1
MONITORING OF ACQUISITION PROGRAMS AND ITS INFLUENCE
ON OVERALL PERFORMANCE: USING AN AGENCY THEORY &
RATIONAL ECONOMIC VIEW
David Awede
S3845729
MSc BA Strategic Innovation Management June 22, 2020 Supervisor Dr. P.J.O. Kuusela Co-assessor P.Arque-Castells, PhD Word Count: 8383
Abstract
2
INTRODUCTION
Over the years, performance of acquisitions and more specifically ‘the performance impact of acquisitions’ has been among one of the most actively studied topics in research on mergers and
acquisitions (Keil et al., 2012). As of now, research on the topic of acquisitions has over time increasingly moved from trying to explain the performance of acquisitions to instead identifying antecedents that would explain the variance in performance (Keil et al., 2012). One argument that has prevailed over time and has received attention is that “firms differ in their abilities to perform acquisitions” (Keil et al., 2012, p.148). As of now, there has been an increasing number of firms that partake in acquisitions (Aktas et al., 2009), but no clear explanation as to how acquisitions should be successfully monitored.
Earlier studies have suggested that the main differences between one’s firm ability to perform acquisitions better than another is due to the firm’s acquisition experience (i.e. CEOs experience) (e.g. Kale et al., 2002; Heimeriks and Duysters, 2007). Over the years, at the forefront of acquisition research, acquisition program perspective has emerged due to the observations that more and more firms have moved from individual acquisitions to being serial acquirers (Keil et al., 2012; Laamanen, 2007; Keil, 2008). Building on the acquisition programs perspective, the accepted notion is that acquisitions programs are done with a specific or common goal in mind. On that note, this paper defines acquisition programs as “sequences of acquisitions initiated by an acquiring firm, with the intention of achieving a specific business goal or market position” (Keil et al., 2012, p.149).
Addressing the growing topic of serial acquirers, Degbey (2015, p.11) defined serial acquirers as “firms which engage in streams of acquisitions to execute their strategies for enhanced value or
performance”. The contrasting difference between acquisition programs and serial acquirers is that while acquisitions programs are done with a shared/specific goal in mind, serial acquisitions don’t require shared goals, and can simply be done for the purpose of increasing one’s firm value.
As of now, many large companies are serial acquirers who partake in several tens of acquisitions each year. However, various studies on serial acquirers show us that there are declining cumulative abnormal returns (CARs) when dealing with acquisitions (Aktas et al., 2009; Fuller et al., 2002). This means that when firms engage in their first acquisitions, they get a positive market reaction, but the acquisitions that follow yield poorer results. Jansen and Ruback (1983) explained that “acquirers’ cumulative abnormal returns around the announcement date are at best equal to zero or, worse, even negative” (Aktas et al., 2009, p.543).
3
perspective (Aktas et al., 2009; Billet and Qian, 2008). Addresing the CEO level perspective, this paper intends to touch upon the role of monitoring by the board in order to explain the relationship between CEOs performance based on the deals conducted and the declining market reaction (i.e. CARs) from deal to deal. Recent empirical research has shown that, from deal to deal, there is a declining CARs for serial acquirers (Aktas et al., 2009). However, Aktas et al., (2009) questioned the notion or perspective that the declining trend of CARs has mostly been attributed to CEOs' hubris perspective, which could be one possible explanation of a negative performance in multiple acquisitions.
Addressing the perspective of monitoring, prior research reports have identified the need for accountability especially in the context of M&A, as various reports have shown that M&As fail to create shareholder value and often destroy it (Goranova et al., 2017; Datta et al., 1992; King et al., 2004). One core finding of Goranova et al., (2017) stated that monitoring is associated with both lower M&A losses as well as gains, hinting that there is a complex relationship between monitoring of acquisitions and its performance, a term they referred to as “dark-side to monitoring”.
Masulis et al., (2007) findings, on the other hand, showed that acquirers with more antitakeover provisions experienced significantly lower announcement abnormal stock returns, but they also found that acquirers operating in highly competitive industries or separating the role of CEO from chairperson of the board (i.e. elimination of CEO duality) experience higher abnormal announcement returns. Hence, this paper will be focusing explicitly on the ICT sector as a highly competitive industry. Some other findings addressing monitoring of acquisition perspective show that CEOs hold greater accountability to firm shareholders especially in the context of acquisitions (Bebchuk, 2005, 2003; Goranova et al., 2017). From all these prior researches, we know that CEOs (and other powerful individuals) personal motives and biases might influence performance in serial acquisitions, and that increased monitoring as a governance solution does influence individual’s behavior.
Using these studies and their findings as the rationale for this paper, this paper will be using the ‘agency theory’ and ‘rational economics view’ to explain the role monitoring (of CEO by the board of directors (BoD) plays in acquisition programs performance. By combining the agency theory with the rational economics view, this paper wants to use CARs as a measure of performance amongst other factors.
4
research question for this paper is: “How does monitoring of acquisition programs influences its overall performance?”.
The paper is structure as follows. Firstly, a theoretical background is presented which gives an insight into the key concepts used in this paper. Secondly, the hypotheses will be presented followed by the methodology section that outlines the data, sample, and variables. The paper will then progress to present the main findings in the result section that will be followed by discussions to make sense of the findings. Lastly, the paper will conclude and finish by providing implications and future recommendations of this paper.
THEORETICAL BACKGROUND
Acquisition Programs: Monitoring
Research conducted on acquisition programs is scarce and most of them originate from as early as the year 1980s and primarily come from the field of finance (Chatterjee, 2009). With the sparseness of information in this field of acquisition programs, the importance of understanding what leads to an overall successful program cannot be overemphasized. As already mentioned, acquisition programs is defined as “sequences of acquisitions initiated by an acquiring firm, with the intention of achieving a specific business goal or market position” (Keil et al., 2012, p.149). In other words, if CEOs conduct multiple acquisitions with a specific goal in mind like the acquisition of new knowledge and resources, such undertaking is referred to as acquisition programs. On the other hand, an acquisition program ends “when the underlying logic is no longer viable” (Chatterjee, 2009, p.138), hence the importance of monitoring. Additionally, acquisitions programs’ may not always become realized as intended by the acquiring firm. Most often, ‘the streams of acquisitions may be discontinued prematurely or receive a new meaning in the acquiring firm’ (Keil et al., 2012, p.153). As such, this paper places great emphasis on the importance of corporate governance from a monitoring perspective by addressing the agency theory and the rationale economics view perspectives.
Monitoring as a form of corporate governance helps the Board of Directors (BoD) mitigate the likelihood of the CEOs conducting deals that benefits them and not the company at large. The role and importance of monitoring has been reviewed by Goranova et al., (2017). According to Goranova et al., (2017), prior research by scholars and practitioners have stressed the importance of placing greater accountability on CEOs to firm shareholders (Bebchuk, 2005, 2013). There is no better place to stress the importance of accountability especially in the part of CEOs that when conducting mergers and
5
shareholders. Good corporate governance or successful monitoring would be achieved when the board of directors’ act in the interests of the shareholders by “reigning in the overt and aggressive risky tendencies of management” (Weitzner and Peridis, 2011, p. 37).
The Agency Theory
The ‘agency theory’ is the most dominant perspective in corporate governance research (Dalton et al., 2007; Goranova et al., 2017, p.2294). The agency theory is the behavior of managers seeking
excessively risky opportunities in pursuit of their own personal gain when they are left to their own devices (Weitzner and Peridis, 2011). This theory recognizes “the differences in interests and risk profiles between the principals of a corporation (i.e. shareholders) and their agents (i.e. managers) and the
implications for managerial behavior (Weitzner and Peridis, 2011, p. 37).
The agency theory suggests that managers will always most likely and explicitly choose risky behavior (Weitzner and Peridis, 2011). Goranova et al. (2017) explain that agency theory emphasizes boards of directors as vigilant monitors who constraints CEOs’ ability to pursue self-serving or hubris-driven strategies. By addressing the agency theory, this study aims to explore whether these different roles support the prevention of value-destroying or promotes value-creating strategies by contributing to a small but growing research stream that questions the “one-size-fits-all” approach to corporate governance (Goranova et al., 2017).
The Rationale Economics View – The ‘Independence’ approach
The independence approach of the rationale economics view suggests that the board of directors which is set up to be independent of management, can monitor managers and assure that their interests do not diverge to a great extent from the owners (Dalton et al., 2007, p.3). The composition of the board may be reasonably independent of firms’ CEOs and the fundamental responsibility of the board is to monitor the management of the firm. (Dalton et al., 2007). Regarding the composition of the board, some have repeatedly argued that the “boards’ willingness and ability to do so are partly or largely related to members’ independence” (Dalton et al., 2007, p.7).
6
Acquisition Programs: Learning Mechanisms
Learning in the context of acquisition programs is defined as “the process by which the CEO incorporates signals sent by the market at each deal announcement” (Aktas et al., 2009, p.553).
Furthermore, addressing CEOs learning mechanism, one claim that has been supported over the years is that firms (and their CEOs) that often undertake acquisition programs learn from each completed deal (Aktas et al., 2009; Schipper & Thompson, 1983; Fuller et al., 2002). In this context, CEOs learning mechanism indicates that the performance of the acquisition program should increase the longer the CEOs experience running the acquisition program, as they will be able to correct any mistakes that might have occurred in previous deals (Aktas et al., 2009). In other words, CEOs that successfully learn from each acquisition deals they partake in are believed to have a higher survival rate as regards job retention as they will be able to accurately self-assess (Aktas et al., 2009). The danger though is that the more successful they are, the likelihood increases for them to become overconfident (i.e. hubris).
Acquisition Programs: Hubris Mechanisms
Roll’s (1986) first introduced the hubris hypothesis or theory. In his theory, he states that
“managers engage in acquisitions with an overly optimistic opinion of their ability to create value” (Billet & Qian, 2008, p.1038). In the context of this paper, supporting Roll’s view, hubris is defined as “a cognitive bias in the CEO’s decision-making process” (Aktas et al., 2009, p.555; Malmendier & Tate, 2008). CEO hubris mechanism can “affect either the CEOs initial perception (the anticipated synergy at the first deal attempt), or his learning process (the interpretation of market reactions to past deals), or both” (Aktas et al., 2009, p.555). In the context of this paper, when talking about hubris, the focus will be its effect on the learning process of the CEO during acquisition programs. To elaborate on this, once the CEO conducts the first deal in an acquisition program, there will be tendency for the CEO (who becomes an experienced acquirer as he/she partakes in more deals) to become overconfident from their past successful acquisitions due to hubris or self-attribution bias, which results to declining performance in the subsequent deals that follow in the acquisition program (Billet & Qian, 2008).
HYPOTHESES
Relationship between CEOs Learning & Hubris mechanisms
7
deals, one growing stream of thought is that hubris and/or learning mechanisms could affect the behavior of the CEO.
Deighton (2006) who focused on CEO learning states that: “CEOs bid more accurately with experience (the cross-sectional variance of CAR decreases with the deal order number in the acquisition programs)” (Aktas et al., 2009, p. 545). Another paper by Hayward (2002) which analyzes the conditions under which experience is transferred from deal to deal at the organizational level also applies to the CEO level. Hayward’s (2002) view regarding CEOs learning experience states that: “Quite similar targets retard/delay learning because there is little new information. But very dissimilar targets also slow down learning since knowledge is not transferable” (Aktas et al., 2009, p.545). To support this claim, key findings from Hayward (2002, p.21) paper addressing when firms learn from their acquisition experience stated that: “a firm’s focal acquisition performance relates to prior acquisitions that are (a) not highly similar or dissimilar to the focal acquisition, (b) associated with smaller losses and (c) not too temporally close to or distant from the focal acquisition”.
To strengthen CEO’s learning mechanism, the consulting firm’s perspective addressing growing hubris views (Billet & Qian, 2008) state that “successful frequent acquirers are on a learning curve: they often start with small, lower-risk deals, and build capabilities in deal making. They institutionalize the processes and create a feedback loop to learn from mistakes” (Aktas et al., 2009, p.544). Aktas et al., (2009) also placed emphasis on “CEOs learning”. They state that: “if acquirer CEOs are learning, they improve their target selection and integration processing abilities from deal to deal” (Aktas et al., 2009, p.544). It is this learning process that impacts the bidding behavior of CEOs regarding what firms they need to acquire initially and the corresponding programs to strengthen their initial choice. Based on these reasonings, CEOs need to be aware that partaking in multiple acquisition programs of similar and
dissimilar nature (could) slow(s) their learning, resulting to higher likelihood of CEOs hubris.
Addressing CEOs hubris and its impact on performance, one theory suggests that what can explain the declining CARs is due to growing hubris or self-attribution bias (Billet & Qian, 2008; Aktas et al., 2009). The growing hubris explanation suggests that (serial) acquirers have the potential to learn from experience (Hayward, 2002; Aktas et al., 2009), from a management perspective. This growing hubris emphasizes that “if the CEO is not fired after completing the first acquisition deal, there is a high
8
From comparing both mechanisms, we could deduce that these two mechanisms could indicate an inverted U-shaped relationship.
Hypothesis 1: The relationship between CEO learning and hubris in acquisition programs has an inverted U-shaped relationship in terms of overall performance.
By monitoring CEOs learning and investigating at what stage the CEO has learned enough (i.e. identify patterns), monitoring would influence the hubris by mitigating its effect rather than prevent it from kicking in. To elaborate on this, this paper takes into the consideration that from the first deal, the CEOs learn something as well as hubris starts building up – so called ‘growing hubris’ (Billet & Qian, 2008). As such, this paper will assume that from all the deals conducted by the CEOs, the first part will be classified as CEO learning, and the later part will be considered as CEO hubris. Hence, should a CEO have participated in 22 deals for example, the first 11 deals will be considered as CEO learning stage, and the later 11 deals will be classified as CEO hubris. Since there is no clear evidence to point out when the CEOs have learned enough or stopped learning, it is the managers (i.e. BoD) responsibility to advice the CEO to focus on other programs in order to prevent the (continual) declining of the overall performance of the acquisitions’ programs. Hence it is important to monitor the decisions maker’s (i.e. CEOs)
learning/and or hubris and behavior as a form of governance.
Board of Directors (BoD) characteristics moderating effect
When dealing with corporate governance especially in the context of acquisitions programs, the relationship between the CEO and the BoD are inseparable. Both stakeholders have their own stake within the company and will be expecting the firm to meet their demands, respectively. Prior research indicates that the BoD, by monitoring the CEO, can reduce the number of value-destroying acquisitions (i.e. mitigating agency theory), however, an excessive control by BoD can also limit value-creation opportunities that acquisitions provide (Goranova et al., 2017; Masulis et al., 2007).
Over the years, the boards’ oversight role has dominated governance research (Tuggle et al., 2010; Goranova et al., 2017), however, Dalton et al., (2007, p.11) stated that: “there is no evidence of systematic relationship between board composition and corporate financial performance”. Various researches on the other hand, have shown that it is the fundamental responsibility of the BoD to monitor the management of the firm (Dalton et al., 2007; Johnson et al., 1969; Hermalin & Weisbach, 2003) as an important internal control mechanism (Masulis et al., 2007), and as such they are indispensable in enforcing corporate governance.
9
focuses on giving the BoD power to ‘self-govern’ by separating the role of CEO from the leader of the BoD. By doing so, CEOs will be able to receive constructive criticism and accurately self-assess and not become overconfident (i.e. hubris/self-attribution) (Billet & Qian, 2008). In a way, this paper is
interlinking the aspect of CEO duality (refer to hypotheses 2c) with BoD independence to investigate the impact both have on acquisition programs performance. Hence, by ensuring that the BoD (who have stake within the firm) are independent by having power to self-govern and thereby monitor the CEOs, this paper considers its influence on deals’ overall performance.
Hypothesis 2a: Board independence (i.e. board characteristic) has a positive moderating effect on overall performance.
Since monitoring by the BoD is an important internal control mechanism, the characteristics of the board will play an important role regarding monitoring acquisitions performance. As such, another
important characteristic is addressing the role of CEO duality. In most firms, there are situations where the CEO of a firm is also the chairperson of the board. This is referred to as CEO duality (Masulis et al., 2007; Goranova et al., 2017). To hold CEOs more accountable (Goranova et al., 2017), there is a need to
separate the role of CEO from the chairperson of the board. By doing so, CEOs will be monitored, and this will mitigate the likelihood of agency cost. Regarding CEO duality, according to Dalton & Dalton, (2011): “there is no evidence of substantive, systematic relationships between corporate financial performance and board leadership structure” (Goranova et al., 2017, p.2287). However, the agency theorists’ views argue that CEOs who serve as board chair weaken their board’s monitoring and control (Morck et al., 1989; Goronova et al., 2017). Since the complexity of a board’s roles becomes apparent or clear especially in the context of mergers and acquisitions, it is important to separate the roles of a CEO from that of the chairperson of the board (Westphal, 1999; Goranova et al. 2017).
Hypothesis 2b: CEO duality (i.e. board characteristics) has a negative moderating effect on overall performance.
The need for monitoring by the BoD to prevent self-interested behavior of CEOs influences the base relation (learning and hubris mechanism). As such, the characteristics of the BoD (i.e. CEO duality) and board independence (i.e. self-governance) play an important role as moderators when addressing their effect on acquisition programs deals.
METHODS
10
Data and Sample
The sample consists of 1603 acquisitions done by 33 firms which are publicly listed U.S. – based companies operating in the sector of energy and ICT. The criteria these companies met were that they were: serial acquirers, classified as acquisition programs, operating in SIC 737; announced between year 2000 and 2017; conducting at least 22 deals within the specified timeframe. The focus of this paper is on the software or ICT industry because firms in theses industries are the most active in terms of acquisition activity and they experience higher abnormal announcement returns. The sample used consists of 509 acquisitions which fit the specified criteria. The extracted acquisition data was taken from the SDC Platinum database. From the SDC Platinum dataset, information such as the parent acquirer, target name, announcement dates, deal value amongst others were available.
Announcement dates were cross-checked and categorized using LexisNexis to identify which deals are classified as acquisition programs and to document the sources of information. During this process, keywords emerged which were then categorized in programs according to one’s understanding. The link to the sources as well as quotes from both the acquiring firm and the target firm were used to comprehend the motives behind the acquisition of the deals when made stated in LexisNexis. During the categorization of the dataset in LexisNexis, deals classified as belonging to acquisition programs were classified as 1, and those not belonging to acquisition programs were classified as 0. Press releases that were untraceable on LexisNexis were classified also classified as not belonging to acquisition programs in this study, hence classified as 0 as well. From each press release, literal quotes from ultimate parent (i.e. acquirer) and target names or firms were deducted to strengthen the classification process, as each quote revealed the motivation behind the acquisition of the target firm and the target firm’s opinion about the acquisition. From all the collected and categorized data, and based on all the criteria mentioned above, for this paper, a total of 509 acquisitions done by 12 firms were categorized.
Dependent Variable
Cumulative abnormal returns (CARs). When focusing on agency theory and a rational
11
and its expected return, during a specific period surrounding the date of the firm’s divesture announcement”. The estimation window of the CARs used in this study are calculated as stated by Laamanen, Brauer & Junna paper (2014, p.917): “250 trading days (one year), measured from 295 to 45 days before each event”. The CARs dataset is extracted from CapitalIQ (WRDS).
Independent Variables
The independent variables with a moderating effect are based on two key board characteristics as specified in the hypotheses. Board independence (i.e. self-govern) and CEO duality are two attributes that previous research has shown to affect the effectiveness of a board function (Masulis et al., 2007). The BoD are responsible for monitoring the CEOs to prevent the likelihood of agency cost on the part of CEOs. As such, they have a key role to play in the relationship between CEOs learning/hubris and the impact they have on the overall acquisition performance. They (BoD) are to encourage CEO learning, as far as they mitigate the likelihood and effect of CEO hubris on overall performance (Aktas et al., 2009).
Board Independence (i.e. Self-governance). According to Weitzner & Peridis (2011, p.39),
“from a purely legal perspective, the board’s fiduciary duty is limited to monitoring the strategy of the firm”. Acquisition of other firms is considered as a firm strategy, and hence, it is the duty of BoD to monitor CEOs as they execute this strategy. By doing so, the BoD will be able to monitor the CEOs and thereby, CEOs will be able to accurately self-assess and not be overconfident (Billet & Qian, 2008). Hence, when addressing the role of BoD monitoring CEOs, this is only possible when the BoD can self-govern and make decisions in accordance to the wishes to the shareholders. To measure board
independence, the dummy variable ‘execdir’ (which represents executive served as a director during fiscal year) is used as the source. As already mentioned before, board independence (expressed as
self-governance in this paper) and CEO duality are interlinked. As such, in the dummy variable ‘execdir’ “1” represents CEO duality, while “0” represents board independence.
CEO Duality. CEO duality refers to the CEO holding a dual role as the CEO as well as the
12
CEO Learning. CEO learning is calculated by the sum of acquisition program deals conducted by
CEOs/Executives (refer to fig.1). A dummy variable was created using categorical variable
‘exec_fullname’ and a dummy variable was encoded and labelled ‘execname_codes’ with each CEO name corresponding to a number (i.e. based on alphabetical order). E.g. 1 in the ‘execname_codes’ corresponds to the name of a CEO/Executive like “Aart J. de Geus” in this case.
CEO Hubris. CEO hubris is expressed as percentage change in profit for each Acquirer per year.
It also represents the price change of a security (i.e. ownership position in a firm), hence the justification of using this variable to represent hubris. Formula used for CEO hubris is: bysort Acquireultimateparent (fyear): gen pchange= 100*(profit[_n]-profit[n-1])/profit[_n-1]. In terms of the variable profit, ‘gp’ which represents gross profit was used
and placed into the equation. Hence, CEO hubris in this paper is expressed as the percentage in profit for each acquirer per year. The variable is represented by ‘CeoH_pchange’ in this paper.
13
Control Variables
Year of acquisition announcement. A year control dummy variable was created for the year
period 2000-2017 which represents the year that corresponds to the announcement of an acquisition program based on variable ‘fyear’ in the datasets (refer to table 7 in appendix).
Firm. As already stipulated, the datasets contain 12 firms that partook in 509 programs deals.
Some firms are more likely to perform better than others in the acquisition of programs deals, as such, this is an important variable to control for. Like with year of acquisition announcement variable, a firm dummy variable based on variable ‘Acquirerultimateparent’ was also created in the datasets (refer to table 7 in appendix).
Acquirer size or Firm size. According to prior studies, large firms make large acquisitions that
results in large loses for the firm in comparison to smaller firms who make acquisitions that are profitable to their shareholders (Moeller, Schlingemann & Stulz, 2004). In this paper, acquirer firm size is measured using the natural logarithm of an acquirer’s total assets in line with Laamanen, Brauer and Junna (2014). The variable is represented by ‘Assets_Log’.
Analytical Model/Approach
The ordinary least squares (OLS) regression is the main approach used to empirically test the proposed hypotheses. Through the OLS regression, the relationship between the dependent variable CARs and the independent variables: CEO learning, CEO hubris, board independence and CEO duality can be tested and observed. The main program to test the models is the software program Stata.
RESULTS
An OLS regression analysis was conducted to explain what the direction of an effect is and whether an effect is statistically significant or not. Table 1 presents the descriptive statistics of the
14
abnormal returns. Table 2 provides the correlations between the variables, thereby outlining the how they relate to each other. The results marked with an asterisk (*) as pointed out in the table represents the significance levels (0.05 significance level), and it outlines the likelihood of rejecting the null hypothesis when it is true. In the table, several correlations are presented as significant, but not to a high standard, with the value being 0.4892. This is implying the relationship between CEO learning (signifined by execname_codes in the table) and its impact on total assets. This reasoning makes sense in the way that if CEOs learn from acquisition deals that they partake, the resulting outcome should be better deals which will result in increase of total assets within the firm. This will be due to the fact that CEOs will be able to select better deals and, in that way, reduce the wastage of firm’s resources in acquisition of bad or costly deals.
Table 1: Descriptive Statistics
Table 2: Correlations CeoH_pchange 496 3.815092 28.03505 -86.43732 560.7183 execname_c~s 509 16.56778 7.940095 1 28 execdir 509 .9332024 .2499168 0 1 Assets_Log 509 10.056 1.345681 7.029005 11.74581 car 509 -.001909 .039184 -.2907885 .1689172 Variable Obs Mean Std. Dev. Min Max
15
Table 3 presents the summary of the regression output and highlights the relationship of dependent variable CARs with independent variable (i.e. CEO learning and CEO hubris). Firstly, from the table, we see that the total observation has now dropped to 496 which is in relation to the 13 missing observations once CEO hubris percentage was calculated and created. Also included in the table are probability for F-test and T-F-test along with its coefficient and the R-squared (i.e. explains percentage of variance
explained). The F-value and Prob(F) statistics measures or test the significance of the whole model – they test the null hypothesis that states that all the regression coefficient (r-squared) are equal to zero. The Prob(F) tests the likelihood that the null hypothesis for the model is true (i.e. referring to regression coefficient being 0). In this table, we cannot fully reject the null hypothesis as Prob(F) values just mildly passed the 0.05 significant level demand. That means that there is a 6% chance that R-squared variable does not explains the variable (i.e. there is some sort of relationship between these variables and overall performance).
As such according to hypothesis 1, this paper claims that the relationship between CEO learning and hubris in acquisition programs has an inverted U-shaped relationship in terms of overall performance. However, the regression analysis does not fully support or reject this claim. From the table 3, only CEO hubris is negatively significant (refer to graph 1 below to see relationship), showing that hubris does negatively impact overall acquisition performance.
Table 3: Results of Regression Model 1: Relationship of CEO Learn & CEO Hubris with CARs
In table 4, the relationship between all independent variables and cars as the dependent variable is tested to investigate the reasoning behind hypothesis 2. Hypothesis 2 claims that CEO duality and board independence have a moderating relationship with overall acquisition programs performance. A
_cons -.0016475 .0039877 -0.41 0.680 -.0094824 .0061875 CeoH_pchange -.0001457 .0000609 -2.39 0.017 -.0002654 -.0000261 execname_codes 4.70e-06 .0002158 0.02 0.983 -.0004194 .0004288 car Coef. Std. Err. t P>|t| [95% Conf. Interval] Total .7192975 495 .001453126 Root MSE = .03798 Adj R-squared = 0.0075 Residual .711032263 493 .001442256 R-squared = 0.0115 Model .008265236 2 .004132618 Prob > F = 0.0579 F(2, 493) = 2.87 Source SS df MS Number of obs = 496 Adding : execname_codes CeoH_pchange
16
multicollinearity test was also conducted to test for variance inflation factors (VIF). Based on the scores, there is no multicollinearity present between the dependent and independent variables, as a score starting from 5 or higher signifies the presence of VIF. Just like in table 3, only CEO hubris hold a negative significance (refer to graph 1 below for scatter graph) and rejects the mean. However, the other variables have weak evidence against the null hypotheses based on T-test and P>T values. However, with a low 12% of the dependent variable not been explained by the independent variable, this paper concludes to mildly reject the null hypothesis.
Table 4: Results of Regression Analysis: All independent variables relationship and VIF Test
Graph 1: Correlation between CARs and CEO Hubris (i.e. medium strength linear relationship)
17
Robustness Checks
A robustness test was also conducted on the findings of table 3 and table 4 to avoid any type 1 or type 2 errors. After executing the robustness checks, the differences noticed were the impact of robustness on standard error, which in turn impacted the t-test values and the P>test values. In the case of t-test, the observed difference was the increase in t-value and P>t of CEO learning in the regression table (see table 5 and 6 below). In the case of CEO hubris and its significance, it remains negatively significant in both cases. The robustness check is meant to offset any deflation in the residual test in order to reduce the likelihood of error.
Table 5: Robustness Check of Table 3 results
Table 6: Robustness Check of Table 4 results
_cons -.0016475 .0036625 -0.45 0.653 -.0088436 .0055487 CeoH_pchange -.0001457 .0000633 -2.30 0.022 -.0002701 -.0000213 execname_codes 4.70e-06 .0001772 0.03 0.979 -.0003435 .0003529 car Coef. Std. Err. t P>|t| [95% Conf. Interval] Robust Root MSE = .03798 R-squared = 0.0115 Prob > F = 0.0707 F(2, 493) = 2.66 Linear regression Number of obs = 496 . regress car execname_codes CeoH_pchange, robust
18
DISCUSSION
Implication to Theory
This paper starts by stressing the importance of monitoring on overall acquisition programs performance. Monitoring, does not mean taking a passive role and leaving the CEOs to continue things as they are in, instead, it entails taking a step back to analyze what might be the root cause of why it is so difficult for firms and mainly the CEOs to maintain their ego as they continue to learn from deal to deal. Many studies have tackled the issue of merger and acquisitions, but as of now, there are still limited information addressing the topic of acquisition programs in terms of technological capabilities, organic growth options, alliance portfolios, stability of program-level acquisition scope and governance of acquisitions programs, just to name a few. Findings from this paper all boils down to one thing: are CEOs unable keep learning without hubris setting it?
By investigating the past CEOs performance in terms of acquisition of programs’ deals, one of the aims of this paper was to understand the relationship CEO learning and hubris had one overall firm performance. One of the findings addressing the hypothesis of an inverted u-shaped relationship between CEO learning and hubris claim that indeed evidence point out to some extent the reality of this relationship. Although, this paper cannot claim to fully reject the null hypothesis in regards to that relationship, findings from past studies supports the claim that indeed, there is a high tendency for CEOs after many years of working in the same industry and same line of work to become influenced by retard and slow learning, thereby resulting to poorer and poorer deals (Hayward, 2002; Aktas et al., 2009). To support this claim, “there is plenty of anecdotal evidence suggesting that hubris-infected or overconfident CEOs do exist (Martin & Davis, 2010, p.80)”, and this buttress the unavoidable trap called hubris that CEOs will eventually have to face from deal to deal.
19
finding was that 475 deals (93%) were led by CEOs who held dual roles as CEO and chairperson of the board, while only 34 deals (7%) were as a result of the a person of the BoD holding the position of
chairperson. Link this to the fact that 50% of the deals of acquisition programs resulted in poor CARs, this strengthens the evidence of CEO hubris, and supports this paper views on giving the BoD the right to self-govern in order to monitor CEOs and thereby increase the likelihood of mitigating the impact of hubris on overall acquisition programs performance.
Limitations
The key limitations of this study are first the initial unfamiliarity with the main software – Stata that was to be used merge and analyze the datasets. Finding the key variable to use to merge, only to be rejected by Stata is one of the highlights of my thesis experience. Majority of the time I had went to experimenting and familiarizing myself with this program that it felt like it took away a great deal of time that could have been spent in strengthening this paper. Support this with the fact that we were unable to meet, work on and talk face to face due to COVID-19 was another limitation. Thirdly, my limited knowledge in writing a quantitative paper (i.e. well-versed in qualitative papers) meant that I struggled in terms of writing my methodological section and interpreting my results. Fourthly, the restrictive nature of the dataset and trying to fit it to my own variable made it a tough and challenging experience, not only to make sense of the data, but also to constantly doubt myself on whether I was going in the right direction. Lastly, another limitation of this paper was the difficult in finding the right variable to attach to majority of the independent variables used in this study may have impacted the outcome and focus of this study. Nevertheless, I was appreciative that I could find justification to support my choices, and broaden my horizon by coming out of my comfort zone and working on quantitative paper as my final thesis as a Strategy and Innovation Management (SIM) master student.
Implication to Practice & Future Research
20
opportunity to grow and develop your personal skills by monitoring your CEO and developing the necessary skills needed to constructively give feedback, so that your CEO will be able to accurately self-reflect as well. All this is for the sake of ensuring an increasing overall performance in acquisition programs.
In terms of future research, this paper to the best of its ability has tried to shed light to the limited availability of research paper addressing acquisition programs. Future researcher can not only investigate the relationship of hubris with CEO learning but can also test out the role of monitoring works in practice. How would you define the stage where CEO hubris set it? What other ways can you tackle the limited stream of research on acquisition programs and how will you go about redefining and proving the validity of your findings? I can only wait in anticipation in the direction this research on acquisition paper will take, so, feel free to take the next step.
CONCLUSION
This paper concludes by saying that acquisition programs although limited in research now, has been gaining popularity over the couple of decades, and not only that, as of now, we have seen
21
REFERENCES
Aktas, N., de Bodt, E., & Roll, R. (2009). Learning, hubris and corporate serial acquisitions. Journal of Corporate Finance, 15, 543–561.
Bebchuk, L.A. (2005). The case for increasing shareholder power. Harvard Law Review, 118, 833-917. Bebchuk, L.A. (2003). The myth that insulating boards serve long-term value. Columbia Law Review,
113(6), 1637-1694.
Bertrand, M., & Schoar, A. (2003). Managing with style: the effect of managers on firm policies. Quarterly Journal of Economics, 118, 1169-1208.
Billett, M.T., & Qian, Y. (2008). Are overconfident managers born or made? Management Science, 54(6), 1037-1051.
Chatterjee, S. (2009). The keys to successful acquisition programs. Long Range Planning, 42 137-163. Croci, E., & Petmezas, D. (2009). Why do managers make serial acquisitions? An investigation of
performance predictability in serial acquisitions. SSRN Working Paper, 1-35.
Dalton, D. R., & Dalton, C. M. (2011). Integration of Micro and Macro studies in governance research: CEO duality, board composition, and financial performance. Journal of Management, 37(2), 404-411.
Dalton, D.R., Hitt, M.A., Certo, S. T., & Dalton, C.M. (2007). The fundamental agency problem and its mitigation. Annals of the Academy of Management, 1(1), 1-64.
Datta, D. K., Pinches, G. E., & Narayanan, V. K. (1992). Factors influencing wealth creation from mergers and acquisitions: A meta-analysis. Strategic Management Journal, 13(1), 67-84. Degbey, Y. W. (2015). Customer retention: A source of value for serial acquirers. Industrial Marketing
Management. 46, 11-23.
Deighton, E. (2006). Patterns in the performance of successive acquisitions: Evidence from individual CEO acquistions track records. Unpublished working paper, Helsinki School of Economics. Fama, E.F., & Jensen, M.C. (1983). Separation of ownership and control. Journal of Law & Economics,
26, 301-325.
Fuller, K., Netter, J., & Stegemoller. M.A. (2002). What do returns to acquiring firms tell us? Evidence from firms that make many acquisitions. Journal of Finance, 57, 1763-1793.
Goranova, M.L., Priem, R.L., Ndofor, H.A., & Trahms, C.A. (2017). Is there a “Dark Side” to Monitoring? Board and Shareholder Monitoring Effects on M&A Performance Extremeness. Strategic Management Journal, 38, 2285-2297.
22
Hayward, M. L. A. (2002). When do firms learn from their acquisition experience? Evidence from 1990-1995. Strategic Management Journal, 23, 21-39.
Heimeriks, K. H. & Duysters, G. (2007). Alliance Capability as a Mediator between Experience and Alliance Performance: An Empirical Investigation into the Alliance Capability Development Process. Journal of Management Studies, 44(1), 25-49.
Hermalin, B. E., & Weisbach, M. S. (2003). Boards of directors as an endogenously determined institution: A survey of the economic literature. Economic Policy Review, 9, 7-26.
Jensen, M.C., & Ruback, R.S. (1983). The market for corporate control: the scientific evidence. Journal of Financial Economics, 11, 5-50.
Johnson, J. L., Daily, C. M., & Ellstrand, A. E. (1996). Board of directors: A review and research agenda. Journal of Management, 22, 409-438.
Kale, P., Dyer, J. H., & Singh, H. (2002). Alliance Capability, Stock Market Response, and Long-Term Alliance Success: The Role of the Alliance Function. Strategic Management, 23(8), 747-767. Keil, T. (2008). Performance of Serial Acquirers: Toward an Acquisition Program Perspective. Strategic
Management Journal, 6(2), 663-672.
Keil, T., Laamanen, T., & Mäkisalo, A. (2012). Acquisitions, Acquisition Programs, and Acquisition Capabilities. In Faulkner, D., Teerikangas, S., Joseph, R.J. (Eds.), The Handbook of Mergers and Acquisitions. (pp. 148-167). Oxford University Press.
King, D. R., Dalton, D. R., Daily, C. M., & Covin, J. G. (2004). Meta-analysis of post-acquisition performance: Indications of unidentified moderators. Strategic Management Journal, 25(2), 187-200.
Laamanen, T. (2007). On the Role of Acquisition Premium in Acquisition Research. Strategic Management Journal, 28(13), 1359-1369.
Laamanen, T, Brauer, M., Junna, O. (2014). Performance of acquirers of diveseted assets: Evidence from the U.S. Software industry. Strategic Management Journal, 28(13), 914-925.
Malmendier, U., & Tate, G.A. (2008). Who makesacquistions? CEO overconfidence and the market’s reaction. The Journal of Financial Economics, 14, 237-250.
Masulis, R.W., Wang, C., & Xie, F. (2007). Corporate Governance and Acquirer Returns. The Journal of Finance, 62(4), 1851–1889.
McNulty, T., Florackis, C., & Ormrod, P. (2013). Board of directors and financial risk during the credit crisis. Corporate Governance: An International Review, 21(1), 58-78.
Moeller, B.S., Schlingemann, P.F., & Stulz, M.R. (2004). Firm size and the gains from acquisitions. Journal of Financial Economics, 73, 201-228.
23
Schipper, K., & Thompson, R. (1983). Evidence on the capitalized value of merger activity for acquiring firms. Journal of Financial Economics, 11, 85-120.
Tuggle, C. S., Sirmon, D. G., Reutzel, C. R., & Bierman, L. (2010). Commanding board of director attention: Investigating how organizational performance and CEO duality affect board members’ attention to monitoring. Strategic Management Journal, 31, 946-968.
Weitzner, D., & Peridis, T. (2011). Corporate governance as part of the strategic process: Rethinking the role of the board. Journal of Business Ethics, 102, 33-42.
Westphal, J. D. (1999). Collaboration in the boardroom: Behavioral and performance consequences of CEO-board social ties. Academy of Management Journal, 42(1), 7-24.
Westphal, J.D., & Zajac, E.J. (2013). A Behavioral Theory of Corporate Governance: Explicating the Mechanisms of Socially Situated and Socially Constituted Agency. The Academy of Management Annals, 7, 607–661.
Westphal, J.D., & Zajac, E.J. (1995). Who shall govern? CEO/board power, demographic similarity, and new director selection. Administrative Science Quarterly, 40, 60–83.
24
APPENDIX
Table 7: Results of Regression: Independent & Control Variable (i.e. showing the control)
_cons .2996218 1.063932 0.28 0.778 -1.79094 2.390184 VeriSign Inc .0197408 .0148402 1.33 0.184 -.0094193 .0489008 Synopsys Inc .0144122 .0126527 1.14 0.255 -.0104496 .0392741 Symantec Corp -.0035393 .0107381 -0.33 0.742 -.0246389 .0175603 Schlumberger Ltd -.0296127 .0195699 -1.51 0.131 -.0680663 .008841 Qualcomm Inc -.0168203 .0139186 -1.21 0.227 -.0441695 .0105289 Motorola Solutions Inc -.0055813 .0147892 -0.38 0.706 -.0346411 .0234786 Motorola Inc -.0091621 .0174609 -0.52 0.600 -.0434716 .0251475 L-3 Communications Holdings .0134841 .0139023 0.97 0.333 -.013833 .0408012 Intuit Inc .0211339 .0113105 1.87 0.062 -.0010905 .0433584 IBM Corp -.0338676 .0199728 -1.70 0.091 -.0731128 .0053777 EMC Corp -.011592 .0126359 -0.92 0.359 -.0364207 .0132367 BMC Software Inc .0072801 .0119125 0.61 0.541 -.0161273 .0306875 AcquirerP_codes fyear -.0002084 .0005475 -0.38 0.704 -.0012843 .0008674 Assets_Log .0127322 .0065154 1.95 0.051 -.0000702 .0255345 execdir -.0017933 .0074358 -0.24 0.810 -.0164041 .0128175 CeoH_pchange -.0001372 .0000627 -2.19 0.029 -.0002603 -.0000141 execname_codes .0001419 .000391 0.36 0.717 -.0006264 .0009101 car Coef. Std. Err. t P>|t| [95% Conf. Interval] Total .7192975 495 .001453126 Root MSE = .03808
