the target can be improved
R. ter Maat
21 June 2013
University of Groningen
Faculty Economics and BusinessMSc. Finance
the target can be improved
ABSTRACT
Keywords: Acquisition announcements, Event study, public targets, private targets, pre-deal profitability, room for improvement, European firms.
JEL codes: G14, G34
R. ter Maat
University of Groningen
Faculty Economics and BusinessMSc. Finance
Supervisor:
Dr. J.H. von Eije*
*
I would like to thank my supervisor Dr. J.H. von Eije for his guidance and advice during the research and writing of my thesis. He helped me to structure my ideas and provided me with useful feedback.
1. Introduction
Acquisitions are prevalent events in corporate finance and investigated extensively by scholars. Many scholars focus on the question what enhances the performance or profitability of acquisitions. Although the number of previous studies is large they hardly address the question whether pre-deal profitability is a variable which impacts the short term shareholder wealth.
Profitability is a broad term which can be measured in multiple ways in an acquisition context. In addition to this, profitability can be seen as the outcome of the whole business process through which a firm goes. Besides this, profitability is a term which is used by many different parties involved in the business environment. For example, in business valuation and in corporate finance often so called “rules of thumb” or relative valuations are used. Rather than sophisticated valuation techniques, these methods are relatively simple methods to determine the value of a firm. Many of these valuation techniques relate the value of the firm to a multiple of a profit related measure. Besides that measures of profitability are often used by practitioners, it is also of major importance for shareholders. Shareholder value maximization is according to the shareholders the major goal of firms. Since profitability is inextricably related to shareholder value, pre-deal profitability could influence the short-term stock market reactions to acquisition announcements.
So probably profitability is the most valued term by all participants in the business process. Hence, this paper investigates the effect of pre-deal profitability of both acquirer and target in the context of acquisition announcements. This paper takes the perspective of shareholders of the acquirer and investigates whether and how pre-deal profitability is valued.
An event study is conducted according to the methodology of Brown and Warner (1980, 1985) to investigate the stock market reaction to acquisition announcements. The cumulative abnormal returns (CARs) estimated by this methodology are used as input to investigate the impact of pre-deal profitability. Furthermore, we differentiate these impacts on the type of target. Both public and private targets are incorporated in this sample, which are all based in Europe. The sample consists of 817 acquisition announcements during the period 2000 until 2012.
Previous event studies on shareholder wealth effects of acquisition’ announcements primarily focus on takeovers of public firms. Various scholars (Chaterjee, 1986; Lubatkin, 1987; Chang, 1998; Capron and Shen, 2007) report that these acquisitions on average result in significant negative CARs. This implies negative short term shareholder wealth effects to takeover announcements of public firms.
targets on average result in positive CARs (Chang, 1998; Capron and Shen, 2007; von Eije and Wiegerinck, 2010).
Thus far only Capron and Shen (2007) investigated the influence of pre-deal profitability of acquirer and target on the CARs. This was done by using a 5 point scale for the profitability’s with anchors ranging from “much less profitable” to “much more profitable” compared to the industry average of the firm. They report a significant positive coefficient for target pre-deal profitability and non-significant coefficient for acquirer pre-deal profitability. However, they did not have a theory and explanation for this positive coefficient of target pre-deal profitability. Contrary to Capron and Shen (2007) this paper provides an extensive analysis of pre-deal profitability of the acquirer and target on the CARs. We do this at least in two ways different compared to Capron and Shen (2007). Firstly, this paper uses the actual reported profits rather than scales since we believe that shareholders do not think scale wise. Secondly, we take the perspective of the new company (Newco) by normalizing the profits by dividing them through total assets of both acquirer and target. This Newco perspective is taken since we believe that shareholders look to the future, rather than to the past.
Hence, this paper contributes to the research on public and private acquisitions since it extensively explores the pre-deal profitability of both acquirer and target. It adds value to both academics as business practitioners. Since profitability is a widely used concept it is useful to know how it is valued by shareholders in an acquisition context. Moreover, this paper investigates whether the effect of pre-deal profitability of acquirer and target differs for public and private targets. Besides the pre-deal profitability variable, this paper contributes to the current literature by investigating whether the documented cumulative average abnormal returns (CAARs) reported by other scholars papers holds in Europe.
2. Literature Review
2.1
Hypotheses regarding takeovers
Although acquisitions are a prevalent phenomenon in corporate finance, it is often a one-time event for parties involved. Especially private firms will not face acquisitions that often. Therefore, it is important to understand the hypotheses established in the literature regarding takeovers before examining its consequences.
There are three major hypotheses established in prior literature for takeovers: The synergy hypothesis, the agency hypothesis and the hubris hypothesis (Berkovitch and Narayanan, 1993). According to these scholars (1993, p. 350) the synergy hypothesis argues that “managers of targets and acquirers maximize shareholder wealth and would engage in takeover activity only if it results in gains to both sets of shareholders.” Therefore, in the context of the event study results synergy could be present when the CARs to both parties are positive. However, the synergy effect will not be studied in this paper because we do not study the abnormal returns here of the many private targets in our sample.
Whereas the synergy hypothesis results in gains to both parties involved, the agency hypothesis is motivated primarily by self-interest of the acquirer management (Berkovitch and Narayanan, 1993). According to this hypothesis shareholder value is decreasing due to the divergence in interest of managers and shareholders. For example, it could be that an acquisition is motivated by managerial enrichment rather than value-creating for shareholders. Managers might acquirer status or private bonuses by acquiring big firms, which simultaneously might destroy shareholder value. Since the agency hypothesis only incorporates the acquirer shareholders this can be measured in the event study executed in this paper.
The last hypothesis assessed in the literature is the hubris hypothesis, which is established by Roll (1986, p.212) who defines hubris as follows: “decision-makers in acquiring firms simply pay on average too much for their targets.” As a consequence, Roll (1986) predicts that the combined value of acquirer and target falls slightly, the value of the bidding firm should decrease, whereas the value of the target firm should increase. In accordance with this, Berkovitch and Narayanan (1993) take the perspective of the acquirer firms shareholders and assume that acquisitions are motivated by managers’ mistakes and that there are no synergy gains. This would be an explanation of the overpayment for the targets acquired. In order to test for hubris the perspectives of both bidder and target shareholders should again be taken into account. Besides that it is beyond the scope of this paper to measure the hypotheses above, it is also impossible to do for the hubris hypothesis due to the private targets in our sample.
following three groups: value conserved, value created and value destroyed. In some way this categorization can be related to the different hypotheses for takeover discussed by Berkovitch and Narayanan (1993). Hence, the outcome of the synergy hypothesis can be categorized as value creating, the outcome of agency hypothesis as value destroying and the outcome of the hubris hypothesis as value conserved.
2.2 Acquisitions of public firms
Traditionally research on mergers and acquisitions focuses on both listed bidders and targets. Chatterjee (1986) for example, examined mergers during the period 1969-1972 which involved merging firms having at least an asset base of 10 million. The primary aim of his research is to examine where the value creation of mergers and acquisitions stems from. This is done by selecting six different portfolios of firms based on different types of mergers and the different parties involved. The types of mergers analysed by Chatterjee (1986) are related (non-horizontal) mergers and unrelated mergers which in addition with three types of firms involved, targets, acquiring and rivals, results in the six portfolios. The results of the event study of Chatterjee (1986) reveals, although not highly significant, negative cumulative average abnormal returns (CAARs) for the shareholders of the acquiring firms. In addition to acquiring firms, rivalry firms who announce an acquisition do report negative CAARs as well. Contrary to acquiring and rivalry firms, target firms experience significant positive CAARs during the announcement period.
In accordance with the results of the event study of Chatterjee (1986), Lubatkin (1987) reports similar results for both acquiring firms as targets. By conducting an event study on both pre-merger as post-merger performance Lubatkin (1987) analysed the relationship of stockholders gains and relatedness of the merging firms. This was done by taking a long term perspective since the event study was conducted on a time interval ranging from 18 months prior to the merger announcement to 64 months after the announcement. Overall the pre-merger CARs for the different types of mergers were positive, whereas most post-merger CARs were negative for the acquiring firms. More recently Fuller et. al. (2001) studied the returns earned by firms making multiple acquisitions during a pre-specified time period. The acquired firms include public targets, private targets and subsidiaries during the period 1990 until 2000. The sample used includes 3,135 take-overs and the results with respect to acquisitions of public firms are in accordance with Chatterjee (1986) and Lubatkin (1987). The event study results of Fuller et. al. (2001) report significant negative CAARs for public targets of -1.00%. Whereas buying private firms and subsidiaries report positive CAARs of respectively 2.08% and 2.75%.
private targets is discussed in the next sections.
2.3 Acquisition of private firms
Capron and Shen (2007) noted that the volume of acquisitions involving privately held targets by far surpasses that of publicly traded firms. However, the studies performed on acquisitions of public targets by far surpasses that of studies on private targets. This raises the question whether the evidence based on acquisitions of publicly traded firms holds in the context of private targets. Starting point is the question why to choose for private targets instead of public targets is partially answered by Capron and Shen (2007). They report that acquirers are more likely to buy private targets when their resources, business, geographic position, and their prior acquisition experience provide them with sufficient capacity to search and evaluate the assets from private firms (Capron and Shen, 2007). Knowing why acquisitions of private targets are executed is helpful, but does it affect shareholder wealth? Several studies have studied the effect on short-term shareholder wealth by examining abnormal returns of acquirers in acquisition announcements of private target. Mostly they report significant positive CAARs on the announcement of acquisitions of a private target (Chang, 1998; Hansen and Lott, 1996; Fuller et. al., 2001; Capron and Shen, 2007). This contrasts, to the negative CAARs reported in previous research on public targets.
Fuller et. al. (2001) argue that this difference in reaction is primarily due to the fact that acquiring firms get a better price when they acquire private targets. According to Fuller et. al. (2001, p.3) this could be due to a liquidity effect; “private firms cannot be bought and sold as easily as publicly traded firms.” Therefore, private firms become less attractive which can result in a discount which is captured by the bidder. This would explain the positive CAARs to the acquisition announcement of private targets. Capron and Shen (2007) in similar sense argue that due to the fact that private firms are less attractive, bidders face less competition. Hence, due to the lack of competition a discount can be earned on buying a private firm which is harder for publicly traded firms.
Private firms might also be less attractive due to the limited information availability compared to public firms. Since private firms do not have to disclose as much information as public firms, information asymmetries are more likely to occur. According to Capron and Shen (2007) acquirers know this and are therefore aware of the risk of overpaying. Consequently, in negotiations it is more likely that acquirers offer lower prices for private target.
(2010) report evidence for a governance effect, which is evident from the fact that private acquisitions by civil law acquirers result in smaller abnormal returns as private acquisitions of acquirers in common law countries.
This paper also uses a sample of listed European acquirers but here we focus on acquisitions of European public and private targets. Nevertheless, we expect results which are in line with the above mentioned scholars.
2.4. Room for improvement
Improvements are likely to occur in the context of acquisitions. For example improvements with respect to governance, synergies and networks. Since in general targets are smaller in size as acquirers, targets could benefit of the superior governance of the acquirer which in the end could result in cost savings. In the same manner potential synergies could exist. One could think of cost savings due to reduced overhead. It is likely that a target, since it is smaller, has a relative larger overhead per unit output. The overhead of the target might relatively easily be carried out by the current overhead of the acquirer. Lastly, the network of the acquirer is likely to be superior to the network of the target. Hence, the target is likely to benefit from this network and consequently the performance of the target in the new company (Newco) could be enhanced.
Besides this, the target could learn from the knowledge that resides within the acquirer firm. This is likely to enhance the performance as well. Moreover, learning could also occur in the opposite direction. As Vermeulen and Barkema (2001) argue that acquisitions allow firms to achieve greater market power, overcome barriers to entry, to enter new markets quickly, and acquire new knowledge and resources.
The major contribution of this paper to the current literature is not to disentangle all the reasons for improvements. Instead we try to gauge the potential for improvement. We do this by examining the impact of pre-deal profitability on the abnormal returns during the event window surrounding an acquisition announcement. Despite the fact that profitability is a widely used concept in a business environment previous literature does not extensively discuss this variable in the context of mergers and acquisitions.
In business valuation it is not unusual to refer to a multiple times the profit as the value of the firm. Bhojraj and Lee (2002) call this “relative valuation” in which firm value is estimated using an accounting-based market multiple of a comparable company times the corresponding accounting number of the target. These relative valuation practices are often used as a cross-check of more advanced valuation methodologies.
value is the sum of the value of all financial claims on the firm- including equity, debt, preferred stocks and warrants.” Profitability can be seen as one way to maximize value of shareholders. However, as Jensen (2000) notes that there is a trade-off between profits and market share. This implies that a firm which is less profitable does not automatically imply that it performs worse, since it may create a larger market share and this might result in better future earnings. However, in general it can be assumed that profitability is valued by shareholders, since continued profitability in the long term is likely to improve the wealth of the shareholders.
Capron and Shen (2007) also incorporate pre-merger profitability in their paper. In their cross-section analyses pre-merger profitability of both acquirer and target is included. These profitability variables are measured on a 5-point scale with anchors ranging from “much less profitable” to “much more profitable.” The firms are classified based on their profitability compared to the average industry profitability. Moreover, these variables are included separately which in our opinion means that a pre-deal perspective is taken.
Capron and Shen (2007) argue that “a highly profitable target is likely to have a valuable resource to be leveraged in an acquisition.” However, they also argued that the higher the profitability, the higher the bargaining power which could lead to higher transaction prices. For the acquirer pre-deal profitability Capron and Shen (2007) suggest that an acquisition could be a way to hide poor results or to find new ways to grow. It can be concluded from these hypotheses that Capron and Shen (2007) did not have a theory for the effect of pre-deal profitability. This is evident from the different directions in which their hypotheses regarding pre-deal profitability of acquirer and target head. In the end the results of Capron and Shen (2007) report a significant positive coefficient for target pre-deal profitability and non-significant coefficient for acquirer pre-deal profitability. However, they did not really have an ex-ante theoretical explanation for this positive coefficient of target pre-deal profitability.
Therefore, we expect that shareholders of the acquirer value “room for improvement” in targets. This implies that shareholders prefer targets with a low profitability or even losses over a highly profitable target. This is due to the fact that targets with low profitability or even losses have considerable room for improvement. In the end, this room for improvement could result in higher post-acquisition profits for Newco and this is likely valued by shareholders at the acquisition announcement. This leads to the following hypothesis regarding pre-deal profitability:
H1: Shareholders of the acquirer react positively to targets which have considerable room for improvement.
In addition to this we examine the impact of the acquirer pre-deal profitability on the CARs to acquisition announcements. The profitability of the acquirer can be perceived by shareholders as the quality of the acquirer. In order to become profitable a firm is likely to possess valuable resources and knowledge. Consequently, this is likely to be transferred to the target in Newco. So this paper expects that the more profitable the acquirer, the more likely the chance that the acquirer is capable to improve the profits of the target in Newco. As a result our second hypothesis is as follows:
H2: Shareholders of the acquirer react positively to pre-deal profitability of the acquirer.
This paper differentiates between public and private targets. With respect to room for improvement we expect that this is generally bigger for private targets as for public targets. This is due to the fact that in general public firms are more professionalized than private targets. In the literature on IPO-related change it is also reported that private firms going public experience organizational changes. According to von Eije et. al. (2004) these organizational changes are mainly in financial reporting and financial management. In addition to this they found that when those IPO-related changes were reported, the long term stock market performance was on average higher than when these changes were not apparent. Hence, IPO-related changes can enhance the performance of firms. In the acquisition context of this paper this has two major implications. Firstly, it can be argued that private firms are less professionalized and once they are acquired by a listed firm this is likely to improve the performance of the target in Newco. Secondly, this also suggests that for poorly performing public firms it is harder to improve the performance since they are already reasonably professionalized. Hence, we expect that shareholders differentiate between the different types of targets as well. Since private targets are in general less professionalized as their listed equivalents it could be that the room for improvement is bigger for private targets. This yields the following hypothesis:
Lastly, this paper examines the influence of the relative profitability of the acquirer and target. This measure gives an indication of the profitability of the acquirer relative to the profitability of the target. For this measure of pre-deal profitability the following is hypothesized:
H4: Shareholders of the acquirer react positively to targets which are relatively small in profitability compared to the acquirer’s profitability.
2.5 Other major determinants
Besides relative pre-deal profitability and the type of target (public or private) more determinants for shareholders reactions are indicated in the literature.
2.513 Relative deal size
The relative deal size is used as a regression variable in the cross-section analyses. Deal value is the amount of money paid by the bidder in order to acquire the target. Hansen and Lott (1996) control for the deal value in their paper by using the absolute dollar amount paid for the target. They report a small significant negative coefficient implying that the larger the amount of money paid, the smaller the CARs surrounding the acquisition announcement. Von Eije and Wiegerinck (2010) incorporated a relative measure of the deal value in their paper, by dividing the deal value by the stock market value of the acquirer. They find significant positive coefficients for U.S. targets which is contrary to the results of Hansen and Lott (1996). A possible explanation for this is according to von Eije and Wiegerinck (2010) that hubris is on average not an issue for European acquirers. Which suggest that shareholders are convinced of the good intentions of the management with respect to the price paid for the acquisition of the target.
In accordance with the pre-deal profitability variable we normalize the deal value using both the total assets of acquirer and target (Newco perspective). So the deal value in absolute euro’s amounts is divided by the last reported total assets of acquirer and target.
2.5.2 Acquirer Size
therefore used as proxy of acquirer size. Acquirer size is incorporated because Moeller et. al. (2003) reports that larger firms are more likely to overpay for their targets and in general overestimate the synergies. Hence, shareholders can be aware of the so-called “size-effect” and could incorporate the acquirer size in their reaction to acquisition announcements.
2.5.3 Method of payment
The implications and possible explanations of the effects of paying with cash rather than stock have been discussed by Myers and Maljuf (1984). According to them paying with stock is more likely in transactions when the board of the firm believes its stocks are overvalued. However, targets are familiar with this strategy and rather do not accept stock offers. Previous literature assessed the method of payment as a possible determinant explaining CARs.
According to Chang (1998) the method of payment does effect the abnormal returns, differentiating between stock and cash offers, only significant positive results are found for stock offers. This could be explained by the theory of Myers and Maljuf (1984) that paying with stock is more likely when the stocks are overvalued. In addition to this Fuller et. al. (2001) also examined the influence of the method of payment on the abnormal returns. They report significantly positive CARs regardless of method of payment for both private targets and subsidiaries. However, for public targets only significantly negative CARs are found for the stock offers (Fuller et. al., 2001).
Faccio et. al. (2006) controlled in their study for method of payments as well. Contrary to Chang (1998) and Fuller et. al. (2001) they report both significant results for stock and cash offers. It is found that paying with cash influences the CARs to shareholders of the acquirer positively, whereas paying with stock has a negative effect. This implies that the findings in the literature are mixed for the influence of the method of payment on the CARs. Hence this paper incorporates the method of payment to check its impact on the CARs to acquisition announcements.
2.5.4 Cross-border acquisitions
In the cross-section analyses this paper controls for cross-border acquisitions. According to Doukas and Travlos (1988) shareholders of acquirers experience significant positive abnormal returns when an international acquisition is announced. In line with these results Eun et. al. (1996) report positive combined wealth gains to international acquisitions, which provides evidence for the synergy hypothesis.
cross-border acquisitions. According to Conn et. al. (2003) the main difference between public and private acquisitions is that glamour acquirers experience negative announcement and long run returns to acquisition of public targets. However, for private targets this does not appear to be the case. 2.5.5 Diversification or focus strategy
It is reported in the literature that diversified firms trade on average at discount compared to focussed firms (Berger and Ofek, 1995; Serveas, 1996). Also in the acquisition context it is found that diversification destroys value. Dos Santos et. al. (2008) report that unrelated cross-border acquisitions lead discounts in the value of the post-acquisitions established firm. This suggest that deviation from “core business” destroys value. However, according to Campa and Kedia (2002) who used alternative econometric models to control for the endogeneity of diversifications decision the discount drops or even becomes a premium. Consequently, due to the mixed findings this paper incorporates the diversification variable as a dummy variable in the regression analysis.
2.5.6 Acquired stake
Lastly, this paper controls for the acquired stake in the acquisition announcement. This could proxy the amount of control the acquirer gains in the target after the transaction is completed. Therefore this could be an important variable in acquisition context. For this a dummy variable is used, which differentiates between majority and minority acquisitions.
3. Data and Methodology
3.1 Data selection
The dataset used in this paper consists solely of firms within Europe that were involved in an acquisition announcement from 2000 until 2012. The announcement dates, as well as other characteristics of the acquirer and target were obtained from the Zephyr Mergers & Acquisitions database of Bureau van Dijk. Furthermore, for the purpose of the event study the acquirer had to be listed in Europe and the target should also be a European firm. Hence, the returns of the acquirer who announced the acquisitions had to be available around the acquisition announcement. The prerequisite for this was 103 trading days before the acquisition announcement until two trading days after the acquisition announcement. Moreover, the returns for the benchmarks used to calculate the abnormal returns had to be available during the same period. In case these data were not available the acquisition announcements were removed from the dataset.
should at least be €5 million. This requirement is set in order to be sure that the acquisition is considered by the shareholders as substantial. This number is based on Faccio et. al. (2006). In case when this value was unknown, the estimated deal value recorded in Zephyr is assumed to be the most reliable estimate. Secondly, the announcement day had to be equal to the rumour date. The reason for this is explained below in the methodology section. Furthermore, information on the acquired stake, the two-digit US SIC code and the method of payment had to be available.
The final dataset used in the event study consists of 817 acquisitions. In total 53 acquisitions announcements involve public firms as target and 764 are acquisition announcements of private targets. For the cross-section analyses some additional information of the acquirer and target is required. This includes the last pre-acquisition reported total assets of both parties. In addition to total assets, pre-deal profitability of targets and acquirers had to be available in Zephyr.
3.2 Descriptive statistics
Table 1 presents the characteristics of the sample used in the event study. From Panel A it is immediately evident that the majority of the acquisitions involves acquisitions of private targets. These 764 acquisitions by far surpass the 53 acquisition announcements of public targets. It follows from Panel A that most acquisitions of public targets are announced in Italy (22.6%). Whereas for private targets the most acquisitions are announced in the United Kingdom, namely 38.7%. Furthermore, it may be noted that the majority of the countries selected in our sample are located in Western-Europe. From Panel B it can be concluded that most acquisitions in this dataset were announced prior to the crisis which started in 2008. In 2007 the total amount of acquisitions announced in our dataset was 175, which equals 21.4% of the total amount of acquisition announcements in our sample.
The minimum deal value for incorporation in the dataset is set at €5 million. This is also evident from panel C which shows the characteristics of the deal values associated with the acquisitions. In addition to this, the pre-deal profitability of the targets is included in panel D. It follows that the average deal value for acquisitions of public targets is approximately twice as high as for equivalent deals with private targets. In addition to this, the average pre-deal profitability of public targets is more than 40 times as high compared to the average pre-deal profitability of private targets. From panel E it follows that most acquisitions are paid with cash. This holds for both public as private targets. However, also for a substantial amount of acquisitions the method of payment is unknown.
industries targets were active in our sample, panel F reports the one-digit US SIC code and the corresponding number of targets belonging to that code.
Number Percentage Number Percentage
Panel A: Acquirer country
Austria 1 1.9% 3 0.4% Belgium 36 4.7% Denmark 7 0.9% Finland 32 4.2% France 11 20.8% 64 8.4% Germany 6 11.3% 34 4.5% Greece 1 1.9% 9 1.2% Ireland 7 0.9% Italy 12 22.6% 85 11.1% Norway 9 17.0% 24 3.1% Portugal 3 5.7% 6 0.8% Spain 4 7.5% 65 8.5% Sweden 1 1.9% 55 7.2% Swiss 1 1.9% 20 2.6% The Netherlands 20 2.6% Turkey 0.0% 1 0.1% United Kingdom 4 7.5% 296 38.7% Total 53 100.0% 764 100.0%
Panel B: Acquisition year
2000 2 0.3% 2002 2 0.3% 2003 1 1.9% 26 3.4% 2004 4 7.5% 45 5.9% 2005 5 9.4% 91 11.9% 2006 10 18.9% 122 16.0% 2007 13 24.5% 162 21.2% 2008 7 13.2% 100 13.1% 2009 3 5.7% 32 4.2% 2010 1 1.9% 61 8.0% 2011 4 7.5% 75 9.8% 2012 5 9.4% 46 6.0% Total 53 100.0% 764 100.0%
Panel C: Deal value (x €1000) N=53 N=764
M ean 184.552 90.989
M edian 20.300 19.600
M inimum 5.167 5.000
M aximum 2.377.676 11.800.000
Panel D: Pre-deal profitability (x€1000) N=53 N=764
M ean 64.425 1.523
M edian 7.554 538
M inimum -312.044 -1.489.515
M aximum 763.071 2.009.117
Table 1 - Dataset
Characteristics of dataset used in event study.
Number Percentage Number Percentage Panel E: M ethod of payment
Cash 33 62.3% 301 39.4% Converted Debt 2 0.3% Debt assumed 1 1.9% 44 5.8% Deferred payment 25 3.3% Earn-out 60 7.9% Loan notes 4 0.5% Other 1 1.9% 4 0.5% Shares 4 7.5% 104 13.6% Unkown 14 26.4% 220 28.8% 53 100.0% 764 100.0%
Panel F: Target industry (First US SIC code)
0 (Agriculture) 3 0.4% 1 (M ining/Construction) 39 5.1% 2 (M anufacturing US SIC 2) 6 11.3% 73 9.6% 3 (M anufacturing US SIC 3) 12 22.6% 143 18.7% 4 (Transportation) 9 17.0% 56 7.3% 5 (Trade) 2 3.8% 74 9.7%
6 (Finance, insurance, real estate) 12 22.6% 72 9.4%
7 (Services) 11 20.8% 195 25.5%
8 (Health services) 1 1.9% 104 13.6%
9 (Public administration) 5 0.7%
Total 53 100.0% 764 100.0%
Panel G: Acquired Stake
0-10% 25 47.2% 0.0% 11-20% 5 9.4% 0.0% 21-30% 3 5.7% 0.0% 31-40% 1 1.9% 0.0% 41-50% 2 3.8% 0.0% 51-60% 3 5.7% 39 5.1% 61-70% 4 7.5% 14 1.8% 71-80% 3 5.7% 24 3.1% 81-90% 0.0% 10 1.3% 91-100% 4 7.5% 677 88.6% Unkown 3 5.7% 53 100.0% 764 100.0% Table 1 (Continued)
T he table present the characteristics of the 817 acquisition annoucements by European acquirers annoucing an acquistion of a European target during the period 2000 until 2012.
3.3 Event study methodology
In order to examine the stock market reactions to acquisition announcements we use the traditional event study methodology. In accordance with earlier research, we calculate the CARs as measure of shareholders reaction to acquisition announcements. As mentioned above, we solely concentrate on the reaction of shareholders of the acquirer. Rather than using the day on which the acquisition is completed, this paper uses the announcement day as the event day. In case the event day was not on a trading day, the first trading day after the announcement is considered as the event day.
Moreover, it is important that no prior leakage of information on the acquisition has become available to the market. Since the event study methodology is built upon the efficient markets hypothesis (Fama, 1970). Which argues that prices at any point in time reflect all the available information. Hence, leakage of information prior to the announcement of acquisitions could already be incorporated in the prices and therefore could bias the event study results. In order to avoid such biases the rumour date should be equal to the announcement date in our dataset.
The event day is considered as day and the event window used in this paper is start two
days prior ( ) to the event day and ends two days afterwards ( ). This means that the CARs are
calculated over an event window of five days ranging from [-2;2]. Before calculating the CARs the abnormal returns of all events have to be calculated. To accomplish this, the returns in the event windows have to be adjusted with the normal returns associated with the acquirer i. Therefore, this paper estimates the normal returns using an estimation window of 100 trading days (starting on
until ).
This paper makes use of three different models based on Brown and Warner (1980, 1985): i) the Mean Adjusted Returns model, ii) the Market Adjusted Returns model and iii) the Market and Risk Adjusted model. These models are the benchmarks which can be used to calculate the normal returns. For models ii and iii local indexes and their characteristics are used to determine the expected returns of the acquirer’s stock. These local indexes are the largest local indexes of the country in which the firm is established. This does not automatically imply that the acquirer is incorporated in this index since the acquirer can also be a Midcap stock or even Smallcap stock. However, this paper assumes that these local indexes are the most accurate estimates of normal country returns.
The actual returns, , adjusted for any dividends paid during period , , in the estimation
and event window surrounding event i are calculated with formula (1):
(1)
Where is the price at trading day , and is the price at the previous trading day and the
The Mean Adjusted model calculates the abnormal returns by subtracting the mean of the estimation period returns of firm i ̅ from the actual returns. Hence the abnormal return in the event window is the difference between actual return during the event window minus the average return during the estimation window.
̅ (2)
Whereas the Mean Adjusted model uses the simplifying assumption that the normal returns are the average past returns during the estimation window, the Market Adjusted Model makes a simplifying assumption as well. It assumes that the normal returns can best be estimated as the return on the market, . This results in the following abnormal return calculation formula:
(3)
The Market and Risk Adjusted model is based on the Market model and uses the following equation:
(4)
is the return on the acquirer’s stocks, which announces acquisition at trading day ,
is the return on the local index at trading day , is the intercept of the Market model. is the systematic risk of the bidders’ security and is the error term. The parameters and are estimated over the estimation window.
Since this paper wants to generalize its conclusions average abnormal returns (AARs) are calculated with the following equation for all N acquisitions on day :
∑ (5)
Finally, the cumulative abnormal returns (CARs) are calculated by summing up the during the event window. These CARs are especially useful since they capture the overall stock market reaction during the whole event window. Equation 6 shows how the CARs are calculated for each acquisition :
For till (where is the event date and is similar to )
Finally, all the can be averaged over all N acquisitions:
∑ (7)
3.4 Cross-section analyses
3.4.1 Pre-deal profitability measures
In order to answer the hypotheses of this paper we use multiple cross-section regression analyses. Firstly, this section will introduce the various measures of pre-deal profitability that are included in the different cross-section analyses.
In order to test our first hypothesis (H1) the following measure of pre-deal profitability is used:
(8)
Where is the profit-after-tax of the target, are the total assets of the acquirer and
are the total assets of the target. The denominator takes the Newco perspective by
normalizing the pre-deal profit of the target by the total assets of acquirer and target, the Newco situation. The first hypothesis will be confirmed if a negative and significant coefficient is found for equation 8. This implies that the lower the pre-deal profitability of the target, the higher the CARs and would provide evidence that shareholders react positively to targets which have room for improvement.
To test the influence of the acquirer pre-deal profitability (H2) on the CARs the following measure is used:
(9)
Where denotes the pre-deal profit-after-tax of the acquirer. Hypothesis 2 will be confirmed
Besides that we test target and acquirer pre-deal profitability in isolation we also use them together in the regression analysis. This is done in order to test our fourth hypothesis (H4) were we use a relative pre-deal profitability measure:
(10)
The last hypothesis (H4) will be confirmed when a significant positive coefficient is reported for the relative pre-deal profitability (equation 10). This would imply that shareholders of the acquirer react positively to targets which are relatively small in pre-deal profitability compared to the acquirers profitability. Moreover, it would provide evidence that shareholders of the acquirer prefer targets which have room for improvement.
Lastly, we use an interaction dummy to check whether pre-deal profitability is especially important for private targets. If the results report a significantly positive coefficient for this interaction dummy than we find results that confirm the third hypothesis (H3). This would imply that shareholders of the acquirer react more positive to private targets than to public targets if both have room for improvement.
For the robustness checks the profit-after-tax symbols, , in equations 8, 9 and 10 are changed for EBITDA.
3.4.2 Regression equations
The following equations are estimated using the Ordinary Least Squares (OLS) regression analysis technique in order to test our hypotheses.
Regression equation A is used to test our first hypothesis (H1): Regression equation A (Target profitability):
Regression equation B is used to test our second hypothesis (H2): Regression equation B (Acquirer profitability):
Regression equation C is used to test our third hypothesis (H3): Regression equation C (Target and Acquirer profitability:
The last regression equation, D, is used to test our fourth hypothesis (H4): Regression equation D (relative pre-deal profitability):
Where is the cumulative abnormal return of the listed acquirer in the event window,
which is calculated according to the three different methods, i) Mean Adjusted ii) Market Adjusted and iii) Market and Risk Adjusted (Brown and Warner 1980, 1985).
The measures to proxy pre-deal profitability of both target (TARGETPROF) and acquirer (ACQUIRERPROF) are discussed above (equation 8 and 9). Furthermore, is also describe above and is estimated by the profit-after-tax of both acquirer and target divided by total assets bidder and target (equation 10). is the relative deal size, calculated by dividing the deal value of the acquisition announcement i by total assets of both bidder and target. is variable which proxies the size of the acquirer and is estimated by the log of acquirer total assets. The log is used in order to normalize the variable.
3.4.3 Multicollinearity
In order to test for multicollinearity two correlation matrixes are provided in table 12 in Appendix F to investigate whether high correlations between the independent variables used in the regression equations above exist. As is evident from both matrixes that none of the major variables are highly correlated. Only the dummy variables that PRIVATE and STAKE report reasonable high correlations. This is due to the fact that all the acquisitions of private targets involve an acquired stake that exceeds 50%. Hence, results for these coefficients have to be interpreted with caution.
4. Results
4.1 Event study results
This section will discuss the results of the event study conducted. First it reports the results of the whole sample. Thereafter, the results of the public and private targets will be discussed separately.
4.1.1 Event study results total sample
Table 2 reports the CARs for the whole sample and for public and private targets according to the Market and Risk Adjusted model (equation 4). These CARs are calculated during the event window starting two days prior to the event day until two days afterwards. The CAARs are the means of the individual CARs. It follows that the CAAR to acquisition announcements for the whole sample is 0.8%. In order to test the significance of this CAAR both the T-test (parametric distribution) and the Wilcoxon Signed Rank test (non-parametric distribution) are used. It is evident that this CAAR is significant at 1% level for both methods, which implies that shareholders of the acquirer react
positively to announcements of acquisitions in the five days event window surrounding the acquisition announcements.
Table 2 - Event study resulst calculated by the Market and Risk Adjusted model
Total Sample Public Targets Private Targets
N 817 53 764 M ean (CAAR) 0.008 0.012 0.007 M inimum -0.505 -0.206 -0.505 M aximum 0.648 0.244 0.648 Standard Deviation 0.082 0.066 0.083 P-value T-test 0.009 0.207 0.016
P-value Wilcoxon Signed Rank test 0.002 0.269 0.003
4.1.2 Public and private targets
In order to investigate whether shareholders differentiate in their stock market reaction between the type of target, either public or private, the sample is split based on the type of target. Column 2 in table 2 reports the CAAR for the public targets and column 3 reports the CAAR for private targets. It appears that the CAAR for public targets is 1.2% which implies that shareholders react positive to an acquisition announcement during the surrounding event window which results in a 1.2% cumulative increase in stock market value. However, according to both the T-test and the Wilcoxon Signed Rank test we cannot reject the null hypothesis that the CAAR is significantly different from zero. Therefore, we cannot conclude that shareholders react positively to acquisition announcements of public targets.
This result is in line with results of previous studies done by scholars based on a sample of U.S. acquirers and targets (Chaterjee, 1986; Lubatkin, 1987; Fuller et. al., 2001) who report zero or negative CAARs to acquisition announcements of public targets.
Column 3 of table 2 reports the CAAR for acquisition announcements of private targets according to the Market and Risk Adjusted model (equation 4) during the specified event window. It appears that the CAAR for acquisitions of private targets, which equal 0.7% in our sample, is lower than for public targets. However, what is noteworthy is that the CAAR of 0.7% is highly significant. This implies that for the T-test we can reject, at 2% level, the null hypothesis that the CAARS are not significantly different from zero. Moreover, according to the Wilcoxon Signed Rank test we can reject the null hypothesis at a 1% level. Hence, it follows that shareholders react positively to acquisition announcements of private targets. Moreover, this is in line with studies done by previous scholars on the difference in stock market reaction to acquisition announcements of different types (public or private) of targets (Hansen and Lott, 1996; Chang, 1998; Fuller et. al., 2001; Capron and Shen, 2007).
4.2 Cross-section analyses
The CARs calculated based on the sample of 817 acquisitions are used as dependent variable to run the cross-section analyses. Due to the additional information necessary as input for the regression analyses, the outputs of the regression analysis may contain less observations.
4.2.1 Pre-deal profitability of the target for the total sample
Table 3 reports the results of the cross-section analysis which uses regression equation A. The coefficients of the independent variables are shown in table 3 in columns 1 until 3 for respectively the whole sample, the public targets and the private targets. The dependent variable, CARs, are calculated using the Market and Risk Adjusted model.
Firstly, this subsection discusses the results for the whole sample. Most important result of regression A is the negative and significant (at 5% level) coefficient for the pre-deal profitability of the target (TARGETPROF). This implies the lower the pre-deal profitability of the target, the higher the CARs. Hence, this negative and significant coefficient provides evidence for our first hypothesis that shareholders react positively to targets which have considerable room for improvement (H1). This room for improvement has a positive effect on the CARs, which could be due to the expected learning and synergies in Newco, which could ultimately results in higher post-acquisition profits. This is in accordance with the shareholder value maximization of Jensen (2000) and therefore valued by shareholders. Consequently, this could explain the negative and significant sign. Another explanation could be that the deal value associated with an acquisition is likely to be lower for a low profitable firm than for more profitable firm. However, due to the incorporation of the relative deal value (RDVAL) variable we control for this explanation. Since this variable reports a positive sign in table 3, column 1 it is not likely that the second explanation holds.
In addition to the target pre-deal profitability, the acquirer size (SIZE) does at a 5% level have a significant negative effect on the CARs in the event window as is evident from table 3. The negative sign of the coefficient provides evidence for the so-called size effect reported by Moeller et. al. (2003, p.226). This size-effect argues that “larger firms pay higher acquisition premiums and enter acquisitions with negative dollar synergy gains.”. The reported results in our study suggest that shareholders are aware of this size-effect and react negatively to acquisition the larger the acquirer size.
Furthermore, the dummy variables that indicate whether the target is public or private (PRIVATE) and whether the acquisition is cross-border (CROSS) or not report significant results for the whole sample. Firstly, the negative and significant at a 10% level private variable indicates that the CARs are smaller for acquisitions of private targets. This is in line with the results of the event study reported above. However, it should be noted that although the CARs to acquisition announcements of public targets are higher, we do not have evidence to reject the null hypothesis that they are significantly different from zero.
4.2.2 Public versus private targets and their pre-deal profitability
Columns 2 and 3 of table 3 report the results for the public and private targets. It is evident from table 3 that for the public targets (column 2) none of the independent variables report significant results. Moreover, as already mentioned above, the F-test does not report a significant results either. This implies that regression equation A is not capable to explain any of the variance in the CARs to
(1) Total
Sample (2) Public (3) Private
TARGETPROF -0.285** -0.384 -0.277* (0.131) (0.253) (0.142) RDVAL 0.025 0.011 0.025 (0.015) (0.164) (0.016) SIZE -0.004** -0.004 -0.004** (0.002) (0.004) (0.002) PRIVATE -0.028* (0.015) PAYM ENT -0.002 -0.017 -0.001 (0.006) (0.019) (0.006) CROSS 0.009* -0.006 0.011* (0.005) (0.021) (0.006) DIVERS 0.005 0.012 0.003 (0.006) (0.019) (0.007) STAKE 0.020 0.020 ¹ (0.018) (0.027) Constant 0.058** 0.067 0.049* (0.028) (0.060) (0.027) Number of observations 815 52 763 Adjusted R-sqaured 0.068 0.103 0.066 F-test 8.448 1.839 9.963 p-Value of F-test 0.000 0.103 0.000
Table 3 - Target pre-deal profitability
Test on the cumulative abnormal returns calculated by the M arket and Risk Adjusted model of European acquisitions with target pre-deal profitability as major independent variable.
Cumulative abnormal returns
The de pe nde nt va ria ble s a re the C AR s c a lc ula te d fo r a n e ve nt windo w o f 5 da ys ba s e d o n the M a rke t a nd R is k Adjus te d m e tho d fo r e ve nt
i fo r whic h da ta wa s a va ila ble o n a ll va ria ble s . TAR GETP R OF is the pre -de a l pro fita bility o f the ta rge t m e a s ure d by la s t-re po rte d pro
acquisition announcements of public targets.
Contrary to this, the private targets sample (column 3) does report significant results. Firstly, we find evidence that shareholders value room for improvement for private targets. This is evident from the negative and significant at a 10% level coefficient for target pre-deal profitability. The difference between the negative significant coefficient for private targets and the non-significant coefficient for public targets is noteworthy. It could either be due to the limited amount of observations of acquisition announcements of public target or due to the fact that shareholders do not have different stock market reactions to pre-deal profitability if the target is listed. The second explanation could possibly imply that shareholders do not believe that acquisitions of public targets yield higher future profits due to learning or synergies. In general a listed firm does already have a professional organization with a decent governance structure. Moreover, the listed firm has already undergone IPO-related changes (von Eije et. al., 2004) and therefore less improvements might be achievable compared to private targets. Hence, acquiring a public target could imply less room for improvement. As a result the expected post-acquisition gains in terms of profits could be less than for less professionalized (private) firms.
In addition to the target pre-deal profitability, the acquirer size (SIZE) and cross-border (CROSS) variables report significant results for private targets. Firstly, the positive and significant at 5% level acquirer size variable reports evidence for the above mentioned size effect. The non-significant acquirer size variable for public target suggests that shareholders perceive that especially acquisitions of private targets are likely to be sensitive to the size effect.
Lastly, the cross-border variable reports a positive and significant result at a 10% level for acquisitions of foreign private targets. This implies that cross-border acquisitions of private targets outperform those of domestic targets. This contrasts the findings of Conn et. al. (2003) who report higher announcement returns to acquisitions of domestic private targets compared to acquisitions of cross-border private targets.
4.3.1 Pre-deal profitability of the acquirer for the total sample
Table 4 reports the results for regression equation B which tests our second hypothesis whether shareholders of the acquirer react positively to the pre-deal profitability of the acquirer (H2). The dependent variables are the CARs calculated using the Market and Risk Adjusted model. The Adjusted R-squared range from 3.4% to 14.1% for respectively the total sample and the public targets. Moreover, the F-tests for the total sample (column 1) and the private targets (column 3) are significant at a 1% level whereas the public targets (column 2) is significant at a 10% level. This subsection will discuss the cross-section analysis results for the total sample.
second hypothesis (H2). It was hypothesized that the more profitable the acquirer is, the bigger the chance that the acquirer is capable to improve the performance and in the end the profits of the target in Newco. Since shareholders of acquirers appear to value room for improvement in targets, this could be a plausible explanation. However, no evidence is provided due to the non-significant coefficients. An explanation for this occurrence could be that shareholders are already convinced of the quality of the acquirer. It is conceivable that they otherwise would not have become shareholder in the first place. This could explain the fact that acquirer pre-deal profitability is not perceived by shareholders as strong determining variable.
4.3.2 Public versus private targets and acquirer pre-deal profitability
In addition to the results for the total sample, columns 2 and 3 of table 4 report the results of the regression equation B by making the distinction between the types of targets. Firstly, the results for the public targets will be discussed.
For the public targets the only significant variable is the pre-deal profitability of the acquirer. This variable reports a large negative, at 10% level significant, coefficient. This contrasts our second
(1) Total
Sample (2) Public (3) Private
ACQUIRERPROF 0.021 -0.421* 0.027 (0.037) (0.180) (0.037) RDVAL 0.025 -0.169 0.026 (0.017) (0.226) (0.018) SIZE -0.004** -0.007 -0.004* (0.002) (0.005) (0.002) PRIVATE -0.038* (0.020) PAYM ENT -0.004 -0.012 -0.003 (0.006) (0.020) (0.006) CROSS 0.010* -0.005 0.012* (0.006) (0.023) (0.006) DIVERS 0.003 0.027 0.002 (0.006) (0.020) (0.007) STAKE 0.030 0.033 ² (0.022) (0.034) Constant 0.059** 0.112 0.048* (0.030) (0.072) (0.028) Number of observations 813 50 763 Adjusted R-sqaured 0.034 0.141 0.035 F-test 4.619 2.149 5.562 p-Value of F-test 0.000 0.059 0.000
Table 4 - Acquirer pre-deal profitability
Test on the cumulative abnormal returns calculated by the M arket and Risk Adjusted model of European acquisitions with acquirer pre-deal profitability as major independent variable.
Cumulative abnormal returns
The de pe nde nt va ria ble s a re the C AR s c a lc ula te d fo r a n e ve nt windo w o f 5 da ys ba s e d o n the M a rke t a nd R is k Adjus te d m e tho d fo r e ve nt
i fo r whic h da ta wa s a va ila ble o n a ll va ria ble s . AC QUIR ER P R OF is the pre -de a l pro fita bility o f the a c quire r m e a s ure d by la s t-re po rte d pro
hypothesis (H2) which argues that shareholders of the acquirer react positively to pre-deal profitability of the target. Hence, from this it appears that shareholders react negatively to acquisition announcements of public targets when their firm is relatively high in pre-deal profitability. A conceivable explanation for this relative large negative coefficient could be that shareholders associate acquisitions of public targets with hubris. According to Roll (1986) this implies that acquirers simply pay too much for their targets. Since paying too much is more convenient when a firm “has much” this a conceivable explanation.
In addition to this the relative deal value variable (RDVAL) reports also a negative coefficient. Although this coefficient is not significant, it does indicate the appearance of hubris. The negative coefficient suggests that the higher the relative deal value, the lower the CARs to acquisition announcements of public targets. This is in line with Roll’s hubris hypothesis.
For private targets (column 3) our major variable in regression equation B gives a positive coefficient. Since it is not significant it gives no reason for accepting our second hypothesis (H2) for acquisition announcements of private targets. As already indicated above, this could imply that shareholders are already convinced of the quality of the acquirer and therefore do not perceive acquirer pre-deal profitability as strong determining variable.
Furthermore, for private targets the acquirer size (SIZE) variable and the cross-border acquisitions (CROSS) variable report significant results. The acquirer size variable is significantly negative at a 10% level for acquisition announcements of private targets. In addition to this the cross-border acquisition announcements of private targets report a positive coefficient at a 10% significance level as well. Those results are in line with the results for the total sample, and with the results of regression equation A for targets pre-deal profitability.
4.4.1 Pre-deal profitability of the target and acquirer for the total sample
Besides that pre-deal profitability of the targets is an important variable for shareholders, acquirer pre-deal profitability appears also to be an important variable. From table 5 column 1 it is evident that shareholders react positively to acquirer pre-deal profitability. This contrasts the findings for regression equation B in table 4 which could indicate that shareholders do not look at acquirer pre-deal profitability in isolation. This suggest that shareholders rather look to the Newco situation and hence incorporate both target- and acquirer pre-deal profitability. The positive and at a 5% level significant coefficient for acquirer pre-deal profitability gives evidence for our second hypothesis (H2). Hence, it suggests that the profitability of the acquirer is an indication of the possible improvements of the target in the Newco situation.
Besides that the pre-deal profitability variables report significant results, also the relative deal value (RDVAL) reports a significantly positive coefficient (at a 5% level). This indicates that shareholders react positively to acquisitions announcements the higher the relative deal value. This might sound somewhat contradicting at first hand. However, it indicates that hubris is on average not an issue in the perception of shareholders. Moreover, this could indicate that shareholders realize the amount of effort and time that the pre-acquisition preparations require. Hence, when the acquisition is announced it appears that all these efforts and associated costs paid-off which otherwise would have been sunk cost.
(1) Total
Sample (2) Public (3) Private
(4) Total Sample TARGETPROF -0.312** -0.179 -0.305** -0.271 (0.129) (0.246) (0.139) (0.194) ACQUIRERPROF 0.057** -0.340* 0.060** -0.235* (0.029) (0.194) (0.029) (0.138) PRIVATE*TARGET PROF -0.035 (0.240) PRIVATE*ACQUIRER PROF 0.295** (0.141) RDVAL 0.030** -0.166 0.030* 0.030** (0.015) (0.180) (0.015) (0.015) SIZE -0.004** -0.006 -0.004** -0.004** (0.002) (0.004) (0.002) (0.002) PRIVATE -0.033** -0.032** (0.016) (0.015) PAYM ENT -0.004 -0.010 -0.002 -0.003 (0.006) (0.019) (0.006) (0.006) CROSS 0.010* -0.007 0.011* 0.010* (0.006) (0.023) (0.006) (0.006) DIVERS 0.005 0.022 0.003 0.005 (0.006) (0.021) (0.007) (0.006) STAKE 0.023 0.034 ³ 0.015 (0.019) (0.032) (0.018) Constant 0.061** 0.102 0.050* 0.068** (0.028) (0.066) (0.026) (0.029) Number of observations 813 50 763 813 Adjusted R-sqaured 0.073 0.131 0.071 0.073 F-test 8.073 1.925 9.317 6.790 p-Value of F-test 0.000 0.082 0.000 0.000
The de pe nde nt va ria ble s a re the C AR s c a lc ula te d fo r a n e ve nt windo w o f 5 da ys ba s e d o n the M a rke t a nd R is k Adjus te d m e tho d fo r e ve nt
i fo r whic h da ta wa s a va ila ble o n a ll va ria ble s . TAR GETP R OF is the pre -de a l pro fita bility o f the ta rge t m e a s ure d by la s t re po rte d pro
fit-a fte r-tfit-a x divide d by to tfit-a l fit-a s s e ts o f bo th fit-a c quire r fit-a nd tfit-a rge t (e qufit-a tio n 8) AC QUIR ER P R OF is the pre -de fit-a l pro fitfit-a bility o f the fit-a c quire r m e a s ure d by la s t-re po rte d pro fit-a fte r-ta x divide d by to ta l a s s e ts o f bo th a c quire r a nd ta rge t (e qua tio n 9). P R IVATE*TAR GETP R OF a nd P R IVATE*AC QUIR ER P R OF a re inte ra c tio n dum m ie s us e d to te s t hypo the s is 3. R DVAL is the de a l va lue divide d by la s t e x-a nte re po rte d to ta l a s s e ts o f bo th a c quire r a nd ta rge t (Ne wc o pe rs pe c tive ). S IZE m e a s ure s the s ize o f the a c quire r a nd is e s tim a te d by the lo g o f its to ta l a s s e ts . P R IVATE e qua ls 1if the ta rge t is a priva te firm a nd 0 if the ta rge t is public firm . P AYM ENT ha s a va lue o f 1is the a c quis itio n is pa id with c a s h, o the rwis e 0. C R OS S ha s a va lue o f 1if the ta rge t is lo c a te d a c ro s s the bo rde r, o the rwis e 0. DIVER S e qua ls 1 if ta rge t is no t in the s a m e indus try a s a c quire r, o the rwis e 0. S TAKE ha s a va lue 1 if the a c quire d s ta ke in the a c quis itio n e xc e e ds 50%, o the rwis e 0 (³ S TAKE is o nly inc lude d fo r public ta rge ts s inc e fo r a ll priva te ta rge ts a c quire d s ta ke e xc e e ds 50%). C o ns ta nt is the c o ns ta nt te rm o f the e qua tio n. White he te ro s c e da s tic ity-c o ns is te nt S ta nda rd Erro rs a re be twe e n the pa re nthe s e s . S ignifa nc e s te s t a re pe rfo rm e d us ing two -s ide d t-te s ts , whe re * de no te s p< 0.1, ** de no te s p< 0.05.
Test on the cumulative abnormal returns calculated by the M arket and Risk Adjusted model of European acquisitions with both target and acquirer pre-deal profitability as major independent variable.
4.4.2 Public versus private target: target and acquirer pre-deal profitability
Columns 2 and 3 of table 5 report the results of regression equation C for public and private targets. For public targets (column 2) the only variable which is significant is acquirer pre-deal profitability. The coefficient is negative and significant at a 10% level which implies that shareholders react negatively to acquisition announcements of public targets when their firm is relatively high in pre-deal profitability. This variable together with the negative coefficient for relative deal value (RDVAL) provides again evidence for hubris. Moreover, this contrasts our second hypothesis (H2) for public targets which suggested positive reactions to acquirer pre-deal profitability.
For private targets (table 5, column 3) evidence is found in support of both our first hypothesis (H1) and second hypothesis (H2). It appears that shareholders react positively to targets which have room for improvement which is evident by the negative and significant at a 5% level coefficient for target pre-deal profitability. This suggests that shareholders of the acquirer especially perceive that private targets have room for improvement. In some way this corresponds with the findings of von Eije et. al. (2004) that private targets professionalize in preparations for IPO’s. Hence, when the acquisition is announced they become part of a listed firm and can benefit of the public acquirer. In accordance with this also the acquirer pre-deal profitability variable (ACQUIRERPROF) reports a positive and at a 5% level significant coefficient for acquisition announcements of private targets. The explanation for this, as already mentioned above, is that acquirer pre-deal profitability could signal the quality of the acquirer. It is interesting that this variable is positive and significant for private targets whereas for public targets this variable reports a negative and significant coefficient (table 5, column 2). This suggests that the quality of the acquirer is especially important in the context of acquisition announcements of private targets. Taken together the fact that private targets are generally less professionalized as public targets and the reported findings of von Eije et. al.(2004) on IPO-related changes for private targets, this is a conceivable explanation.
In addition to this, it appears that for acquisition announcements of private targets hubris is not an issue. This is evident from the positive and at 10% level significant relative deal value variable (RDVAL). The positive and significant sign indicates that shareholders realize the amount of effort and time that is accompanied with the pre-acquisition preparations. Hence, when the acquisition is announced it implies that all effort paid off. Moreover, this is in line with the findings of von Eije and Wiegerinck (2010) who also found that hubris is on average not an issue for European acquirers of private targets.
