“Organizational Ambidexterity – a Question of Control?”

(1)

“Organizational Ambidexterity –

a Question of Control?”

Empirical Analysis of the Impact of the Control Perspective on

Organizational Ambidexterity

A.C. Schneckenreither

S3149447

Supervisor: Prof. Dr. Jana D.R. Oehmichen Co-assessor: Dr. Charlie Carroll

Groningen, June 25th₂₀₁₈

Word count: 10.362

Master Thesis

MSc. BA Strategic Innovation Management Faculty of Economics and Business

(2)

LIST OF FIGURES AND TABLES

Figure 1: Conceptual model ... 11

Figure 2: Interaction effect of model 2 ... 17

Figure 3: Interaction effect of model 3 ... 18

Table 1: Descriptive statistics and correlation matrix ... 016 Table 2: OLS regressions on OA ... 20

(3)

ABSTRACT

In this paper, I examine the effects of duality and independence on organizational ambidexterity (OA) (i.e. simultaneous exploration and exploitation) and the moderating role of human capital. The theory is tested on a longitudinal sample of listed U.S. firms. Drawing on a conjoint view of the agency and resource dependence theories, I claim and confirm that CEO duality affects OA negatively, whereas board independence is positively associated with OA. These effects build on the idea that the dual CEO/chairman position dominates the board of directors by controlling information and the decision-making process and inhibits the function of the board as a monitoring role. Independence however increases efficiency and transparency as outside directors boost the effectiveness of monitoring and control. The strength of the duality effect is contingent upon human capital differences, namely time to retirement and number of qualifications. My results indicate that lower time to retirement and a higher number of qualifications strengthen the relationship between duality and OA. No significant effect was found on the relationship between independence and OA. Thus, the composition of the top management matters greatly when it comes to OA strategies. I conclude with implications for theory and practice, as well as key directions for future research.

(4)

1. INTRODUCTION

To increase the chances of a firm’s survival and to adopt the ability of meeting complex challenges due to globalization and continuous technological changes, management must embrace both exploitation of existing resources and assets as well as exploration of new market opportunities (e.g. Hill & Birkinshaw, 2014; Piao, 2010). Organizational ambidexterity (OA) comprises of exploration and exploitation, an ability a firm needs to effectively renew its knowledge base and compete in the market (O'Reilly & Tushman, 2013). Whereas exploration strategies include risk-taking, flexibility, experimentation and autonomy in behavior, exploitation activities are supported by efficiency-seeking, refinement, control and incremental improvement (March, 1991; O'Reilly & Tushman, 2013). Moreover, exploration helps to align a firm’s focus long-term, in contrast to the exploitation approach which is more predictable in the short-term. Due to the different structure of results of these two learning strategies, Piao (2010) showed that being able to commit to both schemes significantly improves the chances of long-term success and survival as well as being more innovative and competitive. Admittedly, the ability to balance the two divergent approaches of organizational learning can be quite challenging for many firms. This issue derives from the fact that the nature of these activities is paradoxical. Hence, pursuing both strategies could indeed be inflicted with trade-offs. However, recent studies indicate that implementing both learning concepts is both possible and needed (Raisch, Birkinshaw, Probst & Tushman, 2009). In order to seize new opportunities, the successful implementation of exploration and exploitation strategies is becoming increasingly important for companies. As a result, a firm that wants to pursue OA needs to embed it at the highest level of strategy, the top management (Oehmichen, Heyden, Georgakakis & Volberda, 2017).

(5)

1989). Here, the dynamic of the board itself plays a major role as a monitoring and control mechanism. The way that boards are structured, for instance by the ratio of independence, suggests a certain authority to replace executives or to implement major policies. Furthermore, following research of Harris and Helfat (1998), board positions are set by individuals with different characteristics, resources and skills. These individuals (i.e. human capital) affect the process of setting up and implementing an ambidextrous strategy.

Existing research has explored the relationship between the board of directors and performance, but it has often neglected OA (Elsayed, 2007; Hillman & Dalziel, 2003). Since corporate governance can impact a firm’s innovation activities (Zahra, 1996), the lack of research on how CEO duality and board member independence affect OA limits our understanding of a firm’s survival and success. This constitutes my literature gap as the control perspective in this regard has not yet been discussed extensively.

My work contributes to current literature in two ways. First of all, by jointly considering aspects of agency and resource-dependence theory, I give insights in how CEO duality and board independence affect OA. Secondly, this research seeks to provide explanations of how human capital, namely time to retirement (i.e. CEO tenure) and number of qualifications (i.e. education) drives these relationships.

Overall, this paper seeks to fill this research gap by providing a consistent explanation of how the control perspective influences exploration and exploitation and how shareholders can make use of this knowledge. Based on these aspects, my main research question will be:

“How does duality and independence affect organizational ambidexterity?” To further elaborate, I will also answer the following sub-question:

“How is this relationship driven by the human capital of the CEO and board?”

(6)

2. THEORY AND HYPOTHESES

This paper considers the effects of control management on OA and relies on agency theory as well as on resource-dependence literature. These perspectives complement each other when looking at the top management and how their decisions affect a firm’s strategy, and thus OA (Stiles & Taylor, 2001). Furthermore, shareholders can play a major role in the decision-making process by giving a certain structure to the top management. I will review the literature on agency and CEO duality in the subsequent paragraph, followed by a conjoint view of agency and resource-dependence theory on independence and OA. Then, I will look at the influence of human capital on these relationships.

2.1. Agency perspective and CEO duality

Following previous research on OA and the composition of the board (Hillman & Dalziel, 2003; Oehmichen, Heyden, Georgakakis & Volberda, 2017), the agency perspective emphasizes the role of the board as a monitor of management activities to minimize agency costs and therefore maximize shareholder wealth (Stiles & Taylor, 2001). The main idea of agency theory is the relationship between the principal, the board acting in the name of the shareholders, and the agent, the executive manager (Eisenhardt, 1989). This constellation of relationship poses as a central challenge and determines the delegation of control within a firm (Stiles & Taylor, 2001). Due to the agent working on behalf of the principal, tensions may arise owed to a misalignment in interests and goals as well as differences in levels of risk. Such constellation can cause agency problems (Eisenhardt, 1989). When both the principal and the agent aim to maximize their benefits, self-interest and opportunistic behavior become present, thereby increasing the agency problem (Jensen & Meckling, 1976).

(7)

as the linking key holds the authority to share or withhold information which leads to information asymmetry and impedes effective monitoring (Jensen, 1993). This in turn could increase moral hazards and adverse selection problems (Ross, 1973).

Following previous research (Hillman & Dalziel, 2003), lack of independence in the decision-making process can derive from CEO duality as an agency problem. Whereas nondual boards separate decision management (the initiation and implementation of decisions made by managers) and decision control (the approval and monitoring of decisions performed by the board of directors), dual boards have the joint CEO/chairman role for these functions (Fama & Jensen, 1983). Agency theory supports the separation of the decision-making process as otherwise independence is not provided and too much decision-making power is bundled within the CEO/chairman. With a lower board independence, governance control and effective monitoring can also be inhibited by CEO duality (Baliga, Moyer & Rao, 1996; Boyd, 1995). This might lead to less honesty in firm performance evaluations since board members are constantly observed and feel dependent on the CEO/chairman (Baliga et al., 1996). Here, it might also become more difficult for a board to replace a poorly performing CEO as previously mentioned. Overall, a dual structure raises the question of “who monitors the monitor?” (Alchian & Demsetz, 1972, p.782) as more hierarchical power is distributed to one individual. Therefore, one could argue that CEO duality appears to be more effective in promoting exploitative innovation as the person in charge might take a short-term and more opportunistic perspective in decision-making. This leads to my first hypothesis:

(8)

One key challenge of the relationship between the agent and the principal is the so-called “independence paradox” (Aguilera, 2005, p.50). This paradox describes that NEDs are given critical information and are dependent on executives whom the NEDs are supposed to supervise and be independent from (Aguilera, 2005). As previously mentioned, the lack of complete information implicates the difficulty for NEDs to judge appropriately. This could question the quality of board decisions. Furthermore, as independent directors do not operate on a daily basis in a company, a full body of information is needed even more for a high-quality decision-process and for a board to function as an effective control mechanism (Stiles & Taylor, 2001).

Another controversial tool is interlocking directorates (Stiles & Taylor, 2001). These directors are actively engaged in several boards at the same time. On the one hand, agency theorists argue that quality and independence is doubtful (Stiles & Taylor, 2001), however on the other hand, resource-dependence theorists reason that interlock directors create a link to external factors, such as competitors and reduce environmental uncertainty (Zahra & Pearce, 1989).

Involving outside directors, both interlock and non-interlock, is boundary-spanning, and an instrument to knowledge learning and attainment. Thereby, a company can strengthen its board composition due to resource-rich backgrounds of those outside directors (Stiles & Taylor, 2001). As NEDs have access to external information, they can bring novel ideas and a wider perspective to debate for strategies therefore supporting complementarity and breadth in knowledge and experience (Cornett et al., 2008). In that sense, independent directors can also act as a reminder for innovative priority of a firm. Nonetheless, independent directors have, other than their directorship, no links and attachment to the management. This refrains them from being captured or dominated by interests of executives (Becht, Bolton & Röell, 2003).

Correspondingly, agency costs are reduced significantly by the monitoring and control of performance and decision management through outside directors in particular (Roberts et al., 2005). Mace (1971) found that this reduction is further strengthened and even more crucial when a firm is undergoing a crisis. Independence on boards shall bring increased efficiency, transparency and accountability as Aguilera (2005) states these three elements as key factors for successful corporate governance. Stiles and Taylor (2001) explain that objectivity and independence, advice to the board, knowledge of external environment and wider experience account for the most important qualities that NEDs can bring to the board.

(9)

constructive arguments of control support a company in both long- and short-term thinking, leading to the second main hypothesis:

Hypothesis 2: CEO independence has a positive effect on organizational ambidexterity.

2.3. Human capital

Numerous literature streams have devoted their attention to OA, their antecedents, their outcomes and the impact of moderators (e.g. Jansen, George, Van Den Bosch & Volberda, 2008; O'Reilly & Tushman, 2011; Yitzhack Halevi, Carmeli & Brueller, 2015). The common factor of the majority of these scientific papers is human capital, more precisely, originating from the top management level. Since the managerial influence has been determined as a key driver of OA, this resource of each company should be considered more precisely. Barney (1991) classifies human capital as training, experience, expertise, judgment, network and insight of managers in a firm. RDT further views NEDs as resource providers who utilize human capital for counseling, access to new resources and legitimacy to the firm (Pfeffer & Salancik, 2003).

2.3.1. Time to retirement.

(10)

Hypothesis 1a - Duality: Lower time to retirement strengthens the negative impact of CEO duality on organizational ambidexterity stated in H1.

Hypothesis 2a - Independence: Lower time to retirement weakens the positive impact of independent board members on organizational ambidexterity stated in H2. 2.3.2. Number of qualifications.

A board member’s qualifications and education also shape the relationship between CEO duality and independence, and OA. According to Dalziel, Gentry and Bowerman (2011), and based on integrated agency RDT, the higher the education of executives, the less R&D investments are arranged, indicating increased exploitative activities and an unbalanced OA. This leads to the following hypothesis:

Hypothesis 1b - Duality: A high number of qualifications strengthens the negative impact of CEO duality on organizational ambidexterity stated in H1.

However, in the case of independent NEDs, higher education implicates more awareness for the importance of a long-term focus as NEDs tend to analyze and monitor spending (Jensen & Meckling, 1976). Here, absorptive capacity on the individual level comes into play as an essential antecedent for organizational learning (Zahra & George, 2002). The higher the level of individual experience and knowledge, the higher the absorptive capacity (Zahra & George, 2002). Ciampi (2017) argues that a top management team with a diverse background of high-quality education further strengthens the link to simultaneously pursue explorative and exploitative activities. Finally, having attended prestigious educational institutions transfers that prestige to the individual which provides them with certain additional benefits such as access to other individuals (Daily & Johnson, 1997). These aspects provide me to build the following hypothesis:

(11)

2.4. Conceptual model

I integrate agency and resource-dependence theory into my conceptual model. I argue that CEO duality negatively effects OA due to increased agency costs deriving from lack of monitoring and concentration of control among other aspects. For independence, it shows that outside directors are positively related to OA according to attainment of external resources and organizational learning as well as corporate governance components. These relationships are moderated by human capital, more specifically the time to retirement and the number of qualifications. Figure 1 provides a visual representation of the above mentioned theoretical constructs.

(12)

3. METHODOLOGY

The methodology section describes and motivates the methodological choices made in this research. In order to analyze the effects of duality and independence on OA, ordinary least-squares (OLS) regressions were applied. An OLS regression is used to predict values and to explain the strength of the relationships among variables (Hutcheson & Sofroniou, 1999).

3.1. Data and sample

(13)

3.2. Measurement

This part describes the dependent variables (OA), independent variables (duality, and independence), moderators (time to retirement, and number of qualifications), control variables (Tobin’s q, firm size, R&D intensity, return on assets, and long-term debt) and dummy variables (industry, year, and director type) which I selected for this study.

3.2.1. Dependent variable Organizational ambidexterity.

As a way to measure innovation, I retrieved the dependent variable via WordStat and computer-aided text analysis (CATA) based on McKenny, Aguinis, Short & Anglin (2016). These authors demonstrate that CATA can be applied to analyze content. This technique facilitates the measurement of constructs by transforming text into quantitative data based on the frequency of words. CATA would assist in filtering, classifying and processing information by combining the strengths of computer reliability and proficient human reasoning (Krippendorff, 2004). This approach for data measurement provides the specific purposes as a valid and reliable procedure. Specifically, this variable derived from management discussion and analysis (MD&A) text data adjusted by word stems and calculated by the adjusted explorative count times the adjusted exploitative count.

3.2.2. Independent variables. Duality.

Duality is a binary variable that takes the value of 1 if the position of the CEO and chairman is undertaken by the same person and 0 otherwise. A company’s board of directors is supposed to act as a monitoring function. Given that such a role is compromised when the CEO controls the board, my measure is expected to negatively influence OA.

Independence.

(14)

3.2.3. Moderators. Time to retirement.

I measure this variable as the number of years that a manager has to be active in his or her current working position of the CEO until retirement. This number ranges from -25.5 to 41.5 years. Thus, the CEO tenure reaches from relatively low to overdue as the minus indicates the years the CEO could have already been in retirement.

Number of qualifications.

This moderator indicates the level of education the CEO has received, ranging from a minimum of 0 to a maximum of 11 qualifications.

3.2.4. Control variables.

To control for firm effects, I added control variables to my regression analysis. Firm size is measured as the natural logarithm of the number of employees. To proxy for financial firm performance, I included return on assets (ROA) and Tobin's q as control variables. Moreover, I added financial leverage as long-term debt divided by total assets and R&D investments as the ratio to total assets. All control variables were obtained from Thomson Reuters DataStream.

3.2.5. Dummy variables.

(15)

4. RESULTS

4.1. Descriptive statistics and correlation matrix

Table 1 reports the descriptive statistics and correlation matrix for the main variables within the analysis. The sample size comprises a total of 5131 observations of 67 companies. There are no high values of correlations (r < 0.4) presented. Within the sample, 6557 observations were made about OA. The variable for duality is binary with 0.55 as a mean value indicating that there are more dual positions than separated ones on U.S. boards. The average board comprises a ratio of 79% of independent non-executive directors. The time to retirement of a CEO ranges from -25.5 to 41.5 years with a mean of 10 years. Thus, the CEO tenure reaches from relatively low to overdue as the minus indicates the years the CEO could have already been in retirement. The highest number of qualifications a CEO has is 11, the lowest is 0 with a mean of 2. For the control variables, the average firm size is 8.33 (log). Long term debt serves as a financial leverage variable and was calculated as the ratio to total assets. This control variable has a mean of 0.11. R&D intensity is measured by the ratio of R&D investments to total assets and has a mean of 0.00167%. ROA and Tobin’s q are proxies for firm performance. ROA has an average of 266.02, a minimum of 3307.21 and a maximum of 8053.88. Tobin’s q has an average of -3.0, a minimum of 0.51 and a maximum of 11.94. Concerning the dummy variables, 72% of directors are supervisory directors whereas the other 28% are executive directors. Also, my sample indicates 75% service companies with the other 25% being manufacturing firms.

(16)

Table 1: Descriptive statistics and correlation matrix

Variables Obs Mean S.D. Min Max 1 2 3 4 5 6 7 8 9 10

1 Organizational _{ambidexterity} 6,557 0.44 0.46 0 3.60 1.00

2 Duality 27,315 0.55 0.50 0 1 -0.17 1.00

3 Independence (%) 27,315 0.79 0.13 0 1 0.07 -0.26 1.00

4 Time to retirement 26,450 10.03 8.69 -25.5 41.5 0.02 -0.13 -0.02 1.00

5 Number of _{qualifications} 26,511 2.17 1.14 0 11 0.01 -0.04 0.04 0.01 1.00

6 Firm size (log) 6,503 8.33 1.40 4.060 12.31 0.00 -0.08 0.23 -0.13 0.13 1.00

7 Long term debt (%) 6,514 0.11 0.14 0 1.40 -0.14 0.11 0.09 -0.16 -0.06 0.11 1.00

8 R&D intensity (%) 6,514 0.00 0.00 0 0.000 -0.01 0.01 -0.20 0.04 -0.01 -0.61 -0.12 1.00

9 ROA 5,150 -266.02 474.43 -3307.21 8053.88 -0.07 -0.05 0.00 -0.02 -0.01 -0.06 0.04 0.03 1.00

10 Tobin's q 6,487 3.00 2.13 0.51 11.94 0.02 -0.06 -0.14 0.13 0.06 -0.06 -0.14 0.01 -0.03 1.00

Notes: n = 5,131. Effects of director type, time and industry are included but not reported.

(17)

4.2. Regression results

For my empirical research, I calculated an ordinary least-squares (OLS) regression to predict OA based on duality and independence. Table 2 presents my findings on OA. I tested my hypotheses by first analyzing the main effects (model 1 and 5) and then entering each interaction effect separately (model 2, 3, 6 and 7) before finally testing them together in model 7, 8 and 9.

Model 1 shows the base model of duality only including control and dummy variables, testing hypothesis 1. The results show a negative coefficient of -0.131 and strong significant p-value as it is smaller than 0.001. In the following models, these p-values remain significant, therefore supporting these findings. Based on this, I conclude that hypothesis 1 can be supported.

Model 2 represents the findings for hypothesis 1a. Here, the interaction term is positive and highly significant (β = 0.005, p < 0.001), providing support for hypothesis 1a that lower time to retirement indeed strengthens the negative impact of CEO duality on OA stated in hypothesis 1. To gain further insights, I plotted the relationship between CEO duality and time to retirement (Dawson, 2014). Figure 2 shows this supported moderating effect as depicted in table 2 model 2. This plot shows that CEO duality reduces OA as the line declines to the right, but the difference is even larger for lower time to retirement.

(18)

The interaction term between CEO duality and the number of qualifications in model 3 is negative and significant (β = -0.028, p < 0.01), supporting hypothesis 1b. The plot in figure 3 illustrates that a high number of qualifications indeed strengthens the negative impact of CEO duality on OA stated in hypothesis 1.

Figure 3: Interaction effect of model 3

Model 4 incorporates both the direct effect of the independent variable on the dependent variables as well as both interaction effects, confirming the support of hypotheses 1, 1a and 1b. Model 5 shows the base model of independence on OA only including control and dummy variables. It is positive and significant (β = 0.500, p < 0.001), with an adjusted R2_of

0.217, supporting hypothesis 2, the positive effect of a high ratio of independent directors on board to OA.

The findings of model 6 and 7 both show no significance within the interaction terms. These insignificant outcomes are further supported by model 8 where both hypotheses 2a and 2b are tested simultaneously. Thus, hypotheses 2a and 2b cannot be supported.

Finally, model 9 presents all hypotheses tested within one regression and confirms the previously stated outcomes of significance and insignificance.

(19)

The R2_{of all models indicates an average of approximately 23%. After checking for sample}

size, the adjusted R2_{only shows adjustments of less than 1%.}

(20)

Table 2: OLS regressions on OA

Model 1 Model 2 Model 3 Model 4

b (s.e.) b (s.e.) b (s.e.) b (s.e.)

Intercept 0.256 (0.050) 0.327 (0.051) 0.240 (0.077) 0.298 (0.052)

Independent variables

Duality (ECOB) -0.131*** (-0.011) -0.188*** (0.016) -0.068** (0.025) -0.122*** (0.028)

Ratio of independence (RIT)

Moderators Time to retirement (TTR) -0.006*** (0.001) -0.006*** (0.001) Number of qualifications (NQ) 0.009 (0.007) 0.011 (0.007) Interaction effects ECOB x TTR 0.005*** (0.001) 0.005*** (0.001) ECOB x NQ -0.028** (0.010) -0.030** (0.010) RIT x TTR RIT x NQ Control variables Firm size 0.011* (0.005) 0.009† (0.005) 0.011* (0.005) 0.010* (0.005)

Long term debt -0.381*** (0.043) -0.394*** (0.043) -0.384*** (0.043) -0.397*** (0.432)

R&D Intensity -468.68* (227.576) -541.463* (227.039) -457.592* (228.475) -531.976* (227.956) ROA 0.000*** (0.000) 0.000*** (0.000) 0.000*** (0.000) 0.000*** (0.000) Tobin's q -0.007** (0.003) -0.006* (0.003) -0.007** (0.003) -0.006* (0.003) F-statistic 76.81*** 72.39*** 70.53*** 66.83*** R-squared 0.231 0.238 0.233 0.239 Adjusted R-squared 0.228 0.234 0.230 0.236 N observations 5,143 5,131 5,134 5,131

(21)

Table 2: Continued OLS regressions on OA

Model 5 Model 6 Model 7 Model 8 Model 9

b (s.e.) b (s.e.) b (s.e.) b (s.e.) b (s.e.)

Intercept -0.297 (0.065) -0.287 (0.090) -0.261 (0.123) -0.297 (0.118) -0.073 (0.124)

Independent variables

Duality (ECOB) -0.105*** (0.029)

Ratio of independence (RIT) 0.500*** (0.057) 0.523*** (0.095) 0.518*** (0.122) 0.541*** (0.137) 0.449** (0.139) Moderators Time to retirement (TTR) -0.001 (0.005) -0.001 (0.006) -0.007 (0.006) Number of qualifications (NQ) 0.001 (0.397) 0.001 (0.040) 0.530 (0.042) Interaction effects ECOB x TTR 0.006*** (0.001) ECOB x NQ -0.032** (0.010) RIT x TTR -0.002 (0.007) -0.002 (0.007) 0.001 (0.007) RIT x NQ -0.008 (0.048) -0.009 (0.049) -0.051 (0.050) Control variables Firm size 0.011* (0.005) 0.010* (0.005) 0.012* (0.005) 0.011* (0.005) 0.007 (0.005)

Long term debt -0.388*** (0.044) -0.392*** (0.044) -0.389*** (0.044) -0.393*** (0.044) -0.398*** (0.043)

R&D Intensity -162.896 (229.689) -192.332 (229.935) -147.176 (231.290) -170.383 (231.606) -412.458† (229.377) ROA 0.000*** (0.000) 0.000*** (0.000) 0.000*** (0.000) 0.000*** (0.000) 0.000*** (0.000) Tobin's q -0.003 (0.003) -0.003 (0.003) -0.003 (0.003) -0.002 (0.003) -0.004 (0.003) F-statistic 72.23*** 66.48*** 65.91*** 60.98*** 61.03*** R-squared 0.220 0.223 0.221 0.223 0.244 Adjusted R-squared 0.217 0.219 0.218 0.219 0.240 N observations 5,143 5,131 5,134 5,131 5,131

(22)

5. DISCUSSION AND CONCLUSION

In this paper, I examine how CEO duality and board independence affect a firm’s ability to balance exploration and exploitation complemented by the moderating effect of individual human capital. Specifically, this research focuses on the control perspective based on agency and resource-dependence theory as two complementing views on the top management. The findings show that CEO duality entails severe negative consequences on a firm’s organizational ambidexterity which is in line with the provided literature. Another main contribution is the significantly positive influence of board independence on organizational ambidexterity which is also in line with expectations. Regarding the moderation of human capital, this paper provides evidence that lower time to retirement further strengthens the negative relationship between duality and organizational ambidexterity. This strengthened negative relationship also applies for a higher number of qualifications. However, the moderation of time to retirement and number of qualifications to board independence and organizational ambidexterity do not show any significance, leaving opportunities for future research. To test the hypotheses, multiple OLS regression analyses were conducted, using a sample of 67 U.S. firms over a time period from 2004 to 2016. Moreover, long term debt and ROA showed to have a significant effect as control variables. Drawing upon these findings, I continue to provide crucial implications for theory and practice. Additionally, the main limitations of this research will be discussed as well as suggestions for possible directions of future research.

5.1. Theoretical implications

(23)

Taras & Tarba, 2013). The strong detrimental significant result that I found in this study therefore steers the controversies of CEO duality on OA in a certain direction. Support of my line of reasoning and results can be found with Pi and Timme (1993) as well as with Rechner and Dalton (1991). However, the results of those findings impact firm and financial performance instead of OA.

Moreover, I present an alternative, more positive ambidexterity perspective of board independence. Agency theorists claim that the board acts as an alternative monitoring instrument by assessing the impact of the decisions of the top management (Fama & Jensen, 1983). RDT contends that the board is the key function for access to information and resources (Zahra & Pearce, 1989). Thus, both agency and RDT support the view of the board of directors as a control mechanism. While agency theorists traditionally emphasize the negative effects of board independence such as the “independence paradox” (Aguilera, 2005, p.50) and information asymmetry as a cause for lack of monitoring (Aguilera, 2005; Stiles & Taylor, 2001), resource-dependence scholars support my line of reasoning and found results that a high board independence positively affects OA. Hillman and Dalziel (2003) also indicate that outside directors serve as both a managerial monitor and as resource provision. Consistent with my findings of hypothesis 2, numerous studies (Borokhovich, Parrino, & Trapani, 1996; Byrd & Hickman, 1992; Hermalin & Weisbach, 1988) support the view of outside directors being a beneficial function to shareholders. However, the results of those findings impact firm performance instead of OA directly, leaving room for future research.

Furthermore, it can be argued that outside directors serve as a form of exploration of OA equivalent to executive managers as a form of exploitation. Therefore, it would only manifest if there was a balance with a similar number of outside and executive directors in the top management. Stettner and Lavie (2014) affirmed this line of thought as internal exploration depends on the ability of a firm to innovate utilizing its internal knowledge.

(24)

career and their narrowed cognitive schemata (Mom et al., 2009). Aiming the majority of attention on exploitation instead of an equal balance, will most likely be regarded as a detrimental decision from the principal’s point of view.

The investigation of the moderation on duality and OA has further shown that a high number of qualifications (i.e. more education) indeed strengthens the negative impact of CEO duality. Therefore, my line of reasoning and results have been proven correct based on Dalziel et al.’s (2011) argumentation.

However, the results of hypothesis 2a and 2b were not significant. This indicates that the method I used to analyze human capital on the relationship of independence on OA was either incomplete, lacking additional key elements, or possibly due to strong counter-arguments, leaving room for future investigations. Therefore, the following assumptions need to be interpreted with caution.

First, it might be that the argument of increased tenure and board members getting closer to their retirement age rather strive for exploitative activities also acts vice versa. This line of reasoning would then indicate that younger board members tend to give more attention to explorative innovation. In other words, with increasing CEO tenure exploration gradually decreases while exploitation increases (Ferreira, Raisch & Klarner, 2014). Therefore, it might be possible that an intermediate CEO tenure is related to the strongest orientation towards OA. However, it might also be possible that there is no association of time to retirement as a moderator. McKnight and Weir (2009) found that CEO tenure and agency costs are not related which might be applicable in my study as well. They argue that the potential costs of CEOs close to retirement is counterbalanced by the potential benefits gained by more considerable experience (McKnight & Weir, 2009). Another aspect might be, as argued by Cornett et al. (2008), that a CEO with higher tenure and age has an increased effectiveness in management due to greater understanding of the firm and industry. Thus, this could have also been a factor outbalancing the negative elements of opportunism and self-interest.

Second, as the moderation of number of qualifications with independence and OA also resulted in non-significance, long-term focus, absorptive capacity and prestige might not be influential in this case. Perhaps the most serious leverage that my line of reasoning did not take into account and might outbalance this effect is the role of risk. Herrmann and Datta (2002) argue that due to their stronger capabilities, NEDs with a higher number of qualifications prefer high-risk strategies opposed to low ones which might also be influential in this context.

(25)

Kyereboah-Coleman and Biekpe (2006) support my findings on duality and give preference to the two-tier board governance system which splits authority and roles of CEO and chairman inherently. This is due to agency problems and conflicts of interest that top management has to face unless control mechanisms are introduced (Aguilera, 2005). However, as new outside directors generally perceive a split of CEO duality as a sign of weakness (Khurana, 2002), this bears complications for implementations of the two-tier system.

5.2. Managerial implications

This study helps towards a better understanding for practitioners, especially when choosing and appointing an executive in order to achieve innovation goals and OA.

First, I would like to counsel on CEO duality and the intention of a firm to become or maintain ambidextrous. The presented results show that top management shall separate the function of the CEO and chairman. Boyd (1995) argues that fatal consequences can result in case of lack of monitoring or weak board control. These include aspects which are not in the shareholders’ best interest (Fama & Jensen, 1983) such as hostile takeovers or increased opportunism and managerial perks with higher levels of executive compensation (Boyd, 1995; Jensen & Meckling, 1976; Stiles & Taylor, 2001). However, if duality is applicable, a company can take certain measures to decrease the detrimental outcomes of a dual board structure. Here, agency theory suggests monitoring possibilities, such as raising the number of independent directors or increasing the information the board has about managerial behavior (Eisenhardt, 1989; Fama & Jensen, 1983), and introducing incentive alignment between the principal and the agent (Jensen & Meckling, 1976).

Second, the findings indicate that a high ratio of independent directors supports a firm in achieving ambidexterity. Hillman, Withers and Collins (2009) further suggest that directors of high resources should be the focus of board composition, meaning that the type of director might matter more than the number. However, the question arises of how independent a NED truly is. When appointing a manager, awareness must be given to cronyism, the recruitment through personal contacts or friendships (Aguilera, 2005).

(26)

5.3. Limitations and future research directions

I acknowledge that this research has some limitations. However, these could provide scholars with interesting opportunities for future research. First, this study focuses on two complementary theories, namely agency and resource-dependence to describe the impact of duality and independence on ambidexterity. These theories have restrictions within themselves. For instance, agency theory ignores the complexity of organizations or RDT disregards the use of a resource (Stiles & Taylor, 2001). Here, future studies may explore when other theories, such as the upper echelon theory, can be used to extend the explanations of the effects of CEO duality or independence on OA.

Secondly, this study only focuses on a small fraction of human capital. Here, this can be further elaborated as other characteristics of the top management as well as contextual factors like cultural, technological or industrial aspects moderate these relationships. Also, the results show that further work is required to understand environmental factors in this interplay. Plus, within the data of the number of qualifications, the BoardEx variable definition does not fully explain what the eleven qualifications consist of.

A third limitation concerns the used control variables. I controlled only company-level factors (firm size, long-term debt, R&D intensity, ROA, Tobin’s q), but did not control for board-related factors (such as gender, board size or diversity) due to a lack of data availability. Moreover, I could only run one type of regression analysis due to time and scope limitations. This allows for an interesting opportunity to conduct more various regressions in order to investigate the full spectrum of variables I have collected. A final opportunity for future research would be to broaden the geographical scope and also test the hypotheses among foreign, non-U.S. firms. An international comparison would provide more generalizability of the theories regarding, for instance different board systems such as the one- and two-tier system. The forth limitation of this study is the control perspective per se. As corporate governance mechanisms also include the incentive perspective and executive compensation packages to align the interests of executives and shareholders (Eisenhardt, 1989; Hillman & Dalziel, 2003; Hüttenbrink, Oehmichen, Rapp & Wolff, 2014). According to agency theory, incentives and equity compensation need to be considered as another key antecedent of the monitoring function (Hillman & Dalziel, 2003) but needed to be disregarded due to the subject of scope.

(27)

exploit simultaneously, but this leaves the question whether an exclusively independent board has the same positive influence on OA as a highly independent board. Stiles and Taylor (2001) emphasize that social factors between inside and outside directors create potential for both conflict but also creativity, providing opportunities for innovation.

(28)

REFERENCES

Aguilera, R. V. (2005). Corporate governance and director accountability: An institutional comparative perspective. British Journal of Management, 16(s1).

Ahuja, G., & Morris Lampert, C. (2001). Entrepreneurship in the large corporation: A longitudinal study of how established firms create breakthrough inventions. Strategic management journal, 22(6-7), 521-543

Aken, J. E., Berends, J. J., & Bij, J. D. (2012). Problem solving in organizations: a

methodological handbook for business and management students. Cambridge University Press. Alchian, A. A., & Demsetz, H. (1972). Production, information costs, and economic

organization. The American economic review, 62(5), 777-795.

Baliga, B. R., Moyer, R. C., & Rao, R. S. (1996). CEO duality and firm performance: What's the fuss?. Strategic Management Journal, 41-53.

Bantel, K. A., & Jackson, S. E. (1989). Top management and innovations in banking: Does the composition of the top team make a difference?. Strategic management journal, 10(S1), 107-124.

Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of management, 17(1), 99-120.

Becht, M., Bolton, P., & Röell, A. (2003). Corporate governance and control. In Handbook of the Economics of Finance (Vol. 1, pp. 1-109). Elsevier.

Borokhovich, K. A., Parrino, R., & Trapani, T. (1996). Outside directors and CEO selection. Journal of Financial and Quantitative Analysis, 31(3), 337–355.

Boyd, B. K. (1995). CEO duality and firm performance: A contingency model. Strategic Management Journal, 16(4), 301-312.

(29)

Ciampi, F. (2017). The Antecedents and Consequences of Small Enterprise Organizational Ambidexterity: An Empirical Analysis of Italian Manufacturing Firms. International Journal of Business and Management, 13(1), 110.

Cornett, M. M., Marcus, A. J., & Tehranian, H. (2008). Corporate governance and pay-for-performance: The impact of earnings management. Journal of financial economics, 87(2), 357-373.

Daily, C. M., & Johnson, J. L. (1997). Sources of CEO power and firm financial performance: A longitudinal assessment. Journal of Management, 23(2), 97-117.

Dalziel, T., Gentry, R. J., & Bowerman, M. (2011). An integrated agency–resource

dependence view of the influence of directors' human and relational capital on firms' R&D spending. Journal of Management Studies, 48(6), 1217-1242.

Dawson, J. F. (2014). Moderation in management research: What, why, when and how. Journal of Business and Psychology, 29, 1-19.

Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management review, 14(1), 57-74.

Elsayed, K. (2007). Does CEO duality really affect corporate performance?. Corporate Governance: An International Review, 15(6), 1203-1214.

Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. The journal of law and Economics, 26(2), 301-325.

Ferreira, P., Raisch, S., & Klarner, P. (2014, January). Staying agile in the saddle: CEO tenure, TMT change, and organizational ambidexterity. In annual meeting of the Academy of Management, Philadelphia.

Finkelstein, S., & Hambrick, D. C. (1990). Top-management-team tenure and organizational outcomes: The moderating role of managerial discretion. Administrative science quarterly, 484-503.

(30)

Harris, D., & Helfat, C. E. (1998). CEO duality, succession, capabilities and agency theory: Commentary and research agenda. Strategic Management Journal, 901-904.

Hendry, K., & Kiel, G. C. (2004). The role of the board in firm strategy: Integrating agency and organisational control perspectives. Corporate Governance: An International

Review, 12(4), 500-520.

Hermalin, B. E., & Weisbach, M. S. (1988). The determinants of board composition. Rand Journal of Economics, 19, 589–606.

Herrmann, P., & Datta, D. K. (2002). CEO successor characteristics and the choice of foreign market entry mode: An empirical study. Journal of International Business Studies, 33(3), 551-569.

Herrmann, P., & Datta, D. K. (2006). CEO experiences: Effects on the choice of FDI entry mode. Journal of Management Studies, 43(4), 755-778.

Hill, S. A., & Birkinshaw, J. (2014). Ambidexterity and survival in corporate venture units. Journal of management, 40(7), 1899-1931.

Hillman, A. J., & Dalziel, T. (2003). Boards of directors and firm performance: Integrating agency and resource dependence perspectives. Academy of Management review, 28(3), 383-396. Hillman, A. J., Withers, M. C., & Collins, B. J. (2009). Resource dependence theory: A

review. Journal of management, 35(6), 1404-1427.

Hutcheson, G., & Sofroniou, N. (1999). The Multivariate Social Scientist: Introductory Statistics Using Generalized Linear Models. London: SAGE Publications Ltd.

Hüttenbrink, A., Oehmichen, J., Rapp, M. S., & Wolff, M. (2014). Pay-for-performance–Does one size fit all? A multi-country study of Europe and the United States. International Business Review, 23(6), 1179-1192.

Jansen, J. J. P., George, G., Van Den Bosch, F. A. J., & Volberda, H. W. (2008). Senior team attributes and organizational ambidexterity: The moderating role of transformational

leadership. Journal of Management Studies, 45(5), 982–1007.

(31)

Jensen, M., (1993). The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance 48, 831–880.

Junni, P., Sarala, R. M., Taras, V., & Tarba, S. Y. (2013). Organizational ambidexterity and performance: A meta-analysis. The Academy of Management Perspectives, 27(4), 299-312. Khurana, R. (2002). Searching for a corporate savior. The irrational quest for charismatic CEOs. Princeton University Press, Princeton.

Kor, Y. Y. (2006). Direct and interaction effects of top management team and board

compositions on R&D investment strategy. Strategic Management Journal, 27(11), 1081-1099. Krippendorff, K. (2004). Content analysis: An introduction to its methodology. Second

Edition. Thousand Oaks, CA: Sage.

Kyereboah-Coleman, A., & Biekpe, N. (2006). The relationship between board size, board composition, CEO duality and firm performance: Experience from Ghana. Corporate Ownership and Control, 4(2), 114-122.

Lu, X., & White, H. (2014). Robustness checks and robustness tests in applied economics. Journal of Econometrics, 178, 194-206.

Mace, M. L. (1971). Directors Myth and reality Cambridge. MA Harvard University, Graduate School of Business Administration.

March, J. G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2(1), 71–87.

McKnight, P. J., & Weir, C. (2009). Agency costs, corporate governance mechanisms and ownership structure in large UK publicly quoted companies: A panel data analysis. The quarterly review of economics and finance, 49(2), 139-158.

Mom, T. J., Fourné, S. P., & Jansen, J. J. (2015). Managers’ work experience, ambidexterity, and performance: The contingency role of the work context. Human Resource Management, 54(S1).

(32)

O'Reilly, C. A., & Tushman, M. L. (2011). Organizational ambidexterity in action: How managers explore and exploit. California Management Review, 53(4), 5-22.

O'Reilly, C. A., & Tushman, M. L. (2013). Organizational ambidexterity: Past, present, and future. The Academy of Management Perspectives, 27(4), 324-338.

Oehmichen, J., Heyden, M. L. M., Georgakakis, D., & Volberda, H. W. (2017). Boards of directors and organizational ambidexterity in knowledge-intensive firms. The International Journal of Human Resource Management, 28(2), 283–306.

Pfeffer, J., & Salancik, G. R. (2003). The external control of organizations: A resource dependence perspective. Stanford University Press.

Pi, L., & Timme, S. G. (1993). Corporate control and bank efficiency. Journal of Banking & Finance, 17(2-3), 515-530.

Piao, M. (2010). Thriving in the new: Implication of exploration on organizational longevity. Journal of Management, 36(6), 1529–1554.

Poppo, L., & Zenger, T. (2002). Do formal contracts and relational governance function as substitutes or complements?. Strategic management journal, 23(8), 707-725.

Raisch, S., & Birkinshaw, J. (2008). Organizational ambidexterity: Antecedents, outcomes, and moderators. Journal of Management, 34(3), 375–409.

Raisch, S., Birkinshaw, J., Probst, G., & Tushman, M. L. (2009). Organizational ambidexterity: Balancing exploitation and exploration for sustained performance. Organization science, 20(4), 685-695.

Rechner, P. L., & Dalton, D. R. (1991). CEO duality and organizational performance: A longitudinal analysis. Strategic Management Journal, 12(2), 155-160.

Roberts, J., McNulty, T., & Stiles, P. (2005). Beyond agency conceptions of the work of the non-executive director: Creating accountability in the boardroom. British journal of

management, 16(s1).

(33)

Ross, S. A. (1973). The economic theory of agency: The principal's problem. The American Economic Review, 63(2), 134-139.

Smith, W. K., & Tushman, M. L. (2005). Managing strategic contradictions: A top

management model for managing innovation streams. Organization science, 16(5), 522-536. Stettner, U., & Lavie, D. (2014). Ambidexterity under scrutiny: Exploration and exploitation via internal organization, alliances, and acquisitions. Strategic management journal, 35(13), 1903-1929.

Stiles, P., & Taylor, B. (2001). Boards at work: How directors view their roles and responsibilities: How directors view their roles and responsibilities. OUP Oxford. Swanborn, P. G. (1996). A common base for quality control criteria in quantitative and qualitative research. Quality and Quantity, 30(1), 19-35.

Yin, R. K. (2003). Case study research: design and methods, Applied social research methods series. Thousand Oaks, CA: Sage Publications, Inc. Afacan, Y., & Erbug, C. (2009). An interdisciplinary heuristic evaluation method for universal building design. Journal of Applied Ergonomics, 40, 731-744.

Yitzhack Halevi, M., Carmeli, A., & Brueller, N. N. (2015). Ambidexterity in SBUs: TMT behavioral integration and environmental dynamism. Human Resource Management, 54(S1). Yoshikawa, T., Zhu, H., & Wang, P. (2014). National governance system, corporate

ownership, and roles of outside directors: A corporate governance bundle perspective. Corporate Governance: An International Review, 22(3), 252-265.

Zahra, S. A. (1996). Goverance, ownership, and corporate entrepreneurship: The moderating impact of industry technological opportunities. Academy of management journal, 39(6), 1713-1735.

Zahra, S., and George, G. (2002) Absorptive Capacity: A Review, Reconceptualization, and Extension. Academy of Management Review, 17 (2): 185-203.

(34)

APPENDIX

Description of variables

This table presents the set of variables used in my empirical analysis. I collected the ambidexterity data from WordStat and my board data from BoardEx (ECOB, RIT, TTR, NQ and DDSD). All other data (total assets, leverage, tobin’s q, firm size, ROA, R&D investment data, year dummies, industry dummies) are obtained from the Thomson Reuters DataStream database.

Table 3: Description of variables

Variable Description Calculation

Dependent variable

Organizational ambidexterity

MD&A-text files adjusted by word stems

adj_explor_count_mdua_bereinigtws * adj_exploit_count_mdua_bereinigtws

Independent variables

ECOB Duality Binary variable, 1 if the CEO is also

chairman, and 0 otherwise RIT Ratio of independent directors

to the total number of directors on board

IND (independent directors) / TD (total directors)

Moderators

Time to Retirement Time to Retirement for the individual at a selected Annual Report Date assuming a retirement age of 70 Number of

qualifications

The average number of qualifications at undergraduate level and and above forall the Directors at the Annual Report Date selected

Control variables

Firm size Natural logarithm of the number of employees

ln (employees [WC07011])

ROA Financial performance variable Net income before Preferred

Dividends [WC01651] * (1- Tax rate [WC08346]) / Last Year's Total Capital [WC03998] * 100

Leverage Financial leverage ratio Long term debt [WC03251] / Total Assets [WC02999]

R&D intensity Research and development

investments ratio ln(rd_expense_XWC01201) / Total Assets [WC02999]