1
The value of related acquisitions in the European construction industry
Abstract
_____________________________________________________________________________ This paper presents a two-sided view of acquisitions within the European construction industry as it studies the shareholders' response for acquirers from the construction industry as well as for the construction firms acquirers that are no from the construction industry. Moreover, I examine in particular whether related or unrelated acquisitions lead to divergent shareholder reactions. In addition to the regular literature I also incorporate a variable, whether acquiring firms are diversified or not. This reduces the omitted variable bias. This paper shows that firms not from the construction industry generate higher abnormal returns than firms within the construction industry. Furthermore, without the combination with whether the acquirer is focused or
diversified, no significant results can be found for the relatedness of the acquisitions. However, including this combination leads to some evidence that related acquisitions are better rewarded than unrelated acquisitions. Moreover, this paper finds that in particular using a long event window construction focused acquirers are more beneficial than construction diversified acquirers.
______________________________________________________________________________ Studentnr: s1994727
Name: Mark de Jong
Study Program: Msc Finance Supervisor: dr. J.H. von Eije JEL code: G34, L74
2
1. Introduction
Since the burst of the housing bubble, the European construction industry has to cope with difficult economic circumstances and low outputs (www.euroconstruct.org). Currently, there are rumors that the European construction market bottoms out (www.euroconstruct.org) and
subsequently that there will be opportunities to grow.
According to Koller et al. (2010) growth can i.a. be split up into three categories: 1) portfolio momentum; growth through expanding the market segments, which are currently in their portfolio. 2) Market share performance; increasing or decreasing market shares. 3) Mergers and Acquisitions (M&A); growth by acquiring firms or divesting firms. This thesis will focus on the last category and in particular on growth by acquiring firms and it focuses in particular on the M&A activity in the European construction industry. This industry consists of firms which are "primarily engaged in, the construction of buildings or engineering projects, the preparation of sites for new construction and subdividing land for sale as building sites." (www.naics.com). There are several determinants that drive takeover gains, this paper will focus especially on the distinction between a related (diversifying) acquisition or non-related (focused) acquisition. This effect will be gauged through an event study by calculating and comparing the abnormal returns of the acquirers.
This paper differs the existing literature in several aspects. Firstly to my best knowledge, there is no literature that includes the variable which gauges whether or not the acquirer itself is focused. By introducing such a variable I reduce the omitted variable bias. Secondly, it describes the specific situation of the European construction industry, while providing it with a two-sided view: I measure the shareholders' response of firms not from the construction industry which take over a construction industry target and the shareholders' response of firms from the construction industry regardless of the relatedness of their acquisition.
Previous papers discussing this subject show that unrelated acquisitions worsen firms and that related acquisitions strengthen firms (Servaes (1996); Kaplan and Weisbach (1992); Tasi (1994)). However, critics argue that in the construction example, the opposite holds true (Junnonen
(1998); Kim and Reinschmidt (2011); Delaney and Wamuziri (2004)). The latter state that related acquisitions worsen firms and unrelated acquisitions strengthen firms.
3
response, this is in line with Choi and Russell (2004 and 2005) who do not find a significant response between a related and an unrelated acquisition.
This paper does not find a significant effect of the relatedness of the acquisition. However, there is some evidence that related acquisitions are better rewarded than unrelated acquisitions. This evidence is found when the analyses are focused solely on the bidders from the construction industry and the combination of the relatedness of the acquisition and whether the acquirer is focused or diversified is taken into account.
Furthermore, there is evidence that especially for the long event window, the focused construction acquirers are better rewarded by the shareholders than diversified construction acquirers.
Moreover, this paper shows that bidders not from the construction industry generate higher abnormal returns when acquiring a firm from the construction industry than bidders within the construction industry regardless of the relatedness of their acquisition.
The next section gives an overview of the existing literature concerning this subject. Section 3 describes the data and the methodology used to calculate the abnormal returns. Section 4 presents the analyses results of this paper. This paper closes with a conclusion in Section 5. Tables are presented at the end of this paper.
2. Literature review
When and why do acquisitions occur?
Andrade et al. (2001) mention two features about acquisitions: 1)acquisitions come in waves and 2) acquisitions within these waves tend to cluster by industry. These features can be explained by the fact that acquisitions will occur in reaction to industry shocks (Andrede et al., 2001). Andrede et al. (2001) find that in the 1990s, industry deregulation was the main determinant of acquisition activity. Moreover, Maksimovic et al. (2009) find that acquisitions within these waves realize higher gains in productivity.
4
Determinants that drive takeover gains
Lang, Stulz and Walking (1989) find that abnormal returns have a relationship with Tobin's q. They find that the best takeovers with respect to value creating are the ones in which the bidder has a high q ratio and the target has a low q ratio. The takeover in which the situation concerning the q ratios is reversed are considered to be the worst takeovers. However, Servaes (1991) argues that there are more determinants that affect takeover gains, namely the characteristics of the takeover (friendly vs. hostile bidders, one vs. multiple bidders), the form of payment (cash or securities), the time period and the relative size of the target. If the q ratio is correlated with one of the previously mentioned determinants, the findings of Lang et al. (1989) are weakened. Servaes (1991) finds that this was not the case and the results of Lang et al. (1989) thus holds. Furthermore, Gorton et al. (2009) argue that the profitability of an acquisition is negatively related with the acquirer’s size.
Diversification; theoretical
Another well-established determinant in the literature is the difference between a related (focused) or unrelated (diversifying) acquisition. A unrelated acquisitions is defined by Morck, Shleifer, and Vishny (1990) as an acquisition where the bidder does not share the same three or four digit level of its SIC Code.
Servaes (1996) argues that there are two forces that drive diversification. Firstly it is beneficial for shareholders, if there are external capital market imperfections which lead to rejected positive NPV projects. Secondly it could diminish agency problems. For example, managers perform better if they act in other industries and this will make them more valuable for the company. The empirical evidence for the latter is mixed, Denis et al. (1997) show that the diversification
decision may also be positively related with agency problems. However, Hyland (1999) shows no empirical relationship between the diversification decision and agency problems.
5
Diversification discount
Serveas (1996) shows that diversified firms trade at a discount. This phenomenon is called as the diversification discount. Custodio (2014) gives possible explanations for this phenomenon: 1) agency costs. 2) the discount can arise from the self selection bias. Cempa and Kedia (2002) argue that the diversification strategy is negatively related to firm value. 3) endogeneity of diversification decisions. Cempa and Kedia (2002) show that diversifying firms have specific firm characteristics that forces them to diversify or 4) data or measurement issues. Custodio (2014) shows that the diversification is upward biased and doubts the reliability when using q ratios to estimate the diversification discount. However, despite the adjustments made by Custodio (2014) there still remains a (substantially lower) diversification discount.
Campa and Kedia (2002) come up with the following potential gains to diversification: 1) managerial economies of scales 2) increase of debt capacity 3) efficient resource allocation through internal capital markets 4) internalization of market failures 5) reduction of adverse selection problems at equity issuances 6) firm specific assets, which can be used in other markets 7) higher productivity. However, there are also potential costs to diversification: 1) inefficient allocation of capital among divisions 2) The costs of changing the reward system for managers, because they are responsible for multiple divisions. 3) information asymmetries between central management and division management 4) increased incentives for managers to seek rent and opportunities to start value destroying investments. If these benefits outweighs the costs the firms should diversify.
Rajan et al. (2000) splits the rationale for firms to diversify into three models: 1) Efficient Internal Capital Model, which argues that internal capital markets leads to an efficient allocation of rescources. 2) Agency cost models, where agency costs lead to overinvestment and this might explain internal misallocation of funds. 3) Influence cost models, in which lobbying is perceived as an important tool to attract resources.
Empirical evidence of differences between related and unrelated acquisitions
6
capabilities by the acquirer, it could be that that the target no longer adds value to the company and thus may it be wise to sell this company. Kaplan & Weisbach (1992) tackle this problem by examining whether these divestures create a gain or a loss on sale. They find that 60% of unrelated acquisitions were divested, where in the case of related acquisitions only 20% of the acquisitions were divested. Moreover, they show mixed evidence of these divestures. Kaplan & Weisbach (1992) define an unsuccessful acquisition as either a loss on sale or the announcement of management that the acquisition was a mistake. Firstly, they discover that 13% of the related acquisitions and 38 % of unrelated acquisitions were unsuccessful. Secondly, they argue that 40% of related acquisitions and 43% of the unrelated acquisitions were regarded as a gain on sale. Lastly, they do not find a significant difference in shareholder reactions between these two types of acquisitions
Serveas (1996) calculates Q ratios from individual segment firms and multiple segment firms. After comparing these ratios, Servaes (1996) finds evidence that unrelated acquisitions did not benefit U.S. firms. However, Serveas (1996) argues that it still may work for some countries and there is still a puzzle that diversification can be beneficial for some, but value destroying for others.
Construction industry
In order to understand whether construction firms benefit from diversification, the construction industry has to be elaborated and characterized further. In the construction industry, the majority of firms act as contractors which produce materials or services (e.g. architects) at a specified price. The market is thus characterized as a competitive bidding landscape, where firms strive for the optimal mark-up (Mayo, 1992). There are three specific market aspects in the construction industry. 1) The firms do not hold an inventory, however they have a in-process inventory. So they can not anticipate on demand shocks by adjusting their inventories. 2) Marketing is an important function in the construction industry, this is the key tool in obtaining bid opportunities. 3) Growth should be balanced, firms should keep high utilization rates. However, the
7
The degree of risk taking by the mangers influences the dominant strategy, though this leads mostly to a focused strategy. According to an US study by Tasi (1994), it is argued that if the managers strive to maximize shareholder value, the focus strategy is dominant for construction firms.
However, given the specific market aspects and the uncertainty and volatility of the market, Kim and Reinschmidt (2011) argue that diversified construction firms are better able to survive. Because they have multiple sector exposure and are less sensitive of individual market shocks. Moreover, Kim and Reinschmidt (2011) simulate the risk attitudes with diversification of several firms and find that diversification is optimal if firms are moderately risk averse. This paper will not elaborate further on the degree of the risk tolerance of firms. The competition simulation view by Kim and Reinschmidt (2011) shows that diversification is feasible and yields the highest financial performance.
Moreover, Choi and Russell (2005) gauge the profitability of 108 firms during 12 years and find that diversified firms have a higher growth rate than focused firms, but not significantly so. Furthermore, Junnonen (1998) uses a business strategy theory perspective to conclude that the diversification strategy is dominant1. Delaney and Wamuziri (2004) examine the abnormal returns of related acquisitions in the UK construction industry and show significant abnormal returns. However, he argues that the degree of risk taking by the managers influences the dominant strategy.
Russell and Choi (2004) performed an event study to gauge the difference between related and unrelated acquisitions. They studied 181 acquisitions firms and measured their Cumulative Average Abnormal Returns. However they found no significant difference between these two strategies.
Concluding the empirical evidence is also within the construction industry less clear cut. On the one hand, there is evidence that a focus strategy is more beneficial (Tasi, 1994). On the other hand, Junnonen (1998) and Kim and Reinschmidt (2011) argue that a diversification strategy is dominant, while Choi and Russell (2004 and 2005) find no significant difference between those two strategies.
8
Hypothesis formulation
Summarizing the above sections leads to the following results:
Firstly it is heavily debated whether or not the relatedness of the acquisitions has an impact. Moreover, in the special case of the construction industry, Choi and Russell (2004 and 2005) find no significant difference between those two strategies. This gives the following null hypothesis.
Hypothesis I0: The shareholders' response to related acquisitions in the European construction
industry does not differ from the shareholders' response to unrelated European acquisition in the construction industry.
Secondly if there is an effect it is questionable in which direction it goes. It could follow the general theory (Servaes 1996 and Kaplan and Weisbach(1992), which argues that related acquisitions are more beneficial than unrelated acquisitions. This effect is also found for the construction industry by Tasi (1994).
Hypothesis Ia: The shareholders' response to related acquisitions in the European construction
industry is higher than for the shareholders' response to unrelated European acquisition in the construction industry.
However, in the special case of the construction industry there is also evidence (Junnonen (1998) and Kim and Reinschmidt (2011)) which supports the view that the unrelated acquisitions (or the diversification strategy) is more beneficial.
Hypothesis Ib: The shareholders' response to related acquisitions in the European construction
industry is smaller for the shareholders' response to unrelated European acquisition in the construction industry.
It is unclear how the shareholders' reaction differs between a construction firm and a non construction firm. This leads to the following hypothesis:
Hypothesis II0: The shareholders' response to construction acquirers not from the construction
9
It could be that the construction firms know their own industry better than non construction firms, which gives them a better insight in attractive takeover targets. This leads to the following
hypothesis:
Hypothesis IIa: The shareholders' response to construction firms not from the construction
industry does not differ from the shareholders' reaction to non construction firms, regardless of their relatedness of the acquisition.
3.Data and Methodology
Explanatory variables and measures
Related Acquisition: The dummy variable equals one, if the target and the bidder are from the same industry. This is the case when both parties have a SIC Code within the range of 1500-1700, which is the classification of the construction industry. The dummy equals zero, if either the bidder or the target is not from the construction industry. This is the case for bidders not from the construction industry and for construction firms which acquire a non construction firm.
Focused Acquirer: This dummy variables captures whether the acquirer is diversified itself. This effect may blur the effect of the relatedness of the acquisitions strategy. If the acquirer is already diversified and engages in a non related acquisition, the takeover could have less of an impact on their risk profile in comparison with a focused acquirer engaging in an unrelated acquisition. This variable equals one if the acquirer is assigned to one industry on the basis of the SIC code classification2. This variable equals zero and thus is described as Diversified Acquirer, if the acquirer is assigned to multiple industries on the basis of the SIC code classification.
2 The SIC codes can be assigned to ten divisions (https://www.osha.gov). If the acquirer has one SIC code the
acquirer is classified as focused. If the acquirer has more than 1 SIC Codes and these correspond to the same division, the acquirer is classified as focused.
10
Bidder not from the construction industry: In order to give a more detailed view of the European construction industry this variable is included. This variable equals one if the acquirer is not from the construction industry and zero if the acquirer is from the construction industry.
Control variables and measures
Cash acquisition
In corporate finance the signaling theory states that if a firm performs a certain action, this conveys information for shareholders. In acquisitions, firms use different methods of payment in different circumstances. For example, firms will use cash when they believe that their firm is undervalued, but will prefer common stock when they believe that their firm is overvalued (Travlos, 1987). This has, however, consequences for the returns of the shareholders, who will interpret such results.
Travlos (1987) shows that common stock financing has a negative impact on the returns of the acquirer firm (as it provides a sign of overvaluation in the eyes of the mangers) and that cash financing has the opposite effect. In order to account for this effect, the dummy variable “cash” will be included if only cash was used to acquire the target and zero otherwise.
Moreover, if the acquisition is paid with cash, the owners of the private company are confronted with tax implications, when it is paid with stock the owners have the possibility to defer the tax. If this tax option is valuable for the owners, they will accept a lower price, which leads to higher bidder returns for the acquirer (Fuller et al., 2002). These theories thus suggest that the sign of the coefficient will be positive.
Firm Size (Logarithm of Market Capitalization)
Gorton et al. (2009) argue that the size of the acquirer negatively influences the profitability of the acquisition. This can be explained by the fact that large firms, which have high private benefits, tend to engage in defensive unprofitable acquisitions. In contrast, small sized firms which have low private benefits, tend to engage in the profitable positioning acquisitions. The effect on medium sized firm depends on the amount of the managerial private benefits. If private benefits are low, firms make profitable acquisitions. However, if these private managerial
11
There are some theories with plausible explanations which describe the size effect for acquisitions: 1) The incentives of managers of small firms are more in line with those of the shareholders, this can be explained by the fact that mangers of small firms have more firm ownership than the shareholders of the larger firms. 2) Hubris of the management will occur more likely in larger firms, because larger firms have more resources, which makes it easier to do an acquisition. 3) A large firm can owe its size to overvaluation. 4) Larger firms may suffer from exhausted growth opportunities. 5) The arbitrage opportunities in small firms are scarce, because it is too costly and too difficult to take a short position. (Moeller et al. 2004)
Moreover, Moeller et al. (2004) argue that there is a size effect and show that the abnormal returns of small firms are on average 2.25 percent point higher than those of the large firms. Moreover, if they replace this dummy variable with the continuous variable representing the logarithm of the market capitalization of the acquirer firm, they also find a negative significant effect on abnormal returns. In order to take the size effect into account, the logarithm of the market capitalization of the acquirer will be used. According to this theory the expected sign of the coefficient will be negative.
Market to Book of the Acquirer (MTB)
Rau and Vermaelen (1998) use the performance extrapolation hypothesis, which means that the value of the acquisition is extrapolated by the past performance of the bidder. This leads to an indirect judgment by the market of the approvers (management or other shareholders) of the acquisition. Firms with high Market to Book ratios are described by Rau and Vermaelen (1998) as glamour firms, whereas low Market to Book firms are described as value stocks. They argue that management of glamour firms are more likely to be captured by hubris due to high past performance and high past earnings. This leads to a less strict monitoring by the board of
directors and large shareholders in comparison with value stocks. They show that value stocks do not suffer from the hubris of management and therefore have a higher probability to engage in value creating acquisitions rather than value destroying acquisitions. Moreover, they find
12
Listed target
Private targets are by definition more illiquid but less costly than public targets are. In order to compensate for this liquidity effect, acquirers use a discount to price the value of the target. This discount will benefit the acquirers’ shareholders (Fuller et al., 2002).
Moreover, private firms only have a few shareholders, so if the private firm is paid with stock and is moderately sized, the likelihood of the occurrence of block holder formation increases. This results in better monitoring by the acquirers’ management after the deal and in a higher value for the acquirer. Furthermore, Conn et al. (2005) argue that it is easier to end the negotiations with private parties. This is due to a decrease in visibility for shareholders, which leads to a decreased chance of loss of face and value after a continuous bidding process and therefore it decreases the probability of a poor outcome.
Faccio et al (2006) show that there is a long list of empirical US evidence showing the existence of the listing effect, which means that the announcement of taking over listed targets leads to negative or zero abnormal returns in contrast to the announcement of taking over unlisted targets leading to positive abnormal returns. Faccio et al. (2013) find the same effect for acquisitions taken place in 17 Western European Countries. The sign of the coefficient is assumed to be negative.
Cross Border Target
There are several reasons why cross border acquisitions add higher value to the firm than domestic acquisitions. Foreign acquisitions could gain from 1) foreign knowledge 2) increase in demand, resulting in efficiency gains due to economies of scope and scale 3) entry to better corporate control systems (Von Eije and Wiegerinck, 2010). Furthermore, firms can benefit from differences in tax systems. Moreover, cross border acquisitions can give firms the opportunity to gain from exchange rate movements due to possible imperfect capital markets.
However, there are also reasons why the opposite may be true. For example, cross border acquisitions could increase agency problems and there may be additional costs due to the differences between culture politics and local institutions. Moreover, there can be larger information differences, which may lead to overpaying the target. Furthermore, if the
13
international firm (Von Eije and Wiegerinck, 2010).
The empirical evidence of the cross border effect is mixed (Von Eije and Wiegerinck, 2010). However, Feito-Ruiz and Menéndez-Requejo (2009) argue that in the case of the European shareholders, cross border acquisitions are preferred. They find a Cumulative Average Abnormal Return (CAAR) of 1.38% for cross border acquisitions, higher than the CAAR of 0.64% for domestic acquisitions. According to these theories the sign of the coefficient is expected to be positive.
Robustness control variables Before Crisis
In 2008 the European indices suffered from the fall of Lehman Brothers and the subsequent credit crisis (Figure 1). Figure 1 depicts that this also occurred in the construction industry.
Figure 1 : European Market and Construction & Materials total return indices
After the credit crisis, the budget deficits of European countries increased sharply. The result was the European debt crisis, in which markets were questioning the credit worthiness of some European countries.
During a crisis it will be easier to gain abnormal returns due to lower priced targets (Beltrati and Paladino, 2013). Moreover, Beltrati and Paladino (2013) find that abnormal returns during the financial crisis for European Banks are different. They show no significant returns around the announcement, which is contrary to the previous literature that shows negative returns when the acquirer announces his acquisition. Beltrati and Paladino (2013) conclude that in case of the European banking sector, investors argue that there is in particular value when acquisitions are executed during the financial crisis of 2008.
0 2000 4000 6000 8000 10000 12000 3 -1 -2000 3 -1 -2001 3 -1 -2002 3 -1 -2003 3 -1 -2004 3 -1 -2005 3 -1 -2006 3 -1 -2007 3 -1 -2008 3 -1 -2009 3 -1 -2010 3 -1 -2011 3 -1 -2012
EUROPE-DS Con & Mat - TOT RETURN IND
14
However, there is no literature available on the construction industry, which inevitably means that the effects remain unclear as of yet. In order to take this event into account, a dummy will be used to describe the abnormal returns before crises (before 2008) as one, and zero after 2007. According to the general view the coefficient is assumed to be negative, because acquisitions before the crises are classified as one and after the crises are classified as zero. However, to my best knowledge, there is no empirical evidence available which shows how the abnormal returns of acquisitions performed by construction firms are influence by the financial crisis of 2008. Acquired Stake
In order to take the amount of the acquired stake into account, a dummy variable is made which equals one if the acquired stake is greater than 50% and 0 if it is either an unknown acquired stake or if the acquired stake is equal to or smaller than 50%. The effect of this variable is unclear.
Double events
In both samples 'double events' take place, which are distinct acquisitions that take place on the same date and are done by the same company. The idea is to use a dataset that is as clean as possible and therefore I control for these clustered acquisitions. In order take this effect into account, a dummy is included and equals one whenever a double event occurs. The effect of this variable is unclear.
Data
The data are retrieved from:
1) the Zephyr Database of Bureau Van Dijk, which was used to obtain M&A announcements and characteristics.
15
The screening procedure was performed by using each of the following criteria:
1) The announcement date equals the rumor date. 2) The deals are completed.
3) The returns of at least 160 days before the announcement are available in DataStream. 4) The market capitalization and the Market to Book value of the bidder are available in
DataStream. In addition, the market capitalization is not negative, which makes the calculation of the logarithm possible.
5) The SIC codes are available in the Zephyr Database. 6) The bidder is headquartered in Europe
7) Mergers are excluded
These adjustments resulted in sample of 816 announcements.
Table 1 shows the trends and distribution of the M&A announcement during the analyzed period (2002 - 2012). The announcements tend to be equally distributed across the years.
Table 2 depicts that most bidders are from the construction industry (532/816). Possible explanation for this uneven distribution are 1) Bidders from the construction industry knows its industry better and thus face fewer difficulties to engage in an acquisition of a construction target. 2) It is more beneficial for bidders from construction industry than for firms which are not from the construction industry. The latter will be tested in the following section.
Moreover, table 2 shows that approximately 470 of the 816 announcements is characterized as a focused acquirer and the other half is characterized as a diversified acquirer. Furthermore, there are 565 related acquisitions and 251 unrelated acquisitions. Possible explanation are 1) Bidders from the construction industry knows its own industry the best and therefore are more likely to engage in related acquisitions. 2) Related acquisitions are more beneficial than unrelated acquisitions. The latter will be tested in the following section.
Table 2 indicates that the targets are primarily private companies (792/816). Moreover, it shows that the minority of the acquisitions is paid solely with cash (129/816).
Table 3 shows the descriptive statistics of the continuous variables. It shows that the minimum of the MTB is -0.15, which means that the book value is negative since negative market
16
Table 4 gives an overview of the origin of the acquirers. It shows that the most of the acquirers are located in the United Kingdom (17,65%) which is a substantial part of the sample.
Methods of cumulative abnormal return
Event studies are commonly used in the M&A literature as a research method. This method assumes that financial markets are efficient and investors have rational expectations.
The event day is either the day itself, when the acquisition is announced on a trading day or the first trading day after the announcement, when the acquisition is not announced on a trading day. In order to gauge the Cumulative Abnormal Returns (CARs), the normal returns have to be estimated by the following equation (Duso et al., 2010):
(1)
stands for the returns of the acquirer and will be collected for , which is defined as 160 days
to 2,3 or 5 days (depending on the event window length) before the first announcement. The αi
andβi will be estimated by using the Ordinary Least Squares (OLS) method. Sequentially the
abnormal returns will be calculated by using the estimation of the and (Duso et al, 2010).
(2)
is represented by the European Market Index. This paper will calculate two forms of ARs,
the Risk and Market adjusted model as described in equation 2 and the second is the Market Adjusted model where equation 2 is adjusted with = 0 and = 1. According to Ramesh et al. (1990) these two models offer more power than the Mean Adjusted Model.
17
Cumulative Abnormal Return
To examine the different shareholder reactions for the four types, the mean of medians of the CARs will be grouped and divided in related and unrelated acquisitions, where both are further divided in focused and diversified acquirers. In order to check whether the groups are statistically different from each other, the t-test for the means and the Wilcoxon/Mann-Whitney test for the medians will be used. The benefit of the Wilcoxon/Mann-Whitney test over the t-test is that it assumes non-normality, which is in this case more suitable, because the CARs in the groups have a non-normal distribution. However, Brown and Warner (1980) argue that non normality does not have an 'obvious impact' on event studies. They argue that there is a low probability of the
incorrect rejection of a true null hypothesis (Type 1 Error).
Multiple Regression
Furthermore to control for other determinants which affect the CAR the following equation will be estimated using OLS3:
(3)
In order to give more detailed insights in the relatedness of the acquisitions. The combination of the relatedness of the acquisition and whether the acquirer is focused or diversified is included in the following equation and will be estimated using OLS:
(4)
The above equation will be estimated for the announcements of the acquirers from the
3
18
construction industry. Because the announcements of bidders not from the construction industry can (per definition) not engage in related acquisitions in the construction industry.
4. Results
4This section is divided into two subsections. The first section shows the analyses of multiple regression of the whole sample. The second section shows the multiple regression analyses and the CAR analyses of the results when excluding the announcements of non construction bidders. Whole Sample
Whole sample short event window
Table 5 shows the multiple regression, which is estimated according to equation 3 using the short event window and the Risk and Market Adjusted Model.
Firstly, it shows a significant effect of the bidder not from the construction industry of 0.97%. This means that bidders outside the industry are better rewarded by their shareholders when announcing their takeover than the construction bidders. This is a remarkable result, since bidders from the construction industry may have an information advantage over those which are not from the construction industry. Secondly, it shows that whether or not the acquirer is focused does not have a significant effect. Thirdly, Table 5 shows an insignificant result for related acquisitions of 0.28%.
Table 6 depicts the OLS, which is estimated according to equation 3 using the Market Adjusted Model and the short event window. Here I find a significant bigger effect for the bidders not from the construction industry in comparison with the Risk and Market Adjusted Model (1.26% vs. 0.97%). Broadly speaking, the other results are in line with the findings of the Risk and Market adjusted model.
Whole sample long event window
Table 7 demonstrates the estimation of equation 3 using a long event window and the Risk and Market Adjusted Model. However, this estimation does not provide any significant explanatory variables. Table 8 exhibits the estimation of equation 3 using a long event window and the
4 This paper focuses solely on the interpretation of the explanatory variables. The subsequent interpretation of the
19
Market Adjusted Model. However, this estimation does not provide significant explanatory variables either.
Bidder from the construction industry
In this subsection, the analyses focus primarily on the bidders from the construction industry, thus the announcements of the bidders that are not from the construction industry are not
considered. This comes from the fact that these announcements cannot be divided into related or unrelated acquisitions, since per definition all announcements are unrelated acquisitions, which makes a comparison on this ground impossible.
When the multiple regression is estimated for either diversified or focused acquirers the related acquisition variable is not significant for both event windows and both CAR estimation models. However, this evidence is not reported in this paper. In order to analyze this effect more in detail the combination between the relatedness of the acquisition and whether the acquirer is focused or diversified is included in the following analyses.
CAR Analyses
Table 9 show the means and medians of the CARs with different estimation windows and different estimation models. The Market Adjusted model depicts in general the same results as the Market and Risk Adjusted model. According to Table 9, the differences between the groups on the short event window are less clear cut. It seems that focused acquirers which engage in a unrelated acquisition are the most beneficial. However, the differences across the groups are very small and most importantly not significant
Moreover, Table 9 shows that with respect to the Market and Risk Adjusted Model with a 11 day event window. It shows that the highest means are found for focused acquirers engaging in related acquisition (2.58%). However, this effect is less pronounced regarding the medians of the CARs. This mean and median are significantly different from the unrelated acquisitions
20
using the Risk and Market model. Moreover, the mean and median CARs of focused acquirers engaging in a unrelated acquisitions are even negative. The mean and the median are significantly different from the related acquisition regardless of whether the acquirer is focused or diversified.
Table 9 depicts the estimation of the CARs for the long event window using the Market Adjusted Model. This estimation shows some different results than the Risk and Market Adjusted Model, the biggest difference (0.91%) is that of the median of focused acquirer engaging in a related acquisition (1.27% vs. 2.18%). However, the differences between the groups have no impact on the sign of these combinations.
Multiple regression of the Market and Risk Adjusted Model estimated CARs5
Table 10 estimates equation 4 (see previous section) using a short event window. So it combines the related acquisitions or the unrelated acquisitions with whether the acquirer is focused or diversified. Moreover, Table 10 shows that the abnormal returns for the short event window the diversified acquirer which engages in a related acquisition is better rewarded than diversified acquirers which engages in an unrelated acquisitions. It shows that the CAR on this short interval differ 0.37% from each other. So for the case of diversified construction acquires it can be
concluded that they are better rewarded by the shareholders when they announce a related acquisition than a unrelated acquisition .
Table 11 shows the estimation of equation 4 (see previous section) using a long event window. However, the results show that three of the explanatory variables are insignificant. Keeping this caveat in mind, it is shows that there is a difference of at least 1.41% with the other
combinations. The focused acquirer which engages in a related acquisition is the most rewarded by the shareholders.
Focused or Diversified Acquirers
Table 7 shows that for the long event window using the Risk and Market Adjusted Model, there is a positive significant effect of Acquirer Focused (2.06%). This means that focused acquirers
5 The Market Adjusted Model gives similar results both event windows, however this analysis is not reported in this
21
are better off regardless of relatedness of their acquisitions. The Market Adjusted Model shows a smaller insignificant effect (1.20%) which is demonstrated in Table 8. This effect is not
significant and even negative for the short run regardless of the estimation method. (Table 5 & 6). Furthermore, when comparing table 10 and 11, there seems to be a reversal of the effect of
whether the acquirer is focused or diversified. In the short event window, diversified acquirers are most rewarded by the shareholders whereas in the long event window focused acquirers are most rewarded.
In order to increase the insight into the reversal of this effect, the Abnormal Average Returns (AARs) are calculated based on the Risk and Market Adjusted models on the 11 day event window and divided into the groups. The graphical representation in Figure 2 shows this reversal of this effect.
Figure 2: AARs, long event window estimated with the Market and Risk Adjusted Model by event window day
It shows that there are substantial gains, 5 to 1 days before the announcement, with respect to focused acquirers engaging in related acquisitions. This is in contrast with diversified acquirers engaging in related acquisitions which shows substantial negative gains for the same period. A possible explanation of this effect is that there are information leakages before the announcement which creates rumor in the market. Furthermore, the focused acquirers announcing an unrelated acquisition generate substantial gains 1 till 4 days after the announcement.
-0.006 -0.004 -0.002 0 0.002 0.004 0.006 0.008 0.01 -5 -4 -3 -2 -1 0 1 2 3 4 5
Average Abnormal Returns by day
22
5. Conclusion
According to the literature, it is unclear which strategy generates the highest results. The general view is that related acquisitions are most beneficial. However, whether or not this also holds for the construction industry is heavily debated. Some authors argue that in particular the
construction industry benefit more from unrelated acquisitions, because this industry have some special characteristics regarding to its risk profile. In order to gauge the effect of the relatedness of the acquisition more precisely, this paper uses a short event window (3 day interval) and a long run even window (11 day interval).
This paper finds there is no significant effect of the relatedness of the acquisitions, which means that the Hypothesis I0 cannot be rejected.
However, there is some evidence, which is in line with Hypothesis Ia, that related acquisitions are more beneficial than unrelated acquisitions. This evidence is found when the analyses are focused solely on the bidders from the construction industry and the combination of the relatedness of the acquisition and whether the acquirer is focused or diversified is taken into account.
This effect is found in the short event window where the multiple regression shows that
diversified acquirers which engage in related acquisitions are better off than diversified acquirers regardless of their acquisitions. Furthermore, the CAR Analysis shows that in the long event window focused acquirers engaging in a related acquisition are better rewarded than acquirers which engage in unrelated acquisitions.
Furthermore, in the long event window construction bidders which are focused are better
rewarded by their shareholders than those which are diversified (2.06%). An additional result of this paper is that there seems to be a reversal of this variable in the short event window. This could be explained by information leakages before and after the announcement. However, this effect needs further research to be interpreted carefully.
Moreover, this paper demonstrates for the short event window a significant positive effect of 0.97% for Bidder not from the construction industry. This means that the shareholders of non construction firms are more rewarding when announcing the acquisition of a construction firms compared to the shareholders of construction firms, regardless of their relatedness of the
acquisition. This result rejects hypothesis II0 and shows the opposite effect of the Hypothesis IIa.
23
References
- Andrade, G., M. Mitchell, & E. Stafford, 2001, New Evidence and Perspectives on Mergers. Journal Of Economic Perspectives 15, no. 2, 103-120.
- Beltrati, A., & G. Paladino, 2013, Is M&A Different during a crisis? Evidence from the European Banking Sector. Journal of Banking and Finance, 37(12), 5394-5405.
- Brooks, C., 2008, Introductory Econometrics for Finance, Cambridge University Press, Cambridge, Second Edition.
- Brown, S.J., & J.B. Warner, 1980, Measuring security price performance. Journal of Financial Economics, 8, 205–258.
-Campa J., & S. Kedia, 2002, Explaining the Diversification Discount. Journal Of Finance, 57(4), 1731-1762.
- Choi, J., & J. S. Russell, 2004, Economic gains around mergers and acquisitions in the
construction industry of the United States of America. Canadian Journal Of Civil Engineering, 31(3), 513-525.
- Choi, J., & J. S. Russell, 2005, Long-term entropy and profitability change of United States public construction firms. J. Manage. Eng., 21(1), 17–26.
- Conn, R. L., A. Cosh, P.M. Guest & A. Hughes, 2005, The impact on UK acquirers of domestic, cross-border, public and private acquisitions. Journal of Business Finance & Accounting, 32, 815–870.
- Custodio, C. Mergers and Acquisitions Accounting and the Diversification Discount. Journal Of Finance 69(1), February 2014, 219-240.
-Delaney, F. T., & S. C. Wamuziri, 2004, The impact of mergers and acquisitions on shareholder wealth in the UK construction industry. Engineering Construction & Architectural Management (09699988), 11(1), 65-73
- Denis, D. J., D. K. Denis, & A. Sarin, 1997, Agency problems, equity ownership, and corporate diversification, Journal of Finance, 52, 135-160.
- Duso, T., K. Gugler, & B. Yurtoglu, 2010, Is the Event Study Methodology Useful for Merger Analysis? A Comparison of Stock Market and Accounting Data. International Review Of Law And Economics 30, no. 2, 186-192.
- European construction market bottoms out, Euroconstruct,
24
- Faccio, M., McConnell, J. J., & D. Stolin, 2006, Returns to Acquirers of Listed and Unlisted Targets. Journal Of Financial & Quantitative Analysis, 41(1), 197-220.
- Feito-Ruiz, I., & S. Menéndez-Requejo, 2011, Cross-border Mergers and Acquisitions in different legal environments. International Review Of Law & Economics, 31(3), 169-187. - Fuller, K., J. Netter , & M. Stegemoller , 2002, What do returns of acquiring firms tell us? Evidence from firms that make many acquisitions. The Journal of Finance 57(August (4)), 1763– 1793.
- Gorton, G., M. Kahl, & R. J. Rosen, 2009, Eat or Be Eaten: A Theory of Mergers and Firm Size, Journal of Finance 64, 409-419.
- Hyland, D. C., & J. Diltz , 2002, Why Firms Diversify: An Empirical Examination. Financial Management (Wiley-Blackwell), 31(1), 51-81.
- Junnonen, J. M., 1998, Strategy formation in construction firms. Eng., Constr., Archit. Manage., 5(2), 107–114.
- Kaplan, S. N., & M. S. Weisbach, 1992, The Success of Acquisitions: Evidence from Divestitures. Journal Of Finance 47, no. 1, 107-138.
- Kim, H., & K. F. Reinschmidt, 2006, A dynamic competition model for construction contractors. Construction Management & Economics 24(9), 955-965.
- Kim.H, & K. F. Reinschmidt, 2011, Association of Risk Attitude with Market Diversification in the Construction Business. Journal Of Management In Engineering, 27(2), 66-74.
- Koller, T., M. Goedhart, & D. Wessels, 2010, Valuation, Measuring and Managing the Value of Companies, Fifth/University Edition, Hoboken, New Jersey: John Wiley & Sons.
- Lang, L. H.P., R. M. Stulz, & R. A. Walkling, 1989, Managerial performance, Tobin's q and the gains from successful tender offers. Journal of Financial Economics 24, 137-154.
- Maksimovic, V., G. Philips & L. Yang, 2013, Private and Public Merger Waves, Journal Of Finance 68, no. 5, 2177-2217.
- Mayo, R.E., 1992, Improved optimum bid markup estimation though workload related bid distribution functions, PhD thesis, Stevens Institute of Technology, Hoboken, NJ.
- Moeller, S. B., F. P. Schlingemann, & R. M. Stulz, 2004, Firm size and the gains from acquisitions, Journal of Financial Economics 73, 201–228.
25
Return-Generating Models and the Effect of Accounting for Cross-Sectional Dependencies in Event Studies, Journal of Accounting Research, Vol. 28, No. 2 (Autumn, 1990), 398-408. -Rajan, R., H. Servaes, L. Zingales, 2000, The cost of diversity: the diversification discount and inefficient investment, Journal of Finance 55, 35–80.
- Rau, R. P., & T. Vermaelen, 1998, Glamour, value and the post-acquisition performance of acquiring firms, Journal of Financial Economics 49, 223–253.
- Servaes, H., 1991, Tobin’s q and the gains from takeovers, Journal of Finance 46, 409–419. - Servaes, H., 1996, The Value of Diversification During the Conglomerate Merger Wave. Journal Of Finance 51(4), 1201-1225.
- Sic Division Structure, United States Department of Labor, https://www.osha.gov/pls/imis/sic_manual.html, 20-06-14, 14:05.
- Six-Digit NAICS Codes & Titles, NAICS, http://www.naics.com/free-code-search/sixdigitnaics.html?code=23, 02-01-14, 15:45.
- Tasi, C., 1994, Diversification strategies for construction companies in the U.S. M.S. thesis, Massachusetts Institute of Technology,Cambridge, MA.
- Travlos, N.s G., 1987, Corporate takeover bids, methods of payment, and bidding firms’ stock returns, Journal of Finance 42, 943–963.
26
Tables
Table 1: Number of M&A announcements. Time period (2000-2012)Period target 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Total
Announcements 52 46 63 42 50 63 67 75 83 72 83 72 48 816
(per year % of total sample) (6.37%) (5.64%) (7.72%) (5.15%) (6.13%) (7.72%) (8.21%) (9.19%) (10.17%) (8.82%) (10.17%) (8.82%) (5.88%)
This table shows the distribution of the announcements by year Table 2: Statistics Dummy variables
Dummies 0 1
Bidder from construction industry 532 284
Acquirer Focused 470 346 Related Acquisition 565 251 Cash 687 129 Listed 792 24 Before Crisis 358 458 Cross Border 569 247 Stake 205 611 Double 735 81
This table shows the characteristics of the announcements. A detailed description of the above stated dummies can be found in section 3.
Table 3: Descriptive Statistics Continuous Control Variables
Continuous Control Variables Mean Median Maximum Minimum Std. Dev. Skewness Kurtosis Jarque-Bera Probability
Firm Size 5.9299 5.8657 9.9087 3.4880 0.9839 0.6242 4.1157 95 0.0000
Market To Book of the Acquirer 2.4733 1.7250 41.2000 -0.1500 3.4198 5.4296 41.2074 53643 0.0000
27
Table 4: Origin of the Acquirer
Country Target Events %
28
Table 5: Multiple Regression, short event window estimated with the Risk and Market Adjusted Model
OLS: Risk and Market Adjusted Model (-1,+1)
Variables: Whole Sample Bidder from construction industry
Bidder not from construction industry
Bidder not from the construction industry 0.0097
(0.0043) Acquirer Focused -0.0059 -0.0060 -0.0048 (0.0035) (0.0037) (0.0074) Related Acquisition 0.0028 0.0030 (0.0036) (0.0036) Before Crisis -0.0016 0.0040 -0.0106 (0.0038) (0.0039) (0.0096) Cash Acquisition 0.0089 0.0070 0.0122 (0.0063) (0.0053) (0.0152)
Cross Border Target -0.0014 0.0029 -0.0095
(0.0035) (0.0044) (0.0069) Double Event -0.0085 -0.0082 -0.0077 (0.0047) (0.0064) (0.0079) Listed Target -0.0075 -0.0011 -0.0206 (0.0088) (0.0113) (0.0134) Firm Size -0.0051 -0.0063 -0.0059 (0.0021) (0.0027) (0.0039) MTB of the Acquirer -0.0002 0.0004 -0.0007 (0.0005) (0.0007) (0.0009) Stake 0.0039 0.0024 0.0095 (0.0047) (0.0051) (0.0098) Constant 0.0338 0.0358 0.0520 (0.0150) (0.0138) (0.0305) Observations 816 532 284
Significant coefficients at the 5% level are presented in bold and the standard errors between brackets which can be found below the coefficient.
29
Table 6: Multiple Regression, short event window estimated with the Market Adjusted Model
OLS: Market Adjusted Model (-1,+1)
Whole
Sample
Bidder from Construction Industry
Bidder not from construction industry
Variables:
Bidder not from the construction
industry 0.0126 (0.0044) Acquirer Focused -0.0060 -0.0073 -0.0018 (0.0035) (0.0038) (0.0073) Related Acquisition 0.0027 0.0034 (0.0037) (0.0037) Control Variables: Before Crisis 0.0015 0.0084 -0.0102 (0.0039) (0.0041) (0.0095) Cash Acquisition 0.0101 0.0073 0.0137 (0.0066) (0.0057) (0.0158)
Cross Border Target 0.0000 0.0042 -0.0088
(0.0035) (0.0045) (0.0069) Double Event -0.0081 -0.0070 -0.0084 (0.0048) (0.0067) (0.0079) Listed Target -0.0078 0.0035 -0.0301 (0.0106) (0.0125) (0.0178) Firm Size -0.0046 -0.0048 -0.0067 (0.0021) (0.0029) (0.0038) MTB of the Acquirer 0.0000 0.0006 -0.0008 (0.0005) (0.0007) (0.0008) Stake 0.0023 0.0004 0.0097 (0.0048) (0.0052) (0.0100) Constant 0.0289 0.0250 0.0581 (0.0148) (0.0184) (0.0296) Observations 816 532 284
Significant coefficients at the 5% level are presented in bold and the standard errors between brackets which can be found below the coefficient.
30
Table 7: Multiple Regression, long event window estimated with the Market and Risk Adjusted Model OLS: Risk and Market Adjusted Model (-5,+5)
Whole Sample
Bidder from construction industry
Bidder not from construction industry Variables: Bidder not from the construction
industry 0.0017 (0.0067) Acquirer Focused 0.0089 0.0206 -0.0137 (0.0056) (0.0068) (0.0102) Related Acquisition -0.0020 -0.0016 (0.0065) (0.0066) Control Variables: Before Crisis -0.0051 -0.0042 -0.0113 (0.0059) (0.0068) (0.0124) Cash Acquisition 0.0125 0.0115 0.0133 (0.0084) (0.0084) (0.0187)
Cross Border Target 0.0041 0.0049 -0.0020
(0.0061) (0.0077) (0.0111) Double Event -0.0081 -0.0136 0.0107 (0.0080) (0.0100) (0.0132) Listed Target -0.0091 -0.0054 -0.0203 (0.0144) (0.0200) (0.0184) Firm Size -0.0062 -0.0060 -0.0058 (0.0031) (0.0048) (0.0047) MTB of the Acquirer 0.0009 0.0009 0.0018 (0.0010) (0.0015) (0.0012) Stake 0.0046 -0.0045 0.0250 (0.0068) (0.0080) (0.0129) Constant 0.0350 0.0360 0.0308 (0.0217) (0.0296) (0.0360) Observations 816 532 284
Significant coefficients at the 5% level are presented in bold and the standard errors between brackets which can be found below the coefficient.
31
Table 8: Multiple Regression, long event window estimated with the Market Adjusted Model OLS: Market Adjusted Model (-5,+5)
Whole Sample Bidder from
construction industry
Bidder not from construction industry
Variables:
Bidder not from the construction industry 0.0087
(0.0069) Acquirer Focused 0.0046 0.0120 -0.0080 (0.0058) (0.0067) (0.0111) Related Acquisition 0.0002 0.0012 (0.0063) (0.0064) Control Variables: Before Crisis 0.0017 0.0039 -0.0063 (0.0060) (0.0068) (0.0132) Cash Acquisition 0.0162 0.0139 0.0172 (0.0092) (0.0089) (0.0209)
Cross Border Target 0.0031 0.0042 -0.0047
(0.0060) (0.0073) (0.0116) Double Event -0.0159 -0.0183 -0.0038 (0.0080) (0.0101) (0.0136) Listed Target -0.0050 0.0105 -0.0382 (0.0169) (0.0222) (0.0221) Firm Size -0.0057 -0.0040 -0.0076 (0.0031) (0.0046) (0.0049) MTB of the Acquirer 0.0005 0.0006 0.0007 (0.0009) (0.0013) (0.0013) Stake 0.0050 -0.0026 0.0242 (0.0072) (0.0081) (0.0144) Constant 0.0329 0.0236 0.0500 (0.0215) (0.0285) (0.0387) Observations 816 532 284
Significant coefficients at the 5% level are presented in bold and the standard errors between brackets which can be found below the coefficient.
32
Table 9: Analysis of Car, short event window, Risk and Market Adjusted model of bidders from the construction industry
Related Acquisition Unrelated Acquisition
All Focused Acquirer Diversified Acquirer Focused Acquirer Diversified Acquirer Risk and Market
Adjusted Model (-1,+1)
Mean 0.0045 0.0024 -0.0005 0.0097 0.0041
Median 0.0026 0.0046 -0.0009 0.0036 0.0032
Market Adjusted Model (-1,+1)
Mean 0.0042 0.0017 -0.0024 0.0094 0.0049
Median 0.0021 -0.0017 -0.0054 0.0078 0.0047
Risk and Market
Adjusted Model (-5,+5)
Mean 0.0059 0.0258a)b) 0.0108c) -0.0077a)c) 0.0028b)
Median 0.0005 0.0127 a)b) 0.0111c) -0.0159 a)c) 0.0031b)
Market Adjusted Model (-5,+5)
Mean 0.0091 0.0234 a) 0.0088 0.0010 a) 0.0077
Median 0.0050 0.0218 a) b) 0.0056 -0.0025 a) 0.0048 b)
This table shows the means and medians of both models for the different event window of all announcements of the bidders in the construction industry.
The t-test is performed to show whether the means of two groups are statistically different.
The Wilcoxon/Mann-Whitney test is performed to show whether the medians of two groups are statistically different.
a)
related acquisitions of a focused acquirer differ at 5% significance from unrelated acquisitions of a focused acquirer.
b)
related acquisitions of a focused acquirer differ at 5% significance from unrelated acquisitions of a diversified acquirer.
c)
33
Table 10: Multiple Regression, short event window estimated with the Risk and Market Adjusted Model
OLS of Risk and Market Adjusted Model (-1,+1)
Relatedness of the Acquisition: Focused or Diversified Acquirer?
Related Acquisition Focused Acquirer 0.0322 (0.0169) Diversified Acquirer 0.0391 (0.0166) Unrelated Acquisition Focused Acquirer 0.0302 (0.0183) Diversified Acquirer 0.0354 (0.0174) Control Variables: Before Crisis 0.0040 (0.0039) Cash Acquisition 0.0070 (0.0054)
Cross Border Target 0.0029
(0.0044) Double Event -0.0082 (0.0064) Listed Target -0.0010 (0.0113) Firm Size -0.0062 (0.0027) MTB of the Acquirer 0.0004 (0.0007) Stake 0.0024 (0.0051)
This table shows the OLS estimated using equation 4 and the Risk and Market Adjusted model for the 11 day interval.
34
Table 11: Multiple Regression, long event window estimated with the Risk and Market Adjusted Model OLS of Risk and Market Adjusted Model (-1,+1)
Relatedness of the Acquisition: Focused or Diversified Acquirer?
Related Acquisitions Focused Acquirer 0.0644 (0.0301) Diversified Acquirer 0.0308 (0.0286) Unrelated Acquisitions Focused Acquirer 0.0503 (0.0300) Diversified Acquirer 0.0422 (0.0300) Control Variables: Before Crisis -0.0039 (0.0068) Cash Acquisition 0.0107 (0.0083)
Cross Border Target 0.0048
(0.0077) Double Event -0.0144 (0.0098) Listed Target -0.0064 (0.0199) Firm Size -0.0064 (0.0048) MTB of the Acquirer 0.0008 (0.0015) Stake -0.0041 (0.0079)
This table shows the OLS, estimated using equation 4 and the Market Adjusted model for the 11 day interval.
35
Appendix I
Correlations
Bidder not from the construction industry Acquirer Focused Related Acquisitions Before Crisis Cash Acquisition Cross Border Target Double Event Listed Target Firm Size MTB of the Acquirer Stake
Bidder not from the construction
industry 1 0.1280 -0.4870 -0.0073 -0.0063 -0.0446 0.1102 -0.0054 0.0471 0.0419 -0.0098
Acquirer Focused 0.1280 1 -0.0131 -0.0110 0.0496 -0.0148 0.1298 -0.0466 -0.0246 -0.0725 -0.0805
Related Acquisitions -0.4870 -0.0131 1 0.0006 0.0460 -0.0056 0.0096 -0.0217 -0.1232 -0.0492 0.0677
Before Crisis -0.0073 -0.0110 0.0006 1 0.0650 -0.0357 0.1035 0.0370 -0.0495 -0.0724 0.1028
Cash Acquisition -0.0063 0.0496 0.0460 0.0650 1 0.0435 -0.0877 0.0836 -0.0689 -0.0104 -0.0433
Cross Border Target -0.0446 -0.0148 -0.0056 -0.0357 0.0435 1 -0.1206 0.0274 0.1837 -0.0538 0.0987
Double Event 0.1102 0.1298 0.0096 0.1035 -0.0877 -0.1206 1 -0.0335 -0.0709 0.0504 -0.2424
Listed Target -0.0054 -0.0466 -0.0217 0.0370 0.0836 0.0274 -0.0335 1 0.0489 0.0480 -0.2002
Firm Size 0.0471 -0.0246 -0.1232 -0.0495 -0.0689 0.1837 -0.0709 0.0489 1 0.1116 -0.0188
MTB of the Acquirer 0.0419 -0.0725 -0.0492 -0.0724 -0.0104 -0.0538 0.0504 0.0480 0.1116 1 0.0003
Stake -0.0098 -0.0805 0.0677 0.1028 -0.0433 0.0987 -0.2424 -0.2002 -0.0188 0.0003 1
