of acquisitions which involve a minority
of shares.
Mergers and acquisitions that take place in the same industry refer to horizontal mergers and
vertical mergers are M&As of firms producing different services or goods for a specific finished
product. Further, M&As between business unrelated to each other are called conglomerate mergers
(Berk and DeMarzo, 2011).
2.2 Motives of Mergers and Acquisitions Mukherjee et al. (2004) provide insights on the motives for acquisitions based on a survey among 75
CFOs in the United States during 1990-2001. The results show that the essential motive is synergy
(37.3%), the second motivation is diversification (29.3%) and lastly achieve a specific organizational
form as part of an ongoing restructuring program (10.7%). Other reasons that responding CFOs gave
for their M&As are: acquire a company below its replacement cost, use excess free cash and reduce
the tax of the combined company due to tax losses of the acquired company.
Acquirers justify their takeovers with many potential synergy reasons from mergers and
acquisitions while the forecasted benefits are often not realized. The majority of research presents that
there are three major reasons for takeovers: synergy motive, managerial hubris and the agency
problem. I illustrate these reasons now:
1. Synergy motive
The first motive for M&A is synergies, the primary motive for a merger. Synergy is a justification that
acquirers give for the premium they pay for a target. An acquirer could add economic value after the
acquisition and an individual investor could not have done this. Berk and DeMarzo (2007) describe
the most often reasons of acquirers to justify their takeovers: vertical integration, expertise, efficiency
7
gains, monopoly gains, tax savings from operating losses, economies of scale and scope,
diversification and earnings growth.
The synergy motive assumes that managers of acquirers and targets maximize shareholder
value and that they only participate in M&A activity if it results in gains to both of their shareholders.
Thus, when synergy motives takeovers the measured benefits to both acquirer and target shareholders
should be positive.
If the target can resist the acquirer or there is a race among possible acquirers for the specific
target, then the target has some bargaining power. The target gain will increase with the total gain.
Also, the wealth gains to the target shareholders correlate positively with the acquirer shareholders
and the total wealth gain (Goergen and Renneboog, 2002).
2. Managerial hubris
The second, unsuccessful, explanation of a merger or acquisition shows Richard Roll. Roll (1986)
brings forward that managerial hubris is the main reason that leads to a high number of value-
destroying mergers and acquisitions. Managers are likely overconfident in their abilities in general.
Research points out that it takes repeated losses for a manager to change his belief that he is above-
average at some actions (Berk and DeMarzo, 2007). The hubris hypothesis is an explanation of
corporate takeovers, which supports that CEOs that are overconfident seek mergers with a low chance
of creating value. The reason behind this comes from CEOs that believe their ability to manage is big
enough to succeed.
Next, Roll (1986) points out that hubris on individual decision makers in acquirer firms can
explain why bids made even when a valuation above the current market price represents a positive
valuation error. Acquirer firms that are infected by hubris pay too much for their target companies.
Goergen and Renneboog (2004) also show that one-third of the large European takeovers in the 1990s
experience managerial hubris. Other evidence of managerial hubris comes from Malmendier et al.
(2005). They conclude that overconfident managers, who choose to retain from shares in their firms,
make less profitable and diversifying takeovers.
3. Agency problem
The agency problem is the third M&A motive. Managers and shareholders are in possession of a
conflict of interest about payout policies when a firm has a substantial amount of free cash flow.
Managers may be tempted to use free cash flow to increase the size of the company, this is known as
‘empire building’. They use free cash flows for increasing the size of the firm instead of returning
these to the shareholders (Jensen, 1986). Further, managers have the incentive to run a more
prominent company, because of the extra salary and prestige it brings to run such an organization. The
private benefits are increasing for managers in correlation with the firm size (Goergen and
8
Renneboog, 2004).
Conyon and Murphy (2002) show an example from the United Kingdom that size is the
primary explanation of managerial salaries, bonuses and allocation of share options instead of
performance. Moreover, Harford and Li (2007) show that in 75% of the mergers where the acquiring
shareholders lose money, the acquiring CEOs are financially better off. Furthermore, Harford (1999)
indicates that the abnormal return of cash-rich acquiring companies becomes negative as a reaction to
takeover announcements and this is decreasing with the amount of free cash flow held by the
acquiring company. This negative abnormal return points out that acquiring firms with a substantial
amount of free cash flow destroy value by overbidding. As a consequence, cash-rich acquiring
companies that undertake value-decreasing acquisitions have a higher chance of being taken over in
the following years (Lang et al., 1991).
Managers are aware of destroying shareholder wealth under the agency problem, but they are better
off from doing so. The agency problem is different from the hubris hypothesis, because the
overconfident CEOs think they are doing an excellent job for their shareholders. Though, they are
overestimating their capabilities.
In summary, while synergy is the reason for the majority of the takeovers, there is substantial
evidence that agency problems and managerial hubris motivate acquisitions. However, agency
problems, not managerial hubris, seems to be the primary motive in takeovers with total negative
gains (Berkovitch and Narayanan, 1993). The findings of this thesis can suggest whether the bidding
firm’s shareholders experience negative or positive benefits due the synergy motive, managerial
hubris, the agency motive or all.
2.3 Merger wavesMerger waves are periods with increased M&A activity followed by a smaller amount of transactions.
Merger waves characterize the takeover market. During economic expansion, merger activity is higher
than during contractions and correlates with bull markets (Berk and DeMarzo, 2007). For this reason,
merger activity is higher in periods of economic expansion than during a financial crisis. M&As are
increasing significantly since the economy is improving the last years after leaving the financial crisis
behind. The financial crisis started in 2008 after the collapse of Lehman Brothers (Von Hagen et al.,
2011).
Merger waves can originate from market timing or clustering of industry shocks for which
mergers facilitate changes to the new environment. Economic, regulatory and technological shocks
drive industry merger waves. If the shock leads to a wave of mergers, this is due to whether there is
sufficient overall capital liquidity present to accommodate the necessary transactions. The economic
motivation for merger activity is the simultaneous peak in divisional acquisitions for cash (Harford,
2005).
9
The first European merger wave aimed at creating monopolies and took place between 1880
and 1904. Anti-trust regulation forbade monopolies and this formed the second merger wave (1919-
1929) which increased vertical integration (Goergen and Renneboog, 2002). The third merger wave
started in the 1950s and reached its peak in the 1960s. This wave is known as the conglomerate wave,
because firms became large conglomerates and bidding companies acquired firms in an unrelated
business. Hostile takeovers characterized the fourth wave and junk bond financing recognized this
period (Lipton, 2006). The fifth wave was the era of the mega-deal from 1993 to 2002. It ended with
the bursting of the Millenium Bubble, also known as the Dotcom bubble. Only three years later the
sixth wave started and was encouraged by globalization.
Moeller et al. (2005) examined the experience of acquirer shareholders in the fifth merger
wave. They found that acquisition announcements in the 1990s were profitable in the aggregate for
acquirer firm’s shareholders until 1997. But, the losses from the shareholders of acquiring companies
from 1998 through 2001 destroyed all the gains made earlier. Hence, acquisition announcements in
the fifth merger wave were costly for acquiring firm shareholders.
In this thesis, the focus is on the most recent merger wave: the seventh merger wave. There is not
much research done in this period, because the wave is still busy. In 2014, optimism was returning to
the market and the largest volume of transactions occurred since 2007. The environment of M&A
after the financial crisis, characterized by risk aversion and a focus on natural growth by firms, has
disappeared.
2.4 Method of payment
An acquiring company has to choose between three methods of payment in M&A transactions: an all-
stock bid, all-cash bid or a combination of cash, debt and stock. The choice an acquiring company
makes is a meaningful signal of the quality of the target firm. The price that an acquirer offers for a
target firm is the initial value of the target before the acquisition and increased by an acquisition
premium.
Jensen (1986) shows evidence that managers use free cash flow for ‘empire building’ as
mentioned before. Financing with cash has a positive effect on managers to using resources more
efficiently. Cash or debt, chosen as a payment method, will also generate more benefits than with
stock financing.
Furthermore, Myers and Majluf (1984) conclude with the pecking order theory that managers
with superior information first will choose to finance their investment with internally generated cash,
then debt and as third option equity. When an acquirer company issues debt, the stock price will not
fall. If equity is issued to finance an investment, the stock price will drop and this is not favorable for
acquirer stockholders. Asymmetric information between the acquirer’s management and outside
investors can influence the choice of payment.
10
The method of payment in a takeover bid affects the share price of a target firm. All-cash
offers and combined cash, equity and loan notes bids have higher abnormal returns than all-equity
bids. If the real value of bidding firm’s shares is worth more than the current price, bidding managers
will issue equity, because they believe that their firm’s shares are overpriced. Managers issuing equity
corresponds with the theory that they release new shares when the stock market is booming (Goergen
and Renneboog, 2004; Martynova and Renneboog, 2006). Acquiring firms that finance their
investment with stocks are overvalued and they use cash to fund their investment when they think
their shares are undervalued (Shleifer and Vishny, 2003).
In the results, I investigate how the market responds to the announcement of M&A transactions for
acquiring firms by the method of payment used in the transaction. I expect a positive price correction
for all-cash bids, because the market interprets the financing choice as a signal about the firm’s over-
or undervaluation. I predict a negative price correction for all-equity bids and bids with equity
involved. The share price of the bidding firm should increase with an all-cash bid and decrease with
an all-equity bid.
2.5 Bid characteristics
An important factor that defines a transaction is the approach of an acquirer. Either the acquiring
company makes an offer to the Board of Directors of the target company. Or the acquirer ignores the
Board and makes the offer directly to the target shareholders; this is a hostile takeover. The approach
of an acquirer company is important, because the type of takeover tactic has a sizable impact on the
short-term wealth effects for the target firm shareholders.
Acquirer shareholders can have negative returns with a hostile takeover since they would
have to pay a higher premium and target shareholders obtain higher returns due to competitive bids
leading to better offer prices. The impact on short-term wealth effects for target shareholders during a
hostile takeover is substantially higher (announcement effect of 12.6% on day 0) than for mergers and
friendly acquisitions (8% and 22% on day 0).
Hence, the market is expecting that opposition against a bid lead to a change of the offer and
eventually to a higher premium. The study of Goergen and Renneboog (2004) confirm this: the
announcement of a merger or friendly acquisition causes a positive abnormal gain of 2.5% whereas a
hostile acquisition produces a negative abnormal return of 2.5%.
11
3. Data and Methodology In this chapter, first I show how I collect the data for this research and which criteria I set. Then I
discuss the event study of MacKinlay (1997).
3.1 Data In this section, I present the used data sample. The sample includes Western European M&A deals
completed between 01/01/2010 and 01/01/2017. Data of Western European mergers and acquisitions
are from the Zephyr database. Zephyr contains information about 1.2 million transactions and rumors.
The selection criteria for M&A announcements are:
1. Mergers and acquisitions announced between 1 January 2010 and 1 January 2017
(completed-confirmed)
2. The acquirer and target has to originate from Western Europe1
3. Percentage of the final stake is hundred percent
4. The acquirer and target are listed or delisted companies
I collect a sample of 97 M&A announcements and I add an overview of the sample in the appendix as
Appendix 1. However, I exclude 16 announcements due to lack of available information. As a result,
the sample consists of 81 M&A announcements. From the sample, I use the daily return on stock
price. The daily gains of the sample are from the database: DataStream. DataStream is a global
macroeconomic and financial data platform covering stock market indices, currencies, equities and
more.
3.2 Methodology
3.2.1 Event Study
I investigate the stock price reactions of bidding firms to measure the announcement effect of M&A.
Based on the efficient market hypothesis, stock prices fully reflect available information about a
company and its future (Fama, 1970). Event studies, survey data, accounting studies and case studies
are the four primary methods to determine the effect of mergers and acquisitions.
I use an event study to measure the effect of an M&A announcement on the abnormal returns
of the stock. Event studies have a long history. The first published study is the work of James Dolley
(1933). James Dolley examines in his work the price effect of stock splits, studying nominal price
changes at the time of the division (MacKinlay, 1997). An event study measures the impact of an
event on the stock price of a firm with financial data over a time window. Based on the papers of
MacKinlay (1997) and Brown and Waner (1985) the methodology for this study is determined.
1 Zephyr classified Western European countries as: Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Gibraltar, Greece, Iceland, Ireland, Italy, Liechtenstein, Luxembourg, Malta, Monaco, Netherlands, Norway, Portugal, San Marino, Spain, Sweden, Switzerland, Turkey and the United Kingdom.
12
First, I identify the event day; the day a Western European company announces the merger or
acquisition. The event day is the first day that the market is informed about the new information.
Then I should define the event window, the period during which effects of the event are
possible detected. The event window must be long enough to cover the whole event, but unnecessary
days must be excluded that can negatively affect the power of the test. In a perfectly efficient market
it would be sufficient to restrain the event window only to include the event day, but in practice, the
event window is expanded due to possible information leakage of the event days before it. The impact
of the event on the price of the stock is reflected in the days after the event day. I use an event window
of 21 days and note this as [-10, +10]. The event window consists of ten days before the M&A
announcement and ten days after. Also, I analyze other event windows: [-5, +5], [-2, +2] and [-1, +1]
based on the articles of Brown and Warner (1980,1985). 𝐿" = 𝑇" − 𝑇' is the length of the event
window.
After defining the event window, I determine the estimation period. During the estimation
window, no event takes place. Estimating the estimation period is necessary, because I need to decide
how the returns of the stock should be under normal circumstances if no event had taken place. The
estimation period and the event window must not overlap to obtain unbiased estimates. The estimation
period, 𝐿' = 𝑇' − 𝑇(, is determined at 180 days before the start of the event window: 𝐿' =
−170 − −10 . I choose for an estimation period of 170 days.
Figure 2: Event study timeline. Source: MacKinlay (1997)
Normal returns would be the returns on the stock if no event took place. Chatterjee (1986) noted that
monthly data to estimate regression parameters give two fundamental problems: the long estimation
period is relatively sure to pick up non-stationarity in the regression parameters and the Cumulative
Abnormal Returns are probably affected by events which are not related to the merger. For that
reason, I use daily returns.
An estimation of the impact on the event requires a measure of the abnormal return. The abnormal
return for firm i and event date τ is:
𝐴𝑅/,1 = 𝑅/,1 − 𝐸 𝑅/,1 𝑋𝜏)
13
where 𝐴𝑅/,1, 𝑅/,1and𝐸 𝑅/,1 𝑋𝜏)are the abnormal, actual and normal returns for a given security 𝑖and
dayτ.
The market model is a commonly used method for creating the normal return. Also, the constant mean
return model is a method to generate the normal return. In short-term event studies, simple approaches
like the constant mean return model and the market model work adequately (Brown and Waner,
1985). In this thesis, the model used to test the existence of abnormal returns in the event period is the
market model.
The market model, a statistical model, demonstrates the returns of any given security to the
return of the market portfolio. A primary assumption of the market model is the joint normality of
asset returns. The market model for any security i is:
𝑅/,1 = 𝛼/ + 𝛽/𝑅>1 + 𝜀/1
𝐸(𝜀/1 = 0)𝑣𝑎𝑟 𝜀/1 = 𝜎EFG" ,
where 𝛼/, 𝛽/ and 𝜎EFG" are the parameters of the model and 𝜀/1 is the zero-mean disturbance term. 𝑅/,1
and 𝑅>1 are the period-t returns on security i and the market portfolio.
For the market portfolio, a broad-based stock index is used such as the AEX, S&P 500 and the CRSP
Value Weighted Index. Here, I use the Stoxx Europe 600 Index because all firms of the sample are
European companies.
The market model parameters use ordinary least squares (OLS) as a measurement procedure under
common conditions. For an estimation window of observations, the OLS estimators of the market
model parameters are:
𝛽H =(𝑅/I − 𝜇/)(𝑅>I − 𝜇>)
KLIMKNO'
(𝑅>I − 𝜇>)"KLIMKNO'
𝛼/ = 𝜇/ − 𝛽/𝜇>
𝜎EF" =
1𝐿' − 2
(𝑅/I − 𝛼/ − 𝛽/𝑅>1)"KL
IMKNO'
where 𝜇/ ='QL
𝑅/IKLIMKNO' and 𝜇> = '
QL𝑅>I
KLIMKNO' . 𝑅/I and 𝑅>I are the returns is event period 𝜏
for security i and the market.
14
Further, I explain the use of the ordinary least squares estimators to measure abnormal returns and to
develop their statistical properties. First, I present the statistical properties of abnormal returns
followed by the aggregation of abnormal returns.
The abnormal returns are measured and analyzed with the parameter estimates of the market
model. 𝐴𝑅/,1, 𝜏 = 𝑇' + 1, … . . , 𝑇", is the sample of 𝐿" abnormal returns for firm i in the event
window. The abnormal sample returns by using the market model to measure the normal return are:
𝐴𝑅/,1 = 𝑅/,1 − 𝛼/ − 𝛽/𝑅>1
When the length of the estimation window, 𝐿', becomes large the variance of the abnormal return will
be 𝜎EF" .
Second, I consider the aggregation of the abnormal returns. Cumulative abnormal return (CAR) is
needed to modify the multiple period event windows. The CAR of one of the Western European
countries from 𝑡'to𝑡" is:
𝐶𝐴𝑅 𝜏', 𝜏" = 𝐴𝑅/I
IW
IMIL
The variance of CAR is:
𝜎/" 𝜏', 𝜏" = (𝜏'−𝜏" + 1)𝜎EF"
I analyze the Cumulative Average Abnormal returns to study the different abnormal returns for all
Western European countries. The CAAR is:
𝐶𝐴𝐴𝑅 𝜏', 𝜏" =1𝑁
𝐶𝐴𝑅/(𝜏', 𝜏")Y
/M'
𝑣𝑎𝑟 𝐶𝐴𝐴𝑅 𝜏', 𝜏" = 1𝑁" 𝜎/" 𝜏', 𝜏"
Y
/M'
where N is the number of M&A announcements for each Western European country.
15
Finally, I test CAAR for significance. The hypothesis for each Western European country (i) is:
𝐻(:𝐶𝐴𝐴𝑅/ = 0
𝐻':𝐶𝐴𝐴𝑅/ ≠ 0
𝐻( can be tested using:
𝜃' =𝐶𝐴𝐴𝑅 𝜏', 𝜏"
𝑣𝑎𝑟 𝐶𝐴𝐴𝑅 𝜏', 𝜏"'/" ~𝑁(0,1)
4. Results In this chapter, I present the empirical findings from the event and analyze the numbers and their
significance. Also, I answer the central research question of this thesis by rejecting or accepting the
hypothesis made in the last chapter. Furthermore, I discuss the impact of the method of payment on
stock prices.
4.1 Event study resultsAfter applying the event study methodology by MacKinlay (1997), I calculated the Cumulative
Average Abnormal Returns (CAAR) for all Western Europe countries with different event windows.
In this research, I used the Erasmus Event Study Tool and made further calculations with Microsoft
Excel. I report the CAARs of the acquiring countries for the period 2010-2017 and their significance
in Table 1 below. Figure 3 illustrates a plot for the whole sample.
16
Table 1: CAARs of Western European countries for event windows [-10,10], [-5,5], [-2,2] and [-1,1]. ***, ** and *
denote significance at the 1%, 5% and 10% levels, respectively.
Country Number of
observations
CAAR [-10,10] CAAR [-5,5] CAAR [-2,2] CAAR[-
1,1]
Austria 1 -0.0177
-
-0.0207
-
-0.0221
-
-0.0131
-
Germany 6 -0.0329
(-1,0288)
-0.0154
(-0,7319)
0.0031
(0,1711)
-0.0048
(-0,2936)
Denmark 5 -0.0315
(-0,8534)
-0.0526*
(-1,8585)
-0.0351
(-1,13170)
-0.0234
(-0,9789)
Spain 3 0.0431
(-0,0237)
0.0538
(1,4005)
0.0623
(1,3305)
0.0567
(1,2033)
Falkland
Islands (GB)
1 -0.0237
-
-0.0527
-
-0.0274
-
-0.0270
-
France 24 -0.0248**
(-1,7973)
-0.0189*
(-1,4723)
-0.0113
(-1,1908)
-0.0076
(-0,9858)
United
Kingdom
16 -0,0450
(-0,3915)
-0,0388
(-0,3747)
-0,0303
(-0,4157)
-0,0254
(-0,4233)
Greece 2 -0,4925
(-0,7200)
-0.0796
(-0,3012)
0.1089
(0,7420)
0.0120
(0,1473)
Ireland 1 0.0121
-
0.0095
-
-0.0251
-
-0.0217
-
Italy 6 0.0397
(1,0722)
0.0470
(2,1651)
0.0205
(0,8725)
0.0135
(0,5543)
Netherlands 5 -0.1267**
(-2,4978)
-0.0716
(-0,6927)
0.0188
(0,1999)
-0.0384
(-0,9207)
Norway 1 -0.0032 0.0330 -0.0320 -0.0288
Switzerland 10 -0.0419
(-1,1603)
-0.0202
(-1,0187)
-0.0140
(-0,6835)
-0.0043
(-0,2475)
Western
Europe
81 -0.0416***
(-3,3417)
-0.0214**
(-2,3083)
-0.0066
(-0,8808)
-0.0097**
(-1,8230)
17
Figure 3: CARs of Western European countries for event window: [-10,10]
As a result, Table 1 shows that I reject the null hypothesis, the Cumulative Average Abnormal
Returns is zero, for Denmark, France and the Netherlands at 10% significance level. Investors owning
shares in acquirer firms from France and Denmark for a period starting five days prior the event day
and selling their shares after five days of the event day would receive a significant negative return of
2% and 5%, respectively. Also, I reject the null hypothesis that there is no effect of the CAAR at 5
percent significance level for France and the Netherlands over a 10-day event window centered
around the event day. The statistically significant CAAR amounts to -2% for France. The price
reactions for the bidding firm’s shareholders in the Netherlands are large: Table 1 shows a significant
negative CAAR of 13%. These results indicate that the announcement of a merger or acquisition
generates a loss for the acquirer shareholders in Denmark, France and the Netherlands.
However, the CAAR of the residual countries and different event windows is not proven
significant at 1%,5% or 10%. Thus, I cannot reject H( for the remaining countries. There is no effect
of the Cumulative Abnormal Average Return at every significance level. Thus, shareholders of
acquiring firms in the residual countries do not lose or gain.
Also, there is a significant decrease in Western European acquirer share prices over 10-day, 5-day and
1-day window centered around the event day: -4% at 1% significance level, and -2% and -1% at 5%
significance level. Thus, M&A announcements destroyed the wealth of bidding firm’s shareholders in
Western Europe. The results are in line with the literature from Walker, (2002); Mitchell and Stafford,
(2000); Akbulut and Matsusaka, (2010); Healy et al., (1992). Though, the result is not proven
significant for the event window two days before and after the event day.
Additionally, Figure 3 demonstrates a positive abnormal return of 4% for all Western