Master Thesis for MSc BA SB&E

(1)

_______________________________________

Master Thesis for MSc BA SB&E

_____________________________________

Ownership and Firm Performance: How Does Being Manager

and Shareholder Simultaneously Affect Firm Performance?

Student:

Bremer, R. S3209903

Faculty of Economics and Business

University of Groningen

Supervisor: dr. des. Kristalova, M.

Word count: 8922

Groningen, 21-06-2020

Semester 1B

(2)

2

“You don’t build a business, you build people,

then people build the business.”

- Zig Ziglar

1

Abstract

It is commonly known that the manager of a company is one of the most important figures inside the company structure. It is important for companies to know how they can boost the manager dedication for the company as well as the performance of the company. Owning shares result in having more dedication for the firm but does this lead to better firm performance? Therefore it is important to know whether there is a difference between managers which are owning shares of the company and managers who do not own shares of the company. In this paper I am investigating if this difference in managers really holds. The

foundation of this study is based on two opposing theories, the Agency Theory and the Stewardship Theory. The one theory is against giving shares to the manager of the company while the other theory is

in favor of giving shares to the manager. To deal with the research question of this study, I composed a list of more than 7000 Dutch small- and medium sized enterprises together with relevant data out of the year 2018. The data is analysed by using a linear regression analysis. The main relation in this study appeared to be negative. In addition, more interesting results have been found which laid the basis for

further future research between several factors and firm performance.

Keywords

Ownership, shareholder, firm performance, ROA, Agency Theory, Stewardship Theory, Dutch context, firm age, firm size, total assets, number of board directors

(3)

3

Preface

This paper is one of my last contributions for this master program. Hopefully this paper is a valuable contribution to the existing literature. Writing this thesis was very interesting, due to the topic in a SME context and therefore the literature and findings were more recognizable. Writing this thesis was sometimes also challenging when complex factors and regressions needed to be incorporated in order to represent meaningful results but eventually this was really educational.

I would like to thank dr. des. Maria Kristalova for her guidance. With her support and feedback during the process I was able to keep on the right track and to bring my thesis to a higher level.

(4)

4

Introduction

After Bill Gates and Paul Allen founded Microsoft, the company has had two other CEO’s, namely Steve Ballmer and current CEO, Satya Nadella. All these CEOs had a large share in the ownership of the company and Microsoft is still one of the most profitable companies in the world. On the other hand, the founder and CEO of Amazon, Jeff Bezos, owns 12% of the shares of Amazon but he reduces his amount of shares every year. Even though this reduction of ownership, Jeff Bezos manages to let Amazon be one of the most successful companies worldwide. Amazon’s revenue grows with an average of 21% year over year (Market Realist, 2020). Thus, for Microsoft the CEO ownership is high and also having the company it’s performance high. While, for Amazon, the CEO ownership is low and declining but the company 's performance is growing.

The examples Microsoft and Amazon are two of the biggest companies worldwide and numerous studies in the field of operations management literature have been investigating with focusing on such worldwide known companies. Little has been done to investigate how relationships and theories can be applied to the management of small firms. Therefore, the focus is on small and medium sized enterprises (SMEs) in this study. In SMEs the CEO often has a large amount of shares in the company, while it also sometimes occurs that a SME has a CEO without any shares. Therefore, the importance of this CEO-shareholder relationship in small and medium sized enterprises (SMEs) might differ. This leads to the main research question of this study: how does being manager and shareholder simultaneously affect firm performance?

In the beginning, when a company is founded, the ‘founding fathers’ are often also the persons who are having the lead of daily management. In a later stage, it is seen that founders may hire a CEO or that the founders just stay at the managerial position (Hendricks et al., 2019). As mentioned in the first paragraph, hiring of a CEO is also the case at the well known technology company Microsoft. Having the lead over the daily operations seems a logical place for founders due to their influence on the

organization, their professional image and their founder’s vision. On the other hand, it is also seen as a logical choice when founders step aside to hand over the leadership to managers, due to the case that a fast growing firm overgrows the managing capabilities of the founder itself. This leads to a situation where the founder doesn’t have the skills and capabilities to control the company anymore (Hendricks et al., 2019).

(6)

6

The Agency Theory is about hiring an agent who performs a service on principal’s behalf (Jensen & Meckling, 1976 ; Eisenhardt 1989). In other words, when a founder hires a manager to perform a service, which is leading the company, on the founders behalf. The problem in this theory arises when the agent is only motivated to act in its own best interests instead of the interests of the company (Jensen & Meckling, 1976). A recent update on the Agency Theory has to do with the renewed view of Panda and Leepsa (2017). They add the problem of different risk perspectives, where the principal invests in capital, while the agent is working with this capital. Due to this, the agent can work with this capital without any risk and therefore the agent will take more risks with this capital.

The Stewardship Theory, on the other hand, states that managers will act as responsible stewards of the assets they are controlling. The theory assumes that a steward will always choose for behavior that is positive for the organization instead of what is positive for the steward personally. This is because stewards are assumed as collective thinkers and trustworthy (Davis et al., 1997). On the basis, a steward places higher value on cooperation than on desertion (Le Breton-Miller and Miller, 2009). In this paper, both the Agency Theory as the Stewardship Theory are employed for answering the research question. Their application depends on the fact if the manager is also shareholder or not.

In this paper the gap of research about CEOs related to small and medium sized businesses as well as the gap in research in the difference between CEOs which are simultaneously shareholders against CEOs which are not simultaneously shareholders will be tried to get filled. Up till now, there are

numerous articles about CEOs investigating, for example, CEO confidence (Lee et al., 2016), CEO rich versus king dilemma after IPO (Fattoum-Guedri et al., 2016) or about the question if founder-led firms are less susceptible to managerial myopia (Schuster et al., 2018). In addition, other examples of articles are those of Chauhan and Kumar (2017), who examined what consequences non-founder

compared to founder ownership has on performance in terms of foreign investments and Bingham et al. (2019), who studied whether rehiring a former CEO brings positive or negative effects.

The above mentioned studies focus on firm performance in terms of foreign investments and only large companies were used in the sample. According to the article of Fitzal and Tihanyi (2017), it is found that there is limited research about the relationship between forms of ownership and firm performance. This paper will focus on the difference of having a shareholding CEO against a non-shareholding CEO and only small and medium sized businesses will be used in the sample. This difference, of having a shareholding CEO against a non-shareholding CEO, is a topic where not much focus has been laid on yet. Also, this research will focus on what the consequences are on the basis of financial performance only, while most other studies mostly focused on company growth (e.g. Andersén et al., 2020).

(7)

7

now is clear that giving shares to managers will not result in better firm performance for SMEs.

This study delivers several important managerial implications. First, companies will be able to get a better understanding of the relationship between the CEO and the company’s performance. The role of CEO is a job which requires numerous skills and responsibilities and therefore it is important for a company to know whether they have the right person in the right place. Second, companies will understand better whether it is advisable for a company to have a CEO who also owns shares in the company or whether it might be better to have a CEO without shares in the company.

(8)

8

Literature review

Introduction of the literature review

In the next section the contradicting viewpoints of the Agency Theory and the Stewardship Theory will be discussed. Even though both theories are considered as opposing, both theories consist of the same core concepts: individual behaviour and firm governance mechanisms in relation to each other and predicting organizational outcomes. The two theories both focus on how a manager behaves in relation to his/her boss and the company where the manager is working. That makes why these two theories are really suitable for this research paper. The purpose of using these two opposing theories is to reflect the two managerial groups which will be used in this research. The Agency Theory reflects the group of managers who are not shareholder of the company, since the Agency Theory highlights a conflict of the desires or goals of the manager versus the owner of a company. The Stewardship Theory, on the other hand, reflects the group of managers who are also shareholder of the company due to the fact that the Stewardship Theory highlights the behaviour of managers who are performing positively and in the best interests of the organisation. These two theoretical frameworks form the foundation of this literature review section.

Agency Theory

The Agency Theory describes the relationship between two parties, the principal (i.e. the owner of the company) and the agent (i.e. the manager of the company). The relationship can be described as when the agent is allowed to make decisions and actions on behalf of the principal. The theory consists of the agency problem, which arises when there is a conflict of the desires or goals of the principal and the agent. It arises when the agent wants to maximize its own benefit with minimal effort (Davis et al., 1997). A consequence of the conflict between the principal and the agent is that agency costs will arise. These costs are related to when an individual wants to maximize its own profit to the prejudice of the company.

The problem also arises when it is difficult for the principal to control what the agent is really doing (Eisenhardt, 1989). That is, because firms cannot know in advance if the individual will behave in a positive way for the company (Williamson, 1985). According to Agency Theory, it happens that when an individual owns less than 100% of a firm’s equity, it is likely that the individual will act as agent and pursue its own interests, which differs from those of other shareholders.

(9)

9

had other incentives than principals. Mitnick, on the other hand, introduced the agency problem by stating where the imperfection is coming from: an agent behaves never fully in a way which is preferred by the principal, since the agent does not get paid enough to make a fully perfect contribution. After Ross and Mitnick, soon the article of Jensen and Meckling (1976) arised where they further completed the Agency Theory by adding the element of ownership structure.

Due to the negative connotation of the Agency Theory, companies do not want to face behaviour related to this theory. According to Davis et al. (1997), this behaviour can be decreased by imposing control mechanisms on the behaviour inside the company. In that way, it will prevent individuals from showing opportunistic behaviour which is in the detriment of the company. So, when a manager and the firm have contrasting interests, the use of mechanisms which monitor can help bring these interests together. This is confirmed by Fama (1980) who states that, after imposing control mechanisms, the interests will align and that will even increase the performance of the firm.

The Agency Theory believes that the more independent departments are, the better that will be for firm performance. Applied to the research question of this paper, the Agency Theory would suggest a negative relationship between being manager and shareholder simultaneously on firm performance.

Agency Theory and firm characteristics

Firm age is one of the factors that influence the firm’s agency costs. On the one hand it is possible that a young firm is very efficient in e.g.using its resources while, on the other hand, an firm may become more efficient when it gets older due to e.g. experience. In the latter case, monitoring by outside directors is not needed anymore (Ang et al., 2000).

Firm size is another important factor for firm performance. According to Williamson (1967 ; 1985) firms which become large may find problems of coordination, because firms cannot know in advance if any of their individual employees will behave in a positive way for the company. That will eventually negatively influence a firm's performance.

In addition, the more assets a company has the more costs occur. Also, the more assets that need to be controlled, the more supervision is needed. This extra need for supervision makes a more complex job of the manager role.

(10)

10

negative for the response time of a company. The opposite means that the Agency Theory foresees a positive relationship between small and outsider dominated boards. As can be seen in figure 1, this means that the Agency Theory believes that if the board of a company is small and the members of the board are not employed by the company in the daily operations, that this will lead to a positive relationship towards firm performance (Boyd, 1995; Nicholson and Kiel, 2007).

Figure 1 (Nicholson and Kiel, 2007)

Prior empirical studies

As investigated by Dalton et al. (1998), they found little evidence for any relationship between governance structure and financial performance. After executing a meta-analytic review and taking into accounts the factors of board composition, leadership structure and financial performance, they found no evidence that supported the relationship between governance structure and financial performance. This indicates that the structure of the board is irrelevant for the performance. In addition to that, a year later Dalton et al. (1999) executed a study again. In this study the focus was on board size, instead of board structure, in relation to firm performance. Results for this study showed that there was also no evidence for influence of board size on firm performance. Related to these two studies of Dalton et al., Demsetz and Villalonga (2001) studied the relationship between ownership structure and corporate performance. They found no empirical evidence for the relationship between these two variables.

Based on the above mentioned studies, it is found by their authors that the three elements of board structure, board size and ownership do not have any kind of relationship with firm performance.

(11)

11

Hypothesis 1

From the findings mentioned above the following hypothesis can be formed:

Hypothesis 1

(12)

12

Stewardship Theory

The Stewardship Theory is related to organisational psychology and sociology. The origin of the theory lays especially in McGregor’s Theory X and Theory Y (McGregor, 1960 ; Glinkowska and Kaczmarek, 2015). These two theories reflect how managers perceive and behave towards their employees. Theory X minded managers have a pessimistic view about employees, and they assume that employees are naturally unmotivated and dislike work. In contrast, Theory Y minded managers have an optimistic and positive viewpoint about employees and they favor a decentralized, participative management style.

The Stewardship Theory is linked to Theory Y, since according to the Stewardship Theory, managers are satisfied when a job is performed well. That is in line with the optimistic and positive viewpoint of the Theory Y minded managers. So, their behaviour is positive for the organisation and it serves the interests of the organisation. This behaviour is also created due to the psychological mindset of owning a business. Due to that mindset, a strong feeling of commitment is present towards the company. Due to that commitment, there is a bigger chance that the founder will act as a steward and in the best interest of the company. (Davis et al., 1997)

Thus, in the field of the Stewardship Theory, a manager acts in the best interest of the company rather than on the individual interest. The manager chooses for behaviour which results in the best possible outcomes for the company. Even if there might be a situation of a conflict of interests, the manager will choose the best solution for the company rather than for itself (Davis et al., 1997). This pro-organisational behaviour takes place partly because steward-managers strongly identify themselves with the mission of the organization. In addition to that, if the company has success it will contribute to the steward-managers self-image and self-concept (Van Puyvelde et al., 2012).

With the Stewardship Theory, controlling mechanisms are not needed, which is in contradiction to the Agency Theory. With the Stewardship Theory, trust is the main approach for alignment (Van Slyke, 2007). This element of trust can be seen in, among others, family firms. It has been proposed by Corbetta and Salvato (2004) that family businesses create a sort of psychological ownership which motivates all involved family members to behave in the best interest of the family firm. These factors of trust and psychological ownership are related to feelings of responsibility and burden sharing for the organization (Pierce et al., 2001).

Stewardship Theory in relation to firm performance

(13)

13

the company in an effective way. This can be seen in figure 2.

Based on the alinea above, in Stewardship Theory high firm performance positively relates to an inside members board, since they naturally work to maximise profit for shareholders. Applied to the research question of this paper, the Stewardship Theory suggests that firm performance will be higher if the manager is also a shareholder of the company simultaneously. So, Stewardship Theory would suggest a positive relationship between being manager and shareholder simultaneously on firm performance.

Figure 2 (Nicholson and Kiel, 2007)

Stewardship Theory in relation to firm age

Firm performance deteriorates with age. Young firms have higher expected growth rates of sales, profits and productivity. Also, younger firms are more capable to convert employment growth into growth of sales, profits and productivity (Coad et al., 2013).

Stewardship Theory in relation to firm size

It is found that a loss of control occurs when different hierarchical levels of a firm communicate. Due to the so-called ‘simple serial reproduction distortion’, the initial information and communication gets changed along the communication way. And if, in addition, goals differ between hierarchical levels in a firm, the loss in control can be even more extensive (Williamson, 1967). Nevertheless, since the

Stewardship Theory states that there are no goal differences between the principal(s) and the agent(s), the statement of an even more extensive loss in control due to goal differences does not apply. Therefore it can be said that, in relation to the Stewardship Theory, only a minimum degree of control loss occurs if a firm grows in size.

Stewardship Theory in relation to total assets

(14)

14 Stewardship Theory in relation to number of board directors

The more board of directors, the more knowledge in the firm. This extra knowledge is in the positive of the firm, which leads to better management.

Prior empirical studies

Positive results are found for the ownership concentration–performance relationship (Gedajlovic and Shapiro, 1998 ; James et al., 2017 ; Li, 1994). In addition to that, Fitzal and Tihanyi (2017) found that form of ownership is an important factor in relation to firm performance. Based on their study, they state that the need for more studies in the field of this relationship is required in order to find more empirical evidence. It can be stated that, based on the above mentioned studies, there is a relationship between forms of ownership and firm performance, but this relationship is also context dependent.

CEO ownership is examined by Mehran (1994), who states that firm performance is positively related to the percentage of equity held by managers. This is also supported by Nelson (2003). In addition to that, Chen et al. (2012) studied the relationship between insider managerial shareholding (IMS) and directors’ shareholding (DIRS) on firm performance. They found a U-inverted relationship. Both IMS and DIRS had a significantly positive impact on firm performance but up to an optimal point. After that point, when both IMS and DIRS were greater than the both optimal points, firm performance decreased.

Bishop and Kaupins (2006) go even further by stating that not only the CEO should possess shares, but that all employees should possess shares. In this way, the company is better able to meet long-term needs and to attract high-quality employees due to the attractiveness of the company.

Hypothesis 2

From the findings mentioned above the following hypothesis can be formed:

Hypothesis 2

(15)

15

Conceptual model

Hypothesis 1 and Hypothesis 2 both result in the following conceptual model:

Conclusion of literature review

The literature review explained both the Agency Theory and the Stewardship Theory. It tried to explain the opposing viewpoint based on their differences. Furthermore, it is shown that comparing both theories is an interesting field for further research, due to the amount of similarities and differences.

In addition, both theories have opposing assumptions of the relationship between their theory and firm performance. These two opposing assumptions are incorporated in the hypotheses and are underlined by stating the prior empirical studies. The empirical studies form the basis of two opposing theories and viewpoints about CEO shareholding. In the next chapter it will be discussed how these two opposing viewpoints will be tested.

CEO Ownership

(Agency theory viewpoint)

Firm performance

Control variables - Firm size - Firm age - Total assets - Number of board directors - Specific industries - Dutch context

CEO Ownership

(16)

16

Methodology

Introduction of methodology

This research will be done using quantitative data to examine the manager/shareholder relationship towards firm performance.

In the coming section the following elements will be highlighted: the data collection, which variables will be used, how these variables will be measured and how the data analysis will take place.

Data collection

To collect the right data, the Orbis database is used. The Orbis database contains information on approximately 365 million companies worldwide making it an enormous source of business data. The database provides easy ways to compare companies internationally.

The following criteria in the Orbis database are held for the selected companies:

1. The company is currently active.

The companies in the data need to be currently active in order to create a currently representative dataset.

2. The company has a maximum of 250 employees.

In this research the focus is on small and medium sized companies. Therefore companies with a maximum of 250 employees are selected, since that is the limit for being a small/medium sized company (OECD, 2019).

3. The company is located in the Netherlands.

The companies must be based in The Netherlands. In order to select these companies, this selection is made with the help of Orbis.

4. The company has a known value for the financial ratio ‘return on assets’.

(17)

17

database. Otherwise it might happen that also unknown values are used for running the regressions.

5. Financial firms are excluded from the dataset, due to their alternative firm structure (this means that the industry sectors, NACE Rev. 2, with the numbers 64 t/m 69, 85 t/m 88, 91, 94 and 96 t/m 99 are excluded).

Due to the alternative firm structure and way of operating, financial firms are excluded from the dataset. In addition, government regulations often affect firm performance.

6. DM is also a shareholder (Parameter in the Orbis dataset)

This parameter is inserted as a “column” in Orbis. With this parameter the managers which are shareholder simultaneously can be found in the data set.

The final dataset results in 7136 companies. After cleaning the dataset for missing, the remaining amount of companies is 6762.

Data distribution

When looking at the distribution of the industries of the selected companies, it is found that the industry type ‘wholesale and retail trade (4500-4799)’ is strongly present. After that, the industry of ‘activities of head offices; management consultancy activities (7000-7022)’ is the most present. The histogram of this distribution can be found in Appendix A.

The firms are observed by taking their last available year where they presented data. That resulted in a data set where 90% of the data is coming from the year of 2018.

Variable measurements

The dependent variable in this study represents return on assets of the companies selected. The dependent variable is labeled as ‘firm performance’. ‘Ownership’ is the independent variable. This variable

(18)

18 Dependent variable - ‘Firm’s performance’

The variable ‘firm performance’ will consist of the financial ratio ‘return on assets’ - ‘ROA’. It is calculated by dividing the net income of a company by its total assets. ROA is a worldwide used proxy for firm performance (Chen and Church, 1996; Al-Sa'eed, 2018).

Independent variable - ‘Ownership’

For this variable, dummy variables will be used. The one group will get a ‘0’ and will consist of managers who are not shareholders of the company, while the other group will get a ‘1’ and will consist of

managers who are also shareholders of the company.

Control variables

Firm size, firm age, total assets and number of board directors will be added as a control variable in order to relate on firms’ performances (Porter, 1981). From all variables the logarithmic function will be taken, consistent with other empirical studies (Garg et al., 2003).

Firm size

Firm size is determined by the number of employees. It is found that larger firms perform better than small firms, since large firms can utilize economies of scale better than small firms (Azeez, 2015).

Therefore, firm size could affect the relationship between ownership and firm performance negatively and therefore it needs to be controlled for.

Firm age

Firm age is determined by taking the current year minus the firm’s founding year. Younger firms tend to have lower sales than older firms (Cabral and Mata, 2003 ; Coad et al., 2015). This is due to the smaller network and the less experience a younger company has. So, age could affect the ownership -

performance relationship, since sales can impact ROA. This is the case if there is also a change in expenses of the company.

Total assets

Total assets is determined by the amount of assets owned by the firm presented as U.S. dollars in thousands. According to Vintila and Nenu (2015) total assets is negatively correlated with firm

(19)

19 Number of board directors

This variable is determined by the amount of board directors of the firm. It is found that firm performance increases when the number of board directors increases (Bachiller et al., 2015).

Data description

Descriptive statistics

In the following table the basic descriptive statistics of the variables is shown.

Table 1. Range and distributions of variables used in the study

Variables N Min. Max. Mean Standar

d dev. Skewnes s Kurtosis Return on assets(1) 7136 -98.36 93.95 7.02 14.70 -0.67 9.40 DM also shareholder (2) 7136 0 1 0.004 0.06 15.87 250.00 Firm age (3) 6762 0 200 30.58 22.66 1.68 4.32 Number of employees (4) 7136 1 250 79.52 57.58 0.91 0.19 Total assets (5) 7136 2.39 246,440,0 00 142,04 0.12 3197096. 23 66.48 4978.34 Number of board directors (6) 7136 0 124 9.87 8.74 2.75 15.17

For ‘return on assets’, the minimum is -98.36, which means that the company has more invested

capital or that the company earns lower profits. The opposite holds for when return on assets has a positive value. Capital intensive industries will have a lower ROA percentage than labor intensive industries. An example of this is that, normally, the ROA of a software company will be higher than the ROA of a car manufacturer. The average ROA in the dataset is 7%, which means that the companies, on average, have a profit of 7% on every dollar they have spent on their assets.

(20)

20

For ‘firm age’, it can be seen that the youngest firm is 0 year old, which means that the firm has recently started its operations, and that the oldest firm is 200 (!) years old. The average age of the companies in the dataset is around 31 years.

The minimum for ‘number of employees’ is 1 and the maximum is 250, which are the limits for being a small- and medium enterprise. On average the SME’s in this dataset have around 80 employees. This is relatively high since a SME in the Netherlands has 3.2 employees on average (SBA Fact Sheet, 2018).

The ‘total assets’ of the companies in the dataset has a minimum of 2.39, which indicates a possible self-employed home worker with no assets. The maximum is 246,440,000, which indicates a relatively large SME.

For the ‘number of board of directors’, the minimum is 0, the maximum is 124 and the mean is around 10.

For skewness, if its value is positive, the data is positively skewed or skewed to the right. This means that the right side of the distribution is more/longer than the left side. If skewness has a negative value, the data is negatively skewed or skewed to the left. This means that the left side of the distribution is more/longer than the right side. The variable total assets has a relatively high positive value for

skewness, which means that the most companies have total assets which are less than the average value of total assets. This is understandable, since most SME’s will have a total assets value which is less than the average of 142,040 thousand U.S. dollars.

For kurtosis, the rule of thumb is that if the value is greater than 1, the distribution is too peaked. On the other hand, if the value of kurtosis of less than –1 indicates a distribution that is too flat. It can be seen that number of employees is the only variable with a kurtosis which is desirable. All other variables are relatively peaked distributions.

Transforming the variables

All the control variables will be transformed into logarithmic functions. Transforming the control variables ‘firm age’, ‘firm size’, ‘total assets’ and ‘number of board directors’ into logarithmic variables will result in having better distributions, more normality and less outliers.

Difference of means test

The table below shows the difference of means tests between ‘DM also shareholder’ and ‘DM not shareholder’ firms. ‘DM also shareholder’ firms represent 0.39% of the whole sample. The means tests are based on every variable of this study.

(21)

21

significant difference between ‘DM also shareholder companies’ and ‘DM not shareholder companies’, in relation to the number of board directors.

Table 2. Difference of means test

Variable Manager also

shareholder Manager not shareholder t-statistic Number of firms 28 7108 x ROA -0.10 7.06 2.57 Log of age 3.09 3.14 0.28 Log of size 4.02 4.02 0.01

Log of total assets 10.37 9.86 -2.24

Log of number of directors in board 1.62 1.95 2.03**

*p<0.05 (2-tailed) ** p<0.01 (2-tailed)

(22)

22 Correlation matrix

In the correlation matrix below, the correlation coefficients between the variables are shown. Each cell in the table shows the correlation between two variables. With this matrix the data of the variables can be summarized. Also, it is the first step towards a more advanced analysis.

Table 3. Correlation matrix

Variables ROA (1) DM also

shareholde r (2) Log of firm age (3) Log of number of employees (4) Log of total assets (5) Log of number of directors (6) ROA (1) 1 -0.030* 0.044** -0.015 -0.071** -0.088** DM also shareholde r (2) -0.030* 1 -0.003 -0.000 0.026* -0.024* Log of firm age (3) 0.044** -0.003 1 0.129** -0.015 0.225** Log of number of employees (4) -0.015 -0.000 -0.129** 1 0.086** 0.189** Log of total assets (5) -0.071** 0.026* -0.015 0.086** 1 0.316** Log of number of directors (6) -0.088** -0.024* 0.225** 0.189** 0.316** 1 N (number of observations)= 7136 *p<0.05 (2-tailed) ** p<0.01 (2-tailed)

From the matrix it can be derived that ‘ROA’ significantly correlates with ‘DM alsho shareholder’, ‘log of firm age’, ‘log of total assets’ and ‘log of number of directors’. ‘DM also shareholder’ significantly correlates with ‘ROA’, ‘log of total assets’ and ‘log of number of directors’. ‘Log of firm age’

(23)

23 employees’, ‘log of total assets’.

The highest correlation is between ‘log of number of directors’ and ‘log of total assets’. This means that these two variables have the strongest positive correlation among all variables used. These two variables do influence each other the most. This also holds for ‘log of firm age’ and ‘log of number of directors’, but the underlying relation is slightly less strong.

Industry fixed effects

Industry fixed effects will be included in the regressions to capture industry-specific phenomena. The industries are divided over 14 groups, as showed in the table in Appendix C. The dividing of the groups is based on their industry characteristics.

Data analysis

In this study the effects of the independent variables and the control variables on the dependent variable will be examined. This will be done by analyzing the data by executing a linear regression analysis. In order to do so, a selection of steps need to be made first in order to create the right situation to fulfill the analysis.

(24)

24

Results

Testing the assumptions

In order to know whether a linear regression is possible, its assumptions need to be tested. This will be done by taking the following steps in a chronological order: Linear relationship ; Homoscedasticity ; Checking for normality ; Multicollinearity.

Linear relationship

Linearity is tested to see if there is a linear relationship between the dependent variable and the

independent variable. Linearity is checked by means of a scatterchart, which can be seen in Appendix D. For linearity, the dots must be placed in a random cloud around the zero line. This non-systematic deviation from the zero line, can be seen in the scatterplot to a certain extent. Therefore, we can assume linearity.

Homoscedasticity

In order to prove potential homoscedasticity the dots in the scatter plot need to be randomly distributed. As can be seen in Appendix D, the scatter plot of the dependent variable ROA appears to be not

homoscedastic, due to the cluster of dots. The scatter plot shows potential heteroskedasticity, since there is more variation around the zero at lower values of the x-axis.

Checking for normality

For checking the normality the P-P plot will be used. Ideally, the dots around the diagonal line will follow the line closely. If this is not the case, the data is not normally distributed. As can be seen in Appendix E, there is a slightly deviation visible. Based on the P-P plot, the deviation is not drastic and therefore normality can be assumed.

Multicollinearity

(25)

25

Running the regressions

In the coming section several models will be run. This will be done with the use of linear regression. All models will be controlled for industry fixed effects by inserting all industries as independent variables.

The first model (column ‘Model 1’ of Table 4) consists of the dependent variable, ‘firm performance’, and the independent variable of ‘DM also shareholder’.

The second model (column ‘Model 2’ of Table 4) consists of the dependent variable, the independent variable and the control variable ‘firm size’ will be added to this.

The third model (column ‘Model 3’ of Table 4) consists of the dependent variable, the independent variable and the control variables, ‘firm size’ and ‘firm age’.

The fourth model (column ‘Model 4’ of Table 4) consists of the dependent variable, the independent variable and the control variables, ‘firm size’, ‘firm age’ and ‘total assets’.

The fifth model (column ‘Model 5’ of Table 4) consists of the dependent variable, the

independent variable and the control variables, ‘firm size’, ‘firm age’, ‘total assets’ and ‘number of board directors’.

Table 4. Table of different regression models

Model 1 Model 2 Model 3 Model 4 Model 5

(26)

26 Firm age x x 0.601* (0.263) 0.515† (0.263) 0.774** (0.265) Total assets x x x -0.743** (0.149) -0.445** (0.157) Number of board directors x x x x -1.418** (0.231) Industry dummies

Yes Yes Yes Yes Yes

(27)

27

* p<0.05 ** p<0.01

(2-tailed)

Discussion of findings

In this section the different models will be analyzed and also the two hypotheses will be discussed.

Model 1

The first model (column ‘Model 1’ of Table 4) consists of the dependent variable, ‘firm performance’, and the independent variable of ‘DM also shareholder’. The effect size is not significant and therefore this effect size does not hold. Furthermore is the R-squared of this model 0.003, which means that 0.3% of variance of ROA is explained. The F-value is 1.646 and is significant on a 10% level. This F-value compares the current model with zero predictor variables, the intercept only model, and it indicates whether the added coefficients improved the model or not.

Model 2

In the second model (column ‘Model 2’ of Table 4) a control variable is added. The effect sizes are not significant and therefore these effect sizes do not hold. Furthermore, the R-squared of this model 0.003, which means that 0.3% of variance of ROA is explained. The F-value is 1.579 and is significant on a 10% level.

Model 3

In the third model (column ‘Model 3’ of Table 4) another control variable is added. The effect size of ‘log of firm age’ is 0.601 and is significant on a 5% level. This means that if we change ‘log of firm age’ by 1 unit, it is expected that ROA changes by 60.1%. The other effect sizes are not significant and cannot be interpreted. Furthermore is the R-squared of this model 0.004, which means that 0.4% of variance of ROA is explained. The F-value is 1.823 and is significant on a 10% level. Compared to the previous models, the addition of the control variable ‘total assets’ makes the model having a slightly better fit.

Model 4

(28)

28

change by 1 unit, ROA will change with – 74.3%. Furthermore is the R-squared of this model 0.008, which means that 0.8% of variance of ROA is explained. The F-value is 3.263 and is significant. Compared to the previous models, the addition of the control variable ‘total assets’ makes the model having a slightly better fit.

Model 5

In the fifth model (column ‘Model 5’ of Table 4) the control variable ‘number of board directors’ is added. The effect size of ‘DM also shareholder’ is -5.283 and is significant on a 10% level, which means that being both a manager and shareholder will decrease ROA by 5.283 percentage point. For ‘firm age’ the effect size is 0.774 and is significant on a 1% level. For ‘total assets’ the effect size is -0.445 and is significant on a 1% level. For ‘number of board directors’ the effect size is -1.418 and is significant on a 1% level. Furthermore is the R-squared of this model 0.014, which means that 1.4% of variance of ROA is explained. The F-value is 5.314 and is significant on a 1% level. Compared to the previous models, this model is explained the best.

Hypothesis testing

In this section, the hypotheses are tested with the use of the linear regression analysis. The hypotheses are tested by analyzing the fifth model, because this models contains all control variables and had the highest R-square.

The first hypothesis: "Being both the manager as a shareholder negatively relates to firm performance." can be supported, since in the fifth model the dummy variable ‘DM also a shareholder’ is negative and significant on a 10% level.

The second hypothesis: "Being both the manager as a shareholder positively relates to firm

performance.", cannot be supported. Since, as discussed above, the fifth model has a significant negative effect size.

Remarkable results

The control variables firm age, total assets and number of board directors show significant results in relation to firm performance. For total assets and number of board directors this is the most remarkable observation.

(29)

29

This is because the costs for the company increase and also the controllability and the internal

organisation takes more time and costs. While, on first instance, it is commonly thought that more assets will lead to more firm performance.

Focusing on ‘board of directors’, it proves that the Agency Theory holds. For number of board directors, it is also the case that firm performance will decrease when the number of board directors increases. A larger board means more opinions and longer decision time, which is negative for the response time of a company. The longer the response time of a company is, the more probability there is that customers will choose for a competitor. Apparently, the more captains on the ship, the less quality of leadership there is.

Results related to literature

The results of this study are in line with Sanders (2001) and Park and Song (1995). Sanders states that giving ownership to one of the employees of the company will result in an increase in risk aversion among employees. This leads to less risk-taking of the company. According to Park and Song, employee ownership increases the agency costs, which makes the company less efficient.

On the other hand, Kim (2018) found that firms which gave ownership to their employees experienced higher performance. In addition to that, findings that focus on employee ownership and firm performance, are mixed. According to Kim and Patel (2017) this is due to several factors such as cultural factors, between-industry differences and firm-specific components.

Alternative regressions

Additional filters and variables were selected in this original dataset in order to test whether alternative regressions would give different results. It was founded that this did not gave any significant other results. Therefore, it is chosen to use the original dataset and variables.

(30)

30

Discussion and conclusion

Discussion

This study revisited a topic which has interest to many business scholars: the conflict between Agency Theory versus Stewardship Theory. Both theories have their own governance mechanisms for managing a company and this study looks at the conflict in the context of small- and medium-size enterprises. The results of this study provide interesting insights towards the opposing literature of the Agency Theory and the Stewardship Theory. In terms of management, this study is highly relevant for company owners with regard to their choice of making the manager shareholder or not.

The main findings of this study represent the importance of giving ownership to managerial functions in order to create responsibility and leadership.

The first hypothesis stated that being both the manager as a shareholder negatively relates to firm performance. The coefficients from all the models which tested this hypothesis were negative and the last model was significant.

The second hypothesis stated that being both the manager as a shareholder positively relates to firm performance. As stated earlier, all coefficients from all models were negative but and the last was significant, so the second hypothesis cannot be supported.

Implications

Theoretical implications

The existing scientific literature has mainly focused on either the Agency and Stewardship comparison or on the ownership and performance relationship, but this study combined those two elements. By focusing on the combination of both, this study addressed two highly popular topics in the literature of

management.

In addition, the mainstream literature is based on large firms, but for SME’s there is no ‘one thing fits all’ strategy. This study is an addition to the SME focused literature. Even though this study does not have significant results, due to this study it is now also known that this field of research has already been done.

Managerial implications

(31)

31

and, on the other hand, if ownership helps for having better performance.

Another potential implication is that managers of SMEs can relate to this study rather than managers of large companies. This study found significant results for supporting the Agency Theory, which could be used by managers. In addition, this study elaborated on both theories in the literature review section and in that way managers of firms can understand what both theories mean and then they can choose which theory suits their business the best.

Limitations

The findings of this study are subject to some limitations, due to, among others, limited resources.

Validity of performance measurement

For this study ROA is chosen as performance indicator, since it is used in numerous other highly ranked articles and it is regarded as one of the better performance indicators. Even though this is a good firm performance indicator, there are more factors which can measure firm performance, for instance ‘return on equity’ or ‘profit margin’.

Sample size

The overall sample size lays above the 6000 companies. However, taking a closer look at the sample size shows that only a tiny part of the companies have a manager who is also a shareholder of the company. Only 0.39% of the sample companies have a manager who is also shareholder of the company. This unequal balance is not in the better of the study, by having an unequal distribution of the two groups.

Data distribution

The histogram of Appendix A shows that ‘wholesale and retail trade (4500-4799)’ is the main industry which is highly represented in this study. Therefore, there is not an equal balance between the number of companies per industry. This might lead to results which do not represent reality.

Small- and medium-sized enterprises

This study focused on small- and medium-sized enterprises. Therefore, the findings might not be generalizable to larger firms due to the different firm structure.

Representativity

(32)

32

that micro firms don’t need to prepare financial reports and therefore have missings in Orbis.

Significance level

In the fifth model the significance level of the independent variable is on a 10% level. This is a relatively high significance level, since in most studies a significance level of 5% or 1% is used.

Future research

This study can be seen as an addition to the studies which are interested in the ‘management ownership and firm performance relationship’. For this study, the focus on the two theories has been done in a quantitative way but more qualitative research is also needed. This quantitative research focuses on studies with the same topic, while for companies it is also interesting to find out what is playing ‘in the field’. In addition, longitudinal panel studies which will study this topic will bring even more trustworthy results to support the findings of this study.

In addition, the control variables showed several remarkable results. For future research it would be interesting to further study one or more of these control variables in relation to firm performance, since these results indicate that more research will bring promising results.

Conclusion

In conclusion, this study has provided insight on share-owning managers versus non-share owning managers, Agency versus Stewardship and which approach is better. It is a discussion about a contest which is based on two elements. The one contest is about ownership for the manager and the other contest is about the Agency Theory and the Stewardship Theory. As seen in the results, the most important outcome of this study is that there is a relationship between whether or not the manager is a shareholder and the performance of a firm. The difference in having shares or not as a manager apparently does influence firm performance at all. This means owners of companies might want to reduce the share owning of their managers in order to boost firm performance. Other factors do also have significant influence on firm performance rather than only owning shares as a manager. These factors are interesting components for future research. For example, the number of board directors in a company, this finding of more board directors and less firm performance is in line with the Agency Theory.

(33)

33

(34)

34

References

Al-Sa'eed, M. A. A. (2018) The Impact of Ownership Structure and Dividends on Firm’s Performance: Evidence from Manufacturing Companies Listed on the Amman Stock Exchange, Australasian Accounting, Business and Finance Journal, 12(3), pp. 107-126

Andersén, J., Jansson, C. and Ljungkvist, T. (2020) Can environmentally oriented CEOs and environmentally friendly suppliers boost the growth of small firms?, Business Strategy and the Environment, Volume 29, Issue 2, pp. 325-334

Ang, J. S., Cole, R. A., and Lin, J. W. (2000) Agency cost and ownership structures, The Journal of Finance, 55(1)

Arthurs, J. D. and Busenitz, L. W. (2003) The Boundaries and Limitations of Agency Theory and Stewardship Theory in the Venture Capitalist/Entrepreneur Relationship, Entrepreneurship Theory and Practice, Vol. 28, Issue 2

Azeez, A. A. (2015) Corporate governance and firm performance: Evidence from Sri Lanka Journal of Finance and Bank Management, 3, pp. 180-189

Bachiller, P., Giorgino, M.C. and Paternostro, S. (2015) The Relationship Between The Board of Directors and the Performance Analysis of Family and Non Family Firms in Italy, Department of Business and Social Studies School of Economics University of Siena

Begley, T.M. and Boyd, D.P. (1987) A comparison of Entrepreneurs and Managers of Small Business Firms, Journal of Management, Vol. 13, No. 1, 99-108

Bingham, C., Hendricks, B. E., Howell, T. and Kolev, K. (2019) Boomerang CEOs: What Happens when

the CEO Comes Back?, Kenan Institute of Private Enterprise Research Paper, No. 19-27

(35)

35

Boyd, B. K. (1995) CEO duality and firm performance, a contingency model, Strategic Management Journal, 16(3), 301-312

Cabral, L.M.B. and Mata, J. (2003) On the evolution of the firm size distribution: facts and theory, American Economics Review, 93 (4), pp. 1075–1090

Carland, J.W., Hoy, F., Boulton, W.R. and Carland, J.A.C. (1984) Differentiating entrepreneurs from small business owners: A conceptualization, Academy of Management Review, 9, 354-359

Chauhan, Y. and Kumar, S. (2017) Does founder ownership affect foreign investments? Evidence from India, Emerging Markets Review, Vol. 32, p. 116-129

Chen K. C. W., Church B. K. (1996). Going Concern Opinions and the Market's Reaction to Bankruptcy Filings. The Accounting Review, 71(1), pp. 117-128

Chen, M. H., Hou, C. L. and Lee, S. (2012) The impact of insider managerial ownership on corporate performance of Taiwanese tourist hotels, International Journal of Hospitality Management, 31 pp.338– 349

Coad, A., Segarra, A. and Teruel, M. (2013) Like milk or wine: Does firm performance improve with age?, Elsevier, Volume 24, pp. 173 - 189

Coad, A., Segarra, A. and Teruel, M. (2015) Innovation and firm growth: Does firm age play a role?, Elsevier, Volume 45, Issue 2, pp. 387-400

Corbetta, G., Salvato, C. (2004) Self-serving or self-actualizing? Models of man and agency costs in different types of family firms: a commentary on “Comparing the agency costs of family and non-family firms: conceptual issues and exploratory evidence”. Entrepreneurship Theory and Practice, 28, (4), 355– 362.

Cronbach, L. J. (1951) Coefficient alpha and the internal structure of tests, Psychometrika, vol. 16, no. 3

(36)

36 269–290

Dalton, D. R., Daily, C. M., Johnson, J. L. and Ellstrand, A. E. (1999) Number of Directors and Financial Performance: A Meta-Analysis, The Academy of Management Journal, Vol. 42, No. 6, pp. 674-686

Demsetz, H. and Villalonga, B. (2001) Ownership structure and corporate performance, Journal of Corporate Finance, Volume 7, Issue 3, pp. 209-233

Davis, J. H., Schoorman, F. D., & Donaldson, L. (1997). Toward a stewardship theory of management. Academy of Management Review, 22, 20–47

Eisenberg, T., Sundgren, S.; and Wells, M. T. (1998) Larger Board Size and Decreasing Firm Value in Small Firms, Cornell Law Faculty Publications, 393.

Eisenhardt, K. M., (1989) Agency Theory: An Assessment and Review, The Academy of Management Review, Vol. 14, No. 1, pp. 57-74

European Commision (2018) SBA Fact Sheet - Netherlands, EC Europe

Fama, E. F. (1980). Agency problems and the theory of the firm, Journal of Political Economy, 88, 288-307

Fattoum-Guedri, A., Delmar, F. and Wright, M. (2016) The best of both worlds: Can founder-CEOs overcome the rich versus king dilemma after IPO?, Wiley Online Library

Fitzal, M. and Tihanyi, L. (2017) How Much Does Ownership Form Matter?, Strategic Management Journal, 38: pp. 2726–2743

Garg, V., Walters, B. and Priem, R. (2003) Chief executive scanning emphases, environmental dynamism, and manufacturing firm performance, Strategic Management Journal, 24(8): 735–744. Gedajlovic, E. R. and Shapiro, D. M. (1998) Management and ownership effects: evidence from five countries, Strategic Management Journal, Vol. 19, 533–553

(37)

37

Hendricks, B., Howell, T. and Bingham, C. (2019) How much do top management teams matter in founder-led firms?, Wiley Online Library

James, A. E., Jennings, J. E. and Devereaux Jennings, P. (2017) Is It Better to Govern Managers via Agency or Stewardship? Examining Asymmetries by Family Versus Nonfamily Affiliation, Family Business Review, Vol. 30 (3) pp. 262–283

Jensen, M.C. and Meckeling, W.H. (1976) Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3 p. 305-360

Kim, K. Y. (2018). The causal effect of employee ownership on firm performance: Evidence from a natural experiment. Conference of European Academy of Management, Reykjavik, Iceland.

Kim, K. Y. and Patel P. C. (2017). Employee ownership and firm performance: A variance decomposition analysis of European firms, Journal of Business Research, 70, pp. 248–254

Le Breton-Miller, L., & Miller, D. (2009). Agency vs. stewardship in public family firms: A social embeddedness reconciliation. Entrepreneurship Theory and Practice, 33, 1169–1191.

Lee, J. M., Hwang, B. H. and Chen, H. (2016) Are founder CEOs more overconfident than professional CEOs? Evidence from S&P 1500 companies, Strategic Management Journal

Li, J. (1994) Ownership structure and board composition: A multi-country test of agency theory predictions, Management Decision Economics, 15 (4), pp. 359–368

Madison, K., Holt, D. T, Kellermanns, F. W., and Ranft, A. L. (2015) Viewing Family Firm Behavior and Governance Through the Lens of Agency and Stewardship Theories, Family Business Review, Vol. 29, Issue 1

Market Realist (2020) Is Jeff Bezos Going to Dump More Amazon Shares?

https://marketrealist.com/2020/02/is-jeff-bezos-going-dump-more-amazon-shares/

(38)

38

McGuire, J. and Matta E. (2003) CEO Stock Options: The Silent Dimension of Ownership, The Academy of Management Journal, Vol. 46, No. 2, pp. 255-265

Mehran, H. (1995) Executive compensation structure, ownership, and firm performance, Journal of Financial Economics, 38 (2), pp. 163–184.

Mitnick, B. M. 1973. Fiduciary rationality and public policy: The theory of agency and some

consequences. Paper presented at the 1973 Annual Meeting of the American Political Science Association, New Orleans, LA. In Proceedings of the APSA, 1973 (formerly available from Xerox University

Microfilms and, later, UMI Serials).

Nelson, T. (2003) The persistence of founder influence: management, ownership, and performance effects at initial public offering, Strategic Management Journal, 24: pp. 707–724

Nicholson, G. J. and Kiel, G. C., (2007) Can Directors Impact Performance? A case‐based test of three theories of corporate governance, Corporate governance: An International Review, 15(4), 585-608

OECD (2019) OECD SME and Entrepreneurship Outlook 2019, Policy Highlights http://www.oecd.org/industry/smes/SME-Outlook-Highlights-FINAL.pdf

Pierce, J.L., Kostova, T., Dirks, K.T. (2001) Toward a theory of psychological ownership. Academy of Management Review, 26, (2), 298–310.

Porter, M. E. (1981). The contributions of industrial organization to strategic management. Academy of Management Review, 6(4), 609–620.

van Puyvelde, S., Caers, R., Bois, D.U.C., Jegers, M. (2012) The governance of nonprofit organizations integrating agency theory with stakeholder and stewardship theories, Nonprofit Voluntary Section, Q. 41, 431–451

(39)

39

Ross, S. A. (1973) The Economic Theory of Agency: The Principal's Problem, The American Economic Review, Vol. 63, No. 2, pp. 134-139

Schuster, C. L., Nicolai, A. T., and Covin, J. G. (2018) Are Founder-Led Firms Less Susceptible to Managerial Myopia?, Entrepreneurship Theory and Practice, Vol. 44(3) 391–421

van Slyke, D.M. (2007) Agents or stewards: using theory to understand the government-nonprofit social service contracting relationship, J. Public Adm. Res. Theory, 17, 157–187.

Stewart, W. H. and Roth, P. L. (2001) Risk propensity differences between entrepreneurs and managers: a meta-analytic review, Journal of applied psychology, vol. 86, No. 1, 145-153

Vintila, G. and Nenu, A. (2015) An Analysis of Determinants of Corporate Financial Performance: Evidence from the Bucharest Stock Exchange Listed Companies, International Journal of Economics and Financial Issues, 5(3), 732-739

Williamson, O. E., (1967) Hierarchical Control and Optimum Firm Size, Journal of Political Economy, Vol. 75, No. 2, pp. 123-138

Williamson, O. E., (1985) The Economic Institutions of Capitalism, Simon and Schutser, Free Press, New York

(40)

40

Appendices

Appendix A

(41)

41

Appendix B

Table of variables used

Variable Measurement Source of data

Firm performance (DV)

The financial ratio ‘return on

assets’ Orbis

Ownership (IV) Binary variable. Both

manager as shareholder of the company: yes/no.

Orbis

Firm size (CV) Number of employees. Orbis

Firm age (CV) Amount of years since the company was founded.

Orbis

Total assets (CV) Amount of assets owned by the firm.

Orbis

Number of board directors (CV)

Amount of board members of the firm.

(42)

42

Appendix C

Industry classifications for industry fixed effects

Industry number NACE number (2 digit) General description

Industry 1 01 - 02 - 03 Natural extraction

Industry 2 05 - 06 - 07 - 08 - 09 Crude industry

Industry 3 10 till 32 Manufacturing

Industry 4 33 till 39 Basic facilities services

Industry 5 41 - 42 - 43 Construction

Industry 6 45 - 46 - 47 Wholesale and trade

Industry 7 49 - 50 - 51- 52 - 53 Transportation

Industry 8 55 - 56 - 58 - 59 - 60 Entertainment

Industry 9 61 - 62 - 63 IT

Industry 10 70 - 71 - 72 - 73 - 74 Professional activities

Industry 11 75 - 77 - 78 - 79 Travelling

Industry 12 80 - 81 - 82 - 84 Office activities

Industry 13 85 - 86 - 87 - 88 Social activities

(43)

43

Appendix D

(44)

44

Appendix E

(45)

45

Master Thesis for MSc BA SB&E

_______________________________________