ABSTRACT - This paper analyzes the likelihood of corporate fraud occurring post-acquisition due to simultaneous acquiring behavior and links it to the characteristics of the acquisition. From an agency perspective, hypotheses are built to test whether a higher degree of information asymmetry caused by an increased acquisition rate increases the likelihood of corporate fraud occurring post-acquisition. By using a sample of 27,079 acquisitions and 408 corporate fraud filings, it is found that the amount of simultaneous mergers during any given period is positively related to the likelihood of corporate fraud occurring post-acquisition. Furthermore, the results indicate that the average level of post-acquisition integration negatively moderates this relationship. Theoretically, this contributes to the development of agency theory into the mergers & acquisitions theoretical field. This study is, however, also interesting to managers, as it indicates they can reduce the likelihood of corporate fraud occurring post-acquisition by focusing their resources on a fewer amount of simultaneous acquisitions and by integrating the simultaneously acquired firms more extensively.
Keywords: corporate fraud, post-acquisition integration, agency, acquisition rate, M&A
I. Introduction
Mergers and acquisitions are still the most popular research areas in finance, and have been for quite some time (Haleblian, Devers,
McNamara, Carpenter, & Davison, 2009). However, despite the popularity both in research and as a growth strategy, two-thirds of mergers & acquisitions are regarded as a failure (Hitt, Hoskisson, & Ireland, 1991). Some authors have argued that this failure stems from agency problems, whereby the interests of the management of acquiring firm acts in ways undesired by the shareholders (Amihud & Lev, 1981; Comment & Jarrell, 1995). This undesired behavior in some instances can go as far as corporate fraud, resulting in extreme negative abnormal returns for the defendants (Johnson, Ryan, & Tian, 2009; Dyck, Morse, and Zingales, 2010). It is estimated that organizations lose an average of 5% of their revenues to fraud, even when efforts such as increased legislation and stricter enforcement are institutionalized to prevent it from occurring (Murphy & Dacin, 2011). Most researchers studying corporate fraud, have studied antecedents at an individual level (Gillett & Uddin, 2005; Carpenter & Reimers, 2005; Ajzen, 1991). However, recently some studies have started to study corporate fraud at the organizational level (Nguyen, Yung & Sun, 2012). Nguyen, Yung & Sun (2012) have argued that the opportunity to commit corporate
Predicting Corporate Fraud in M&A: An Agency Perspective
Aligning the interests of stakeholders to reduce the likelihood of corporate fraud occurring post-acquisition
Esteban Barnhardt (s2504707)
Rijksuniversiteit Groningen
Supervisor: Dr. K.J. McCarthy
Second Supervisor: Prof. Dr. D.L.M. Faems
fraud during an acquisition stems from information asymmetries. However, although these authors have made a start, so far, no research has studied causes of corporate fraud from an organizational level in an M&A setting. Considering the high costs associated with corporate fraud (Johnson, Ryan, & Tian, 2009), it is worthwhile to study what some causes could be, and what can be done to decrease the likelihood of it occurring. The development an understanding regarding these causes of corporate fraud will contribute to further develop an understanding on why so many mergers fail to reach desired goals set at the time – or prior to – the acquisition. This is particularly interesting for senior managers engaging in M&A, as it will intend to help them reduce the amount of cases of corporate fraud against their firm post-merger, as well as reduce the amount of failed M&A attempts. Theoretically, this study will further develop the extension of agency theory into the M&A field. Furthermore, this study contributes to an improved understanding of factors that influence the likelihood of a firm being alleged of having committed corporate fraud, all in an M&A setting, by focusing on two aspects. First, it is analyzed whether an increased acquisition rate enhances the likelihood of corporate fraud occurring at a firm. Second, the moderating effect of the average level of post-merger integration is analyzed on the relationship between the acquisition rate and likelihood of corporate fraud occurring. Using a sample of 27,079 mergers and 408 cases of alleged fraud support is found for the hypothesized positive relationship between a higher acquisition rate
and an increased the likelihood of corporate fraud occurring, and partial support is found for the hypothesized negative moderating influence of the average level of post-acquisition integration. These findings contribute to both research on acquisition programs as well as on the stream of literature analyzing the complexities involved in an M&A setting. For acquiring managers, it confirms that they should place more focus on a lower amount of simultaneous acquisitions in their acquisition program such that they can reduce the likelihood of corporate fraud occurring at their firm.
This study continues with a review of relevant theoretical concepts used (Chapter II), followed by an explanation of the methods (Chapter III). After that the results (Chapter IV) will be thoroughly discussed before they are compared to relevant previous literature in the discussion (Chapter V). Next, Concluding remarks and implications are given (Chapter VI). Finally, the limitations are given and it is discussed how they provide a pathway for future research (Chapter VII).
II. Theoretical Section
Literature ReviewAgency Theory in M&A
Agency theory is a stream of literature that focuses on the difficult relationship in which ‘one party (the principal) delegates work to
another party (the agent), who performs the work’ (Eisenhardt, 1989, p. 58). More
ensure that the agent behaves in appropriate ways due to knowledge asymmetries that result from the agent being more involved with the organizational processes (Eisenhardt, 1989). The focus of agency theory is, thus, to analyze appropriate ways to align the interests of the principals in agents. This perspective was taken in the M&A field by Amihud & Lev (1981), who focused on the understanding of why some organizations engage in conglomerate mergers. They argued that this is done particularly to diversify and thereby reduce risk to managerial human capital. Schleifer & Vishny (1989), studied agency in M&A in a similar manner and argued that acquiring managers engaged in acquisitions to make organizations dependent on their skills. Morck, Shleifer, & Vishny (1990) argued that these agency problems become apparent when managers’ interests are not in line with those of the firm’s owners. These are of particular interest as there are differences in interests here between shareholders and managers; M&A is a higher risk strategy for shareholders, but can be attractive for managers ’who have fewer
avenues available to diversify their own risk’
(Eisenhardt, 1989, p. 68).
Although the level of analysis in agency has often been the individual manager within an organization, Eisenhardt (1989, p. 58) argues that the unit of analysis – ‘contract governing
the relationship between the principal and the agent’ - can also be used at organizational
level. At this level of analysis, the agency problem in question is related to the asymmetry in information between the acquiring firm and its shareholders.
Acquisition Rate
Acquisitions are often part of an acquisition programs including multiple (simultaneous) acquisitions over a certain period of time, as opposed to isolated events (Schipper & Thompson, 1983). This study takes that into consideration by analyzing the acquisition rate of firms, defined as “the number of
acquisitions that acquirers carry out over a given time period” (Laamanen & Keil, 2008, p.
664).
Schipper & Thompson (1983) argued that firms obtain substantial benefits from an increased acquisition rate due to the spread of risk, a higher degree of efficiency, and an increased possibility of generating synergies amongst other things. Other authors take a learning perspective and argue that acquisition capabilities are improved through acquisition experience, which can be enhanced by an increased acquisition rate (Haspeslagh & Jemison, 1991; Haleblian & Finkelstein, 1999). According to Haspeslagh & Jemison (1991), acquisition capabilities include “organizational
skills such as the ability to identify suitable acquisition targets, negotiate the deal, and manage the integration process” (p. 371).
negative relationship is related to corporate indigestion, which arises in those situations of excessive acquisition behaviour. Hitt, Harrison, Ireland, & Best (1998) found similar results in their multiple-case study. More specifically, they found that in those situations where firms entered an acquisition in the same year as another large acquisition, the result was “chaos,
as opposed to systematic implementation” (p.
108). Kusewitt (1985) even went as far as to argue that firms should have no more than a single merger being processed at any given time so as to prevent time and resource constraints in assimilation and integration phases of the acquisition.
Finally, other authors argue that (e.g. Steger & Kummer, 2007) argue that the M&A process becomes incrementally more difficult as it moves from the phases prior to the acquisition to the integration phase. They further argue that, it is logical that many managers feel confident in the initial phases of acquisitions, however as they move through the process, it becomes increasingly demanding and, thus, more difficult to manage. So, whilst in the early phases of the acquisition process a higher acquisition rate is feasible, later on in the process when the multiple acquisitions become increasingly complex, it can lead to insufficient management of each individual acquisition (Shrivastava, 1986)
Post-acquisition integration
When a firm acquires another company, the acquirer has the choice with regards to the degree to which it will integrate the organization and its processes into the parent
practices. This decision concerns the level of integration, which is defined following Zollo & Singh (2004, p. 1236) as the “extent to which
the functions of the acquired unit are linked to, aligned with, or centralized in, the equivalent functions of the acquiring organization”. This
definition is particularly useful in this situation, as it does not focus on the integration at individual level, but at organizational level. In line with the agency perspective, they further find a positive significant relationship between the level of integration and acquisition performance, indicating that a higher level of integration aligns management goals and provides control mechanisms to ensure that mutual goals are pursued. Furthermore, Zollo & Singh (2004) and Datta, Deepak, & Grant (1990) argue that a higher level of integration is necessary to be able to realize the potential value that emerges from the synergies between the two firms. Steger & Kummer (2007) find support for this notion with their results suggesting that, in the banking industry, the costs incurred with a higher degree of integration are outweighed by the benefits that occur from cost efficiencies.
Other studies show that a higher level of integration results in a higher level of disruption of the resources, which in turn leads to a negative performance (Marks & Mirvis, 1985;; Buono & Bowditch, 1989; Empson, 2001). Agreeing with this, another stream of studies indicate that a higher level of integration leads to a more complex integration process (Kitching, 1967; Jemison & Sitkin, 1986). This, because it brings with it ‘a large
simultaneous decision-making processes, involving increasing levels of interaction between parts and functions of the two organizations’ (Zollo & Singh, 2004, p. 1240).
Finally, Christensen & Overdorf (2000) take a contingent view and argue that the appropriate level of integration should depend on the reason for the acquisition. More specifically, it is argued that when the motivation for the acquisition is acquiring capabilities, it would be contradictory to fully integrate the organization into the parent organization, as this would mean stirring up processes and thereby diminishing those capabilities.
Corporate Fraud
Previously, much of the M&A-related fraud literature has focused at individual level (Gillett & Uddin, 2005; Carpenter & Reimers, 2005; Ajzen, 1991). More specifically, it has focused on motives for CEOs to acquire organizations at a premium, whilst linking it to the post-acquisition compensation increases that CEOs obtain (Haleblian, Devers, McNamara, Carpenter, & Davison, 2009). However, as this study is at organizational level – rather than individual – fraud is taken from the organizational perspective. Fraud is defined here Following Baucus & Near (1991, p. 11) as “violations in which the law assumed an
organization acted with knowledge or intent and the courts ruled that organization was guilty of illegal behavior”. The reasoning by
Baucus & Near (1991) was followed to focus this study on actual illegal behavior, as opposed to corporate wrongdoing. The main theoretical framework used in this study to build theory on
corporate fraud is the theory of fraud triangle, which argues that the possibility of corporate fraud occurring is a dependent on of (1) incentives, (2) opportunities, and (3) attitudes. (Cohen, Ding, Lesage, & Stolowy, 2010). This theory states that when all three of these elements are present, they can predict the existence of fraudulent activity within an organization (Murphy & Dacin, 2011). It follows that if each one of the elements is decreased, the likelihood of corporate fraud occurring will lower. As this study analyzes the corporate fraud at organizational level, only the first two elements of the fraud triangle are taken into account. The attitude element is a construct measured at the level of the individual, thereby differing substantially from the organizational perspective in this study (Cohen, Ding, Lesage, & Stolowy, 2010). Opportunity is defined following Cohen, Ding, Lesage, and Stolowy (2012, p. 604) as “the
perceived opportunity that one can perpetrate the fraud while not getting caught”. Cohen,
Ding, Lesage, and Stolowy (2012, p. 604) then continue by defining incentives as “a reason to
commit fraud”. Finally, they broaden incentives
it to ‘motives’, such that it also includes ‘social
pressures, such as how individuals wish to be seen by others’.
the possibility of corporate fraud to be committed is due to asymmetric information that exists between shareholders and acquiring managers, such that managers can engage in acquisitions which are not in line with those interests of the shareholders. Hitt, Harrison, Ireland & Best (1998) further argued that although not all were directly related to the acquisition, the majority of these cases had a profound impact on the outcome of the acquisitions involved and that the amount of unethical practices exposed was particularly high in the cases where the acquisition was unsuccessful. Corporate fraud related to M&A takes many forms such as blocking other parties from making bids, failing to release relevant information during an acquisition, and artificially inflating stock value to acquire using common stock (Nguyen, Yung & Sun, 2012). Figure 1 shows a conceptual model of all the concepts discussed in the theoretical review of this paper, with the expected relationships that are hypothesized in the next section.
Hypotheses
Acquisition Rate
Miller, Bretton-Miller & Lester (2010) argue that agency creates the opportunity for acquisition managers to pursue acquisitions that risk shareholder value through the power advantage they possess caused by information asymmetry. This information asymmetry is common in all corporate investments as “managers can continually observe changes in
investment productivity on an individual asset bases, whereas outsiders obtain only highly aggregated information on investment productivity at discrete points of time”
(Aboody & Lev, 2000, p. 2749). Aboody & Lev (2000) further argue that uniqueness of each acquisition contributes to an increased degree of information asymmetry. However, although this discusses individual acquisitions, many mergers are often a part of an acquisition program (Schipper & Thompson, 1983). When multiple acquisitions are put into effect simultaneously, the performance may be measured not by each single acquisition, but by entire program, thereby further expanding the information asymmetry – and thus opportunity to commit corporate fraud – between
organization and its shareholders (Kummer & Steger, 2008).
Previous research has argued that a higher acquisition rate can lead to negative performance outcomes (Harrison, Ireland, & Best, 1998; Kusewitt, 1985). It has further been found that, as acquisitions progress through the acquisition process, the complexity increases exponentially (Kummer & Steger, 2008). Hitt, Harrison, Ireland, & Best (1998) even argued that when firms engaged in an acquisition at the same time as a large acquisition, the management process was often extremely chaotic. According to Kummer & Steger (2008, p. 11) “successfully mastered first
(and easy) phases of M&As foster the illusion that people are competent enough to succeed in the following phases as well”. However, they argue
that in many cases the expected synergies are not realized. In situations with a high acquisition rate and dissatisfying results from one or more mergers, managers could be tempted to use their information asymmetry to their advantage and engage in fraudulent activities in order to inflate the acquisition program results. This notion is supported by the findings of Johnson, Harley, Ryan, & Tian (2008), who find that the motive of those firms engaging in corporate fraud is to prevent stock price declines. More specifically, they argue that firms in which corporate fraud has been committed often anticipate price declines if they would not engage in fraudulent behavior. Therefore, in the context of an acquisition portfolio, preventing shareholders from seeing negative acquisition performance as a result of inflated acquisition rates is therefore expected to provide the motive to engage in fraudulent activities.
Overall, it argued that, as firms engage in more than one acquisition at the same time, they find themselves in situations in which a lack of control exists and in which a large degree of information asymmetry exists between the shareholders of the firm and its acquiring management. It is further argued that the large level of information asymmetry increases the opportunity for management to engage in corporate fraud, whereas the negative results in individual acquisitions within an acquisition program provides the motive to engage in corporate fraud to prevent negative overall acquisition program performance. The first hypothesis is thus as follows:
H1: A higher acquisition rate positively influences the likelihood of corporate fraud occurring post-merger within a given organization.
Average Level of Post-Merger Integration
asymmetry advantage that managers have over their shareholders.
Thus, an increased average level of integration thereby reduces the information asymmetry that becomes particularly prevalent in cases of an increased acquisition rate (Kummer & Steger, 2008). This, in turn, would diminish the opportunity element of the fraud triangle, thereby lowering the likelihood of fraudulent activity. Using these author’s findings, it is argued that an increased level of average post-acquisition integration lowers the opportunity element of the fraud triangle, thereby contributing to a lower likelihood of corporate fraud. Besides the reduction of the information asymmetry, an increased level of post-merger integration also has the potential to diminish the chaos that exists when firms engage in multiple simultaneous mergers (Eisenhardt, 1989), for the distance between shareholders and acquiring managers is lessened as the targets become more integrated with the focal firm in question.
Other authors have argued that one of the core motives of acquiring an organization is to improve badly run practices and better utilize the human resources and capital resources (Varaiya, 1987; Sales & Mirvis, 1984). It has further been argued that in order to obtain the benefits expected from an acquisition, a thorough degree of integration is required (Trautwein, 1990). An increased performance through the realization of synergies (Cassiman, Colombo, Garrone, & Veugelers, 2005) could limit the negative impact of an increased acquisition rate as some difficult to imitate synergies could compromise losses in other
underperforming acquisitions (Barney, 1991). A decreased amount of losses would, in turn, reduce the motive aspect of the fraud triangle thereby reducing the likelihood of corporate fraud occurring.
Overall, it is argued that an increased average level of average integration of those acquisitions comprising the acquisition rate reduces the motive element of the fraud triangle by increasing the synergies that are generated from the acquisition portfolio, and reduces the opportunity element of the fraud triangle by increasing the amount of control mechanisms and reducing the information asymmetry advantage that acquiring managers have over their shareholders. The second hypothesis therefore argues the following:
H2: An increased degree of average post-acquisition target integration negatively moderates the relationship between the acquisition rate and likelihood of corporate fraud occurring post-merger.
III. Methods
Research SettingIn this study three sources of data were used to obtain the necessary information, each of which is briefly discussed in this section. Once the data of each of the three independent databases was retrieved, it was merged into one dataset from which the analyses were performed.
Thomson Reuters’ SDC
the acquiring and target firms at the time of the simultaneous acquisitions. It provides a wide array of vari ables along the following topics: M&A transactions, Debt Capital Markets; bond deals, Equity Capital Markets transactions, Syndicated Loans transactions, Public Financings data, Project Finance data, Joint Ventures and Repurchases, Corporate Takeover Defense: Takeover defense profiles, and Shareholder Activism: Detailed histories (Thomson Reuters). As Kim & Song (2007) argued that the SDC database commenced systematically gathering their data in 1988, no data prior to that year was used in this study: only information after 1 996 was used, thereby ensuring that the information used will be systematically gathered and complete.
SCAC filings Database
In order to obtain information on cases of corporate fraud, the Stanford Lawschool’s Securities Class Action Clearinghouse (SCAC) was used. This database has previously been used for studying corporate fraud in several other studies (e.g Deng, Willis, & Xu, 2012;
Walker, Turtle, Pukthuanthong, & Thiengtham, 2015). The SCAC database contains extensive information on complaints, filings, briefs, and other litigation-related materials filed in the cases.
Thomson Reuter’s DataStream
The dataset was further complemented with that of Thomson Reuters’ DataStream, which offers a combination of strategic and economic research (Thomson Reuters). More specifically, the DataStream database was used to gather data on the economic status for each of the years through the Morgan Stanley Capital International (MSCI) World index.
Sample
An initial sample was generated from the SDC database. In doing so, several criteria were to be fulfilled for an acquisition to be included in the sample. First, the sample only included those acquisitions in which the acquirer was publicly listed. This was done because the dependent variable – fraud – is likely to become more prevalent in this section due to
Table 1
Variables and data source
Variables Description Database
Dependent Amount of Corporate Fraud Filings SCAC
Independent Amount of Simultaneous Mergers SDC
Moderating Average % of outstanding shares owned by acquirer SDC
Controls Difference SIC SDC
Total Amount of Acquisitions by Acquirer SDC
Difference Country SDC
Economic State DataStream
Market Value Acquiring Firm during Acquisition SDC
Transaction Value SDC
the fact that shareholders exist, who are likely eager to sue the firm for substantial losses due to the case of corporate fraud (Johnson, Ryan, & Tian, 2009). This reduced the database to 402,266 observations.
The second criterion was to include only acquisitions that occurred between 01/01/1996 and 31/03/2015 to ensure that old acquisitions with incomplete data were excluded from the sample. This reduced the sample to 320,825 mergers. Next, the deal value had to be upwards of 50 million US Dollars. This was done to exclude small acquisitions due to the same reason as the first criterion, to exclude those organizations in which corporate fraud would not be brought out. This reduced the sample to 53,931.
Third, as this study analyzes acquisitions as opposed to investments, a minimum ownership threshold was set to 5% of outstanding shares at the day of the acquisition. This reduced the sample to 53,706 acquisitions.
The Fourth criterion applied excluded spinoffs, recapitalizations, self-tenders, and repurchases. This was done prevent the consideration of departments or repurchased organizations as acquisitions. This reduced the sample to 43,087.
The final criterion excluded those acquisitions that had not been completed at the time the data was extracted. This was required so all information regarding the information was present. The last criterion reduced the final sample to 37,716 acquisitions.
To complement the dataset generated from the SDC database, a second sample was generated from the SCAC filings database. In doing so,
initially, 3,786 filings were extracted from the database. The filings were merged with the SDC dataset using both the ticker/listing, and the firm name to check for erroneous matching. This yielded a match in 405 cases. For each of these cases, additional data was extracted from the SCAC filings database. This included the total amount of filings against any given defendant, the status of the filing (ongoing, dismissed, or settled), and, in case there was a settlement, the settlement amount in US Dollars. It is important, however, to note that non-settled filings were also included.
Finally, Thomson Reuter’s DataStream was used in order to obtain information regarding the economic state for each of the range of years included in the sample. This was done with the MSCI World Index. This data was merged with the complete dataset and a value was assigned to each acquisition, depending on the year in which it was performed.
Measures
Dependent variable
Number of Corporate Fraud Filings – As the
event studies of Johnson, Ryan, & Tian (2009), and Dyck, Morse, & Zingales (2010) found that cases of alleged corporate fraud filed against a firm generates significant negative abnormal returns for the defendant, filings were used as a proxy for corporate fraud. More specifically, the amount of filings registered in the SCAC database against a given defendant in a given year was used in this case.
Independent variable
Acquisition Rate – first a variable was
computed containing the total amount of mergers performed per company. However, this depicts more the experience that firms have with acquiring, not the acquisition rate. Therefore, a proxy for this concept was generated, following Zollo & Singh (2004), by computing the amount of acquisitions that were performed within the same year of the focal firm’s acquisition. Due to the high right-skewed nature of the variable, a natural logarithm was used to improve the distribution.
Moderating variables
Average Post-acquisition integration – To
measure the average level of post-acquisition integration, the average level of ownership on the day of the acquisition was identified. This was done following, Mahoney (1992) and Fowler & Schmidt (1989), who argued that that financial ownership of one organization in another in cases of vertical relationships can be seen as an extreme level on the level of
integration continuum, with on the other side markets and contracts. Following Mahoney (1992) and Fowler & Schmidt (1989), the percentage of outstanding shares owned by the acquirer on the day of the transaction was used as a proxy for the level of post-acquisition integration. This variable is termed in the SDC database as PCTWON. This proxy was used as it represents the level of commitment that a firm makes to acquire the target organization, which seems an adequate indication of the degree to which it will intend to merge post-acquisition. Once this variable was constructed, the average was taken over the simultaneous acquisitions that a given firm engaged in at a given point in time.
Control variables
Amount of unrelated simultaneous acquisitions - Chaterjee, Lubatkin, Schweiger, & Weber
the same and they are therefore active in the same industry. Following that, the total amount of unrelated acquisitions for a given firm during a given year were summed.
Economic State – Next, as has been shown in
the literature review, fraud activities require incentives to be performed. These incentives can stem from negative organizational performance during economic downtimes (Soltani, 2014). Therefore, this study controlled for whether the acquisitions – and related fraudulent activity – were within a period of economic crisis. The MSCI World Index was used to measure the economic state following. This index was assigned to each of the years in which the acquisitions of any given firm occurred.
Amount of International Deals – Further, this
study controlled for differences in host country of target and acquirer. Following, Gertsen & Söderberg (1998), it is argued that a larger geographical distance increase the complexity of succeeding with an acquisition. They argue that this is the case due to a lower degree of socialization between managers that occurs when distance is increased, as well as due to national identities that enhance confusion. Initially, a dummy variable was created and coded as 1 if the country in which the target and acquiring firms were based were unequal. Consequently, the variable was coded 0 if the target and acquiring firms’ base countries were the same. Once each acquisition was coded, like the amount of unrelated acquisitions, this variable, was aggregated by summing the
amount of international deals a given firm engaged in during any given year.
Average Firm Size – firm size was also
controlled for in this study following Haleblian, Kim, & Rajagopolan (2006). They used a logarithm of the acquirer’s total assets at the time of the acquisition. Therefore, this study also included the logarithm of the value of the acquiring firm in US Dollars at the time of the deal. The natural logarithm was based on the average value of total assets during the time of the acquisitions within a given year.
Another way of measuring firm size was done by Puranam & Srikanth (2007), and involved the number of employees the acquiring firm had at the time of the acquisition. This study also included the measure of employees, however, due to the skewness of this variable in our data, a natural logarithm was used. As with the value of the acquisition, the average of the number of employees during the acquisitions was calculated and used in this study.
Average Target Value – Finally, following
Statistical Methods
An initial variance inflation factor (VIF) test was performed in order to test for multicollinearity. The resulting VIF for each of the explanatory variables were lower than or equal to 2. 44. The resulting average VIF was 1.72, indicating that multicollinearity was not found to be an issue in this study. Therefore, we proceeded by determining the adequate statistical tests in order to test the hypotheses. Due to the excess amount of zeros in the sample, a zero-inflated model was used in this study. A zero-inflated model deals adequately with excess zero’s by using both a Logit regression and a Poisson or Negative Binomial model (Lambert, 1992). As over-dispersion in the dependent variable was not an issue, the Zero-Inflated Poisson (ZIP) was used. A further reasoning for the Zero-Inflated Poisson versus the regular Poisson was the statistically significant (P < 0.03) Vuong statistic (Vuong, 1989). The ZIP in this study was performed in a hierarchical manner. In the hierarchical ZIP, model 1 included the control variables. Following that, in model 2, the independent variable – acquisition rate (ln) – was
introduced. Finally, model 3 included the interaction effect of level of integration.
Due to the fact that there were multiple different-year observations for the same firms, interclass correlation could be an issue in this study (Moulton, 1986). Therefore, to correct for that, the robust standard errors were clustered by firm id in all the models. This was done using STATA’s VCE function.
IV. Results
Descriptives & Correlations
An initial tabulation of the corporate fraud filings indicates that there were an average of 20 filings per year of those years studied and in 2001 and 2008 there were both peeks of 39 and 36 filings, respectively. When compared to the amount of mergers in those years respectively it can be seen that, on the contrary, this is below average in 2001 with 1,162 acquisitions, and only slightly above average with 1,416 acquisitions in 2008. What is interesting to note, however, is that each of these years in
Table 2
Descriptive Statistics
Variables Mean SD Min Max
1 Amount of Filings 0.015 0.126 0 2
2 Amount of International Deals 0.403 0.713 0 12
3 Amount of Unrelated Deals 0.820 0.835 0 19
4 Average Transaction Value ($mln) 627.5 3,214 50 202,785
5 Amount of Employees 20,333 58,953 1 2,100,000
6 Firm Value ($mln) 7.623 2.283 -6.908 15.142
7 Amount of Other Mergers 4.921 7.378 0 110
8 Economic State 3,707.6 1,118.2 1,920.4 6,192.7
9 Acquisition Rate 1.301 0.781 1 22
10 Average Ownership (%) 79.3 30.1 5 100
which many filings are omitted, are preceded by peaks in acquisition behavior. In 2000, 1,774 acquisitions were completed and in 2007 this number was as high as 1,954.
Table 2 shows descriptive statistics of the variables used in this study. As can be seen, the average firm in the sample acquires 1.3 firms per year. Interesting is that the average level of ownership of those simultaneous acquisitions in any given year is as high as 79.9% of outstanding stock at the day of the acquisition. This indicates that the vast majority of the acquisitions in the sample are fully acquired (100%). Finally, it can be seen that the maximum amount of filings in a given year against a given firm in this sample is 2.
Table 3 shows the Pearson correlation matrix. Interesting is to see that the acquisition rate is strongly correlated to all the other variables used in this model, but is negatively correlated with the state of the economy at the time of the acquisitions. Furthermore, the amount of corporate fraud filings is significantly correlated to both the acquisition rate and average level of ownership, where acquisition rate shows a positive correlation and average level of ownership shows a negative correlation. A firm’s acquisition rate, further, shows significant correlations to all other variables in the model. It is also interesting to note that both proxies for firm size (number of firm’s employees and value of total assets in US Dollar) are both positively correlated with all other variables in the model. Finally, it is worthy to point out the strongly significant correlations between average ownership and all
other variables, except the average value in millions of US Dollars of the transactions.
Hypotheses
Variables 1 2 3 4 5 6 7 8 9 10
1 Amount of Filings 1.00
2 Amount of International Deals -0.0002 1.00
3 Amount of Unrelated Deals 0.0073 0.1619*** 1.00
4 Average Transaction Value ($mln) 0.0138** 0.0105 -0.0183*** 1.00
5 Amount of Employees (ln) 0.0493*** 0.2343*** 0.1064*** 0.1335*** 1.00
6 Firm Value ($mln) (ln) 0.0508*** 0.2105*** 0.1244*** 0.1695*** 0.7085*** 1.00
7 Amount of Other Mergers 0.0598*** 0.1849*** 0.2328*** 0.0670*** 0.3592*** 0.4767*** 1.00
8 Economic State -0.0149** 0.0118* -0.0049 0.0096 -0.0261*** 0.0280*** -0.0363*** 1.00
9 Acquisition Rate (ln) 0.0271*** 0.3320*** 0.5134*** 0.0338*** 0.1608*** 0.2562*** 0.3262*** -0.0479*** 1.00
10 Average Ownership -0.0196*** -0.2844*** -0.3439*** -0.0111 -0.2090*** -0.3188*** -0.2908*** 0.0151*** -0.6507*** 1.00
Pearson's correlation. * are significant at .10 level, ** at .05, and *** at .01 levels.
the average ownership of the simultaneous acquisitions and the likelihood of corporate fraud occurring. However, in line with hypothesis 2, a significant negative interaction effect (P < 0.01) is found between the average ownership of simultaneous acquisitions and the relationship between the acquisition rate and
the likelihood of corporate fraud occurring. In line with the hypothesis, this indicates that when a firm chooses to fully integrate their simultaneous acquisitions, the likelihood of corporate fraud occurring post-acquisition decreases. Hypothesis 2 is, thus, supported.
Poisson Model Amount of Filings Amount of Filings Amount of Filings
(1) (2) (3)
Total International -0.126 -0.176 -0.209**
(0.109) (0.108) (0.0973)
Total Unrelated -0.00518 -0.0970 -0.121
(0.0712) (0.0808) (0.0789)
Average Transaction Value ($USD) 6.15e-06 5.81e-06 3.57e-06
(1.13e-05) (1.14e-05) (1.21e-05)
Amount of Employees 0.140** 0.145** 0.152**
(0.0591) (0.0583) (0.0674)
Average Acquiring Firm Value 0.0785** 0.0668* 0.0740*
(0.0383) (0.0390) (0.0411)
Amount of Other Acquisitions 0.0214*** 0.0190*** 0.0200***
(0.00634) (0.00612) (0.00582)
Economic State -3.38e-05 -2.80e-05 -3.03e-05
(7.74e-05) (7.74e-05) (7.81e-05)
Acquisition Rate (ln) 0.438*** 2.259***
(0.157) (0.490)
Average Ownership (ln) 0.0121***
(0.00424)
Avg. Ownership (ln) * Acquisition rate (ln) -0.0206***
(0.00529) Constant -4.955*** -4.926*** -6.145*** (0.646) (0.465) (0.931) Zero-Inflation model Amount of Employees (ln) -0.093 0.090 0.096 (0.066) (0.064) (0.071) Constant -0.550 -0.557 -0.614 (0.548) (0.547) (0.531) Nonzero Observations 301 301 301 Zero Observations 15,956 15,956 15,956 Wald's Chi-square 47.98 54.73 80.79 P > χ2 0.000 0.000 0.000
To further analyze the interaction relationship more thoroughly, a split test was performed including the control variables, and the acquisition rate using the moderator as conditions for the ZIP regression. First, the average ownership was transformed to a dichotomous variable, coded as 1 in the cases of a full 100% average ownership and 0 in other cases. The results, shown in table 5, are consistent with those of the ZIP regression in table 4, and indicate a negative relationship
between the average level of ownership and the relationship between acquisition rate and the likelihood of corporate fraud occurring post-acquisition. More specifically, it can be seen that the coefficient of acquisition rate in those situations of low average ownership is higher than the coefficient in the model conditioned by a high average level of post-acquisition integration of simultaneous acquisitions. This relationship is graphically depicted in figure 2, and clearly indicates that firms engaging in
Table 5
Robustness Zero-Inflated Poisson
Poisson Model Amount of Filings
Low Average Ownership High Average Ownership
Total International -0.0693 -0.408**
(0.120) (0.192)
Total Unrelated -0.0567 -0.189*
(0.110) (0.102)
Average Transaction Value ($USD) -4.20e-05 7.30e-06
(4.87e-05) (1.41e-05)
Amount of Employees 0.367 0.117***
(0.238) (0.0452)
Average Acquiring Firm Value 0.181* 0.0536
(0.0934) (0.0443)
Amount of Other Acquisitions 0.00452 0.0219***
(0.0172) (0.00576)
Economic State -3.21e-05 -1.12e-05
(0.000216) (8.11e-05) Acquisition Rate (ln) 0.807** 0.387* (0.331) (0.198) Constant -8.087*** -5.027*** (2.234) (0.496) Zero-Inflation model Amount of Employees (ln) 0.328 0.0282 (0.298) (0.0423) Constant -2.105 -1.621 (3.308) (2.867) Nonzero Observations 57 244 Zero Observations 3,326 15,956 Wald's Chi-square 32,66 54.73 P > χ2 0.000 0.000
multiple simultaneous acquisitions with a lower level of post-acquisition integration have greater likelihood of corporate fraud occurring post-acquisition than when a high average level of post-acquisition integration is used. A further result that is of particular interest in this model is that while no significant relationship exists between the amount of employees and the likelihood of corporate fraud occurring post-acquisition in the model with low integration, but when high integration is used post-acquisition, the amount of employees an acquiring firm possesses at the time of simultaneous acquisitions significantly (P < 0,01) p ositively influences the likelihood of corporate fraud occurring post-acquisition. Further results indicate that, in case of high average ownership, the total amount of simultaneous international deals significantly negatively influences (P < 0.05) the likelihood of corporate fraud occurring post-acquisition,
whereas this variable is insignificant with low average ownership. Finally, the total amount of mergers a given firm in this dataset completed between 1996 and 2015 significantly positively (P < 0.01) influences the likelihood of corporate fraud occurring post-acquisition, whereas this variable is insignificant in the condition of low average ownership.
Robustness
Although significant negative cumulative abnormal returns were identified by Johnson, Ryan, & Tian (2009) and Dyck, Morse, & Zingales (2010), we performed a robustness check to test the proxy used to measure the likelihood of corporate fraud occurring post-acquisition. This robustness check was done using the amount of settled filings a focal firm had in a given year as a dependent variable. As with the main model, the analysis started with a VIF test to test for multicollinearity issues. In
Figure 2: Interaction Effect
line with the results for the main model, multicollinearity was not found to be an issue. The highest VIF was 2 .44, and the mean was 1.68. Further, as with the main model, excess zeros were present in the database, favoring a Zero-Inflated model. Finally, over-dispersion was not found to be an issue, thereby making
the Zero-Inflated Poisson an adequate test. To maximize the resemblance of the robustness check to the main model, the infla ted variable used was the natural logarithm of the average amount of employees a firm had in a given year during the acquisitions.
Table 6
Robustness Amount of Settled Filings
Poisson Model Settled Filings Settled Filings Settled Filings
(1) (2) (3)
Total International -0.275 -0.306* -0.317**
(0.171) (0.164) (0.147)
Total Unrelated 0.0461 -0.0379 -0.0975
(0.0936) (0.107) (0.0930)
Average Transaction Value ($USD) 1.93e-05** 1.84e-05** 1.77e-05** (8.36e-06) (8.37e-06) (8.58e-06)
Amount of Employees 0.0253 0.0310 0.0325
(0.0492) (0.0495) (0.0497)
Average Acquiring Firm Value 0.0689 0.0594 0.0691
(0.0479) (0.0485) (0.0491)
Amount of Other Acquisitions 0.0164* 0.0145 0.0131
(0.0094) (0.00952) (0.00981)
Economic State -0.000222** -0.000216** -0.000217**
(9.75e-05) (9.81e-05) (9.81e-05)
Acquisition Rate (ln) 0.374* 1.249***
(0.224) (0.353)
Average Ownership (ln) 0.00495
(0.00364)
Avg. Ownership (ln) * Acquisition rate (ln) -0.0215*
(0.0116) Constant -4.829*** -4.805*** -5.255*** (0.479) (0.476) (1.592) Zero-Inflation model Amount of Employees (ln) -0.00725 -0.00881 0.0265 (0.00610) (0.00652) (0.0182) Constant -12.20*** -11.63*** -16.10*** (0.202) (0.200) (0.288) Nonzero Observations 146 146 146 Zero Observations 16,273 16,273 16,273 Wald's Chi-square 34.14 40.38 65.25 P > χ2 0.000 0.000 0.000
The test was built up in a hierarchical manner, with the first model including only the control variables. In line with the main model, the results indicate a significant positive relationship (P < 0.05) between average transaction value and the likelihood of a filed case of corporate fraud being settled post-acquisition. It further indicates that the economic state negatively influences the likelihood of a filed case corporate fraud being settled post acquisition.
Model 2 introduced the independent variable – acquisition rate. In line with the main model, a marginally significant positive relationship (P < 0.1) was found between the acquisition rate and the likelihood of a filed case of corporate fraud being settled post-acquisition. The marginal significance, as opposed to stronger significance in the main model could be attributable to the smaller size of sample of cases of settled corporate fraud, as opposed to allegations. Nonetheless, the result regarding hypothesis 1 is regarded moderately robust. Finally, the third model introduced the interaction effect of the average level of ownership of those simultaneous acquisitions a firm engaged in in a given year. Unlike the main model with alleged cases of corporate fraud, the average level of ownership does not directly significantly influence the likelihood of corporate fraud occurring when it is measured only by settled filings. However, in accordance with the main results, a marginally significant negative interactions effect (P < 0.1) is found on the relationship between the acquisition rate and the likelihood of settled corporate fraud cases occurring at a given firm post-acquisition.
Again, this marginal significance might be attributable to the smaller sample size used in this robustness check. However, when considering the overall results of the robustness check, it appears that no contradicting results were identified, thereby making the results presented in the main model with alleged fraud moderately robust.
V. Discussion
While previous studies have often analyzed corporate fraud from an individual perspective, this study has focused on the firm level characteristics that enhance the likelihood of corporate fraud occurring. More specifically, this study has a firm level perspective in an M&A setting, by analyzing characteristics surrounding one – or more – simultaneous acquisitions that influence the likelihood of corporate fraud occurring post-acquisition. The findings of this study show multiple novel and interesting insights.
First, a consistent positive relationship was found between firm size and the likelihood of corporate fraud occurring post-acquisition. This finding could be related to the notion that the control of firms because increasingly complex as they increase in size (Demsetz, 1988), thereby increasing the opportunity element of the fraud triangle. This, in turn, increases the likelihood of corporate fraud occurring post-acquisition,
occurring post-acquisition. This result was significant even when controlling for firm size and the total amount of acquisitions, and was moderately robust. This indicates that, as hypothesized, both increased complexity due to an increased acquisition rate, and decreased performance, increases the likelihood of corporate fraud occurring post-acquisition. Third, the average level of integration had a significant direct positive relationship, and a negative interacting relationship with the likelihood of corporate fraud occurring post-acquisition. The findings regarding the direct positive relationship are in accordance with several different literature streams. First, scholars studying the human reactions to changes post-acquisition have taken into consideration the manner in which the level of integration affects individuals in an organization (Cartwright & Cooper, 1993). From that perspective, researchers have argued that as a more thorough level of integration can generate a high level of resistance leading to poor integration results (Haunschild, Moreland, & Murrell, 1994).
A stream related to the human resource perspective is that of organizational culture, in which several authors argued that differences in organizational cultures cause the integration process to become complex (Olie, 1994; Sales and Mirvis, 1984). In line with the marginally significant direct positive relationship between level of integration and likelihood of corporate fraud, Gertsen, Soderberg, & Torp, (1998) have argued that greater cultural differences can generate lower post-acquisition performance. However, in line with the negative interaction
effect found between the level of integration and the relationship between the acquisition rate and the likelihood of corporate fraud occurring, others have argued that cultural differences can be sources of synergies and, therefore, are not necessarily value-destroying (Chaterjee, Lubatkin, Schweiger, & Weber, 1992; Datta, 1991). Others have argued that the goal of integration is to reduce the edge that managers can have over their owners by more closely working together (Miller, Bretton-Miller & Lester, 2010).. Therefore, in the context of likelihood of corporate fraud occurring at a given firm post-acquisition, the most important role of integration is reducing the information asymmetry that exists between shareholders and acquiring managers. The identified negative interaction effect regarding the average level of ownership of the simultaneously acquired firms also supports the notion that the information asymmetry between a firms’ acquiring managers and shareholders is reduced as a new acquired firm becomes more integrated with the acquiring firm.
According to Agnew (2001), the likelihood of corporate fraud occurring increases significantly in situations in which those negative emotions hold strong for a longer period of time and, thus, remain unresolved. This finding is also consistent with arguments brought forward in section II that corporate fraud occurs mainly in order to prevent the stock prices from dropping, as opposed to intending to increase the stock prices through corporate fraud (Johnson, Ryan, & Tian, 2009).
VI. Conclusions
This study made a start in enhancing the understanding why firms find themselves being charged for committing corporate fraud whilst engaging in acquiring behavior. In doing so, this study analyzed post-acquisition cases of corporate fraud filed against the acquiring firm from a firm-level perspective. Using 27,462 acquisitions and 405 cases of alleged corporate fraud, a positive significant relationship was found between the acquisition rate and the likelihood of corporate fraud occurring at a given organization post-acquisition. This means that, as firms engage in more simultaneous acquisitions, the likelihood that cases of corporate fraud will be filed against that firm increases. Furthermore, the average level of post-acquisition integration was found to negatively moderate the relationship between acquisition rate and the likelihood of corporate fraud occurring post-acquisition, indicating that, the more firms integrate their simultaneous acquisitions, the lower the likelihood of
corporate fraud occurring as caused by the firms engaging in multiple acquisitions simultaneously.
Theoretical Implications
The findings of this study have several implications for research. First, the results confirmed the adequate use of an agency lens to build theory in an M&A setting. More specifically, the use of information asymmetry concepts in an M&A setting was particularly novel and fruitful. To our knowledge, this is the first study that has taken such a perspective to study the relationship between corporate fraud and M&A strategies. Second, analyzing corporate fraud at organizational level, as opposed to individual level, this study has linked certain organizational features and configurations to the corporate fraud literature. In doing so, this study has shown that it is fruitful to study antecedents to corporate fraud from the perspective of organization theory, focusing on firm configurations.
Practical Implications
fraud. Third, managers working on the prevention of corporate fraud should consider the influence of economic state when making strategic decisions. More specifically, they should take into consideration that the likelihood of corporate fraud is significantly higher during economic downtimes, and adjust their control mechanisms accordingly.
VII. Limitations & Future
Research
As is the case with most other studies, this study also has several limitations that provide a pathway for future research. First, although studying corporate fraud from an organizational perspective can be particularly interesting, it is always individuals – or a group thereof – that engage in fraudulent behavior, and therefore analyzing such a matter from an individual perspective can provide a more thorough understanding of the motives behind the engagement in corporate fraud. Although, this is indeed the case, the individual perspective was not taken into consideration in this study. This was because the main purpose of this study was to analyze certain post-acquisition organizational configurations that reduce the likelihood of corporate fraud occurring at any given organization. As such, it would be particularly fruitful for future research to expand this study by including individual characteristics or pressures in a contingent manner. This would give a better understanding of, not only which organizational
characteristics reduce the likelihood of corporate fraud, but also of the difference of these organizational characteristics per person or team of individuals.
Second, in this study many alleged cases were taken into account, for they were shown to have a significant negative effect on the defendant in question. However, it could be argued that including also those filings that were dismissed could give a somewhat biased result. Although robustness checks were included to test for the deviance between those that were dismissed and those that settled, future research could focus only on those cases of corporate fraud that were found guilty and therefore had to settle. More specifically, a cross-case analysis of several corporate fraud cases would provide particular fruitful in-depth understanding of the antecedents to corporate fraud, such that future research could determine which what can be empirically proven to most effectively reduce those antecedents and thereby the likelihood of corporate fraud occurring.
differences in the likelihood of corporate fraud occurring. Furthermore, this would substantially increase the generalizability of the study.
Fourth, in this study only the acquisition rate of any given firm was taken into consideration. However, other studies (Kusewitt, 1985) have taken into consideration the variance from the standard amount of acquisitions at any given point in time. Although the acquisition rate shows the amount of simultaneous mergers a firm is engaging in at any point in time, it does not include to which degree they are accustomed to engaging in such an amount of simultaneous mergers. Thus, the arguments relating to chaos due to a higher acquisition rate could become irrelevant. Future research could include the variance of the acquisition rate for a more accurate understanding of the relationship between the chaos and information asymmetry caused by such an abnormally high acquisition rate and the likelihood of corporate fraud occurring at a given firm.
Fifth, in this study, only the years were taken into account in which firms acquired targets. However, no data was included for the years in which firms did not engage in acquisitions. Therefore, the study does not analyze the cases of corporate frauds that are filed against a given firm during years in which they did not engage in acquiring behavior. It could be particularly fruitful for future research to isolate the effect of the acquisition rate on the likelihood of corporate fraud occurring post-acquisition through a complete set of panel data, including both the amount of cases filed against firms during years in which they engaged in
(multiple) acquisitions, as well as in years in which they did not acquire any firms.
Sixth, no differentiation was made between the gravity of a variety of different cases of corporate fraud. Some authors have done so by discussing low, middle, and complex large scales of corporate crimes (Langton & Piquero, 2007). By studying not just the presence of corporate fraud, but also the gravity of each of these cases, researchers can develop a better understanding of where the differences lie between minor cases of corporate fraud versus large corporate scandals.
Seventh, although it was controlled for in this study, no differentiation was made between the targets, whereas other authors have argued that the integration process becomes more difficult in those cases in which the acquiring and target firm are in different industries or countries (Olie, 1994; Miller, Bretton-Miller, & Lester, 2010; Zollo & Singh, 2004). Adding such contingency factors in this study could be particularly fruitful to understand whether a high level of integration uniformly lowers the likelihood of corporate fraud occurring, or whether this is the case in situations in which targets are easier into integrate in the acquiring firm.
the strength of the arguments regarding information asymmetry in this study could be somewhat diminished. It would be particularly fruitful for future research to study the difference in information asymmetry during the period prior to the 1990s in relation to that after.
Acknowledgements
I would like to pay a special thanks to Dr. Killian McCarthy for the supervision and advice provided during the entire process of this thesis. Next I would like to thank Prof. Dr. Dries Faems for the valuable feedback provided on the first versions of this thesis, and thereafter. I would furter like to give a special thank to Marc Donders, Jork Netten, and Timko Ogink, who made the process of writing a thesis team-like. Finally, I would like to thank Prof. Dr. Hossein Dadfar, for steering me in the right direction during the initial phases of the thesis.
References
Aboody, D., & Lev, B. (2000). Information Asymmetry, R&D, and Insider Gains. The
Journal of Finance , 55 (6), 2747-2766.
Agnew, R. (2001). Building on the foundation of general strain theory: Specifying the types of strain most likely to lead to crime and delinquency. Journal of research in crime and
delinquency , 38 (4), 319-361.
Agnew, R., & White, H. (1992). An Empirical Test of General Strain Theory. Criminology, 30 (4), 475-500.
Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human
decision processes , 50 (2), 179-211.
Amihud, Y., & Lev, B. (1981). Risk reduction as a managerial motive for conglomerate mergers. The bell journal of economics , 605-617.
Barney, J. (1991). Firm Resources and Sustained Competitive Advantage. Journal of
Management , 17 (1), 99-120.
Baucus, M., & Near, J. (1991). Can Illegal Corporate Behavior Be Predicted? An Event History Analysis. Academy of Management
Journal , 34 (1), 9-36.
Carleton, R. (1997). Cultural due diligence.
Training , 34, 67-80.
Carpenter, T., & Reimers, J. (2005). Unethical and fraudulent financial reporting: Applying the theory of planned behavior. Journal of Business
Ethics , 60 (2), 115-129.
Cartwright, S. (2006). 30 Years of Mergers and Acquisitions Research: Recent Advances and Future Opportunities. British Journal of
Management , 17.
Cartwright, S., & Cooper, C. (1993). The psychological impact of merger and acquisition on the individual: A study of building society managers. (347, Ed.) Human relations , 46 (3), 327.
Cassiman, B., Colombo, M., Garrone, P., & Veugelers, R. (2005). The impact of M&A on the R&D process: An empirical analysis of the role of technological-and market-relatedness.
Research Policy , 34 (2), 195-220.
Chaterjee, S., Lubatkin, M., Schweiger, D., & Weber, Y. (1992). Cultural differences and share-holder value in related mergers: linking equity and human capital. Strategic Management Journal , 13, 319-334.
Cohen, J., Cohen, P., West, S., & Aiken, L. (2013). Applied multiple regression/correlation
analysis for the behavioral sciences. Routledge.
Cohen, J., Ding, Y., Lesage, C., & Stolowy, H. (2010). Corporate Fraud and Managers' Behavior: Evidence from the Press. Journal of
Business Ethics , 95, 271-315.
Comment, R., & Jarrell, G. (1995). Corporate focus and stock returns. Journal of Financial
Economics , 37 (1), 67-87.
Datastream Thomson Reuters . (2015).
Datastream Professional. Retrieved 2015, from
http://thomsonreuters.com:
http://thomsonreuters.com/en/products-
services/financial/investment-management/datastream-professional.html
Datta, D., Deepak, K., & Grant, J. (1990). Relationships between type of acquisition, the autonomy given to the acquired firm, and the acquisition success: An empirical analysis.
Journal of Management , 16 (1), 29-44.
Demsetz, H. (1988). The Theory of the Firm Revisited. Journal of Law, Economics, &
Organization , 4 (1), 141-161.
Deng, S., Willis, R., & Xu, L. (2012). Shareholder litigation, reputational loss, and bank loan contracting. Journal of Financial and
Quantitative Analysis .
Dyck, A., Morse, A., & Zingales, L. (2010). Who Blows the Whistle on Corporate Fraud?
The Journal of Finance , 65 (6), 2213-2253.
Eisenhardt, K. (1989). Agency Theory: An Assessment and Review. The Academy of
Management Review , 14 (1), 57-74.
Empson, L. (2001). Fear of exploitation and fear of contamination: Impediments to knowledge transfer in mergers between professional service firms. Human relations ,
54 (7), 839-862.
Fowler, K., & Schmidt, D. (1989). Determinants of Tender Offer Post-Acquisition Financial Performance. Strategic Management
Journal , 10, 339-350.
Gertsen, M., Soderberg, A., & Torp, J. (1998).
Cultural dimensions of international mergers and acquisitions (Vol. 85).
Gillett, P., & Uddin, N. (2005). CFO intentions of fraudulent financial reporting. Auditing: A
Journal of Practice & Theory , 24 (1), 55-75.
Haleblian, J., & Finkelstein, S. (1999). The Influence of Organizational Acquisition Experience on Acquisition Performance: A Behavioral Learning Perspective.
Administrative Science Quarterly , 44, 29-56.
Haleblian, J., Devers., C., McNamara, G., Carpenter, M., & Davison, R. (2009). Taking Stock of What We Know About Mergers and Acquisitions: A Review and Research Agenda.
Journal of Management , 35 (3), 469-502.
Haleblian, J., Kim, J., & Rajagopolan, N. (2006). The Influence of Acquisition Experience and Performance on Acquisition Behavior: Evidence from the U.S. Commercial Banking Industry. The Academy of Management Journal , 49 (2), 357-370.
Halpern, P. (1983). Corporate acquisitions: A theory of special cases? A review of event studies applied to acquisitions. Journal of
Finance , 297-317.
Hansen, R. (1987). A theory for the choice of exchange medium in mergers and acquisitions.
Journal of Business , 60, 75-95.
Harrison, J., Hitt, M., Hoskisson, R., & Ireland, R. (1991). Synergies and Post-Acquisition Performance: Differences versus Similarities in Resource Allocations. Journal of Management ,
17 (1), 173-190.
Haspeslagh, P., & Jemison, D. (1991). Managing Acquisitions: Creating Value Through Corporate Renewal. Academy of
Management Review , 18 (2), 370-374.
Haunschild, P., Moreland, R., & Murrell, A. (1994). Sources of Resistance to Mergers Between Groups. Journal of Applied Social
Psychology , 24 (13), 1150-1178.
