• No results found

The announcement effects of failed mergers & acquisitions on target company shareholder wealth

N/A
N/A
Protected

Academic year: 2021

Share "The announcement effects of failed mergers & acquisitions on target company shareholder wealth"

Copied!
31
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The announcement effects of failed mergers &

acquisitions on target company shareholder wealth

Faculty of Economics and Business

Msc International Financial Management

(2)

2

Abstract

This study examines the relationship between merger and acquisition (M&A) failures, and relative size. As well as the relationship between M&A failures and the cause of M&A failures. The focus is centered on the event window surrounding the M&A failures, and the effects for target company shareholder wealth are examined, for the period from 2012 to 2016. An international sample is used, which consists of 71 companies from four different countries: the Netherlands, the United Kingdom, the United States, and Germany. This study finds that stand-alone M&A failures are value destroying for target company shareholders. Evidence shows that shareholder value is enhanced when a bid is rejected, contrary to withdrawn and terminated M&A offers, with effects that last up to two years following an M&A failure. The relative size of the firm does not have a significant relationship with shareholder value enhancement surrounding M&A failures.

Keywords: Merger and acquisition failures; Shareholder value; Event study; Firm size; Cause of merger and acquisition failures

JEL classification:

(3)

3

Table of contents

1. Introduction………..………...p.4

2. Literature review………...………….p.7 2.1 M&A announcement effect………..………p.7 2.2 M&A failures.……….……….…p.8 2.2.1 Withdrawal………..……….p.8 2.2.2 Termination………..……….…..p.9 2.2.3 Rejection………..………..……….p.10 2.2.4. Company size……….……….…….p.11 3. Hypotheses………...p.13

4. Data and methodology………..……….……….………..………….p.15 4.1 Data……….p.15 4.2. Variables….………...p.17 4.3. Methodology………p.18 4.3.1. Short-term event study……….…p.18 4.3.2. Long-term event study………..………p.19

5. Results………p.20 5.1 Stand-alone M&A failures..……….………p.20 5.2 Shareholder wealth and cause of M&A failures……….……….p.22 5.3 Long-term revaluation………..……….p.23 5.4 Relative target size……….………..p.25

(4)

4

1. Introduction

Failures in merger & acquisition (M&A) is an understudied area in the financial literature (Neuhauser, Davidson, and Glasock, 2011). Understanding M&A failures is important because both bidding company shareholders and target company shareholders are impacted by this event (Bettinazzi and Zollo, 2017). Shareholders of both bidding company and shareholders of the target company face the risk of value destruction, or have the opportunity to see their stock value enhanced. Gaining insights in the factors that explain value enhancement, or being aware of value destructive factors, can assist shareholders to maximize stock value, or minimize risk of value destruction.

The volume of M&A is substantial, in 2016 there were more than 49,000 announced transactions with a value of over $3.5 trillion (IMAA, 2017). The magnitude of M&A illustrates the importance of understanding as many aspects and related topics of M&A. M&A failures represent bids that do not materialize after bid announcement, and in other cases, bids that become deals that are later terminated. Meaning that both bids and deals do not necessarily result in a combined firm of the target and the bidding company. Various reasons for bid failures and deal failures exist, from management issues to company underperformance. A limited number of reasons for M&A failures have been studied in financial literature (e.g. Jacobsen, 2014; Bates and Becher, 2017)

In the event of failed M&A, either the target company or the bidding company, or both the target and the bidding company can have their own reasons. A reason for the targeted firm to reject an offer is not seeing the proposed value creation of the combined companies. Fig. 1 Value of M&A transactions in trillions of Dollars and number of M&A transactions, from 2012 till 2016. Adjusted from

the “Institute of Mergers, Acquisitions and Alliances” (IMAA), 2018.

40754 38886 43138 47313 49180 2.532 2.533 3.963 4.766 3.639 0 1.000 2.000 3.000 4.000 5.000 6.000 0 10000 20000 30000 40000 50000 60000 2012 2013 2014 2015 2016 TR ILLI ON S OF D OLLA RS AN N O U N CE D M & A TR ANSAC TI O NS YEAR

(5)

5

This means that the offer for M&A is not in the interest of the owners of the targeted company and the bid is therefore rejected (Bates and Becher, 2017). On the other hand, the bidding firm can withdraw the bid because, for example, it feels their company ends up paying too much (Jacobsen, 2014), and could destroy their own shareholders’ wealth. Hence, the issue is value creation or value destruction, where value creation is the reason why firm engage in M&A (Seth, 1990), and value destruction is unwanted. Thus, if value destruction can be avoided, management of either firm should try and do so, to act in the interest of the owners of the company.

Earlier studies, regarding M&A, have examined the potential reasons for M&A failures and the effects of M&A failures. Some studies have revealed that withdrawing a bid is an indication of CEO quality, from a bidding firm perspective (Alexandridis, Petzemas and Travlov, 2010; Jacobsen, 2014). It is explained that these CEOs notice potential overpaying, and are not afraid to safeguard their companies by terminating a deal. Another cause of M&A failures is target firm rejection of the bid. The reason for rejecting a bid can be that it is regarded as too low, hence the bid is not value creating for target company shareholders (Bates & Becher, 2017).

(6)

6

cash offer termination, compared to stock offer termination. Another factor that can influence abnormal returns surrounding M&A is relative company size, as examined by Asquith, Brunner, and Mullins (1983). The relative size of the target company influences the abnormal returns that bidding companies receive after M&A attempts that have failed (Asquith, et al., 1983).

Literature about M&A failures attempts to provide reasons for the value enhancement surrounding M&A failures. This study will add to literature by examining potential explanatory factors of value enhancement surrounding M&A failures, with an international perspective. This is achieved by studying two potential influences on target company shareholder abnormal return, surrounding failed M&A. First, the cause of M&A failures, which company ended the M&A talks, the bidder or the target company. Defined are bidding company withdrawal, target company rejection, and mutual termination. Second, the relative size of the target company. An international sample is used that consists of 71 companies from the United States (US), the United Kingdom (UK), Germany, and the Netherlands. The impact of the stock value will be examined for the target-firm through the use of an event study. Announcement effect is often studied by using an ‘event study’, in this framework share price on the days around the announcement are observed (MacKinlay, 1997; Seghal, Banerjee, and Deisting, 2012) and are compared to the performance of the market. Inferences are made for target companies, based on their cumulative abnormal return (CAR).

I find evidence that stand-alone M&A failures have an overall negative effect on shareholder wealth, and is value destructive in the short-run. Shareholders experience significantly different abnormal returns with various causes of M&A failures: the abnormal returns between termination, rejection and withdrawal are significantly different from one another. The same conclusion holds for relative company size, relative smaller companies (<30%) experience significant different returns from relative larger companies (>30%).

(7)

7

2. Literature review

M&A is an intensely studied topic in economics and finance. The dominating methodological approach in M&A literature is an event study (Bruner, 2001). This study examines M&A failures, with special interest in company size, and the cause of M&A failures. Neither are studied into great extent by other authors. The reason why companies engage in M&A is studied quite extensively, Seth (1990) mentions: market power, economies of scale, economies of scope, coinsurance, diversification of risk, and non-value maximizing. The reasons mentioned are all part of the idea that the combined companies create value, more value than the two stand-alone companies.

Whether a combination of two firms results in value enhancement, or in value destruction, has divided authors. Some authors argue that the combined companies are value enhancing, whereas others authors show that the combined companies destroy value (see for instance Seghal et al., 2012 for an overview). The literature on M&A wealth creation is more alike, in general, shareholders of the bidding firm do not earn a high return, and in some cases they do not earn any return at all (Wu, Yang, Yang, and Lei, 2016; Bruner, 2001). The shareholders of the target firm experience wealth enhancement surrounding M&A (Campa and Hernando 2004; Bruner, 2001), sometimes even up to 30% (Wu et al., 2016). Bruner (2001) shows that combined, shareholders of bidding companies and shareholders of target companies experience value enhancement in M&A.

2.1 M&A announcement effect

(8)

8

returns on dates surrounding the announcement (Keown and Pinkerton, 1981), and others do not report a significant change at all, at best zero (Alexandridis et al., 2010).

The semi strong efficient market hypothesis states that a market securities’ price conveys all public and private information (Keown and Pinkerton, 1981). This implies that new information will have an effect on the price of the security. Hence, the reason for the abnormal return is the newly available information, the result of the information, in this case excess return, depends on the perception of the investors. Another reason for abnormal return can be insider information (Keown, Pinkerton, and Bolster, 1992), evidence for insider information is found if abnormal return is reported in the days prior to the event date. Abnormal returns can also be caused by general market reaction, which could make a takeover more or less likely, however trading behavior is hard to prove (Keown et al., 1992).

2.2 M&A failures

This study focusses on M&A failures, in other words, a deal (agreement) or offer made by the bidding firm has been withdrawn, rejected, or terminated, for whatever reason. It is interesting to see the differences between the causes of M&A failures since this has not been studied in a similar way. Previously, authors have focused on either failed takeovers (Malmendier et al., 2016) or failed mergers (Sullivan et al., 1994). The exception is Neuhauser et al. (2011), they study differences between takeover and merger attempts, and divide into termination, withdrawal, and share repurchases. Previous studies have identified various causes of M&A, yet, to the best of my knowledge, the different causes of M&A failures have not been combined altogether. In addition, the effect of relative firm size is examined as a possible explanatory factor of abnormal return surrounding M&A failures. Asquith et al. (1983) show the relevance of relative target company size, in their study the effects for merger and acquisition gains for the bidding company are shown. Finding that relative target company size has a significant effect on the returns of the bidding company surrounding M&A.

2.2.1 Withdrawal

(9)

9

shareholder value (Bates & Becher, 2017). On the other hand, Jacobsen (2014) finds that CEOs who are unwilling to increase an M&A offer, for price reasons, are better in valuation of a target company than other CEOs. Alexandridis, Mavis, Terhaar, and Travlos (2013) add that CEOs who withdraw bids are more likely to make better acquisitions compared to other CEOs. This implies that superior CEOs feel that if the ‘maximum’ price is offered to the target company, and this bid does not materialize, the offer could better be withdrawn. Hence, an excessive and potentially value destructive premium is not offered, thus their valuating skills exceed those of other CEOs. CEOs that do not increase the price are more capable to valuate a company correctly. This can be interpreted as a sign of a CEO quality, and CEOs who have this quality are better CEOs (Alexandridis et al. 2013). These CEOs are less likely to be replaced (Jacobsen, 2014), and have a higher chance of moving to a larger company.

The fact that bidding companies withdraw their bids can thus be because of over-valuation of the target company (Jacobsen, 2014). Lee and Opp (2005) show that a withdrawn merger can still have a positive impact on target company share prices. This depends on the way the M&A offer has been constructed. If the proposed bid is to be financed with cash, positive returns to target shareholders have been found by earlier studies (Huang and Walkling, 1987; Malmendier et al., 2016). Yet, if the bid is financed with shares this effect does not exist, and share prices return to their initial value, the value prior to the M&A bid (Lee and Opp, 2005). Liu (2018), points out that bidding companies do normally experience value enhancement when M&A withdrawal has been announced. Signaling that the market did not see the value creation at the proposed takeover price for the combined firms, and values the individual companies higher than the combination of the two companies. On the other hand Neuhauser et al. (2011) show that bid withdrawal, as a stand-alone event, has a negative announcement impact on shareholder value. Lastly, Malmendier, et al. (2016) find that companies who have experienced a bid withdrawal are more likely to be purchased in the eight years following the withdrawal, compared to companies who have not experienced a takeover attempt.

2.2.2 Termination

(10)

10

1994). Lee and Opp (2005) argue the same, but for withdrawn M&A bids. Sullivan et al. (1994) find no significant returns or impact for bidding companies after termination of a merger. Davidson, Dutia and Cheng (1989) also examined the revaluation of shares affected by terminating a merger, and found that target companies who are involved in subsequent M&A activity experience positive abnormal returns after failed M&A. In comparison, target companies who do not experience subsequent M&A activity do not experience enhanced value, they return to pre M&A announcement share prices (Davidson et al., 1989). Also, studies support the possibility of abnormal return following M&A failures (examples include: Lee and Opp, 2005; Dodd and Ruback, 1977). More recently Malmendier et al. (2016) found evidence for the re-valuation effect, following M&A failures. However, Malmendier et al. (2016) find that subsequent M&A activity does not explain the abnormal returns after termination, as Davidson et al. (1989) find, but that a cash or stock bid influences the returns after M&A failures. Finally, if an M&A deal is terminated, termination fees can be imposed on the company who is ‘at fault’ (Liu, 2018). These fees negatively impact the stock performance of the firm who needs to pay the termination fee. The impact holds for both the bidding company and the target company. The variety in explanatory factors, and the fact that some authors contradict one another implies that value enhancement surrounding M&A failures is not a fully established field.

2.2.3 Rejection

(11)

11

realization of an anticipated follow-up bid. Initial announcement gains refer to the enhanced value the shares received upon announcement of the bid. In addition, companies who rejected a takeover proposal do not improve performance in the long-term (Bradley et al. 1983). This implies that rejecting a takeover bid is not always sensible, since rejecting a takeover bid has not created value for the stand-alone company and their shareholders. Safieddine and Titman (1999), however, find that companies that reject a takeover bid and afterwards increase leverage improve their performance. Other target companies that experienced a failed takeover and increase leverage experience similar results. Safieddine and Titman (1999) show that offers are rejected because target firm experiences them as inadequate, pointing towards undervaluation of the company. Bates and Becher (2017) examine the motives of management for resisting a take-over bid, and study the consequences for shareholder value. The main reason for rejection, Bates and Becher (2017) find, is an offer that undervalues the company in the eyes of the target shareholders, and is associated with higher bid rejection. However, if a CEO cannot close the deal on an appropriate bid, the chance of a forced turnover increases (Bates and Becher. 2017). Hence, bid rejection should not be taken lightly by the board and the CEO.

2.2.4 Company size

Moeller, Schlingemann, and Stulz (2004) examine bidding firm offered premiums, finding that large firms that are acquirers pay higher premiums than small firms that are acquirers. Because of the higher paid premium it is argued that, overall, large companies are bad acquirers, and their acquisitions result in losses. Since small companies pay lower premiums, they are good acquirers, and their acquisitions do not result in a loss. This implies that premium paid is dependent on the size of the bidding party. Moeller et al. (2004) study the aforementioned size effects using absolute market values of equity, small companies are the ones being smaller than the 25th percentile of companies on the New York Stock Exchange. They find that the larger the acquirer the higher the premium paid. Hence, more time is needed for the added value of the transaction to materialize for the bidding company shareholder. Target companies experience enhanced value through M&A, when premium is high a mere transition of wealth from bidder to target shareholders takes place.

(12)

12

the bidding company. This indicates that relative firm size impacts the potential added value for bidding companies. For target companies, the effect of their relative size on M&A failures has not been studied. Asquith et al. (1983) examined the impact of relative target size, and found that target companies who are half the bidder companies’ size yield 1.84% higher cumulative abnormal return than target companies who are only one-tenth the size of the bidder. This means that abnormal return of a bidding firms depends on the relative size of the target company (Asquith et al., 1983). When the target equity value is 10% or more of the bidders’ equity value, the average abnormal return is 4.1%. For mergers in which the relative target size is less than 10%, the average abnormal return is 1.7% (Jensen and Ruback, 1983).

(13)

13

3. Hypotheses

Previously, literature has shown various outcomes when studying announcement effects in M&A. Some argue that announcement effect is significant and exists, others argue that there is no more than a zero return (see Seghal et al., 2012, for an overview), for bidding firms. In M&A, announcement effects have been observed for target firms when M&A materializes (Bruner, 2001; Campa and Hernando, 2004), resulting in positive abnormal returns. In case of M&A failures, such as withdrawal the opposite is expected. Neuhauser et al. (2011) show that failed takeover attempts impact the target company, and that the stand-alone effect on shareholder value is negative. Most authors report negative stand-stand-alone impact as a result of M&A failure (Malmendier et al., 2016). Hence the first hypothesis is:

H1: M&A failure has a negative impact on target company shareholders value

The causes of failures in M&A, in this study, are to be sorted in three categories, withdrawal, termination, or rejection. Bates and Becher (2017) provide reasons for rejecting a take-over bid, rejection takes place when the target company receives a bid with poor quality. In other words, the target company feels undervalued and the bid, even if a premium is offered, is not value enhancing in the eyes of the target company. This could signal investors that the target company should be valuated at a higher stock price. Other authors (Dodd and Ruback, 1977; Dodd, 1980; Desai et al, 1983) argue that rejected companies will trade above the share price prior to the M&A offer.

M&A failures through withdrawal can be a signal of bidding company CEO quality (Jacobsen, 2014; Alexandridis at al. 2013). Not willing to raise the bid, could be a sign of the ‘right’ value. Hence, bid withdrawal could signal the appropriate value of the target firm, or even overvaluation.

(14)

14

Because of the different causes of M&A failures, and their own reasons. It is expected that the cause of M&A failures can result differences in abnormal returns compared to withdrawal. The second hypothesis states:

H2: Shareholder wealth of target firms is influenced differently by the various causes of failures in M&A.

Malmendier et al. 2016 show the revaluation effect of failed M&A, with regard to medium of exchange. Stating that failed cash offers are value enhancing for target companies, and that failed stock bids are not. In addition, Neuhauser et al. (2011) argue that the cumulative effect of a failed takeover is positive for target company shareholders. Meaning that even when an announced takeover or merger does not go through, the target shareholders have benefitted from the process. Davidson et al. (1989), and Sullivan et al. (1994), found that terminated mergers and terminated merger proposals yielded positive abnormal returns for target company shareholders compared to the pre-announcement date. Asquith et al. (1983) showed that, relative bigger target companies generate a higher abnormal return for bidding companies than smaller target companies. This provides evidence that firm size has an impact on abnormal return. Malmendier et al. (2016) show that relative target company size is negatively related with abnormal return of the target company after M&A failures. Combining the insights on company size and M&A failures implies that relative size of the target company can potentially influence target company abnormal return. It is expected that relatively bigger target companies experience more negative returns than relatively smaller target companies. The following hypothesis is proposed:

(15)

15

4. Data and methodology

In this study the focus is centered on the impact of M&A failures on the target company’s shareholders value. An event study uses financial data to measure the impact of a specific event on the value of the firm (MacKinlay, 1997). An international sample is used to examine short-term and long-term effects of M&A failures

4.1 Data

The initial sample includes 201 companies from the US, the UK, Germany, and the Netherlands. The event, M&A failures, is identified through ThomsonOne. To conduct an event study information about the share price of a company is needed, meaning that I impose some selection criteria, to delete observations. The first criteria relates to stock information, hence all unlisted target companies, and target companies without stock data have been deleted, since they can’t be used. This resulted in a sample of 94 companies. A second criteria considers clarity of information, to make sure the event dates of both M&A announcement and M&A failures were accurate, the results from ThomsonOne have been verified with online news sources. If unclear dates have been reported the companies in question have been deleted. A third criteria has been suggested in literature (Malmendier et al., 2016), which suggests to delete cases with contested M&A proposals, I will follow this approach. This yields a sample of 71 companies, for Sample A. The sample size is comparable to the sample size of Sullivan et al. (1994)1. Sample A will be called “failed M&A attempts” from here on. This sample will be used to test hypothesis 1 and hypothesis 2.

To test the third hypothesis, one needs information on the market capitalization of the bidding company, to calculate the relative size of the target company. Hence an additional sample, sample B, is constructed which builds upon “failed M&A attempts”. Adding the market capitalization of the bidding company as a criteria reduced the sample to 35 observations. Following the approach of Malmendier et al. (2016), companies that report less than five trading days between M&A announcement and M&A failure are excluded. This

1 The sample of Sullivan, Jensen, and Hudson (1994) consists of 84 observations with an interest period from

(16)

Table 1

Summary statistics

The table reports statistics for the two main samples, which are described in section 4. Time to failure is measured in trading days from M&A announcement till failure announcement. Target

market cap refers to the market capitalization of the target company in billions of dollars. Target q is the market value plus the company assets minus book value of equity, divided by the total

assets. Cash/assets is the cash holdings of the target company relative to the total assets the company holds. % premium offered reflects the premium the bidding company offered to the target company. Cash bid in % shows the amount of the received bid as a cash percentage. Target relative size is the target company market capitalization divided by the bidding company market capitalization. The p-value refers to a two-sided difference in means test, for US companies and European companies. All values have been obtained at the end of the preceding fiscal year of the target company.

Sample A: failed M&A attempts

Cause of M&A failures European companies US companies

Mean Median Std.Dev Min Max Mean Median Std.Dev Min Max Mean Median Std.Dev Min Max p-value

Time to failure 90 55 108 0 593 70 53 67 0 209 99 69 121 0 593 0.20

Target market cap 8.956 0.940 19.023 0.008 118.254 6.255 0.538 11.716 0.014 35.002 10.168 1.147 21.514 0.008 118.254 0.32 Target q 1.860 1.348 1.521 0.661 9.900 1.997 1.215 1.672 0.883 7.446 1.797 1.381 1.461 0.661 9.900 0.63 Cash/ Assets 0.149 0.114 0.143 0.002 0.687 0.151 0.093 0.117 0.003 0.379 0.148 0.116 0.154 0.002 0.687 0.92 Market/book 4.439 1.841 11.082 -15.942 64.882 5.118 1.777 12.579 -7.764 58.543 4.128 1.449 122.947 0.000 593.00 0.75 Offer premium in percent 32.66 23.68% 36.11 -4.32 206.01 34.44 28.46 38.83 -1.08 127.00 32.06 23.37 35.60 -4.32 206.01 0.84 Cash bid in percent 67.70 100.00 39.71 0.00 100.00 72.93 100.00 36.76 0.00 100.00 65.78 100.00 41.00 0.00 100.00 0.54

N 71 22 49

Sample B: M&A failures with relative size Relative target size <30% Relative target size >30%

Mean Median Std.Dev Min Max Mean Median Std.Dev Min Max

Time to failure 91 50 148 8 593 148 135 90 20 350

Target market cap 6.148 1.916 9.515 0.070 34.551 24.881 17.929 30.351 0.139 118.254 Target relative size 0.102 0.076 0.091 0.002 0.264 1.255 0.741 1.527 0.344 6.500 Target q 1.681 1.287 0.877 0.883 4.092 2.176 1.467 2.251 0.999 9.900 Cash/ Assets 0.144 0.058 0.196 0.002 0.687 0.122 0.089 0.120 0.005 0.396 Offer premium in percent 57.41 45.07 37.27 5.92 127.00 22.31 18.87 16.26 0.00 50.61 Cash bid in percent 62.36 75.00 42.22 0.00 100.00 40.93 33.39 40.61 0.00 100.00

N 14 15

(17)

17

reduced the sample for Sample B to 29 companies. Sample B will be named “M&A failures with relative size” from here on.

The period of interest has been set to be 2012-2016, for US, UK, German and Dutch companies, with 49, 14, 5, and 3 observations, respectively. Meaning 49 US observations, and 22 European observations. Because of the reduction of the initial sample, “failed M&A attempts”, “M&A failures with relative size” consists of 19 US observations, 7 UK observations, 2 German observations , and 1 observation from the Netherlands. Which is equal to 19 US and 10 European observations, respectively. The individual company share prices at end of day, have been gathered through Yahoo! Finance (2017) for all companies. As well as the information about market indices. Company variables for both “Failed M&A attempts” and “M&A failures with relative size” are gathered through Compustat. Information about the cause of M&A failure (withdrawal/ rejection/ termination) is gathered through various news sources. Table 1 reports the descriptive statistics of the samples, it should be pointed out that in some cases the mean and median are quite different from one another. For example, the variable “Target market cap”, median and mean are different which signals skewness.

4.2 Variables

The variables used for hypotheses testing are relative company size and cause of M&A failures. Where the cause of M&A failures refer to one of the following: rejection, termination, or withdrawal.

(18)

18

Relative company size refers to the relative size of the target company. The relative size is the market capitalization of the target company over the market capitalization of the bidding company, and is expressed as a percentage of the bidding company’s market capitalization.

4.3 Methodology

An event study uses financial data to measure the impact of a specific event on the value of the firm (MacKinlay, 1997). In accounting and finance an event study is a well-known method to study stock price changes (Corrado, 2011; MacKinlay, 1997) to examine the possible enhanced value of an event. I use two types of event studies of event studies, the short-term event study, which uses the cumulative abnormal returns (CAR) as a measurement. As well as the long-term event study, which uses buy-and-hold abnormal returns (BHARs) as a measurement.

4.3.1 Short-term event study

Since the event study method is well-known in financial literature, and common in similar studies, this study will take a similar approach. In this study abnormal return (ar) is defined as:

ari,j = ri,j – E(ri,j), (1)

where ri,j is the return on security i at time j, with time in days, and E(Ri,j) is the expected return on security i at time j, in days. To make inferences about company performance, the expected return will be calculated by using the market model. With

E(ri,j) = αi + βi Rm,j + ε, (2)

(19)

19

Note that the cumulative abnormal returns can be compared across deal failures, with different event window lengths. For instance, Schwert (1996) examined that the announcement effect suffers from a pre-bid run-up. By constructing an event window with 25 days pre-bid in mind, the pre-bid run-up is accounted for, as do Malmendier et al. (2016). The test statistic for the cumulative abnormal return is:

Z= CARt /(𝜎𝜎𝜎𝜎𝜎𝜎𝜎𝜎𝑡𝑡√𝑛𝑛

),

(4)

4.3.2 Long-term event study

In order to examine potential abnormal returns in the long-run, a change in approach is needed. Barber and Lyon (1997) describe the differences between cumulative abnormal returns (CARs) and buy-and-hold abnormal returns (BHARs). For instance, in a long-term event study cumulative abnormal returns tend to report higher abnormal returns than buy-and-hold returns do (Barber and Lyon, 1997). In addition, the authors mention that BHARs are best used in a one to five year (long-run) window. BHAR is defined as:

with

BHARi,j =

where Ri,j is the return on security i at time j, in months. And Rm,j is the return on the market index, m, at time j, in months. Again, market indices from the US, the UK, The Netherlands, and Germany are used. Kothari and Warner (1997) use time in months for their examination into long-horizon security performance, this approach is followed in this study. The test statistic for buy-and-hold abnormal returns is similar as for cumulative abnormal returns. The test statistic is defined as:

Z= BHARt /(𝜎𝜎𝜎𝜎𝜎𝜎𝜎𝜎𝜎𝜎𝑡𝑡√𝑛𝑛

),

(7)

(20)

20

5. Results

5.1 Stand-alone M&A failures

It is argued that the stand-alone effects of M&A failures have a negative impact on shareholder value. Table 2 shows the results of stand-alone M&A failures for both “failed M&A attempts” and “M&A failures with relative size”. All estimates are statistically significant at the 1% level, except for rejecting an M&A offer, at t(-1,+1). It is shown that both the “failed M&A attempts” and “M&A failures with relative size” experience negative cumulative abnormal returns in various event windows, statistically significant at the 1% level. Which is in line with hypothesis 1, hence this study fails to accept that M&A failure does not impact shareholder wealth. Interestingly, rejecting an M&A offer results in positive cumulative

abnormal returns in all three different event windows. Yet, for t(-1.+1) results are not significantly different from the normal performance. When comparing the CARs of event windows, for rejected M&A offers, it becomes obvious that CAR doubles from 1,+1) to t(-5,+5). Suggesting that investors are processing the information and act on the information after the announcement. Or, investors already acted on insider information before the

Table 2

Abnormal returns for stand-alone failure announcement

The table reports statistics on the stand-alone cumulative abnormal returns at the time surrounding M&A failure. With t(-5,+5) indicating an event window of 5 trading days before failure announcement until 5 trading days after failure announcement. The same holds for t(-1,+1) and t(-3,+3). CAR, shows the cumulative abnormal return in percentages, with the respective Z-score in the column next to it. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively.

t(-5,+5) t(-1,+1) t(-3+3)

N CAR Z-score CAR Z-score CAR Z-score

Failed M&A attempts 71 -5.80 -10.95*** -3.73 -4.79*** -3.93 -6.87***

Rejection 20 2.98 3.47*** 1.44 1.12 3.55 3.57*** Termination 14 -13.11 -7.81*** -9.19 -3.59*** -10.07 -5.52*** Withdrawal 37 -8.08 -12.60*** -4.63 -5.32*** -5.98 -9.13*** European companies 22 -8.58 -10.27*** -3.71 -3.56*** -5.03 -6.29*** US companies 49 -4.55 -7.104*** -3.74 -3.48*** -3.44 -4.49***

(21)

21

announcement, as suggested by Keown et al. (1992). Data reveals positive returns for trading days prior to the announcement that M&A has been rejected, hence insider trading might explain why t(-1,+1) differs from other event windows. Alternatively, the overall value enhancement for rejected M&A can be explained by undervaluation of the target company. Bates and Becher (2017) examine that perceived low M&A offers are rejected more frequently, implying that M&A offers were lower than the actual value of the target company. The results of Table 2 suggest the same: on average, after rejecting a takeover bid, the value of a company increases slightly adding value to the shareholders. Meaning that the perceived value of the company has increased to a more appropriate level.

For completeness, the observations from US companies and European companies are compared, to see if investors react differently to M&A failures in different continents. Results show that this is not the case. The US and European companies have been compared in each event window used, in Table 3. Results show that there is no statistically significant difference between US and European companies’ cumulative abnormal return for M&A failures, with p-values of 0.32, 0.99, and 0.64, for t(-5,+5), t(-1,+1), and t(-3,+3), respectively.

In both samples, “failed M&A attempts” and “failed M&A with relative size”, the cumulative abnormal return (CAR) is negative. Along with the inability to accept that M&A

Table 3

Difference in means

The table reports various cumulative abnormal returns over different event windows. Statistics are presented on the one-sided difference in means of cumulative abnormal returns for different ‘causes of M&A failures’. The P-value shows the ANOVA single factor values. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively.

(22)

22

failure has no impact on shareholder wealth, hypothesis 1 is confirmed. Hence, M&A failures have a negative impact on shareholder value. The stand-alone impact of M&A failures adds to the financial literature, and is in line with other authors such as Malmendier et al. (2016), and Neuhauser et al. (2011).

5.2 Shareholder wealth and cause of M&A failures

The discussion on the impact of the cause of M&A failures, on shareholder value has not been studied into great extent in the financial literature. In the past, authors have identified explanatory factors to describe shareholder wealth effects following M&A for different causes of M&A failures (Bates and Becher, 2017; Sullivan et al., 1994; Davidson et al., 1989; Safieddine and Titman, 1999). These previous studies generated different explanations for positive and negative wealth effects.

Table 3 shows that the causes of M&A failures differ in their CARs, and that there is evidence that shareholders experience most negative impact from termination, and the least negative impact from rejection. Moreover, rejection of an M&A bid generates a positive CAR

in the calculated event window, earlier examination into M&A rejection has reached similar results (Dodd and Ruback, 1977; Dodd, 1980). Table 3 provides evidence that rejection is significantly different from both termination and withdrawal. The means of termination and withdrawal are not significantly different from one another. The means are, however,

Figure 2

Buy-and-hold abnormal returns of failed M&A through: rejection, termination, and withdrawal.

-0,1 0 0,1 0,2 0,3 0,4 0,5 0,6 3 6 9 12 15 18 21 24 27 30 33 36 BU Y A N D HO LD A BN O RM AL R ET U RN

MONTHS AFER FAILURE

(23)

23

significantly different at t(-1,+1), at the 10 percent level. The expected negative relationship, that termination has a more negative impact than withdrawal, is present. Since CARs of withdrawal and termination are not significantly different from one another through all event windows, hypothesis 2 is not supported. The results are in line with Neuhauser et al. (2011), for event windows t(-3,+3) and t(-5,+5), who argue that there is no difference between termination of M&A and withdrawal of M&A. Yet, this study observes a statistical significant difference in means between termination and withdrawal at t(-1+1), at the 10 percent level. In almost all event windows a significant difference in means between rejection and both termination and withdrawal is found, which adds to literature.

5.3 Long-term revaluation

In stand-alone M&A failures the CARs of various causes of failures differ significantly from one another, as showed in Table 3. Besides the stand-alone effects of M&A failures it is interesting to know whether these differences persist over a longer period of time. Therefore, estimations about the long-run abnormal return are examined, up to three years. A few

Table 4

Long-term persistence of differences between cause of M&A failures

Buy-and-hold estimates for one to three years portfolio return after M&A failure date. N is the amount of companies in the portfolio. Standard deviation is reported in parenthesis. *,**, and *** denote significance at the 10%, 5%, and 1% level, respectively

Sample A

Rejected Terminated Withdrawal

(24)

24

remarks are in order before the results of the long-run test are discussed. In the period between gathering daily data for stand-alone testing and the monthly data for long-run tests, some information has become unavailable. A total of eight companies had to be removed for the long-run tests, resulting in a total of 63 observations. This leaves the sample to be divided in the following way: Rejection: 19 observations (-1), termination: 12 observations (-2), and withdrawal: 32 observations (-5). Since the data is relatively young, inferences cannot be made for the full test period of three years. Simply because not all event dates are three years in the past. This means that the number of observations will vary per year. The horizon of three years is significantly longer than the event windows used previously, hence reliance on CAR is no longer suitable. As discussed in section 4, the long term event will follow the buy-and-hold approach of Barber and Lyon (1997).

Figure 2 illustrates the evolution of the buy-and-hold abnormal returns in the months after M&A failure. Differences between the causes of M&A failures seem to exist in the first two years. Returns for “failed M&A attempts” in year 1 are positive on average, about 13%. A remarkable difference is presented in Figure 2; between 12 and 15 months after M&A failure, for terminated and rejected M&A attempts. Rejected M&A offers more than double their abnormal return in this three month period, and terminated M&A offers lose half of their enhanced value. One should be aware of the change in the sample size, as reported in Table 4, since this can be a, partial, explanation for the sudden change in buy-and-hold abnormal returns (BHAR). A similar change, for withdrawn M&A, is also visible between months 24 and 27, at this time the sample size changes again.

Table 4 shows the BHARs for each cause of M&A failures. Two years after M&A failure, companies that have rejected an M&A offer still outperform the market significantly at the 10 percent and 5 percent level respectively. Bradley et al. (1983) find similar revaluation results for companies that rejected a takeover bid. In the third year however, this significant difference disappears. For terminated and withdrawn M&A failures, results are non-significant. In an efficient market, one would not expect abnormal returns months after failed M&A (Fama, Fisher, Jensen, Roll, 1969).

(25)

25 5.4 Relative target size

Figure 3 underlines the suggested evidence that relative size has an impact on shareholder value. The figure plots the evolution of the cumulative abnormal returns from 25 trading days before the bid announcement to 25 trading days after the announcement of failure. In accordance with Malmendier et al. (2016) the interim period trading days, the period between the announced bid and the announced M&A failure, have been normalized.

Whereas Malmendier et al. (2016) interpolated each percentile, this thesis has

interpolated each 10th percentile. For instance, if a bid fails after 40-days the 50th percentile relates to 20-days, and the 10th percentile to 4-days, after bid announcement. In addition, if a bid fails after 100-days the 50th percentile relates to 50-days, and the 10th percentile relates to 10-days, after bid announcement. Observed are the cumulative abnormal bid announcement returns of approximately 20% and 10% for companies with a relative size smaller than 30% and relative size larger than 30%, respective. These magnitudes are similar to earlier studies regarding M&A announcement effect (e.g. Sullivan et al., 1994; Huang and Walkling, 1987). Target companies with a relative size of over 30% lose the initial gains from bid announcement before M&A failure has been announced. 25-days after M&A failure the shares of companies with a relative size larger than 30% of the target company trade around 5% lower than pre M&A bid announcement. Target companies with a relative size smaller than

Fig. 3 Announcement effects at bid and failure (25 days). The figure depicts the cumulative average abnormal returns

(CAARs) from 25 trading days before the initial bid, to 25 days after M&A failure. See panel B, Table 1.

(26)

26

30%, lose about half of the initial gains from bid announcement. However, 25-days after M&A failure, their shares still trade at a premium of about 8%, compared to the period before the bid announcement.

The event window provides evidence of a significant difference in means between the two samples, at the 5% level respectively. Target companies smaller than 30% of the bidding company have significant different CARs from target companies with a relative company size bigger than 30% of the bidding company. The cumulative abnormal returns of relatively smaller companies is bigger than of relatively bigger companies. This implies that the relative size of the target company is negatively related to shareholder wealth.

Table 5 presents the multivariate regression. The test has the purpose to estimate differences in relative size, as hypothesized in section 3, hypothesis 3. First, target CAR is regressed on relative size without control variables and fixed effects. This is done in order to replicate the evidence of Figure 3. However, the results are different from the expectation, and no significant influence of relative size is found, the hypothesized negative relationship,

Table 5

Relative target size regressed on target CAR

The table reports ordinary least squares regressions with target cumulative abnormal return (CAR) from 25 days before announcement (B) to 25 days after failure announcement (F) as the dependent variable. “M&A failures with relative size” has been used for models 1 to 3, where missing values are excluded. For models 4 to 6 missing values are replaced with the means from “M&A failures with relative size”. Relative target size is the market capitalization of the target divided by the market capitalization of the bidder. Offer premium is expressed as the percentage of the bid divided by the market capitalization. Cash is expressed as the part of total payment. A constant term is always included in the absence of fixed effects. *, **, *** denote significance at the 10%, 5%, and 1% level, respectively.

Variable Target CAR (B-25, F +25)

1 2 3 4 5 6

Relative target size -0.058 -0.110 -0.084 -0.058 -0.040 -0.040 (0.04) (0.14) (0.14) (0.04) (0.04) (0.04)

Offer premium in percent 0.169 0.239 0.234 .274

(0.18) (0.20) (0.16) (0.17)

Cash bid in percent 0.018 0.019 0.002 0.026

(0.15) (0.13) (0.14) (0.12)

Country fixed effects No No Yes No No Yes

Year fixed effects No Yes No No Yes No

(27)

27

(28)

28

6. Conclusion

6.1 Conclusion

The purpose of this thesis is to examine two potential explanatory factors of value enhancement surrounding M&A failures: cause of M&A failures and company size. Three hypotheses have been tested to examine whether the proposed explanatory factors indeed have an influence. The first hypothesis tested the stand-alone effect of M&A failures. Results show that stand-alone effects of M&A failure are significant (see Table 2), and value destructing, for both “failed M&A attempts” and “failed M&A with relative size”. No evidence is found that shareholders of European (UK, Germany, and the Netherlands) and US companies experience different wealth effects in case of M&A failures. This holds for various event windows for stand-alone M&A failures for t(-1,+1), t(-3,+3), and t(-5,+5).

The second hypothesis examines the impact of the cause of M&A failures. The difference between termination and withdrawal shows the expected negative sign, with termination having a bigger impact on shareholder wealth than withdrawal. However, for two out of three event windows the mean differences between withdrawal and termination are not statistically significant. Only at t(-1,+1) a statistically significant mean difference is observed for termination and withdrawal, at the 10 percent level. The differences with rejected M&A proposals, however, are significant for all event windows. In the long-run companies who rejected a takeover offer experience abnormal returns up to two years after bid rejection. For terminated and withdrawn M&A, no long-term evidence is found. Summarizing, if the cause of M&A failures is rejection shareholder value is impacted differently than if the cause of M&A failures is either withdrawal or termination.

(29)

29

relatively larger than 30 percent of the bidding company experienced negative returns in the same event window. Finally, a multivariate regression was conducted to examine whether size explained the differences in cumulative abnormal returns, and results were insignificant.

6.2 Limitations

Findings cannot be easily generalized, this is due to the endogenous nature of deal failures (Malmendier et al. 2016). Besides, the problem is a subtle selection bias; of cause of M&A failures, and relative size. Categorization involves a certain amount of judgement which is inevitable.

This categorization makes that inferences about the sample can only be applied to the sample itself. Fig. 2 suggests a difference between the relative sizes of companies, yet the test statistics in Table 5 doesn’t point out that relative size is the explanatory factor. This could be due to the fact that “failed M&A attempts with relative size” has a rather small sample size. Because of the small sample size, the outliers within the sample can disturb a potential relationship, and have a large influence on the test results.

6.3. Future research

Rejecting an M&A offer has proven to be value enhancing in this study. This implies that target companies who have rejected an M&A offer have been undervalued. The value enhancing effects of rejecting an M&A bid can last up to two years after M&A failures. However, as Bates and Becher (2017) point out, rejecting an appropriate bid could lead to forced CEO turnover. This means that company valuation skills are important for managers. Future research could be conducted to examine the relationship between valuation skills and M&A failures.

(30)

30

References

Alexandridis, G., Petmezas, D., & Travlos, N., 2010. Gains from mergers and acquisitions around the world: new evidence. Financial Management 39, 1671–1695.

Alexandridis, G., Mavis, C., Terhaar, L., and Travlos, N., 2013. Can failure signal success? Evidence from withdrawn M&A deals. Unpublished working paper. Henley Business School, Reading

Asquith, P., 1983. Merger bids, uncertainty, and stockholder returns. Journal of Financial Economics 11, 51–83.

Barber, B.M., Lyon, J.D., 1997. Detecting long-run abnormal stock returns: the empirical power and specification of test statistics. Journal of Financial Economics 43, 341–372.

Bates, T., Becher, D., 2017. Bid resistance by takeover targets: managerial bargaining or bad faith? Journal of Financial and Quantitative Analysis 52, 837-866.

Bettinazzi, E.L.M., Zollo M., 2017. Stakeholder orientation and acquisition performance. Strategic Management Journal 38, 2465-2485.

Bradley, M., Desai, A., Kim, E.H., 1983. The rationale behind interfirm tender offers: Information or synergy? Journal of Financial Economics 11, 183-206.

Bruner R.F., 2001. Does M&A pay? A survey of evidence for the decision-maker. Journal of Applied Finance 12, 48-68.

Campa J.M., Hernando, I., 2004. Shareholder value creation in European M&As. European Financial Management 10(1), 47-81.

Corrado, C.J., 2011. Event studies: a methodological review. Accounting and Finance, 51: 201-234. Davidson W.N., Dutia, D., and Cheng, L., 1989. A re-examination of the market reaction to failed mergers. Journal of Finance 44, 1077-1083.

Dodd, P., 1980. Merger Proposals, management discretion and stockholder wealth. Journal of Financial Economics 8, 105-137.

Dodd, P., Ruback, R., 1977. Tender offers and stockholder returns. Journal of Financial Economics 5, 351-373

Fama, E.F., Fisher, L., Jensen, M., and Roll, R. (1969). “The adjustment of stock prices to new information.” International Economic Review 10, 1–21.

Huang, Y.S., Walkling, R.A., 1987. Target abnormal returns associated with acquisition announcements: payment, acquisition form, and managerial resistance. Journal of Financial Economics 19, 329–349.

Imaa.com https://imaa-institute.org/mergers-and-acquisitions-statistics/ [accessed at 27-09-17] Jacobsen, S., 2014. The death of the deal: Are withdrawn acquisition deals informative of CEO quality? Journal of Financial Economics 114, 54-83.

(31)

31 Keown, A.J., Pinkerton, J.M., 1981. Merger Announcements and Insider Trading Activity: An Empirical Investigation. Journal of Finance, 36, 855-869.

Keown, A.J., Pinkerton, J.M., Bolster P.J., 1992. Merger announcement, asymmetrical information and trading volume: an empirical investigation. Journal of Business Finance & Accounting 19, 901-910.

Lee, M.I., Opp, M.M., 2005. Private Information in M&A Transactions: Evidence from Withdrawn Deals. Unpublished working paper. Stockholm School of Economics, Sweden. Available at SSRN: https://ssrn.com/abstract=2565608 [accessed at 12-03-2018]

Liu, Y., (2018) Shareholder wealth effects of M&A withdrawals. Available at:

https://acfr.aut.ac.nz/__data/assets/pdf_file/0006/29724/581149-Y-Liu-Shareholder-Wealth-Effects-of-M-and-A-Withdrawals-_Yue-Liu.pdf [accessed at 25-02-18]

MacKinlay, C., 1997. Event studies in economics and finance. Journal of Economic Literature 35, 13-39.

Malkiel, B.G. & Fama E.F. 1970 Efficient Capital Markets: A review of theory and empirical work. Journal of Finance 25, 383-417.

Malmendier, U., Opp, M.M., Saidi, F., 2016. Target revaluation after failed takeover attempts: Cash versus stock. Journal of Financial Economics 119, 92-106.

McNichols, M. & Manegold, J.G., 1983. The Effect of the information environment on the relationship between financial disclosure and security price variability Journal of Accounting Economics 5, 49-74. Moeller, S.B., Schlingemann, F.P., and Stult, R.M., 2004. Firm size and the gains from acquisitions. Journal of Financial Economics 73, 201-228.

Neuhauser, K.L., Davidson, W.N., and Glasock, J.L., 2011. An analysis of failed takeover attempts and merger cancellations. International Journal of Managerial Finance 7, 347-376.

Safieddine, A., Titman, S., 1999. Leverage and corporate performance: Evidence from unsuccessful takeovers. Journal of Finance 54, 547-580.

Schaik, D.V., Steenbeek, O.W., 2004. Price and volume effects of merger bids in Japan. Available at SSRN:http://ssrn.com/abstract=592081 [accessed at 01-03-2018]

Seghal, S., Banerjee, S., Deisting, F., 2012. The impact of M&A announcement and financing strategy on stock returns: evidence from BRICKS markets. International Journal of Economics and Finance 4, 76-90.

Seth, A., 1990. Value creation in acquisitions: a re-examination of performance issues. Strategic Management Journal 11, 99-115.

Sullivan M.J., Jensen, M.R.H., Hudson C.D., 1994. The role of medium of exchange in merger offers: examination of terminated merger proposals. Financial Management 23, 51-62.

Schwert, G.W., 1996. Markup pricing in mergers and acquisitions. Journal of Financial Economics 41, 153–192.

Referenties

GERELATEERDE DOCUMENTEN

(Color online) Fraction of (a) weak sliding contacts (wsl) and (b) strong sticking contacts (sst) with respect to the total number of weak (  w ) and strong (  s )

This effect relies on the matching of the standing wave field within the multilayer stack with the structure: the minima of the wave field intensity are placed in the center

Time (in seconds) to set up Q as a sparse matrix and the time to evaluate the quasi-stationary distribution using eigs (with m = 20) in the n-patch metapopulation model for

Want het uiterlijke heeft zijn begrip en betekenis niet meer in zich en op zichzelf, zoals bij de klassieke kunstuiting, maar in het gevoel (‘Gemut’).. En het gevoel vindt zijn

This special issue is based on selected papers and extended papers of the Fifth International Conference on eHealth, Telemedicine, and Social Medicine, held on February 2013, in

tiek, wat verskillende gegewens statisties verwerk het ten einde n redelike basis vir prognose te voor- sien; die Hoof en personeel van die Tegniese De-

More specifically, three studies found that the presence of traffic-calming measures was associated with an increased use of active transport, three studies reported that a per-

Gesamentlike skoolopvosding van blank en nie-blank en 'n En5slse skoolopvoeding vir nie-blankes op hulle eie skole .hat die hels