Master Thesis

The Signalling Effect of M&A

Announcements on
Performance
Findings on the signalling effects of M&A
announcements given the moderating effect of
payment method, reputation and experience on
performance of acquiring companies

Master Corporate Finance & Control

Name S. Luijkx (Sam)

Student number 4608631
Date 15 august 2021

Name of assigned supervisor Dr. A.G. Kopányi-Peuker

Radboud University Nijmegen

Abstract

This thesis investigates if and to what extent announcements of mergers and acquisitions act
as a signal to the market, thereby influencing the short-term performance of acquiring
companies. Also the effects of the method of payment, the experience of the acquirer and the
reputation of the target is analyzed as they can either enhance or diminish the effects of the
announcement as a signal. Due to the inherent uncertainty of the M&A landscape due to
information asymmetry, investors rely on observable characteristics to evaluate a company.
The assumption here is that the announcement provides a particular signal to the market about
the quality of the venture.

Where previous studies describe the effects of signals, there is little empirical evidence
on the effects of an M&A announcement as a signal. This thesis tries to combine signalling
theory with the stream of literature on M&A performance. This thesis also contributes to the
stream of literature by analyzing under what circumstances or preconditions, signal are
received that either improved or worsened the effect of M&A announcements on acquirer
performance.

Using a sample of 85 transactions in Europe over the period of 2017 to 2019, no

significant results were found when looking at the performance enhancing or diminishing
effect of announcements. Moreover, an additional multiple regression analysis showed no
significant moderating effects for the method of payment, the experience of the acquirer or the
reputation of the target. Thus, the effect of a M&A announcement as a signal is somewhat
limited, and that acquiring companies should not extensively concern themselves with the
preconditions of signals surrounding M&A’s. This study contributes to the stream of literature
on signalling theory, and especially on its relation with M&A events. Implications and
directions for future research are provided.

Keywords: M&A, Acquirer Performance, Reputation, Signalling

2
Table of Contents

1. Introduction ............................................................................................................................ 4
2. Theoretical framework ........................................................................................................... 7
2.1 M&A performance ........................................................................................................... 7
2.2 Signaling theory ............................................................................................................... 8
2.2.2 Mode of payment .................................................................................................... 10
2.2.3 Reputation ............................................................................................................... 11
2.2.4 Acquirer experience ................................................................................................ 12
2.3 Conceptual model:.......................................................................................................... 13
3. Methodology ........................................................................................................................ 14
3.1 Research method ............................................................................................................ 14
3.2 Measuring constructs – Event Study .............................................................................. 16
3.2.1 Time parameters identification................................................................................ 16
3.2.2 Normal return calculation........................................................................................ 17
3.2.3 Abnormal return calculation and analysis ............................................................... 18
3.3 Measuring constructs – Multiple Regression ................................................................. 18
3.4 Sample ............................................................................................................................ 20
4. Results .................................................................................................................................. 22
4.1 Event study ..................................................................................................................... 22
4.2 Correlation analysis ........................................................................................................ 23
4.3 Regression analysis ........................................................................................................ 26
5. Discussion ............................................................................................................................ 30
5.1 Theoretical implications ................................................................................................. 30
5.2 Practical implications ..................................................................................................... 32
5.3 Limitations and future directions ................................................................................... 32
5.4 Conclusion ...................................................................................................................... 34
References ................................................................................................................................ 35
Appendix .................................................................................................................................. 40

3
1. Introduction
In order for efficient trade to take, all parties involved should have the same information and
equal availability to it. The price in this case fully reflects the value that it actually represents,
leaving no room for misinterpretation or error. However, this situation is rarely present in
practice, with buyer and seller having a mismatch in available information. Often the seller
withholds information leaving the buyer at risk to adverse selection, or overpayment risk. The
buyer however has no other option than to deal with this adverse selection problem, or cancel
the deal. This situation is also not profitable with regard to the seller side as price discounting
occurs by buying (Wu, Reuer & Ragozzino, 2013).

The information problem described above is thoroughly researched in practice, with

the M&A context offering an interesting research landscape for economics, finance, and
strategy literatures. Often involving a great deal of resources, M&As provide companies with
the opportunity to deploy their own assets to exploit external growth opportunities. However,
in the transaction process of M&A there is information asymmetry between the buying and
selling side. After thorough valuation processes, which often involves a lot of intangible
assets, the buying side turns to a due diligence process just to make sure that they value the
company for what it is actually worth. Especially for private companies, which have fewer
obligations to provide information to the market in contrast to public companies, the valuation
process can be very lengthy and difficult. Overall, M&A transactions are not that efficient due
to this information asymmetry problem (Wu & Reuer, 2021).

Originally being a theory related to human capital (Spence, 1973), signalling theory
has moved beyond the scope of human capital, and entered management literature, where it
also is applied to M&A related literature (Riley, 2001). In M&As the effectiveness of signals
can reduce the adverse selection. For example signals can reduce costs of buyers as they
signal the quality of good potential sellers (Pollock & Gulati, 2007), or signals can reduce the
offer price discounting that would arise from asymmetric information between sellers and
buyers (Reuer, Tong & Wu, 2012). Moreover, it can even be the signal of hiring a
management consultant that improves the value of a company significantly (Bergh &
Gibbons, 2011). Signalling theory therefore can provide a solution to the adverse selection
problem and information asymmetry that is described above.

One important factor that has been the topic of much event-study related research in the
realm of M&A literature, is the announcements of M&As. Most studies on mergers and

4
acquisitions reveal mixed results concerning abnormal return during announcements periods.
In this study, I propose that the announcement of an M&A can provide a signal of the
profitability and future prospect to the market, to which it affects the short-term performance
of the acquiring company, which is also touched upon in recent literature (Bessler, Drobetz &
Zimmerman, 2011; Colombo, 2021; Yang & Lander, 2018;). These announcement signals are
beneficial to investors as they often do not have access to direct information, and therefore
rely on observable characteristics or signals.

The extent to which performance is affected depends upon multiple factors, including
the method of payment used, the target reputation and acquirer experience. Previous studies
find that the method of payment chosen by the acquirer has an negative impact on the
announcement returns when funded with equity and a positive effect when funded through
cash (Bessler, Kruizinga & Westerman, 2020). The reputation of the target is important as
information asymmetry is less severe in these cases as the company has proven to others that
it provides a profitable investment (Colombo, 2021). Also, more experienced acquirers tend to
focus more on their own skills and therefore rely less on upon signals of performance (Wu &
Reuer, 2021). This results in several questions that are central to this study:

Research question:

• To what extent do M&A announcements, through signal, affect short term

performance of the acquiring company?

Sub questions include:

- What is the moderating effect of the payment method used on acquirer performance?
- What is the moderating effect of target reputation on acquirer performance?
- What is the moderating effect of acquirer experience on performance?

This study offers two primary theoretical contributions. First, this study tries to combine the
stream of literature related to M&A performance to signalling theory to further explore the
effectiveness of signals that signals can have on their receivers, something that has been
called upon in recent literature (e.g. Connelly et al., 2011; Park & Patel, 2015; Wu & Reuer,
2021). Analyzing a sample of 85 transactions in Europe over the period of 2017 to 2019, this
study showed that the performance that there is no significant effect announcements as a
signal on the short term performance the acquiring company.

5
 Second, I furthermore analyze signals by looking into important factors that could
influence the perception of their receivers, which is also highlighted by previous research (e.g.
Colombo, 2021). I find that combining the method of payment, the experience of the acquirer
and the reputation of the target in one overarching model did not show any significant results.

For practitioners, findings of this thesis could help managers in determining the best
way to go about when dealing with signals in potential M&A deals. Although targets’ signals
are publicly available to all prospective acquirers, not all acquirers will naturally perceive and
act on them (Pollock & Gulati, 2007).

This thesis consists of five chapters. Chapter two outlines the theoretical framework
(§2) where chapter three describes the methodology (§3). In chapter four the quantitative
analysis of the data and the findings of the analysis is described (§4) and finally, in chapter
five, the research findings, its implications, limitations, future research directions and the
conclusions are outlined (§5).

6
2. Theoretical framework
This chapter first elaborates on the stream of literature surrounding M&A and M&A
performance (§2.1). The possible effects that signals can have are then touched upon in the
next section (§2.2). I then also turn to the potential signaling effects of the mode of payment
in M&A announcements (§2.2.2). I also elaborate upon the reputational effects of
announcements (§2.2.3) followed by the effects of the experience of the acquirer (§2.2.4).

2.1 M&A performance

The performance of mergers and acquisitions provides an interesting field of study, which has
been dealt with extensively over the past few decades. Although one of the most studied fields
of research, the academic community is divided on whether M&As provide real benefits to
acquiring firms with it being a favorite growth strategy for businesses across the globe. A
surprising and quite shocking 44 to 50 percent of the M&As do fail, with a large stream of
literature proposing that M&As do not pay off (Cartwright & Schoenberg, 2006; Das & Kapil,
2012). Some research however point out that there are unidentified variables that might play a
big role in explaining the variance in acquisition performance (Das & Kapil, 2012), or that
there are problems with empirically measuring acquisition performance as it is a complex
phenomenon (Zollo & Meier, 2008). Furthermore, there is research which is actually in favor
of the effectiveness and profitability of M&As as shareholders and companies can benefit
from M&As. Empirical studies show that they often increase market share and market power,
provide economics of scale and scope, lower cost of capital, and alleviate redundant corporate
costs (Ross, Westerfield & Jordan 2009; Ma, Zhang & Chowdhury, 2011). Overall, the
literature on M&A performance seems divided.

The performance in M&As can be affected in many ways. Common determinants of

acquirer performance in M&As are for example pre-acquisition performance and proportion
of the state shares has a positive impact on performance of acquiring companies (Changqi &
Ningling, 2010). Acquirer size and political connections can also be determinants for good
M&A performance (Colombo, 2021). If an acquisition is made by diversifying
geographically, the shareholder value and long term performance is increased (Doukas &
Lang, 2003). Industry related variables such as industry growth opportunity, industry
concentration and cash flows are important determinants of merger activity as well (Agrawal
& Sensarma, 2007).

7
 Certain events can also be a determining factor for performance of companies. Such
events are earnings announcements, dividend announcements, and in case of this thesis it can
also be M&A announcements. M&A announcements are extensively examined as well in
research surrounding deals. When M&A’s are announced, which is different from the deal
actually be completed, this is reflected in the stock price of the acquiring company. The extent
to which announcements impact acquirer performance seem to be divided, just as the overall
discussion about the effectiveness of M&A’s, with research showing positive and negative
results (e.g. Bao & Edmans, 2011; Yang & Lander, 2018; Zhao, Ma & Hao, 2019; Wu &
Reuer, 2021). The relation between announcements and acquirer performance has been
further explored with regard to other moderating variables. For example acquirer size plays a
significant negative role on announcement returns (Zhao, Ma & Hao, 2019).

Investigating events such as the announcement of an M&A is relevant as they can be

beneficial to companies in terms of influencing their stock price in both the short and long
run. This effect on the stock price ultimately also influences performance of the acquiring
company. Determining under what circumstances these events actually are beneficial to the
acquiring company is therefore an interesting field of research. I will now argue on how
signalling theory can be a theoretical framework for further examining M&A announcements
as signals can provide a vehicle to communicate future prospects and companies potential to
the market.

2.2 Signaling theory1

Traditionally signaling theory stems from work in social capital (Connelly et al., 2011;
Spence, 1973). The work of Spence (1973) has added additional insights to the information
asymmetry problem. In his article, Spence argues that education can function as a signal of an
employee’s productivity, as education is costly and is positively related to an individual’s
unobservable productive potential. This causes more productive recruits to signal their value
to employers by earning an education and stand out with regard to the other potential
employees. In general, an action qualifies as a signal and helps differentiate a seller’s product
or resource when (1) the action is readily observable to other market participants, (2) the
action is positively related to the unobserved quality of the product or resource being
transacted, and (3) it is more costly for other sellers lacking the same quality level to imitate

1
This section is partially based on Luijkx (2020), where an overview of the signalling theory was also given

8
the action (Bergh, Connelly, Ketchen, & Shannon, 2014; Spence, 1974). Over the last few
years, signalling theory has moved beyond the scope of human capital, and entered
management literature, where it also is applied to M&A related literature.

In the M&A landscape, signalling theory essentially builds on the idea that when there
is a lot of information asymmetry in the market, there is a distinction between high quality
and low quality senders. This distinction originates from the fact that signals are costly and
therefore only a subset of senders can transmit signals due to costs associated with sending
these signals (Bergh et al., 2014). Entrepreneurial companies and mature companies alike aim
to convince or persuade investors of the merits of their firm. In other words, the decision of an
investor to invest is dependent on what the funding receivers consider the basis of the
judgment and what constitutes relevant evidence for consideration (Chen, Yao, & Kotha,
2009). This evidence is being sent in the form of these costly signals by the companies as
described above. These signals are beneficial to investors as they often do not have access to
direct information, and therefore rely on observable characteristics or signals (Kaplan &
Stromberg, 2003). A signal related to quality can overcome or reduce the information
asymmetry which is inherently present in deals surrounding M&As (Amit, Brander, & Zott,
1998; Li & Chi, 2013), which helps investors in mitigating adverse selection or overpaying
for the target (Ragozzino & Reuer, 2007).

In the previous section was mentioned that determining under what circumstances
events, or M&A announcements more specially, is relevant as it can have a performance
enhancing effect. Signalling theory provides a theoretical framework for answering this
question as it proposes the announcement of the M&A, the signal in this case, as a
communication vehicle that the acquiring company can use in order to influence their stock
price, and therefore their performance. The extent to which the announcement of an M&A
influences performance has been thoroughly researched, with performance being measured in
multiple ways (for a meta-analysis see Zollo, 2008).

The announcement of an M&A provides a signal about the future prospects of an

acquirer and influences the value accordingly. Related research for example shows how
simply being in the news as a company can positively impact acquirer performance in the
short run (Yang & Lander, 2018). Similar findings also suggest that investment banks have a
significant effect on M&A announcements and the returns of M&A deals following those
(Bao & Edmans, 2011). Moreover, when an acquisition is announced, it signals the potential
for future growth in the acquirer‘s industry to the market, resulting in positive stock market
9
reactions to rivals of the acquiring firms (Gaur, Malhotra & Zhu, 2013). A meta-analysis by
Yasar, Martin and Kiessling (2020) receivers do react to positive signals from a credible
insider signaler to obviate information asymmetry, and also show that receivers react much
stronger to negative signals. Overall, related literature seems fairly convinced about the
significant effect an announcement as a signal can have on performance. Following the lines
of communicating performance to investors with regard to signalling theory, the signalling of
information should affect future performance positively resulting in the following hypothesis:

Hypothesis 1: The announcements of an M&A has a positive effect on the performance of the
acquiring firm.

2.2.2 Mode of payment

There are however, some influencing factors that could be of interest when considering the
announcement of an M&A as a signal. These signals could either enhance or dimmish the
effect of the signal that is being sent, as it influences the way that the signal is perceived by its
receivers.

The first factor that is of interest is the mode of payment that an acquiring company
chooses to use. The most commonly used methods of payment are full cash, full share-
exchange or a combination of both. As a result of information asymmetry, firms acquire
targets through equity if they believe that their firms’ shares are overvalued however if they
know that the firms’ shares are undervalued they prefer cash offers when acquiring target
instead (Leland & Pyle, 1977; Myers & Majluf, 1984). Therefore, the method of payment can
act as a signal about the acquiring firm’s value with cash offers being interpreted as good
news while equity offers are interpreted as bad news. Consequently, cash offer M&A
proposals should trigger positive market reaction while share offer M&A proposals are
expected to have a negative market reaction impact on the acquiring firm’s share price.
Overall, cash offers acquisitions should generate more abnormal return than equity offer
acquisitions (Bessler, Kruizinga & Westerman, 2020).

Hypothesis 2a: If the method of payment is focused on cash, this has a positively moderating
effect on the relation between announcements and performance

10
Hypothesis 2b: If the method of payment is focused on equity, this has a negatively
moderating effect on the relation between announcements and performance

2.2.3 Reputation
Corporate reputation affects the corporate value, performance, and risk of firms, so their
reputation may be considered as a strategic intangible asset (Veh, Göbel & Vogel, 2019). If
the target company already has been funded in multiple rounds, and is backed by multiple
companies and investors, it is considered to have a good reputation. The information
asymmetry is less severe in these cases as the company has proven to others that it provides a
profitable investment, thereby reducing the extent to which new investors should rely on
signals about performance to overcome the information asymmetry (Colombo, 2021).
Political connections, for example, are related to target reputation and previous research has
shown a positive effect on mergers and acquisitions announcement returns (Zhao, Ma & Hao,
2019). When firms with reputation that is considered good send out signals to the market,
these are considered to be more trustworthy and realistic to their receiver, which in turn
causes these receivers to act more on these signals.

Besides that, research has shown that forming alliances with poor quality ventures will
present more risk to well-established firms’ accumulated reputations, therefore more
prominent alliance partners tend to ally with higher quality ventures (Gulati & Higgins, 2003).
Acquirers will be inclined to seek out targets who signal a better reputation than their peers as
these companies pose less risk to the acquirers reputation, thereby increasing their short term
performance more often. Similarly, recent research has shown that acquirers can rely on
signals coming from the affiliation that a company has to gauge the venture’s resources and
future prospects (Wu & Reuer, 2021). This results in the following hypothesis:

Hypothesis 3: The reputation of the target company has a positive moderating effect on the
relationship between M&A announcements and performance

11
2.2.4 Acquirer experience
Acquisitions are complex events that fail for numerous reasons. Some of these possible
reasons might be that acquiring firms may select the wrong target, pay too much for it and
poorly integrate it (Colombo, 2021). It might also explain why acquisitions yield inadequate
returns for acquiring firms. Acquisition experience is described as a “principal mechanism by
which firms attain extraordinary skills” (Hayward, 2002). There is however mixed evidence
that acquirer experience is sufficient to ensure superior acquisition performance (e.g. Zollo &
Singh, 2000). Research shows how acquirer’s general M&A experience makes it more likely
to act on signals and learning from prior experience in acquisitions might help enhance
acquisition performance (Barkema & Schijven, 2008).

However, although targets’ signals are publicly available to all prospective acquirers,
not all acquirers will naturally perceive and act on them (Pollock & Gulati, 2007). Moreover,
meta-analysis has not found a consistent and significant effect for acquisition experience
(King, Dalton, Daily & Covin, 2004). Also, experienced acquiring companies tend to rely less
on signals about performance, and rely more on their own experience in the field. (Wu &
Reuer, 2021). Also, targets with a patent portfolio which has the potential to block other
patents are of high value to the acquiring firm. So in the situation of the acquirer having a
great deal of experience, in a certain field, may greatly positively impact the actual deal value
(Grimpe & Hussinger, 2008). As the research is mixed the following hypothesis is
constructed:

Hypothesis 4: The experience of the acquirer has a negative moderating effect on the
relationship between M&A announcements and performance.

12
2.3 Conceptual model:
Based on the previous sections and its hypotheses the following conceptual model can be
presented:

Figure 1 – conceptual model

13
3. Methodology
In this section, the methodological approach of this research is delineated and justified. First,
the research method is described, as are the reasons for choosing this method (§3.1.). Second,
it is explained how the constructs are going to be measured in relation to the event study
(§3.2.) and the regression(§3.3). Finally, in this chapter the sample is presented (§3.4.).

3.1 Research method

The main research question is whether the M&A’s affect the value of acquiring companies.
Measurement of a company’s value however is not easy when certain events occur over time
To measure the performance of M&A, empirical researcher uses two studies mainly: one is
accounting studies and another one is event studies. The post-operating performance of the
M&A is investigated by the accounting studies. The post-operating performance for the long
term period will be compared to the industry, size or performance with the benchmark of a
group of non-acquired firms, but this approach has several limitations, some of them being
that companies have different accounting rules and it is hard to measure the direct economic
impact of M&A over a timeframe. (Adnan & Hossain, 2016). Besides that, measuring over a
given timeframe would be difficult given the limited time and resources of writing a master
thesis, as a certain group under investigation should be compared to a reference group, which
would include data that is hard to come by.

Alternatively, in this thesis I will use an event study analysis, a common-used method
in corporate finance and investment analysis that measures the change in stock price at the
time when the event decision becomes public. A conventional event study methodology is
used that measures cumulative abnormal returns (Haleblian & Finkelstein, 1999; Hayward,
2002; Sudarsanam & Mahate, 2006).

There are some different alternatives to choose from when considering event study
related research. These alternatives mainly differ on how the abnormal returns are calculated,
and these different calculations have developed over the year. The work of Mackinlay (1997)
initially showed the number of different empirical models that have been employed in the
literature to estimate abnormal performance around the event. These include the market
model, market adjusted model and the mean adjusted model. The mean adjusted model is the
simplest method used to predict a normal return is to simply subtract a security’s time series
average from an event date return. The most commonly used prediction method is the market

14
model, where firm returns are regressed on a constant term and a market index. The market
adjusted model subtracts the market index from an event date security return. Overall, the
popularity of the market adjusted model has increased significantly over the years (Ahern,
2009). They involve no estimation process and no estimation period, making it a very helpful
model for when there is no data prior to the event (Henderson, 1990), what surely has
contributed the fact that is now the most used model in event study related research.
Comparing the market model to the mean adjusted model, multiple comparison tests were
conducted and it was found that the mean-adjusted returns model did not work as well as the
market model (Henderson, 1990).

However, the mean adjusted model does not differ from the market adjusted model
with regard to their returns displaying no significant mean bias (Ahern, 2009). Although
perceived more simplistic than the other models, simulations indicate that the technique works
relatively well, producing results that are comparable to the more complicated regression
models (Brown & Warner, 1985). Similarly, researchers who use the market adjusted model
need to choose a market index for their calculations and because the criteria for choosing
these similar or related market indexes are not well defined, biases may arise (Ahern, 2009).
For the mean adjusted model an estimation of the expected return is used in the model which
is based on the previous performance of the company itself, therefore no such biases can arise
(Mackinlay, 1997).

For this thesis, I will use the (comparison period) mean adjusted model. The main
reason for this choice is that choosing related market indexes is hard as the dataset contains a
wide range of different companies over different industries and countries, so in this way I
prevent the bias described above. Moreover, I can easily access the past performance of the
companies through the database, thereby making it easier and less time consuming to
calculate the cumulative abnormal returns for all the companies in the sample.

The mean adjusted model is in line with the Capital Asset Pricing Model (CAPM), in
that it assumes unsystematic, firm-specific risk to be zero and the systematic, market-specific
risk is equal to one (Mackinlay, 1997). There is no difference between the anticipated
expected return and the ex-post expected return. Resulting in the mean adjusted model being
generated by the following process, with 𝐸(𝜀𝑛𝑡 ) being equal to 0:

𝑅𝑛𝑡 = 𝐸(𝑅𝑛𝑡 ) + 𝜀𝑛𝑡

15
 This model will be used to analyze the main question in this thesis which is related to
what the effects are of an M&A announcement on acquirer performance, in which the
cumulative abnormal return (CAR) of the stocks is an important factor. In order to calculate
the CAR of all the transactions under consideration, the daily stock returns, normalized
returns and the abnormal returns for each of the stock will have to be calculated in this model.
I will touch upon these calculation in a later stage.

Sequentially, after calculating the cumulative abnormal returns, the analysis will
include a regression to take into account the potential effects of the other factors of this study.
In order to analyze relation of this in the constructed sample, I will use a multiple regression
analysis where the dependent variable is the calculate CAR of the stocks, where the
independent variables being the method of payment, target reputation and acquirer
experience.

3.2 Measuring constructs – Event Study

In this paragraph it is described how the construct in both the event study as well as the
regression analysis is measured. Central to the event study is the calculation of the cumulative
abnormal returns, which are also vital to the regression as it is the dependent variable in the
analysis.

3.2.1 Time parameters identification

The first step in order to calculate the cumulative abnormal returns is to precisely identify
when the actual event takes place. The event in this study is when mergers and acquisitions
are announced, as this is the moment upon which the market would receive the new of the
event. Shown in figure 2 below, I define the event date as t = 0. As the market gradually picks
up the news of the announcements, I am interested in the period around the event date (t = 0).
This period is called the event window indicated by [t1, t2]. Thus the test period is very small
and gives an impression of the immediate effect caused by the announcement of the
acquisition around the event.

According to Brown & Warner (1985) a parameter estimation period of 120 days is
adequate since daily returns data for the 120 days prior to the event date are sufficient in

16
formulating a benchmark for normal returns. The estimation period [T1, T2] is shown in the
figure below.

Figure 2 – examined window

3.2.2 Normal return calculation

The normal return of a company is referred to as the stock return of that particular company.
If a special event, such as an M&A, has not taken place in that company, an estimation of the
normal return requires a definition of the estimation period and an event period. In this study
the estimation window is 120 days and starts 130 days before the precise day of the event
([T1, T2] = [-130, -10]) with an event period of 10 days ([t1, t2] = [-5, 5]). The 5 day gap
between the estimation window and event window is there in order to negate potential effects
that the event window can have on the estimation window, for example insider trading. Each
M&A deal in the dataset has its own estimation window and event window. To calculate the
normal return of a stock, I first need to calculate the daily stock return. The daily stock
return𝑅 is calculated using the stock prices𝑃𝑡 retrieved from the database with the following
formula:

𝑃𝑡 − 𝑃𝑡−1
𝑅=
𝑃𝑡−1

To calculate the normal return of a stock I use the mean-adjusted return model. The normal
return 𝑁𝑅𝑖 is defined as the average return over the estimation period, where 𝑖 is the stock
index and T=T2-T1+1, which is equal to the number of days over the estimation window of
that particular stock:

17
 𝑇2
1
𝑁𝑅𝑖 = ∑ 𝑅𝑖𝑠
𝑇
𝑠=𝑇1

3.2.3 Abnormal return calculation and analysis

Now the normal return of each stock is calculated, I need to compare it with the actual return
that has taken place. This of course in order to determine the potential effects the event has
had on one particular stock. The abnormal return 𝐴𝑅𝑖,𝑡 is the difference between the actual
return and the normal return of stock 𝑖 on day 𝑡, where𝑅𝑖,𝑡 is the actual return of stock 𝑖 on
day 𝑡 and 𝑁𝑅𝑖,𝑡 is the normal return of stock 𝑖 on day 𝑡:

𝐴𝑅𝑖,𝑡 = 𝑅𝑖,𝑡 − 𝑁𝑅𝑖,𝑡

For every company in this sample the abnormal return is calculated for each day over the
entire observation period. After calculating the abnormal returns I aggregated these from
period [t1, t2] as I am interested in the changes over event window period. These aggregated
returns are the cumulative abnormal returns:

𝑡=𝑡2

𝐶𝐴𝑅𝑖 = ∑ 𝐴𝑅𝑡
𝑡=𝑡1

3.3 Measuring constructs – Multiple Regression

The valuation change of a company is represented by the abnormal return, as described earlier
in this chapter. The different companies are described over a particular event period, therefore
I use the cumulative abnormal returns over this event period as the quantitative representation
of the valuation change of a company. To test the constructed hypotheses, the CARs are used
as the dependent variable in this case, where the relationship with the moderator variables will
be tested through a regression analysis.

Using a regression analysis allows for a statistical method to form a relationship

between the dependent and independent variables as a model function. Again, in my research
the dependent variable is the cumulative abnormal returns of the transactions in the dataset.
The independent variables are the three factors of the payment method, target reputation and
acquirer experience.

18
 The amount of funding rounds that a target company has been involved in is often
used as a proxy for the reputation of targets in terms of their profitability (Amor & Kooli,
2020). Highly reputable targets have proven to be a good investment by others in previous
rounds, making new investors less dependent on signals about their potential profitability.

As I am aware of the fact that the proxy proposed to measure the reputation of targets
does not fully reflect the whole reputation of the targets only their financial reputation.
Signalling related research often uses broader proxy’s for reputation, such as the ESG index
(Maung, Wilson & Yu, 2020). However, reputation in this study is confined only to the
financial reputation of a company, which would make such indexes not useful. Furthermore,
one could also argue for the use of credit ratings as a proxy for reputation. For most of the
targets in the constructed dataset no information is available on such credit ratings, so I am
condemned to use the amount of funding rounds in this study.

In measuring the acquirer experience, this thesis will follow research that has been
done in the past on related topics. Acquirer experience is measured through the number of
deals previously conducted by acquirer, which is a proxy that has been used in related
research on signaling theory (e.g. Hayward, 2002; Wu & Reuer, 2021).To measure the
acquiring firm’s prior M&A experience it was determined whether the acquiring firm had
previously completed a merger. The number of previous M&A’s was obtained by looking at
the company’s previous years’ records in the Capital IQ database. As the dataset contains only
public companies information is available of the companies past, therefore I have include all
the deals that were made throughout the history of the company. For some transactions there
are several acquiring companies that each acquire a stake. Luckily, the Capital IQ database
show what is called the ‘lead investor’ that initiates the deal and which is involved for the
Largest percentage in the deal. Further the variableEXP_AcqTotalTransactions, a squared
term of the number of previous M&A’s, is used to examine the potential curvilinear
relationship between experience and acquirer performance.

The method of payment will be measured by which percentage of the deal that has
been funded by either equity or cash. An option in the Capital IQ database exists where the
percentage amount of cash consideration can be retrieved when constructing a dataset. As
most of the transactions show either 100% of cash, equity or other constructions, I will use
dummies in my analysis. One dummy will indicate a 1 if a deal is funded through cash and 0
otherwise. Another dummy will show a 1 if the deal is funded through equity. The group with
other constructions or unknowns is set as the reference category.
19
 For the control variables, I will first take into account the size of both the target and
the acquirer in the model. The size of the target is measured by looking at the market cap,
where the size of the target is measured by looking at the amount of employees of the
company at the time of the acquisition. The reason for choosing different measures for size
between acquirer and target has to do with data availability, as less information is available on
the market cap of target in the database. I also include a variable that indicates whether the
acquirer is related to the industry of the target. Business relatedness is measured by looking at
whether the first two digits of the SIC-code of the acquirer matches with that of the target. If
these two digits are the same, then the business relatedness variable shows a one, and if not, a
zero. Another control variable looks at the attitude of the deal, and shows a one if the
acquisition was friendly and a zero it that is not the case. The next control variable takes into
account the profitability of the acquiring company at the time of the deal. This is measured by
dividing the EBIT with the sales at the time of the announcement, which can easily be
extracted from the Capital IQ database. Lastly, normal return (NR) may have an influence on
CAR, so although I am not interested in if it really affects CAR or how it affects CAR, to
make the formulation more precise, I will also include NR as a control variable.

3.4 Sample
Collecting the data is a two-step process. First I collect the information necessary of the
transactions that are used in the analysis, and secondly I collect the stock price of the
companies around the time of the merger to be able to calculate the CAR of each given stock.
After some calculation and transformation as described earlier, this results in a dataset that is
useable for analysis.

As the Capital IQ database is extensive and enormously detailed, certain filters are
used. Such filters will exclude other types of transactions that are not relevant to this study. I
also filter out any other types of transactions that have not been closed and actually
materialized. For data availability purposes, I also exclude any targets or acquirers which are
labelled as private. So for reasons to limit the amount of transactions that are in the model, I
focus only on acquirers that are active in Europe, and focus only on announcements that were
made between 1/1/2018 and 1/1/2019.

When initially analyzing the data, I noticed that there were serial acquirers in the data
present. These acquiring companies announced M&A’s shortly after the previous one. This

20
could be problematic for calculating the cumulative abnormal returns as the short timeframe
could mean that the previous announcement has been made in the estimation period of
another announcement. In order to prevent the normal return calculation from being biased by
other announcements, I decided to delete these announcements from the sample when they
were in the estimation period of the next announcement.

Besides that, I also deleted the acquiring companies in the sample of which daily
returns over the estimation period were not fully accessible or unknown. Due to the fact that a
lot of transactions were deleted, I decided to also include data from the year before, so it
includes announcements from 1/1/2017 until 1/1/2019. See appendix 1 for an overview of all
acquiring companies and announcement dates in the sample.

Analyzing missing values is done quite quickly, as there are no missing values for the
dependent as well as all the independent variables. For the control variables of acquirer size
and acquirer previous performance, there were some missing values. In total there were 8
observations with missing values for both of these variables. As the other variables of these
observations were known and useful for the analysis, the missing values are replaced with the
mean of the other observations combined when running the regression. Unfortunately, the
variable of deal attitude is deleted, because my sample surprisingly only contains friendly
deals. I deleted this variable entirely as it might affect the analysis due to multicollinearity
reasons (Field, 2013).

21
4. Results
This chapter contains a detailed description of the results of this research describing the event
study analysis, correlation analysis and regression analysis.

4.1 Event study

The collected data has been analyzed using Statistical Package for the Social Sciences (SPSS
version 23.0, IBM, Chicago, Illinois, USA) to find support for the hypotheses. After
calculating the cumulative abnormal return as described in the previous section, a one sample
t-test is used in order to analyze whether the mean of the sample deviates from zero. The null
hypothesis H0 here is that the cumulative abnormal returns do not deviate significantly from
zero, which means that the event has no influence on the performance of the acquirer. The
alternative hypothesis H1 is that the means are significantly different from zero:

𝐻0: 𝐶𝐴𝑅 = 0

𝐻1: 𝐶𝐴𝑅 ≠ 0

In order to run a one sample t-test on the cumulative abnormal return, I first need to
calculate this using the formulas described in the previous chapter. Appendix 1 shows the
acquiring companies of the transactions in the sample and their respective cumulative
abnormal returns. For more information or a more detailed view of the calculations on the
normal returns, abnormal returns or the cumulative abnormal return, I refer to the attached
excel document containing calculations.

Before running the test I checked the assumptions of running a t-test (Field, 2013).
The assumptions of using continuous variables and the data not being correlated or related are
not violated with regard to the data. There are however significant outliers, which also cause
the sample not being normally distributed. This is confirmed when comparing the 5% trim
mean and the mean values, where a large discrepancy is present between the two. Outliers are
identities by looking at the output window in SPSS, where significant outliers are indicated in
the boxplot output. Outliers of the cumulative abnormal returns are shown in appendix 2.
These significant outliers are deleted until no significant outliers can be identified when
analyzing the data. In total, 8 observations have been deleted. Having no significant outliers
is an important assumption for running a proper one sample t-test, as they tend to increase the

22
estimate of sample variance, thus decreasing the calculated t-statistic and lowering the chance
of rejecting the null hypothesis (Field, 2013).

With regard to the normal distribution, as the significant outliers have been removed, I
can assume that the sample is normally distributed. The central limit theorem states that the
distribution of sample means approximates a normal distribution as the sample size gets
larger. Sample sizes that are equal to or greater than 30 are considered to be sufficient for the
central limit theorem to hold (Field, 2013). As the sample in this study is greater than 30, it is
considered large enough to predict the characteristics of a population accurately. The average
of the sample means and standard deviations will equal the population mean and standard
deviation.

Now the assumptions are met and the data is transformed, I can run the t-test on the
cumulative abnormal returns. The mean of the cumulative abnormal return (M= 3.925, SD=
.148) is not significantly different from zero with t(84) = 1.479, p =.143. The test value in this
one sample t-test is zero. As a check, the t-test is also analyzed with the previous model which
did include significant outliers which were also significantly different from zero t(93) =
1.650, p =.102, which confirms that removing outliers was useful as there is a large difference
between the results of the tests. There was no statistically significant difference between
means (p =.143). Therefore, I cannot reject the null hypothesis and the alternative hypothesis
cannot be accepted.

4.2 Correlation analysis

In order to analyze the data, a correlation matrix of this research was made which is shown in
Table 1. In this table the means, standard deviation and correlation of all variables are
exhibited. Where using Pearson correlation measure assumes linearity and homoscedasticity,
Kendall’s tau correlation does not (Field, 2013). As these assumptions are met in my dataset
as described earlier, I therefore use Pearson in my correlation matrix.

Table 1 shows several correlations between the variables in the model. For the
baseline effect, Pearson correlation showed no correlation between CAR and all the relevant
variables, with ρ= .111(p = .156) for the cash payment method, ρ= -.178 (p = .052) for the
equity payment method, ρ= .083 (p = .226) for the experience of the acquirer and ρ= .098(p =
.185) for the reputation of the target. This can be a problem in running the models. Finding a
significant correlation is however no prerequisite for running regression (Field, 2013). Once

23
variables are controlled for, there might be a strong bivariate correlation, but this is a
limitation of the dataset and surely something that is taken into account when interpreting
results. For 48% of the transactions in the sample the cumulative abnormal returns were above
zero, for 52% of the transactions the cumulative abnormal returns were below zero.

The bivariate analysis for the different payment methods showed a significant relation
ρ= -.410, p = .000, which makes sense as they are mutually exclusive. For all the transactions
for which the cumulative abnormal returns were negative over the event period 78% was
funded with cash, for the cases in which the transaction was funded with equity the
cumulative abnormal returns were positive in 93% of the cases. The control variable of
Normal Return showed a strong correlation with the dependent variable CAR ρ= .314 (p =
.002), which makes sense as the CAR variable is partially calculated using the normal returns
of the transactions in the sample.

24
 Variables
M S 1 2 3 4 5 6 7 8 9 10

CAR
1 1.25% 7.76% 1.000

Payment Cash
2 0.5294 0.50210 0.111 1.000

Payment Equity
3 0.0941 0.29373 -0.178 -0.342** 1.000

Acquirer Experience 1.3170 0.48547

4 0.083 0.047 -0.225* 1.000

Target Reputation
5 1.1302 0.54227 0.098 -0.129 -0.114 0.451** 1.000

Size Acquirer
6 13250 18147 0.100 0.060 0.010 0.552** 0.294** 1.000

Size Target
7 32.89 33.268 0.145 -0.055 -0.122 0.438** 0.593** 0.452** 1.000

8 Business Relatedness 0.4118 0.49507 -0.004 -0.025 0.140 -0.010 -0.008 0.029 -0.018 1.000

9 Acquirer Profitability 0.0829 0.48345 -0.021 0.131 0.007 0.242* 0.134 0.145 0.135 -0.156 1.000

10 Normal Return 0.059% 0.325% 0.314** 0.254** -0.075 0.016 -0.099 -0.017 -0.119 -0.179 0.141 1.000

* Correlationissignificantatthe0,05level(2-tailed).
** Correlationissignificantatthe0,01level(2-tailed).

Table1:CorrelationMatrix(Means,StandardDeviationsandPearsonCorrelations)

25
4.3 Regression analysis
After graphically checking the distribution of the variables, highly skewed variables were
normalized before running the regression. The reason is the presence of highly skewed
variables that can influence the distribution of residuals and make these not normally
distributed (Field, 2013). The variables that were adjusted are the total number of previous
acquisitions an acquirer has made and the total amount of previous investments that a target
has received. The total number of previous acquisitions of the acquirer improved when
looking at the Shapiro-Wilk test from p =.000 to p =.088, and the total amount of previous
investments improved from p =.000 to p =.208 , where both are now significant with regard to
being normally distributed. For a graphical representation of the transformation, see appendix
3 and 4.

To check for heteroscedasticity I look at plots in SPSS. The residuals should be

independent from the response variable, from all of the predictors as well as from the
predicted value of response variable. Looking at the plots, I conclude that the assumption of
homoscedasticity is not violated. As SPSS unfortunately does not include any formal tests of
heteroscedasticity, I used macros and codes within the program in order to run a formal test,
which is the Breusch–Pagen test. Results of this test show insignificant results (p >.05),
meaning that the null hypothesis of homoskedasticity cannot be rejected and homoscedasticity
is assumed.

To check for normality, I look at the distribution of the residuals after running the
regression. Plotting the standardized residuals and the standardized predicted residuals, it
shows me that the normality assumption is met. Also running test of normality on the
residuals, it does not show a significant result on the Shapiro-Wilk test for both the
unstandardized residuals (p =.246) and the standardized residuals (p =.246), meaning that the
null-hypothesis is that they are normally distributed cannot be rejected.

Multicollinearity refers to when your predictor variables are highly correlated with
each other. To check for multicollinearity, a correlation matrix is constructed. When values
are greater than .80, there might be a risk of multicollinearity (Field, 2013). As no values
exceed this threshold, I can assume that multicollinearity is not a problem in the data sample.
This is confirmed by looking at the collinearity statistics when running the regression, where
all of the variables are well below a VIF score of 10, and all tolerance scores are well above
0.2 (Field, 2013).

26
 In order to check whether the values of the residuals are independent, I look at the
Durbin-Watson test. With a score of 1.780, this value is close to the value of 2, so I can
assume that this assumption is met. In order to check if there are no influential cases in the
data, I checked the Cook’s Distance statistic foreach transaction. No value is greater than 1, so
I can assume that there are no influential cases. A final assumption with regard to the
regression analysis is linearity. If your residuals are normally distributed and homoscedastic,
you do not have to worry about linearity (Field, 2013), which is the case in this study.

Now assumptions are checked, I can run the regression. The effects of the method of
payment, target reputation and acquirer experience on acquirer performance was analyzed by
running a binary multiple regression analysis. In the first model the control variables are
added, which are the size of the target, size of the acquirer, the business relatedness of the
target and acquirer, the profitability of the acquirer at the announcement and the normal return
used in the calculation of the CARs. These variables statistically significantly predicted CAR,
F(5, 79) = 2.672, p < .05, R² = .145.

In step 2, a model is analyzed which includes all the control variables and the two
dummies which indicate the method of payment, which are the payment in cash and payment
in equity variables. These variables statistically significantly predicted CAR, F(7, 77) = 2.152,
p < .05, R² = .164. Step 3 tests the effect of the experience of the acquirer on the cumulative
abnormal returns. These variables did not statistically significantly predict CAR, F(6, 78) =
2.199, p = .052, R² = .145. Step 4 only focusses on the potential effects the reputation of
target has on the cumulative abnormal returns. These variables statistically significantly
predicted CAR, F(6, 78) = 2.218, p < .05, R² = .146. Step 5 is the overarching model where
not only all control variables are included, but also all other factors which have been tested in
previous steps. These variables did not statistically significantly predict CAR, F(9, 75) =
1.662, p = .113, R² = .166.

Where in the last model the independent variables explain the most of the variance in
the cumulative abnormal returns (16.6%), note that the model is not significant. This indicates
that the relationship between these variables is not significant on population levels. In a
population where there is no relationship between the two variables, the probability of finding
the expected results (or more extreme results) is 13.3 percent. Also note that there are no
significant findings in all of the models, except the effect of the control variable normal
return. For an overview of the results of the multiple regression, see table 2 below.

27
Robustness checks

To check the robustness of the model, first the estimation period is altered for each of the
cumulative abnormal return calculations. Related studies use different estimation periods for
their normal return calculations (Mackinlay, 1997; Corrado, 2011), each with their own
argumentation for doing so. To prevent potential methodological shortcomings of these
previous studies, I changed the estimation period for each of the transaction in the sample by
-60 days. Also, the potential curvilinear relationship between experience and acquirer
performance was examined by using squared term of the number of previous M&A’s as
explained in the previous chapter. Changing the estimation period and changing the
measurement for acquirer experience did not change the outcomes of the multiple regression
analysis. The predictors and their values remain largely the same.

28
 Dependent variable: Step 1 Step 2 Step 3 Step 4 Step 5
CAR
Controls
Size Acquirer 0.034 0.046 0.034 0.033 0,071
(0.000) (0.000) (0.000) (0.000) (0,000)
Size Target 0.186 0.161 0.186 0.163 0,148
(0.028) (0.028) (0.029) (0.033) (0,033)
Business 0.048 0.067 0.048 0.048 0,068
Relatedness (1.676) (1.698) (1.697) (1.686) (1,720)
Acquirer -0.095 -0.087 -0.095 -0.097 -0,080
Profitability (1.730) (1.749) (1.769) (1.744) (1,809)
Normal Return 0.358** 0.349** 0.358** 0.360** 0,352**
(1.730) (2.647) (2.588) (2.587) (2,681)
Independent
variables
Payment Cash - -0.008 - - -0,006
(1.803) (1,851)
Payment Equity - -0.144 - - -0,154
(3.023) (3,185)
Acquirer Experience - - 0.000 - -0,062
(2.113) (2,289)
Target Reputation - - - 0.041 0,046
(1.866) (1,974)
F 2.672 2.152 2.199 2.218 1.662
R² 14.5% 16.4% 14.5% 14.6% 16,6%
# of Transactions 85 85 85 85 85

* Correlation is significant at the0,05 level (2-tailed).

** Correlation is significant at the0,01 level (2-tailed).

Table 2:Hierarchical Overview of Multiple Regression Analysis (Standardized Beta,

Standard error)

29
5. Discussion
In this section theoretical implications, limitations & future directions, practical implications
and conclusions of the study are described.

The purpose of this study was to contribute to literature by exploring the consequences
and effects of an M&A announcement on the short-term performance of the acquiring
company including the moderating effects of the method of payment, the experience of the
acquirer and the reputation of the target. Where there was an expected positive effect of an
announcement on the performance of an acquirer, through an event study methodology no
significant effects were found. Moreover, no significant effects were found for the moderating
variables.

To summarize, hypothesis 1 is not supported due to not being statistically significant.

No support is found for hypotheses 2a, 2b, 3 and 4 which are therefore also rejected. Other
significant correlations and findings of the control variables were found in all the included
models. Only the finding that the higher the normal return of a particular acquiring company
over the estimation period the more likely that the companies would report better short term
performance, which is expected as the normal return is used in the calculation of the
cumulative abnormal returns.

5.1 Theoretical implications

The study makes a number of theoretical implications. Through quantitative research, this
study cannot confirm that the effects of announcing a merger or acquisition actually
contributes either positively or negatively to the short term performance of the acquiring
company. As highlighted earlier, the literature on the performance effects of such events is
rather contracting. Comparing the outcomes of this study with other research results, I find
similar studies that argue on the limited effects of events and announcements surrounding
M&As on performance (Connelly et al., 2011; Long, 2002; Pollock & Gulati, 2007).

However, there are also studies that find contradicting results (Bessler, Drobetz &
Zimmerman, 2011; Cohen & Dean, 2005; Sanders & Boivie, 2004; Stuart, Hoang & Hybels,
1999; Yang & Lander, 2018). Also, characteristics of pre-acquisition resources might not
necessarily predict post-acquisition performance (Zollo & Singh, 2000), which includes the
moderating factors that are proposed in this study.

30
 There are several reasons for why the results of this study are not in line with the
expectations. The main reason has to do with the methodology of event studies. The choice of
the estimation window, event window, the size of the sample and choosing a different model
than the mean adjusted model can have effects on the actual results of the event analysis
(Ahern, 2009; Corrado, 2011). Altering these could in fact lead to different results with the
sample that has been used in this study. Robustness checks in this study were only focused on
changing the estimation period, but not the other factors mentioned above due to time
constraints.

Also, another explanation for insignificant results of this study is that signal will not
be perceived uniformly across firms. Previous research has shown how these signals can be
perceived differently because of the great deal of time and effort that is required of managers
and investors to seek out potential acquisition targets as well as differences in their
understanding of a target’s resources and prospects (Trichterborn, Knyphausen & Schweizer,
2016).

Nevertheless, this thesis does contribute on the overall knowledge of M&A

performance as it takes a signalling theory perspective, something that has been called upon in
recent literature (Colombo, 2021; Wu, Reuer & Ragozzino, 2013). Most of M&A related
research takes a similar approach in deriving at their hypotheses, mainly using for example
agency theory or resource based view (Wu, Reuer & Ragozzino, 2013). This thesis also adds
upon earlier work regarding the sample that has been selected. Overall, work on the effects of
M&A announcements on acquirer performance have mostly been studies that focused on the
US market. This study adds to previous literature on M&A performance and signalling, by
shedding light on the effects of it in Europe alone. Results of studies that focus on the
difference between US and the European market show different results by taking a more
institutional perspective on signaling theory and argue that prospective investors do not
evaluate signals in isolation, but rather, they do so in the institutional context (e.g. Bell,
Moore, & Al-Shammari, 2008; Moore, Bell & Filatotchev, 2012).

Also, this study extends upon methodological shortcoming of other studies by taking a
broader perspective in terms of the industries that are analyzed. This limitation has been
called out in several studies with related research questions (Wu & Reuer, 2021). Analyzing
all public transactions from multiple industries offers a broader, more overarching perspective
of the dynamics between the signaling effects of M&A announcements.

31
5.2 Practical implications
Managers are faced with the difficult task of searching and selecting acquisition targets while
dealing with high levels of information asymmetry. Previous studies have established that
information can be drawn from signals. However, where I propose in this study that the
method of payment, the reputation of the target and the experience of the acquirer are
characteristics of signals that can influence performance of companies, the results show
otherwise. Again, this might be due to different investors interpreting signals and their factors
differently (Trichterborn, Knyphausen & Schweizer, 2016), for example that different types
of acquirers would affect the degree to which they act upon signals (Wu & Reuer, 2021). The
main implication that this study proposes for managers and investors alike, is that the use of
M&A announcements as potential signals to the market cannot be used to boost short term
performance and that the extent to which acquirers should concern themselves with the
preconditions of signals surrounding M&As is limited.

5.3 Limitations and future directions

One of the main limitations of this study is that no significant correlation with the dependent
variable was found, and also the regression analysis showed no significant results. In
particular the last step in the multiple regression model, which included all the variables in
this study, was found to be not significant on population levels. Finding these insignificant
results could be a problem due to the sample size, because it will be difficult to find
significant relationships from the data when having a small sample as statistical tests normally
require a larger sample size to ensure a representative distribution of the population (Field,
2013).

With regard to the event study, the CAPM which is in line with the mean adjusted
model, can produce different results compared to the market adjusted model and should
ideally both be used in the analysis (Armitage, 1995). Due to having limited time and resource
available in writing this thesis, I deliberately picked the mean adjusted model for this study.
The limitation of this study is that calculating abnormal returns using the market adjusted
model, it might produce different results.

Also, the event study methodology in this research depends CAPM and on the
assumption of an efficient market. This assumption is not valid in many situations. The length
of time that potential acquirers respond to event signals is random which implies that that

32
markets could exhibit market inefficiencies because prices do not instantly or fully reflect all
available information (Mackinlay, 1997). This is particularly troublesome for this study as it
might be that abnormal returns might be spread out over such a long period of time that we
are unable to see any significant ‘spike’ in the AR graph. This could be a reason for not
finding significant results in this study.

Another methodological limitation is that the event study methodology provide

estimates of the short-run impact on shareholders only and fail to consider many other effects
of the event. Events in the stocks of competitors might for example be reflected in the stock
price of a company, which can impact the abnormal returns that are not caused by any
announcements or events which include the target company (Ahern, 2009).

Moreover, the event study methodology is inherently vulnerable in providing different

results when the estimation or event periods are changed. A choice of sample size can also
potentially change the results of the study. The sensitivity of event studies is certainly
problematic and different conclusions can be drawn by different researchers studying the
same event, thereby making it hard for us to choose which result to believe in (Ahern, 2009;
Corrado, 2011). Future research could extend upon the sample used in this study, or change
the estimation periods and use a market adjusted model in order to check the results found in
this study.

From a more theoretical perspective, one of the limitations of this study is that the
signalling effects between different industries could have been very different. Not only the
effects of M&A announcements as a signal, but also the potential moderating effects of
payment method, target reputation and acquirer experience could actually be different across
different industries. Research on M&As in multiple industries shows that, although a certain
deviation from the industry norm may be beneficial for the acquirer to achieve a competitive
advantage in the context of signaling, such deviations may make it more difficult for acquirer
to clearly predict the future prospects of the firm based on the information signals (Coff,
2002). In this study I did not control for industry, mainly due to the fact that not all industries
are represented equally in the sample. Potential future research could actually analyze these
differences between industries more thoroughly, and possibly find what industries are more
susceptible to signals and which industries are not.

Similarly, signals could also be interpreted differently between countries. A recent

study by Li, Shenkar, Newbury and Tang (2021) for example, show that firms from countries

33
with better reputations are perceived as having superior skills/capabilities. Investors in this
case react more positive to these signals when the reputation of the acquirer’s country is better
than that of the targets. An interesting future research direction could be focused on what
exactly enhances signals in such countries, and also how firms themselves could exploit these
country reputations with regard to sending signals.

5.4 Conclusion
In this quantitative study the consequences and effects of M&A announcements on short-term
performance of acquirers were analyzed. This thesis tries to combine signalling theory with
the stream of literature on M&A performance. An analysis of collected data on 85
transactions showed, through an event study analysis, no significant results when looking at
the performance enhancing or diminishing effect of announcements. Moreover, an additional
multiple regression analysis showed no significant moderating effects for the method of
payment, the experience of the acquirer or the reputation of the target. Findings of this study
showed the limited effect of announcements as signals in the European market, and also the
limited extent to which acquirers should concern themselves with the preconditions of signals
surrounding M&As. This study contributes to the stream of literature on signalling theory, and
especially on its relation with M&A events. Additional research will be required to
understand more about when and under what circumstances signals work with regard to their
value enhancing capabilities, but on how acquirers could potentially precondition these
signals to work in their favor.

34
References
• Adnan, A., & Hossain, A. (2016). Impact of M&A announcement on acquiring and
target firm’s stock price: An event analysis approach. International Journal of
Finance and Accounting, 5(5), 228-232.
• Agrawal, M., & Sensarma, R. (2007). Determinants of merger activity: evidence from
India. International Journal of Financial Services Management, 2(4), 277-288.
• Ahern, K. R. (2009). Sample selection and event study estimation. Journal of
Empirical Finance, 16(3), 466-482.
• Amit, R., Brander, J., & Zott, C. (1998). Why do venture capital firms exist? Theory
and Canadian evidence. Journal of business Venturing, 13(6), 441-466.
• Amor, S. B., & Kooli, M. (2020). Do M&A exits have the same effect on venture
capital reputation than IPO exits?. Journal of Banking & Finance, 111, 105704.
• Armitage, S. (1995). Event study methods and evidence on their performance. Journal
of economic surveys, 9(1), 25-52.
• Bao, J., & Edmans, A. (2011). Do investment banks matter for M&A returns?. The
Review of Financial Studies, 24(7), 2286-2315.
• Barkema, H. G., & Schijven, M. (2008). How do firms learn to make acquisitions? A
review of past research and an agenda for the future. Journal of Management, 34(3),
594-634.
• Bell, R. G., Moore, C. B., & Al–Shammari, H. A. (2008). Country of origin and
foreign IPO legitimacy: Understanding the role of geographic scope and insider
ownership. Entrepreneurship Theory and Practice, 32(1), 185-202.
• Bessler, W., Drobetz, W., &Zimmerman, J. (2011). Financing corporate mergers and
acquisitions. Capital Structure and Corporate Financing Decisions, 419-474.
• Bessler, W., Kruizenga, D., & Westerman, W. (2020). Stock Market Reactions to
Financing and Payment Decisions for European Mergers and Acquisitions. The
Central European Review of Economics and Management, 4(2), 41-89.
• Bergh, D. D., & Gibbons, P. (2011). The stock market reaction to the hiring of
management consultants: A signalling theory approach. Journal of Management
Studies, 48(3), 544-567.
• Bergh, D. D., Connelly, B. L., Ketchen Jr, D. J., & Shannon, L. M. (2014). Signalling
theory and equilibrium in strategic management research: An assessment and a
research agenda. Journal of Management Studies, 51(8), 1334-1360.

35
• Bonacchi, M., Cipollini, F., & Zarowin, P. (2018) Parents’ Use of Subsidiaries to
“Push Down” Earnings Management: Evidence from Italy. Contemporary Accounting
Research, 35, 1332-1362.
• Brown, S. J., & Warner, J. B. (1985). Using daily stock returns: The case of event
studies. Journal of Financial Economics, 14(1), 3-31.
• Cartwright, S., & Schoenberg, R. (2006). Thirty years of mergers and acquisitions
research: Recent advances and future opportunities. British journal of
management, 17(1), 1-5.
• Changqi, W., & Ningling, X. (2010). Determinants of cross-border merger &
acquisition performance of Chinese enterprises. Procedia-Social and Behavioral
Sciences, 2(5), 6896-6905.
• Chen, X.-P., Yao, X., & Kotha, S. (2009). Entrepreneur passion and preparedness in
business plan presentations: A persuasion analysis of venture capitalists’ funding
decisions. Academy of Management Journal, 52, 199-214.
• Coff, R. W. (2002). Human capital, shared expertise, and the likelihood of impasse in
corporate acquisitions. Journal of Management, 28(1), 107-128.
• Cohen, B. D., & Dean, T. J. (2005). Information asymmetry and investor valuation of
IPOs: Top management team legitimacy as a capital market signal. Strategic
Management Journal, 26(7), 683-690.
• Colombo, O. (2021). The use of signals in new-venture financing: A review and
research agenda. Journal of Management, 47(1), 237-259.
• Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011). Signaling
theory: A review and assessment. Journal of management, 37(1), 39-67.
• Corrado, C. J. (2011). Event studies: A methodology review. Accounting & Finance,
51(1), 207-234.
• Das, A., & Kapil, S. (2012). Explaining M&A performance: a review of empirical
research. Journal of Strategy and Management.
• Doukas, J. A., & Lang, L. H. (2003). Foreign direct investment, diversification and
firm performance. Journal of International Business Studies, 34(2), 153-172.
• Gaur, A. S., Malhotra, S., & Zhu, P. (2013). Acquisition announcements and stock
market valuations of acquiring firms' rivals: A test of the growth probability
hypothesis in China. Strategic Management Journal, 34(2), 215-232.
• Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage.

36
• Grimpe, C., & Hussinger, K. (2008). Pre-empting technology competition through
firm acquisitions. Economics Letters, 100(2), 189-191.
• Gulati, R., & Higgins, M. C. (2003). Which ties matter when? The contingent effects
of interorganizational partnerships on IPO success. Strategic Management Journal,
24(2), 127-144.
• Haleblian, J. & Finkelstein, S. 1999. The influence of organizational acquisition
experience on acquisition performance: A behavioral learning perspective.
Administrative Science Quarterly, 44(1), 29-56.
• Hayward, M. L. (2002). When do firms learn from their acquisition experience?
Evidence from 1990 to 1995. Strategic Management Journal, 23(1), 21-39.
• Henderson Jr, G. V. (1990). Problems and solutions in conducting event studies.
Journal of Risk and Insurance, 282-306.
• Isa, M., & Lee, S. P. (2011). Method of payment and target status: announcement
returns to acquiring firms in the Malaysian market. International Journal of
Economics and Finance, 3(3), 177-189.
• Jandik, T., & Makhija, A. K. (2005). Debt, debt structure and corporate performance
after unsuccessful takeovers: evidence from targets that remain independent. Journal
of Corporate Finance, 11(5), 882-914.
• Luijkx, S. (2020). Consequences of Venture Capital Withdrawal. Radboud University
Master Thesis.
• Kaplan, S., & Stromberg, P., (2003). Financial Contracting Theory Meets the Real
World: An empirical analysis of venture capital contracts. Review of Economic
Studies, 70(2), 281-315.
• King, D. R., Dalton, D. R., Daily, C. M., & Covin, J. G. (2004). Meta‐analyses of
post‐acquisition performance: Indications of unidentified moderators. Strategic
management journal, 25(2), 187-200.
• Long, C. (2002). Patent signals. The University of Chicago Law Review, 625-679.
• Leland, H. E., & Pyle, D. H. (1977). Informational asymmetries, financial structure,
and financial intermediation. The journal of Finance, 32(2), 371-387.
• Li, Y., & Chi, T. (2013). Venture Capitalist’ Decision to Withdraw: The Role of
Portfolio Configuration from a Real Options Lens. Strategic Management Journal
34(1), 1351- 1366.

37
• Li, C., Shenkar, O., Newburry, W. E., & Tang, Y. (2021). How Country Reputation
Differentials Influence Market Reaction to International Acquisitions. Journal of
Management Studies.
• Ma, Q., Zhang, W., & Chowdhury, N. (2011). Stock performance of firms acquiring
listed and unlisted lodging assets. Cornell Hospitality Quarterly, 52(3), 291-301.
• MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of
Economic Literature, 35(1), 13-39.
• Maung, M., Wilson, C., & Yu, W. (2020). Does reputation risk matter? Evidence from
cross-border mergers and acquisitions. Journal of International Financial Markets,
Institutions and Money, 66, 101204.
• Moore, C. B., Bell, R. G., Filatotchev, I., & Rasheed, A. A. (2012). Foreign IPO
capital market choice: Understanding the institutional fit of corporate governance.
Strategic Management Journal, 33(8), 914-937.
• Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions
when firms have information that investors do not have. Journal of financial
economics, 13(2), 187-221.
• Park, H. D., & Patel, P. C. (2015). How does ambiguity influence IPO underpricing?
The role of the signalling environment. Journal of Management Studies, 52(6), 796-
818.
• Pollock, T. G., & Gulati, R. (2007). Standing out from the crowd: The visibility-
enhancing effects of IPO-related signals on alliance formation by entrepreneurial
firms. Strategic Organization, 5(4), 339-372.
• Reuer, J. J., Tong, T. W., & Wu, C. W. (2012). A signaling theory of acquisition
premiums: Evidence from IPO targets. Academy of Management Journal, 55(3), 667-
683.
• Riley, J. G. (2001). Silver signals: Twenty-five years of screening and signaling.
Journal of Economic literature, 39(2), 432-478.
• Ross, S., R. Westerfield, and B. Jordan. 2009. Fundamentals of corporate finance. 9th
ed. New York: McGraw-Hill.
• Sanders, W. G., & Boivie, S. (2004). Sorting things out: Valuation of new firms in
uncertain markets. Strategic Management Journal, 25(2), 167-186.
• Spence, M. (1973). ‘Job market signaling’. Quarterly Journal of Economics, 87, 355–
74.

38
• Stuart, T. E., Hoang, H., & Hybels, R. C. (1999). Interorganizational endorsements
and the performance of entrepreneurial ventures. Administrative science quarterly,
44(2), 315-349.
• Sudarsanam, S. & Mahate, A. (2006). Are Friendly Acquisitions Too Bad for
Shareholders and Managers? Long‐Term Value Creation and Top Management
Turnover in Hostile and Friendly Acquirers. British Journal of Management. 17(1). 7-
30.
• Suwanna, T. (2012). Impacts of dividend announcement on stock return. Procedia-
Social and Behavioral Sciences, 40, 721-725.
• Trichterborn, A., Zu Knyphausen‐Aufseß, D., & Schweizer, L. (2016). How to
improve acquisition performance: The role of a dedicated M&A function, M&A
learning process, and M&A capability. Strategic Management Journal, 37(4), 763-
773.
• Veh, A., Göbel, M., & Vogel, R. (2019). Corporate reputation in management
research: a review of the literature and assessment of the concept. Business research,
12(2), 315-353.
• Wu, C. W., Reuer, J. J., & Ragozzino, R. (2013). Insights of signaling theory for
acquisitions research. In Advances in mergers and acquisitions. Emerald Group
Publishing Limited.
• Wu, C. W., & Reuer, J. J. (2021). Acquirers’ reception of signals in M&A markets:
Effects of acquirer experiences on target selection. Journal of Management Studies.
• Yang, J., & Lander, M. W. (2018). Bearers of (good) news: The impact of business
news reports on acquirer short-term performance. In Advances in mergers and
acquisitions. Emerald Publishing Limited.
• Yasar, B., Martin, T., & Kiessling, T. (2020). An empirical test of signalling theory.
Management Research Review.
• Zhao, X., Ma, H., & Hao, T. (2019). Acquirer size, political connections and mergers
and acquisitions performance. Studies in Economics and Finance.
• Zollo, M., & Meier, D. (2008). What is M&A performance?. Academy of management
perspectives, 22(3), 55-77.
• Zollo, M., & Singh, H. (2000). Post-acquisition strategies, integration capability, and
the economic performance of corporate acquisitions. INSEAD.

39
Appendix

Appendix 1 – Overview sample

Acquirer name Date CAR Acquirer name Date CAR Acquirer name Date CAR
Ellaktor S.A. 12-28-2018 -14,15% Class Editori S.p.A. 03-17-2018 4,64% Aeroporto Guglielmo 08-03-2017 2,86%
Marconi di Bologna
S.p.A
GrupaKapitalowa 12-19-2018 6,65% E.ON SE 03-11-2018 1,75% French Connection 07-27-2017 7,76%
IMMOBILE S.A. Group PLC
Eiffage SA 12-17-2018 -3,31% NTS ASA 03-05-2018 -1,97% Frendy Energy S.p.A. 07-27-2017 -5,59%
(ENXTPA:FGR)
CCC S.A. 12-07-2018 -1,09% AtlantiaSpA 03-02-2018 9,40% FnacDarty SA 07-26-2017 2,58%
ZEAL Network SE 11-19-2018 -14,44% UBS Group AG 02-21-2018 2,22% Andes Energia PLC 07-24-2017 -22,47%
Equinor ASA 11-15-2018 -5,25% RELX PLC 02-15-2018 4,09% Halcor Metal Works 07-19-2017 -0,37%
S.A.
Sistema Public Joint 11-11-2018 -0,91% Vicore Pharma 02-08-2018 6,58% Pt Wicaksana Overseas 07-11-2017 20,63%
Stock Financial Holding AB International Tbk
Corporation
Reworld Media Société 11-08-2018 4,76% Danske Bank A/S 01-25-2018 -0,98% Public Joint Stock 06-28-2017 3,44%
Anonyme Company PIK-
specialized homebuilder
BNP Paribas SA 11-05-2018 0,22% ENGIE SA 01-24-2018 -3,69% PJP Makrum S.A. 06-23-2017 17,84%
Raiffeisen Bank 11-02-2018 7,00% Carrefour SA 01-11-2018 -1,51% Poslovnisistem Mercator 06-20-2017 13,57%
International AG
VTB Bank 10-23-2018 -5,61% Tele2 AB 01-10-2018 -6,24% National Central Cooling 06-19-2017 11,34%
Company PJSC
Spar Nord Bank A/S 09-26-2018 0,87% NN Group N.V. 01-09-2018 4,06% PGS ASA 06-09-2017 -12,10%
Vifor Pharma AG 09-18-2018 -2,78% Delta Drone SA 01-03-2018 53,18% Public Joint Stock 06-08-2017 11,44%
Company Rusolovo
(MISX:ROLO)
Sampo Oyj 09-14-2018 1,79% Public Joint Stock 12-29-2017 6,89% Satis Group S.A. 05-30-2017 -24,71%
Company Inter RAO
UES
Brack Capital Properties 08-17-2018 1,50% Teliani Valley Polska 12-21-2017 8,34% Drillisch AG 05-12-2017 9,38%
NV S.A.
Anglo Pacific Group plc 08-16-2018 8,89% Vergnet SA 12-15-2017 -3,35% Medistim ASA 05-12-2017 -0,09%
Public Joint Stock 07-13-2018 -9,68% SparekassenSjælland- 12-14-2017 4,46% Senterra Energy Plc 05-09-2017 -23,20%
Company ALROSA Fyn A/S
Diageo plc 06-25-2018 -2,37% AS Trigon Property 12-12-2017 -6,91% Primetech S.A. 05-09-2017 9,87%
Development
A2A S.p.A. 05-16-2018 -10,09% Netia S.A. 12-05-2017 22,72% Teliani Valley Polska 04-05-2017 -1,64%
S.A.
Holding Varna PLC 05-10-2018 -3,08% Bioorganic Research 11-30-2017 58,77% EDP Renováveis, S.A. 03-27-2017 11,34%
and Services S.A.
DürrAktiengesellschaft 05-04-2018 3,17% Parallel Media Group 11-29-2017 63,51% MVV Energie AG 03-16-2017 -7,71%
plc
Airtificial Intelligence 04-26-2018 -2,86% Sport1 Medien AG 11-27-2017 8,38% Telefónica Deutschland 03-13-2017 2,42%
Structures, S.A. Holding AG
VTB Bank 04-25-2018 -6,04% Option NV 11-27-2017 -3,35% Public Joint Stock Oil 02-28-2017 -0,12%
Company Bashneft
Holding Varna PLC 04-25-2018 -2,00% Luka Rijeka d.d. 11-09-2017 0,02% International 02-24-2017 1,96%
Consolidated Airlines
Group, S.A.
Polski Bank 04-16-2018 -6,22% asknet Solutions AG 11-08-2017 38,70% TurkiyeGarantiBankasi 02-21-2017 -2,86%
KomórekMacierzystych A.S.
S.A.

40
Holding Varna PLC 04-13-2018 1,27% Krakchemia S.A. 10-04-2017 -2,39% TNS energo Rostov-on- 02-17-2017 -4,48%
Don
International 04-12-2018 -0,39% Volcan Compañía 10-03-2017 29,46% Polimex-Mostostal S.A. 01-23-2017 -2,76%
Consolidated Airlines Minera S.A.A.
Group, S.A.
Auplata Mining Group 04-05-2018 -3,08% Softmatic AG 09-26-2017 -41,74% Intesa Sanpaolo S.p.A 01-23-2017 -10,19%
BERG Holding S.A. 04-02-2018 -11,60% Polymetal 09-13-2017 6,76% Public Joint Stock 01-17-2017 0,52%
International plc Company Group of
Companies TNS energo
Getin Noble Bank S.A. 03-26-2018 -3,91% Svenska 08-29-2017 5,11% TNS energo Kuban 01-10-2017 4,87%
Handelsbanken AB
Highlight Event and 03-22-2018 3,62% Medard S.A. 08-08-2017 38,56%
Entertainment AG
Deutsche Telekom AG 03-21-2018 0,97% China Modern Dairy 08-06-2017 -4,46%
Holdings Ltd.

41
Appendix 2 - Frequencies of cumulative abnormal return with outliers

42
Appendix 3 – Transformation number of total investments target

Frequencies of total number of investments of the target before transformation.

Frequencies of total number of investments of the target after transformation

43
Appendix 4 – Transformation

Frequencies of total number of transactions of the acquirer before transformation.

Frequencies of total number of transactions of the acquirer after transformation.

44