Uncovering the Secrets of a Happy Marriage;
The Influence of Interorganizational Diversity on Innovation Performance in
Mergers and Acquisitions
Master Thesis
MSc BA Change Management Faculty of Economics and Business
University of Groningen Rianne Lukkien S2768968 b.r.lukkien@student.rug.nl Supervisor: M. Hanisch Co-assessor: prof. dr. ir. D.J. Langley
Date: 20th of June 2020
2
ABSTRACT
In this study, we research the influence of interorganizational diversity on post-merger innovation performance. By approaching the concept of interorganizational diversity as entailing three different concepts of diversity; corporate cultural distance, power asymmetry and knowledge variety, this study attempts to uncover the underlying mechanisms that influence innovation performance. The goal of this study is to provide more insight on the differing effects of interorganizational diversity found in previous research. We test our hypotheses using a dataset of 210 M&A deals in the biopharmaceutical industry. The results demonstrate that we do not find evidence to support our hypotheses. However, we do find marginally significant effects for firm characteristics; previous M&A experience of the acquirer, and both absolute and relative size of the acquired knowledge base are found to influence post-merger innovation performance. This study advances the existing literature by testing the influence of diversity on innovation performance as entailing three different concepts and argues for future research on the found marginally significant effects of the firm characteristics of both the target and acquirer.
3
Table of content
1. Introduction 4
2. Theoretical Framework and Hypotheses 6
2.1 Mergers, acquisitions and innovation 6
2.2 Interorganizational diversity 6
2.3 Corporate cultural distance 7
2.4 Organizational power asymmetry 8
2.5 Knowledge variety 10
2.6 Conceptual model 11
3. Methodology 12
3.1 Empirical Setting 12
3.2 Data Collection and Sources 13
3.3 Sample 13 3.4 Measures 14 3.4.1 Dependent variable 14 3.4.2 Independent variables 14 3.4.3 Control variables 16 3.5 Method of analysis 18 4. Results 18
4.1 Preliminary data analysis 19
4.1.1 Descriptive Statistics 19
4.1.2 Correlations 20
4.1.3 Multicollinearity 20
4.2 Regression Results and Hypotheses Testing 22
4.3 Alternative analysis 25
5. Discussion and Conclusion 25
4
1. Introduction
A key motivation for mergers and acquisitions is gaining access to new resources and knowledge to improve innovative performance. When two firms decide to engage in such a merger, they sign up for a marriage that will bond their organisations and take the first step in a complex, yet promising process. By combining resources, knowledge and experience of both firms, synergies can be created with the goal to create a better market position of the merged firms then when the firms would not have merged, a principle known as 1+1 > 2 (Brock, 2005), which is in line with the resource based view of a firm (Wernerfelt, 1984).
Even though this might sound promising, the expected increased performance often faces impediments and many mergers ultimately fail. Numbers show that only 23 percent earn their cost of capital, 47 percent of executives leave within the first year and productivity is reduced by up to 50 percent in the first eight months (Huang & Kleiner, 2004). The incompatibility between two merging firms' management systems, management philosophies, values and ethics is a major cause why marriages between two firms often turn out to be unsuccessful (Franck, 1990). The business synergies that served as motivation for the merger may still exist, yet difficulties in information systems, human resource policies and decision-making processes may arise, of which the costs can outweigh the intended synergies.
Interorganizational diversity is often seen as one of the main reasons for the failure of mergers and acquisitions, however, the M&A literature on the influence of diversity between organizations seems to be contradicting in findings. On the one hand, cultural differences between the two firms are found to negatively affect performance as high cultural differences may lead to friction, affecting the integration process, which in turn affects synergy realization and shareholder value creation (Stahl & Voigt, 2008). On the other hand, differences between merging organizations, rather than similarities, are seen as a source for value creation as it expands the knowledge base of the organization and fosters learning and innovation (Vermeulen & Barkema, 2001). These conflicting findings on diversity between organizations cause that the literature is lacking clear insight on the relationship between interorganizational diversity and innovation performance in mergers and acquisitions. Therefore, the goal of this study is to refine the current understanding of the relationship between interorganizational diversity and innovation performance by approaching diversity as entailing different concepts and accordingly test the separate influences on post-merger performance.
5 concentration of valued social assets or resources, and diversity in terms of variety represents the differences in knowledge, information and experience between the members of the organizations. With the concepts coming from different perspectives such as similarity attraction, the principle of requisite variation and distributive justice theory, the concepts each influence organisational performance in different ways.
Therefore, in considering the influence of interorganizational diversity on post merger innovation performance, we propose that diversity concepts should be clearly separated as being distinct types of diversity in order to uncover the underlying mechanisms that might explain the relationship of interorganizational diversity with innovation performance in mergers and acquisitions. Taking the proposed diversity concepts of Harrison and Klein (2007) as a base, in this study, the influence of interorganizational diversity on innovation will be researched with the following research question:
RQ = How does interorganizational diversity in terms of separation, disparity and variation influence post-merger innovation performance?
To answer this research question, we test three hypotheses, which entail the influence of corporate cultural distance as separation, power asymmetry as disparity and knowledge variety as variation between two merging organisations. First, we propose that increased corporate cultural distance is associated with decreased innovation performance as cultural distance between the firms enhances mistrust, affects the integration and is likely to cause conflict (Ahern et al., 2015; Stahl & Voigt, 2008; Weber & Menipaz, 2003). Second, we prose that high power asymmetry is also associated with decreased innovation performance as loss of autonomy amongst management reduces boardroom effectiveness due to stress, conflict and loss of commitment (He & Huang, 2011; Weber & Schweiger, 1992). Third, we propose that, in terms of variety, knowledge variety has an inverted U-shaped relationship with post-merger innovation performance. We argue that knowledge variety enhances innovation performance due to combinatorial opportunities, yet only up to a certain point. After this point, too much variety causes complications in understanding, communication and learning, consequently causing innovation performance to then decrease for high knowledge variety.
We test the proposed hypotheses with a dataset containing 210 merger and acquisition deals in the biopharmaceutical industry, ranging in period from 1997 to 2016. The results of the Ordinary Least Squares regression demonstrate that we do not find support for the proposed hypotheses. Yet, we do gain additional insights from the control variables; a positive relationship for pre-merger innovation performance and marginally significant relationships for pre-merger innovation performance, deal complexity, absolute and relative size of the acquired knowledge base.
6 arguments, this study can serve as a starting point for future research for testing diversity as entailing different concepts. Furthermore, the additional insights gained from the control variables both advance and challenge the existing literature; the findings on the negative relationship between absolute size of the targets’ knowledge base and innovations performance complements the study of Cloodt et al. (2006) and further challenges the findings by Ahuja and Katila (2001). In addition, the findings for relative size of the knowledge bases are in contrast with both previous mentioned studies. These findings therefore encourage future research for further testing on these influences on post-merger innovation performance.
2. Theoretical Framework and Hypotheses
2.1 Mergers, acquisitions and innovation
Taking the perspective of the resource-based view, an organization is seen as the function of its resources and capabilities, which enable an organization to create value. The market position of a firm is determined by the access to resources and use of these resources, which enables the organization to distinguish themselves from competitors and create competitive advantage (Wernerfelt, 1984). Engaging in mergers and acquisitions allows an organisation to gain access to new resources, which originally were non-marketable, or context dependent (James, 2002), creating opportunities for innovation by combining resources and knowledge of both firms. According to Gantumur and Stephan (2012), this combination of two sets of resources and capabilities allows firms to implement expansion strategies quickly and efficiently. In case the merging firms are complementary, the acquiring firm can enter new markets, expand its product portfolio and access new distribution channels (Wernerfelt, 1984). Also, in striving for innovation, acquiring a firm that possesses advanced technology for innovation is often preferred over internal technology development to overcome time pressure (Sears, 2018). However, despite these market-based motivations, success or failure of a merger often depends on the ability of the firms to integrate effectively and act as one, coherent entity (Caroll & Harrison, 2002).
2.2 Interorganizational diversity
7 Arguing from the transaction costs economics theory (Williamson, 1979), Weber and Mayer (2014) proposed that in situations of uncertainty, people make use of their cognitive frames; cognitive structures that allow us to give meaning to and interpret situations. As this entails how people determine their focus, define concepts and interpret information, misalignment of the cognitive frames of two parties can cause disagreement on how to interpret situations (Weber & Mayer, 2014). In turn, on an organizational level, diversity in cognitive frames between two merging parties can cause disagreement on how to act upon occurring events. Furthermore, misalignment of the cognitive frames causes interpretive uncertainty which can result in bargaining costs to come to an agreement, damage the trust relationship between firms and can damage the reputation of the firms when there are many conflicts (Weber & Mayer, 2014). Here, diversity between firms is mainly associated with increased costs due to conflict solving, lengthy negotiation processes or reputation damage.
These findings from previous studies seem to be inconsistent on the influence of interorganizational diversity on post-merger performance as diversity between firms is a major reason for the M&A in order to gain access to new resources, knowledge and capabilities (RBV). Yet, this diversity between firms might cause conflict, hindering the integration process, with the result that additional costs arise from problem solving and prolonged decision making processes (TCE).
According to Harrison and Klein (2007) the conflicting findings in diversity literature can be explained by viewing the current diversity literature as entailing multiple types of diversity under the same label. In M&As, interorganizational diversity on different aspects of the organization can therefore also have differing influences on the innovative performance. As mentioned earlier, Harrison and Klein (2007) propose that diversity entails three concepts: separation, disparity and variety. In this research, interorganizational diversity will be studied following this distinction of diversity by Harrison and Klein (2007), namely: 1) corporate cultural distance as separation, 2) organizational power asymmetry as disparity and 3) knowledge variety.
2.3 Corporate cultural distance
8 affect the socio-cultural integration of the firms (Stahl & Voigt, 2008) and are related to decrease of participation and conflict (Weber & Menipaz, 2003). Such post-acquisition conflict entails that there are intergroup tensions, originating from an “us” versus “them” mentality between the acquiring and the target firm during the process of integration (Sarala, 2010). This especially becomes salient when the acquiring firm does not tolerate the target firm's culture and adaptation is forced through a control mechanism, which can lead to conflict between both top management teams (Weber & Menipaz, 2003).
Also in the decision making processes of a merged top management team, cultural distance is a source for conflict. According to Weber and Menipaz (2003), the top management team holds a set of shared assumptions, on which they rely in the process of solving problems in order to ensure firm survival. Coping with organizational problems involves interpreting information and acting upon this information. In case of cultural distance amongst the acquiring and target organisation, top management of both firms are likely to interpret information differently due to cultural differences, which are an accumulation of past experiences and interpretation of knowledge. This results in differing definitions of the problem and different preferred strategies for action between the two firms. In addition, Pablo (1994) found that managers find it hard to detach themselves from these underlying beliefs and values of their own organization, which in turn can generate friction and conflict when two top management firms need to engage in joint decision making for innovation after a merger.
Aside from generating conflict, cultural distance between firms is found to also increase resistance. In case of a merger with two cultures that are dissimilar, uncertainty may arise on what culture will be prevalent. Gunkel et al. (2015) found that, in change processes, feelings of insecurity of employees were related to active resistance behaviour as well as feelings of dissatisfaction of employees, which was associated with turnover intention and both active and passive resistance. Such resistance can occur in the form of withholding participation or influencing the decision making process by arguing change is not necessary (Armenakis & Bedeian, 1999). Consequently, a reduced level of commitment influences psychological attachment to the firm and in turn affects job performance and turnover (Armenakis & Bedeian, 1999).
Following this reasoning, we propose that high corporate cultural distance between the merging firms influences innovation performance as the separation between the target and the acquiring firm may cause friction leading to resistance and conflict. This leads to the following hypothesis:
H1: High corporate cultural distance is negatively associated with post-merger innovation
2.4 Organizational power asymmetry
9 resulting in disparity in power; power asymmetry. Stahl et al. (2011) define power asymmetry, as “the extent to which there can be an unidirectionality in influence from acquirer to target”. In mergers, the autonomy of the target firm is often reduced as the acquiring firm from then on interferes with the decision-making for future directions for the merged firm (Weber et al., 1996). More specifically, power asymmetry amongst the two merged management teams refers to imbalance in the power-dependence relations between and among group members as they, after the merger, perform separate but interdependent tasks (Van der Vegt et al., 2010).
Changes in power distribution due to interference of the acquiring company may cause feelings of injustice, as it might not align with the perception of fair or expected power distribution of the target company. With the removal of autonomy and power, feelings of procedural injustice may arise as the target management has reduced influence on the decision making process. When these power differences are significant, the dominating management is often found to make decisions unilaterally, which could be considered as unjust by the management team of the target company. This perception of procedural injustice is found to influence job commitment (Klendauer & Deller, 2009), which in turn leads to increased top management turnover (Krug & Aquilera, 2004).
Determining the new status quo, competition for status as a result of autonomy loss amongst the directors is found to undermine effective boardroom interactions and thus influence firm performance (He & Huang, 2011). Van der Vegt et al. (2010) found that high power asymmetry negatively influenced team learning, which in turn influenced team performance. Moreover, loss of autonomy for the target management shows to evoke stress and negative attitudes towards the acquiring firm amongst the top management of the target firm, resulting in loss of commitment and cooperation (Weber & Schweiger, 1992). In addition, high level of power removal of the target executives has previously led to increased job dissatisfaction and increased executive turnover, which in turn influences firm performance (Krug & Aquilera, 2004). Thus, the interference of the parent company’s top management, especially with large power asymmetry, might cause stress and conflict amongst the target management team as they might feel threatened, resulting in reduced cooperation and increased executive turnover.
10 with large post-merger power asymmetry, distrust can jeopardize the intended performance increase as collaboration and knowledge sharing is complicated.
Taking this all together, we propose that organizational power asymmetry is expected to have a negative influence on innovation as it can lead to friction and conflict, affects the integration process and team performance of top management, which in turn affects post-merger innovative performance of the firm. This leads to the following hypothesis:
H2: Organizational power asymmetry is associated with a decrease in post-merger innovation
2.5 Knowledge variety
Diversity in terms of variety represents the difference in knowledge, resources, skills and background between units (Harrison & Klein, 2007). On an individual level, variety between members of a unit is associated with higher levels of creativity, quality of decision-making and performance (Williams & O’Reilly, 1998) due to the heterogeneity of information resources (Harrison & Klein, 2007). On the organizational level, mergers and acquisitions enable organizations to access additional resources and skills (Bena & Li, 2014), which can enhance innovation.
This diversity in knowledge bases is desirable because when they are too similar, the target's knowledge bases are likely to have little to contribute to the acquirers' already existing knowledge base (Ahuja & Katila, 2001). In addition, organisational learning is increased when a firm is exposed to new and diverse ideas based on differences in capabilities between target and acquiring firm and therefore, acquiring diverse external knowledge is seen as a relevant contribution to the firms’ innovative performance (Cloodt et al., 2006). This is especially the case when the merger’s partner technology can fill the gaps of knowledge and capabilities in the acquirers’ portfolio (Bena & Li, 2014), which is known as one of the major motivations for mergers. Therefore, it is expected that innovation performance increases as knowledge variety increases.
11 seen as an essential condition in order to be able to assimilate new knowledge with the acquired knowledge base. From this, we argue that knowledge variety increases innovation performance due to new combinatorial opportunities, yet, beyond a certain point, where knowledge variety further increases, this effect decreases as too much diversity between the two firms troubles communication as there is a lesser shared understanding. This is in line with Lodh and Battaggion (2014), who argue that the knowledge of both firms should be similar enough to understand each other and enable learning, however, should be complementary in order to add value, positioning the optimal level in the middle.
In their study for the relationship between heterogeneity in technological knowledge between firms and innovation, Nooteboom et al. (2007) found an inverted U-shaped relationship between cognitive distance between the parties and the innovative performance of the firm. They argue that similarity in knowledge structures comes with familiarity, which facilitates collaboration. However, too much familiarity takes out the innovativeness of the collaboration of both parties, as parties will not be able to share new knowledge. In addition, Sampson (2007) found that partner technological diversity in alliances contributes most to innovative performance when the technological diversity is moderate, rather than low or high. In cases where technology was too similar, partners had little to learn from each other, yet, when technology bases were too diverse, the partners had difficulties in learning from each other (Sampson, 2007). This is in line with research by Ahuja and Katila (2001), who also propose an inverted U-shape relationship between knowledge base variety and innovation. They argue that there is little impact for highly diverse or highly related knowledge bases, yet a strong effect on innovation performance when the knowledge bases are to some extent related, yet diverse enough to create new knowledge, suggesting the optimum lies in the middle. In addition, Cohen and Levinthal (1990) found that there should be a sufficient level of overlap in knowledge to ensure effective communication, still, knowledge structures should be diverse enough a enhances to create new linkages and associations and by that enhances the organization's innovative capability through the merger, more than it would have independently. Following these arguments, we propose that diversity between merging firms in terms of knowledge variety has an inverted U-shaped relationship with post-merger innovation performance as the increasing effect of knowledge variety only prevails up to a certain level of variety, after which the effect decreases as knowledge variety increases.
H3: Knowledge variety has an inverted U-shaped relationship with innovation in such a way that both high and low variety are negatively associated with post-merger innovation, whereas moderate variety is associated with increased post-merger innovation.
2.6 Conceptual model
12 cultural distance, power asymmetry and knowledge variety following the distinction in types of diversity. We propose that these separate concepts of diversity differ in their influences on post-merger innovation performance; through mechanisms originating from the resource-based view and transaction costs economics theory. The proposed hypotheses of this study are graphically displayed in the figure below.
Figure 1. Conceptual model
3. Methodology
3.1 Empirical Setting
13
3.2 Data Collection and Sources
The data is collected by using a database of 1093 contracts of mergers and acquisitions from the biopharmaceutical industry, originating from the Securities and Exchange Commissions (SEC). Together with five fellow students, an extensive dataset was created on the M&A information of these 1093 contracts; the contracts are used as a base for the dataset, further complemented with publicly available information. Amongst others, the final data set contains data on deal and firm characteristics, financial data, patent data and CEO information. This information is collected for both the target and the acquiring firm. For the acquirer, the data was collected for the year 1-3 before the deal, the year of the deal itself and for 1-3 years after the merger. For the target firm, the data was collected for the year 1-3 before the merger.
For the dependent variable of this study, Innovation Performance, data was collected on the patent history of both the acquirer and the target by using the patentsview.org. This database contains patent data for all US firms. The patent data is also used for the independent variable Knowledge Variety, as it contains also the patent classes per patent, which indicate the technological grouping of the patents and thus represent the knowledge base of the firms. For the independent variable Corporate Cultural Distance, the firms’ mission statements are collected by taking the “about us” section from the website of both the acquirer and the target. Lastly, the deal contracts and press releases were used to collect the data on equity share of the acquirer to measure Power Asymmetry, which were publicly available on the Internet and via SEC.org.
Data on the control variables Deal Type and Contract Length was derived from the deal contract, data on the Acquisition method (hostile vs friendly) was collected through analysing press releases and data on Deal Size was also taken from press releases and Zephyr. Furthermore, ROA and Firm Size data were obtained from financial reports, accessible via SEC.org. Data on Pre-Merger Innovation Performance and Absolute and Relative Size of the Knowledge Base are obtained from the patent dataset. CEO information was collected through BoardEx, LinkedIn and financial reports.
3.3 Sample
14 distribution in type of firms. These firms contain multiple nationalities; yet, most firms have their headquarters in the US (Acquirer 79%, Target 84%).
Figure 2. Type of firms Acquirer and Target
3.4 Measures
3.4.1 Dependent variable
Innovation performance. To measure the dependent variable, post merger innovation performance, we use the number of granted patents per year. According to Comanor and Scherer (1969), patents statistics are highly correlated with the number of new products a firm releases and thus are a good indicator of innovation. Furthermore, patents represent technological novelty and, because of the property rights associated with granted patents, are found to have economic significance (Ahuja & Katila, 2001). We measure post-merger innovation performance as the average number of granted patents of the acquiring firm over the years t+1 to t+3.
3.4.2 Independent variables
15 The mission statements are collected from the “about us” section from the websites of both the acquirer and the target company. For the textual analysis of the mission statement, we follow the method by Hanisch et al. (2018), who developed a dictionary (Appendix B) with words that were reoccurring in mission statements and are related to a certain themes; time, responsibility, motivation, innovation and strategy. The occurrence of the words in a mission statement represents the firms’ focus on a theme and indicates a firms’ cognitive frame (Hanisch et al., 2018). In our study, we use this dictionary to perform the textual content analysis of the mission statements of the target and the acquiring firm. From the results of the similarity analysis we use the cosine similarity measure, which indicates the similarity between the two “about us” sections and represent how similar both companies score on the five dimensions, which in turn indicates how similar they are in their values. This is equal to the level of cultural fit between the two companies. In accordance, corporate cultural distance then is equal to 1 - cultural fit.
Power asymmetry. The independent variable power asymmetry will be measured by the percentage equity share of the target owned by the acquirer after the merger or acquisition. According to Lloyd et al. (1987), ownership of shares comes with control as large shareholders have a substantial influence on the decision-making and performance of the firm. Therefore, the distribution of ownership between the target and the acquirer determines how much control and power the acquirer has over the target firm. In addition, equity is generally the dominant rule for determining and perceiving distributive fairness, following the distributive justice rules (Meyer, 2001). Therefore, we use the equity share of the acquirer after the merger as a measure for power asymmetry, which is retrieved from publicly available documents such as the press releases or the merger contracts itself. Equity share here is mentioned as the percentage owned by the acquirer after the merger, which was transformed to values between 0 and 1. Consequently, we argue that the bigger the equity share of the acquirer is, the smaller the equity share and thus power of the target, which represents power asymmetry.
Knowledge variety. For measuring the knowledge variety between both firm's knowledge bases, we compare the patent base of the firms for the years t-1 to t-3. According to Park et al. (2013) patents represent the result of R&D activities and therefore entail technological insights. To calculate the knowledge variety between the target and the acquiring firm, we follow the method used by Sampson (2007), which in turn is based on research by Jaffe (1986). In measuring knowledge variety, we first determine the technological position of one firm, relative to the other firm (Sampson, 2007). In order to create this variable, like Sampson (2007), we created a patent portfolio per firm for the years before the merger: t-1 to t-3. This portfolio contains all patents and corresponding patent classes, categorized per year granted. Consequently, this patent portfolio represents the knowledge base of a firm before the merger. The overlap between the knowledge bases per year is calculated using the method of Jaffe (1986):
16 Here, F represents a multidimensional vector in which Fi = (Fi1 … Fis) in which Fis represents the number of granted patents and i ≠ j. Consequently, knowledge variety can vary in value from 0 to 1, where 1 indicates the highest possible variety and 0 the lowest possible variety in knowledge between the firms. As we expect a non-linear relationship, we also create the squared variable from knowledge variety, which will be included as an interaction term in the regression analysis.
3.4.3 Control variables
To account for alternative explanations that might influence post-merger innovation performance or the independent variables, we include control variables on the firm, deal and individual level. In this, we have included variables that are found to influence innovation performance in previous research or that we expect to influence results.
Deal level
Deal type. As our sample entails multiple types of deals (e.g. mergers, acquisitions, stock for stock transactions), we include Deal Type as control variable to test if effects significantly differ for different types of deals.
Deal size. According to Uddin and Boateng (2009), deal size is important as this size has several effects on performance. First, deal value may indicate the size of the target and therefore influence performance. Second, larger deals are more likely to be financed with stock and cash financed deals appear to perform better than stock financed deals. Third, smaller deals face less integration issues, which in turn influences performance (Uddin & Boateng, 2009). To control for the effects of deal size on post-merger performance, we include the logarithm of the deal value as control variable.
Acquisition method. Atanassov (2013) studied the threat of hostile takeovers and its effect on innovation performance. He found that the threat of hostile takeovers prevents managers to shrink, causes the to stay focused and to keep innovating in order to maintain high shareholder value. In contrast, Schleifer and Summers (1988) argue that hostile takeovers reduce incentives amongst managers to invest in human capital and innovation, due to power asymmetry compared to shareholders. To account for these effects, we include the binary variable acquisition method in our model for which 1 = hostile, 0 = friendly.
Deal complexity. Deal complexity is found to negatively influence the time for deal completion as there is more disagreement between stakeholders (Lupyeart & Maeseneire, 2015). This disagreement between stakeholders can indicate a difficult future collaboration which in turn can affect innovation performance. The measurement of contract length (in words) is included to account for deal complexity as complex deals encompass more elaborate clauses in their contract, which lengthens the contract.
Firm level
17 resources, which in turn can drive innovation performance. In addition, patenting activity is found to be larger for firms that are bigger in size (Cohen & Levinthal, 1989). To account for such effects in our model, we include the firm size of the acquirer as control variable, which is measured as the natural log of the assets of the acquirer for t-1
Previous M&A Experience. When the acquiring firm has engaged in previous mergers or acquisitions, this can increase organisational learning with regard to the integration process, which allows a firm to become more efficient in defining problems in the merger or acquisition process (Hayward, 2002). This learning could influence innovation performance as a firm might face less integration issues which in turn no longer affects the innovation performance. Therefore, we include the amount of times the acquirer has engaged in a merger or acquisition before the current deal. Previous M&A experience is calculated by counting the occurrence of a firm in the original dataset of 1093 deals.
Pre-merger Innovation Performance. To test for the influence of pre-merger innovation performance, we include the pre-merger innovation performance of the acquiring firm as control variable. This is measured by calculating the average number of granted patents over the years t-1 to t-3, before the merger.
ROA Acquirer. When a firm's financial performance is poor, the firm might be less eager to invest in R&D as not all R&D expenditures pay off (Lin et al., 2006), which affect innovation performance. To account for the influence of previous financial performance of the acquirer, we calculated the return on assets for acquirer in t-1.
Absolute Size Knowledge Base Target. According to Ahuja and Katila (2001), in technological sectors, the absolute size of the acquired knowledge bases influences innovation. They argue that, from a combinatorial perspective, the larger the acquired knowledge base is, the more new combinations (of knowledge) can be generated. To account for this effect, we include the absolute size of the knowledge of the target as control variable in the regression model. For this, we take the calculated sum of patents granted over the previous ten years for t-1.
Relative Size of the Knowledge Base. Ahuja and Katila (2001) argue a different effect for the relative size of the knowledge base of the target compared to the acquirer; a relatively big acquired knowledge base is associated with decreased innovation as the integration of this knowledge is more complex. Therefore, in addition to the absolute size, we also include the relative size of the knowledge base. This is calculated by taking the absolute size of the knowledge base of the target (sum of granted patents over the previous ten years in t-1) divided by the absolute size of the acquirer's knowledge base, calculated in the same way.
Individual level
18 2006). Founders in particular are found to have their personal values and characteristics embedded in the culture (Giberson et al., 2009). Cultural change and integration after the merger could therefore be affected by a CEO conflict as the CEO who is also the founder might not accept changes with regard to management practices or changes in culture. Therefore, we include a binary variable that indicates whether the CEO is also the founder for the target firm (1 = yes, 0 = no), to account for an effect on innovation performance.
CEO Tenure. Musteen et al. (2006) found that CEO’s tend to become more conservative as their tenure increases. In addressing the attitude of CEO’s towards change, Hambrick and Fukutomi (1991) found that a longer tenure led the CEO’s to be extremely committed to their way of management, are resistant to information that is not in line with their own beliefs and tend to have a conservative attitude towards change. As mergers and acquisitions typically entail organizational change, we control for CEO tenure of the acquiring firm, as a longer tenure is be associated with resistance towards diversity, which affects integration of the target and consequently affects innovation performance. CEO tenure is measured in the number of months the CEO is in the current position of CEO.
3.5 Method of analysis
To analyse our data, we make use of the statistical software Stata, version 16.0 (StataCorp, 2019). We conduct an Ordinary Least Squares (OLS) regression to uncover expected linear relationships between the independent and the dependent variable. The Ordinary Least Squares regression enables us to predict values of a continuous response (Hutcheson, 2011): in this study the post-merger innovation performance of the acquiring firm. To statistically confirm that we were allowed to perform an OLS regression, we tested the assumptions that are fundamental to OLS. In testing the assumption for homoscedasticity, the results (Appendix C) showed that the sample was heteroscedastistic. To prevent this from generating biased results, we used the robust standard errors in performing the regression analysis. In the regression analysis, we test three models of which the first one will test the effect of the control variables on the dependent variable innovation performance. This model will function as the baseline model. In the second model, the independent variables will be added to test for linear effects. Lastly, we include the interaction effect for knowledge variety (knowledge variety squared) to test for a curvilinear relationship.
4. Results
19
4.1 Preliminary data analysis 4.1.1 Descriptive Statistics
Table 1 shows the descriptive statistics of the dependent, independent and control variables. The dependent variable Innovation Performance shows an average of 41.61 (SD = 101.89), meaning that the average post-merger innovation performance in the years t+1, t+2 and t+3 is 41.61 granted patents. For the independent variable Corporate Cultural Distance we find an average of 0.73 (SD = 0.14), indicating that on average, the target and acquiring firm are more distant than close to each other. For Power Asymmetry the sample shows an average of 0.96 (SD = 0.13), which was expected, as the majority of the contracts in the sample were acquisition contracts in which the acquirer owns 100% of the equity of the target after the acquisition. For the variable knowledge variety, the sample shows an average of 0.54 (SD = 0.40). The control variables Acquisition Method, and CEO is also Founder are binary variables. In the sample, 4 out of 210 deals were characterized as hostile (1.90%). Amongst the acquiring firm, there were 32 observations in which the CEO was also the founder (15.24%). Looking at the CEO tenure at the acquiring firms, we find the average tenure is 74.17 months (SD = 71.64).
Descriptive Statistics
Variable Mean S.D. Min Max
Post-merger Innovation Performance 41.61 101.89 1.00 787.00 Corporate cultural distance 0.73 0.14 0.40 0.99 Power asymmetry 0.96 0.13 0.05 1.00 Knowledge variety 0.54 0.40 0.00 1.00
Deal size 19.33 2.26 13.12 24.94
Acquisition method (1 = hostile; 0 = friendly) 0.02 0.14 0.00 1.00 Deal complexity (contract length) 39724.57 13607.20 10161.00 103534.00 Firm size - Acquirer 20.82 2.61 12.73 25.49 Previous M&A experience - Acquirer 1.45 1.87 0.00 11.00 Pre-merger Innovation Performance 43.14 113.32 0.67 949.33 Return on assets - Acquirer -1.95 26.03 -376.83 0.78 Absolute size knowledge base - Target 53.45 221.52 0.00 2583.00 Relative size of the knowledge base 0.90 3.34 0.00 36.50 CEO is also founder – Acquirer 0.15 0.36 0.00 1.00 CEO tenure - Acquirer (in months) 74.17 71.65 1.00 386.00 Note: number of observations: 210
20
4.1.2 Correlations
Table 2 demonstrate the correlations between the variables. The majority shows no significant correlation. In total, the results show 41 significant correlations (p < 0.05), of which most demonstrate a weak correlation (r < 0.3). Then, there are some significant correlations that show a moderate correlation (r > 0.3). We find that Corporate Cultural Distance negatively correlates with Deal Size and with Firm Size of the Acquirer (p < 0.01), indicating that increases in deal size and firm size are both associated with reduced corporate cultural distance.
Furthermore, Deal Size also moderately correlates with Knowledge Variety (p < 0.01), Deal Complexity, (p < 0.01) and Absolute Size of the Knowledge Base Target (p < 0.01) and shows a strong positive
correlation with Firm Size (r = 0.67, p < 0.01). In addition, we find that Firm Size of the Acquirer is positively correlated with both Previous M&A experience (p < 0.01) and Pre-merger Innovation Performance, indicating that bigger firms are associated with increased pre-merger innovation performance and more previous M&A experience. Looking at the dependent variable Innovation Performance we find a strong positive correlation with Pre-merger Innovation Performance (r = 0.95, p <
0.01). This means an increase in pre-merger performance is associated with increased post-merger innovation performance. Furthermore, we see that increased Absolute Size of the Knowledge Base of the Target is also associated with increased Pre-merger Innovation Performance of the Acquirer (p < 0.01). In addition, we find a moderate correlation for Innovation Performance with Absolute Size of the Knowledge Base of the Target (p < 0.01) and with Firm Size of the Acquirer (p < 0.01). Deal Complexity is negatively correlated with Knowledge Variety (p < 0.01) and lastly, the results demonstrate a moderate positive correlation between CEO is also founder Acquirer and CEO tenure of the Acquirer (p < 0.01) indicating that when a CEO is also the founder of a company, the CEO generally has a longer tenure.
4.1.3 Multicollinearity
21
Correlation matrix
Variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
(1) Post-merger Innovation Performance 1.00
(2) Corporate cultural distance -0.23** 1.00
(3) Power asymmetry -0.11 0.00 1.00
(4) Knowledge variety -0.01 0.20** 0.04 1.00
(5) Deal size 0.21** -0.38** 0.06 -0.40** 1.00
(6) Acquisition method (hostile vs. friendly) 0.05 -0.15* 0.04 -0.17* 0.20** 1.00 (7) Deal complexity (contract length) -0.04 -0.20** 0.07 -0.34** 0.35** 0.13 1.00
(8) Firm size - Acquirer 0.43** -0.37** 0.00 -0.21** 0.67** 0.16* 0.10 1.00
(9) Previous M&A experience - Acquirer 0.14* -0.21** 0.06 -0.14* 0.30** 0.00 0.08 0.38** 1.00
(10) Pre-merger Innovation Performance 0.95** -0.23** -0.13 -0.04 0.20** 0.04 -0.02 0.41** 0.09 1.00
(11) ROA - Acquirer 0.03 0.06 -0.01 0.07 0.01 0.01 -0.12 0.23** -0.01 0.03 1.00
(12) Absolute size knowledge base - Target 0.30** -0.25** -0.20** -0.19** 0.37** 0.08 0.09 0.25** 0.23** 0.35** 0.02 1.00 (13) Relative size knowledge base -0.09 0.05 -0.02 -0.18** -0.01 -0.02 0.20** -0.18** -0.06 -0.09 -0.00 0.06 1.00 (14) CEO is also founder - Acquirer -0.12 0.07 -0.07 0.05 -0.15* -0.06 0.01 -0.25** -0.06 -0.12 -0.17** -0.09 0.01 1.00
(15) CEO tenure - Acquirer -0.10 0.00 -0.07 -0.04 0.04 -0.10 0.06 0.00 0.08 -0.11 -0.02 -0.08 -0.08 0.43** 1.00 Note: Number of observations: 210. Significant levels at * p < 0.05; ** p < 0.01
22
4.2 Regression Results and Hypotheses Testing
To test the proposed hypotheses, we conducted an Ordinary Least Squares (OLS) regression. Table 3 displays the results of the regression, with the variable Innovation Performance as dependent variable and the standardized values of the independent and control variables. To judge the goodness of fit of the models, we look at the Log Likelihood values, which display a slight increase over the models. In addition, looking at the AIC and BIC values to determine the probability of a model to minimize information loss, the models show an increase over the models, with model 3 having the highest values. The R-squared is the same for all three models (R2 = 0.92), indicating that the model is able to explain a relatively large part of the variance in the model.
First, in Model 1, we included all control variables and tested the effect on the dependent variable Innovation Performance, which represents the post-merger innovation performance. The results demonstrate a significant effect for Pre-merger Innovation Performance (b = 84.02, p < 0.001), indicating that when the acquiring firm has higher pre-merger innovation performance, their post-merger innovation performance will also be higher. Furthermore, we find some of the control variables to be marginally significant (p < 0.1). We find that Previous M&A Experience Acquirer is marginally significant (b = 2.50, p < 0.1) and shows a positive effect, meaning that when the acquirer has more previous experience, their post-merger innovation performance increases. Also, Absolute Size of the Knowledge Base of the Target is marginally significant (b = -7.11, p < 0.1) indicating that increases in the knowledge base of the target are associated with decreases in post-merger innovation performance. Looking at the Relative size of the knowledge base, we find a positive marginally significant effect (b = 15.32, p < 0.1), indicating that the smaller the size difference between the higher post-merger innovation performance is. Lastly, we find a marginally significant effect for Deal Complexity (contract length) (b = -4.30, p < 0.1) indicating that more complex deals would have a lower post-merger innovation performance.
Second, in Model 2, we added the independent variables Cultural Distance and Power Asymmetry. Looking at the effect of cultural distance on innovation performance, we do not find a significant effect (b = -0.43, p > 0.05) and therefore we do not find support for hypothesis 1. Looking at the effect of power asymmetry on innovation, we also do not find a significant effect (b = -0.99, p > 0.05), indicating that there is no support for hypothesis 2. To account for a possible linear effect of knowledge variety (non-squared), we also included the variable Knowledge Variety in Model 2. The regression results demonstrate that also for knowledge variety there is no significant effect on innovation performance (b = 2.52, p > 0.05).
23 OLS Regression Results
(1) (2) (3)
VARIABLES Model 1 Model 2 Model 3
Control variables
Deal type = All Stock Transaction -4.36 -5.27 -5.38
(4.35) (7.19) (7.20)
Deal type = Majority Stock Purchase 6.38 4.98 4.74
(6.23) (7.16) (7.16)
Deal type = Merger -5.38 -5.14 -5.20
(6.66) (7.06) (6.93)
Deal type = Minority Stock Purchase 1.92 4.51 4.21
(8.27) (12.47) (13.45)
Deal type = Reverse Merger -0.43 -0.31 -0.23
(4.36) (4.32) (4.56)
Deal type = Stock for Stock -5.76 -4.34 -4.51
(4.87) (5.84) (5.44)
Deal type = Stock for Stock Merger -27.19 -26.08 -26.03
(20.12) (20.34) (20.53)
Deal size ($) 3.75 4.67 4.66
(4.50) (4.97) (5.01)
Acquisition method (hostile vs. friendly) 7.66 9.29 9.57
(12.00) (12.07) (12.29)
Deal complexity (contract length) -4.30+ -3.74+ -3.74+
(2.53) (2.26) (2.25)
Firm size - Acquirer 1.84 1.68 1.66
(4.64) (4.75) (4.75)
Previous M&A experience - Acquirer 2.50+ 2.60* 2.60*
(1.28) (1.28) (1.27)
Pre-merger innovation performance 84.02*** 83.88*** 83.82***
(4.41) (4.39) (4.37)
ROA - Acquirer -0.26 -0.38 -0.38
(0.74) (0.83) (0.83)
Absolute size of knowledge base - Target -7.11+ -7.28+ -7.27+
(4.16) (4.03) (4.02)
Relative size knowledge base 15.32+ 18.25+ 18.22+
(9.11) (9.95) (9.99)
CEO is also founder - Acquirer 2.18 1.81 1.79
(5.13) (5.32) (5.35)
CEO tenure (in months) - Acquirer -0.01 -0.01 -0.01
Independent variables (0.02) (0.02) (0.02) Cultural distance -0.43 -0.41 (2.30) (2.25) Power asymmetry -0.99 -1.02 (2.38) (2.41) Knowledge variety 2.52 3.64 Interaction term (2.19) (9.47)
Knowledge variety x Knowledge variety -1.22
(11.50)
R-squared 0.92 0.92 0.92
AIC 2056 2060 2062
BIC 2119 2134 2139
Log-likelihood -1009 -1008 -1008
24 To test the third hypothesis; the inverted U-shaped relationship between knowledge variety and innovation performance, we included Knowledge Variety Squared as an interaction term in Model 3 (knowledge variety * knowledge variety). To test for an (inverted) U-shaped relationship, there is a three-step procedure one should follow (Lind and Mehlum, 2010; Haans et al., 2016). First, the β2 coefficient should be significant and negative in order to indicate an inverted U-shape. Second, the slopes of the U must be steep enough at both sides to exclude other curvilinear relationships and third, the turning point of the (inverted) U-shape should be well located within the data range (Haans et al., 2016). The results of the regression analysis (Model 3) show that there is no significant effect for the interaction term (b = -1.22, p > 0.05), which means that the model already cannot conform to the first requirement for the inverted U-shape. Even though the effect is negative, it is not significant. Therefore, the second and third step in the process of testing for an inverted U-shape are irrelevant for this study. However, to visualize the relationship between knowledge variety and innovation performance and to exclude a non-linear relationship, we plotted the marginal effects of the interaction term (see Figure 3). The marginal effects plot shows the marginal effects for different values of the interaction term Knowledge Variety Squared. As this study proposes an inverted U-shaped relationship between innovation performance and knowledge variety, we would expect the marginal effect to first increase for low values of knowledge variety, yet later to decrease for higher values of knowledge variety squared. However, the marginal effects plot visualizes only decreasing effects for higher values of knowledge variety squared, which confirms that we do not find support for our third hypothesis.
Figure 3. Marginal Effects of Knowledge Variety Squared
25 Acquirer shows a significant effect in this model (b = 2.60, p < 0.05). Both Absolute Size of the Knowledge Base (b = -7.27, p < 0.10) and Relative Size of the Knowledge Base (b = 18.22, p < 0.10) remain marginally significant as well as Deal Complexity (b = -3.74, p < 0.10).
4.3 Alternative analysis
As alternative analysis we performed another regression analysis in which we used a different measure for our dependent variable Innovation Performance. In the initial analysis, we used the average number of granted patents for 1 to 3 years after the mergers. Yet, in order to determine the difference between pre
and post merger performance, an alternative measurement of Innovation Performance is generated by calculating the difference between pre and post merger performance (post minus pre). In this, post-merger performance is the number of granted patents of the acquirer in the years t+1 to t+3 and pre-merger performance is calculated by summing the number of granted patents of the target and acquirer for the years t-1 to t-3 and taking the average of this. The results (Appendix E) demonstrate no significant results for the independent variables Cultural Distance, Power Asymmetry and Knowledge Variety (p > 0.05). Also the interaction term Knowledge Variety Squared does not show significant results in Model 3 (p > 0.05). Therefore, with this analysis we still do not find support for our hypotheses. Furthermore, we find that Premerger Innovation Performance still has a significant effect, yet now has a negative effect (b = -14.66, p < 0.001), which is expected as a higher pre-merger performance reduces the relative difference post-merger performance. In addition, the Relative Size of the Knowledge base also shows a significant effect (b = 41.68, p < 0.05), as well as the Absolute Size of the Knowledge base (b = -52.84, p < 0.001),
which both were only marginally significant in our initial analyses. Furthermore, we find a marginally significant effect for Deal Complexity (b = -7.92, p < 0.10), which was also the case in our initial analysis. Looking at the R-squared, we find this alternative analysis to have a lower R-squared across all models (R2 = 0.69 in Model 1 and 2, R2 = 0.70 in Model 3) compared to our initial analysis (R2 = 0.92).
5. Discussion and Conclusion
The following section discusses the main findings resulting from the analyses. Furthermore, we provide both theoretical and managerial implication for these findings to elaborate on what the findings mean for researchers and the use for managers in practice. Lastly, the limitations of this study will be discussed as well as the suggestions for future research.
5.1 Main findings
26 The purpose of this study was to provide more insight in the underlying mechanisms that cause the differing effects of interorganizational diversity previous studies found on post-merger performance of firms. More specifically, we examined the effects of interorganizational diversity on post-merger innovation performance, because acquiring innovative capabilities is an important motivation to engage in mergers or acquisitions (Hagedoorn & Duysters, 2002). With the distinction made between different concepts of diversity, following Harrison and Klein (2007), we tested diversity in terms of separation (corporate cultural distance), disparity (power asymmetry) and variety (knowledge variety), in which both corporate cultural distance and power asymmetry were expected to be negatively associated with innovation performance and knowledge variety was proposed to have an inverted U-shaped relationship with innovation performance.
An Ordinary Least Squares regression, using 210 merger and acquisition deals, was conducted to test this studies’ hypotheses, however, showed no significant effect for both corporate cultural distance and power asymmetry. Also, for the proposed inverted U-shape, that would represent the relationship between innovation performance and knowledge variety, we did not find a significant effect. Therefore, we conclude that we did not find evidence to support our hypotheses.
However, we did find significant effects for the certain control variables. We found that pre-merger innovation performance of the acquiring firm has a positive influence on post-pre-merger innovation performance, which indicates that higher pre-merger innovation performance is associated with increased post-merger innovation performance. Furthermore, we find marginally significant effects for both the absolute and relative size of the knowledge base of the target as well as for deal complexity (contract length). Lastly, we find a marginally significant effect for Previous M&A Experience of the Acquirer in the baseline model and a significant effect in the full specification model, indicating more previous M&A experience of the acquiring firm is associated with increased innovation performance.
5.2 Theoretical implications
Even though we did not find support for our hypotheses, this study contributes to the existing M&A literature by addressing the conflicting previous results of diversity and testing the types diversity following the distinction as proposed by Harrison and Klein (2007). Previous studies that used such a distinction were often focused on within organization diversity and address diversity on team and individual level (e.g. Van Knippenberg & Schippers, 2007; Bear et al., 2010; Kearney & Gebert, 2009). Moreover, these studies did not address interorganizational diversity in the context of mergers and acquisitions. For research that did specify their study on diversity in the field of M&A’s, the studies entail often only one construct of diversity (e.g. Van de Ven et al., 2008, Sampson 2007). Thus, this study distinguishes itself within the existing literature by researching three conceptualizations of diversity on an interorganizational level, in the context of mergers and acquisitions and more specifically, the constructs are tested for their influence on innovation performance rather than the market-based performance.
27 positive relation between previous M&A experience of the acquirer and post-merger innovation performance. This is in line with previous research on the influence of M&A experience on merger successes (Barkema & Schijven, 2008; Hayward, 2002). They argue that the organisational learning, from both negative and positive experience on the integration process, enables the acquiring firm to be more efficient in detecting and addressing integration problems. However, Hayward (2002) and Barkema and Schijven (2002) both mainly focus on acquisition performance in terms of market-based performance. The results of our study display that previous M&A experience also influences the innovation performance of the acquiring firm. As the results were only significant for model 2 and 3, and not in the baseline model, further research is needed to provide more evidence on the relationship between previous M&A experience and post-merger innovation performance.
Second, this study found marginally significant effects for both absolute size of the knowledge base of the target firm and relative size of the knowledge base of the target compared to the acquiring firm. The found negative effect of the absolute size of the knowledge base is in contrast with findings by Ahuja and Katila (2001) who argue a positive relationship between absolute size and innovation performance, but are in line with Cloodt et al. (2006) who also found a negative effect. Even though only marginally significant, our findings contribute to the existing literature by questioning the earlier found positive relationship (Ahuja & Katila, 2001) with innovation performance. In addition to the research by Cloodt et al. (2006), our findings signal the need for further examination on the absolute size of the knowledgebase of the target firm and its influence on post-merger innovation performance.
Third, looking at the relationship between innovation performance and relative size of the knowledge base of the target, we find a marginally significant positive relationship, which is in contrast with previous findings on influence of relative size of the knowledge base (Ahuja & Katila, 2001), and is also in contrast with findings by Cloodt et al. (2006). We found that when the acquired knowledge base is relatively large with regard to the acquirers knowledgebase, this increases innovation performance, where previous have found the opposing effect. Even though the effects were only marginally significant, these results contribute to the existing literature by providing a starting point for future research that might further test the relationship between both size of the knowledge base (both absolute and relative) and innovation performance.
28
5.3 Managerial Implications
In addition to theoretical implications, this study also is of value for managers, consultants or directors who come in contact with merger and acquisition decision-making or are involved in the post-merger integration process. First of all, this study claims attention for the need to distinguish diversity as entailing different concepts. In the process of assessing a firm as a potential target for a merger or acquisitions, directors need to be aware that diversity should be seen as entailing different concepts, which can influence the innovative performance in different ways. The theoretical framework of this study synthesizes important streams of literature on diversity in mergers and acquisitions, which can be used as insights for the target selection process. Second, the effects of absolute and relative size of the knowledge base provide managers with two more aspects that need to be taken into consideration when selecting a target firm as these factors may influence post-merger innovation performance.
Furthermore, the notion that previous M&A experience influences innovation performance can be used to create awareness amongst managers that, may their firm have not been previously engaged in a merger or acquisition process, there is a risk of lacking the sufficient knowledge and skills to enhance the success of the merger. The insights from this study can therefore function as a motivation for managers to include factors as diversity, but also pre-merger performance, previous M&A experience and absolute and relative sizes of the acquired knowledge bases, in their evaluation process for target selection and development of integration strategies.
5.4 Limitations
This research was subject to certain limitations. First of all, the field of M&A’s is considered to be complex in nature due to the many contextual and process factors that influence research results. This wide span of contextual factors, which can be interrelated or have conflicting findings, is seen as one of the causes that M&A research often shows insignificant and inconsistent findings (Weber, 2012).
29 A third limitation of this study concerns the measurement of cultural distance. For the measurement of corporate cultural distance, we calculated similarity on the basis of mission statements collected from the (historical) website of the firms. As our sample contains deals ranging from 1997 to 2016, for the older deals many mission statements of the target were not accessible anymore as they have been integrated into the acquiring company for years already. This led to a great decrease in the sample size due to the missing values for corporate cultural distance. Future research might consider using other measures or corporate cultural distance, or should limit their span of deals to more recent years. Even though we consider the measurement of cultural distance the most suitable given the size of the dataset, fact remains that the concept of culture is a hard construct to quantify and measure. The literature on culture entails many different definitions, yet have in common that they consider culture as a complex multi-level construct (Taras et al., 2009). In terms of culture defined by Schein (1990), it could be that the similarity measurement of mission statements only captures the artefacts and values, yet fails to capture the underlying assumptions from mission statements as they entail feelings, perceptions and thoughts.
A fourth limitation regards measurement for innovation, more specifically; the consideration of granted versus filed patents. As we retrieved our patent data from patentsview.org, we only included granted patents, which are a good indication of innovative performance of a firm (Sampson, 2007). However, sites such as lens.org also display the number of filed patents, which quite often differs from the number of granted patents in a year. A firm could have filed 20 patents, which would be representative of a good collaboration with no conflicts regarding diversity, yet; only two patents are being granted, indicating disparity in generated innovative performance. This is in line with Han et al. (2018) who argues that a major limitation of using patent data for measuring innovation is that patents do fully equal innovation as not all innovations are patented. Consequently, Lanjouw and Schankerman (2004) made a distinction between research productivity and patent quality, where productivity represents the R&D output and quality entails the generation of new technology. Including this distinction between productivity and quality could possibly further uncover the mechanisms that drive the relationship between interorganizational diversity and innovation performance and therefore forms an interesting topic for future research.
5.5 Future Research
Even though this study was not able to confirm the proposed hypotheses, we believe that this study can function as a starting point for future research as it addresses the need to uncover the underlying mechanisms that cause the differing effects of diversity.
30 might be seen as the main motivation for the merger. Given these industry characteristics such as competitiveness and pressure for innovation might influence the relationship between interorganizational diversity and post-merger innovation performance. Therefore, it can be interesting for future research to study how merger behaviour and in particular, the diversity between organisations and its influence on performance, might differ across different industries.
31
References
Ahern, K. R., Daminelli, D., & Fracassi, C. (2015). Lost in translation? The effect of cultural values on mergers around the world. Journal of Financial Economics, 117(1), 165-189.
Ahuja, G., & Katila, R. (2001). Technological acquisitions and the innovation performance of acquiring firms: A longitudinal study. Strategic management journal, 22(3), 197-220.
Armenakis, A. A., & Bedeian, A. G. (1999). Organizational change: A review of theory and research in the 1990s. Journal of management, 25(3), 293-315.
Atanassov, J. (2013). Do hostile takeovers stifle innovation? Evidence from antitakeover legislation and corporate patenting. The Journal of Finance, 68(3), 1097-1131.
Babnik, K., Breznik, K., Dermol, V., & Trunk Širca, N. (2014). The mission statement: organisational culture perspective. Industrial Management & Data Systems, 114(4), 612-627.
Barkema, H. G., & Schijven, M. (2008). How do firms learn to make acquisitions? A review of past research and an agenda for the future. Journal of Management, 34(3), 594-634.
Bear, S., Rahman, N., & Post, C. (2010). The impact of board diversity and gender composition on corporate social responsibility and firm reputation. Journal of business ethics, 97(2), 207-221.
Bellinger, L., & Hillman, A. J. (2000). Does tolerance lead to better partnering? The relationship between diversity management and M&A success. Business & Society, 39(3), 323-337.
Bena, J., & Li, K. (2014). Corporate innovations and mergers and acquisitions. The Journal of Finance, 69(5), 1923-1960.
Brock, D. M. (2005). Multinational acquisition integration: the role of national culture in creating synergies. International business review, 14(3), 269-288.
Carroll, G. R., & Harrison, J. R. (2002). Come together? The organizational dynamics of post-merger cultural integration. Simulation Modelling Practice and Theory, 10(5-7), 349-368.
Cloodt, M., Hagedoorn, J., & Van Kranenburg, H. (2006). Mergers and acquisitions: Their effect on the innovative performance of companies in high-tech industries. Research policy, 35(5), 642-654.
Cohen, W. M., & Levinthal, D. A. (1989). Innovation and learning: the two faces of R & D. The economic journal, 99(397), 569-596.
Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative science quarterly, 35(1), 128-152.
Comanor, W. S., & Scherer, F. M. (1969). Patent statistics as a measure of technical change. Journal of political economy, 77(3), 392-398.
Comanor, W. S., & Scherer, F. M. (2013). Mergers and innovation in the pharmaceutical industry. Journal of health economics, 32(1), 106-113.
32 Datta, D. K. (1991). Organizational fit and acquisition performance: Effects of post‐acquisition integration. Strategic management journal, 12(4), 281-297.
Deshpande, R., & Webster Jr, F. E. (1989). Organizational culture and marketing: defining the research agenda. Journal of marketing, 53(1), 3-15.
Ferris, S. P., Jayaraman, N., & Sabherwal, S. (2013). CEO overconfidence and international merger and acquisition activity. Journal of Financial and Quantitative Analysis, 48(1), 137-164.
Franck, G. (1990). Mergers and acquisitions: competitive advantage and cultural fit. European management journal, 8(1), 40-43.
Gantumur, T., & Stephan, A. (2012). Mergers & acquisitions and innovation performance in the telecommunications equipment industry. Industrial and Corporate Change, 21(2), 277-314.
Giberson, T. R., Resick, C. J., Dickson, M. W., Mitchelson, J. K., Randall, K. R., & Clark, M. A. (2009). Leadership and organizational culture: Linking CEO characteristics to cultural values. Journal of Business and Psychology, 24(2), 123-137.
Gleadle, P., Parris, S., Shipman, A., & Simonetti, R. (2014). Restructuring and innovation in pharmaceuticals and biotechs: The impact of financialisation. Critical Perspectives on Accounting, 25(1), 67-77.
Gunkel, M., Schlaegel, C., Rossteutscher, T., & Wolff, B. (2015). The human aspect of cross-border acquisition outcomes: The role of management practices, employee emotions, and national culture. International Business Review, 24(3), 394-408.
Haans, Richard FJ, Constant Pieters, and Zi‐Lin He. "Thinking about U: Theorizing and testing U‐and inverted U‐shaped relationships in strategy research." Strategic Management Journal 37.7 (2016): 1177-1195.
Hagedoorn, J., & Duysters, G. (2002). External sources of innovative capabilities: the preferences for strategic alliances or mergers and acquisitions. Journal of management studies, 39(2), 167-188.
Hair, J. F., Jr., Anderson, R. E., Tatham, R. L. and Black, W. C. (1995) Multivariate Data Analysis, 3rd ed, Macmillan Publishing Company, New York.
Hambrick, D. C., & Fukutomi, G. D. (1991). The seasons of a CEO's tenure. Academy of management review, 16(4), 719-742.
Han, J., Jo, G. S., & Kang, J. (2018). Is high-quality knowledge always beneficial? Knowledge overlap and innovation performance in technological mergers and acquisitions. Journal of Management & Organization, 24(2), 258-278.
Hanisch, M., Haeussler, C., Graf-Vlachy, L., Konig, A. S., & Cho, T. (2018, July). The Influence of Cognitive Frame Similarity on Contingency Planning in Strategic Alliances. In Academy of Management Proceedings (Vol. 2018, No. 1, p. 12398). Briarcliff Manor, NY 10510: Academy of Management.
Harrison, J. S., Hitt, M. A., Hoskisson, R. E., & Ireland, R. D. (1991). Synergies and post-acquisition performance: Differences versus similarities in resource allocations. Journal of management, 17(1), 173-190.
