Military CEOs: discussing battle plans, or investing in CSR activities?

Military CEOs: discussing battle plans, or

investing in CSR activities?

MSc Thesis Accountancy

MSc Accountancy University of Groningen Faculty of Economics and Business

Alien Waaijman S2891662 Plutolaan 100 9742 GT Groningen +316 30 703 835 a.c.waaijman@student.rug.nl Supervisor S. Wang Word count: 8,520

2

Abstract

In the corporate world, CEOs run the business. Their view on CSR will highly influence the CSR strategy of the entire firm. This research encounters the military background of the CEO and investigates the influence of this military service on the CSR investments of the company. In this thesis, the influence of CEO military background on CSR net score will be examined. The sample data was aquired from the KLD database and uses 15,985 unique observations from US companies in the years 2000-2016. Besides this relation, the influence of the political view from the state of the company’s headquarters on the CSR net score is investigated as well as the difference between a CEO with a long or short military service tenure and the amount of years that it has been since the CEO was in the military. A relation between the military CEO and the CSR net score was expected, nonetheless no significant relation was found between these variables by means of a fixed-effect panel regression analysis. A possible reason for this finding could be that the variable CSR net score is too broad and should be further divided into multiple factors.

Keywords: Corporate Social Responsibility, Military CEO, Military Background, Military

3

Table of contents

4.2 Pearson correlation matrix ... 19

4.3 Regression analysis ... 21

4.4 Additional regression analyses ... 23

5. Discussion and conclusion ... 26

5.1 Limitations and future research ... 28

4

1. Introduction

hat do Clayton M. Jones, Michael G. Morris and Steven R. Loranger have in common? They are all successful businessmen and CEOs. Besides that, they all served in the military (Duffy, 2006). Only 30 years ago, it was very common for veterans to go into business after serving in the military. In fact, in 1980, 59% of the male CEOs of large organizations had served in the military (Benmelech and Frydman, 2015). Today this share is much smaller, but still, it is stated that military CEOs can be beneficial for a company’s performance (Benmelech and Frydman, 2015). Multiple studies have already investigated the influence of veterans in corporations with business outcomes. For instance, Benmelech and Frydman (2015) discovered that military CEOs are less likely to commit fraud and that firms with a military CEO have better financial results during industry downturns. Overall, they showed that military backgrounds of CEOs influence the managerial decision making and with that, the financial results of the company. Another study that found a negative influence of military CEOs on fraud were Koch-Bayram and Wernicke (2018). They found that CEOs with a military background were less tended to be involved in fraudulent financial reporting and backdating stock options. This influence is, according to Koch-Bayram and Wernicke (2018), moderated by board oversight.

Veterans are often considered to be hardworking, loyal, honest and ethically correct people (Duffy, 2006). CEOs that gain these characteristics during their military service, take these with them to the company that they work at after their time in the army (Benmelech and Frydman, 2015). These characteristics are often linked to less fraudulent behavior, which negatively influences the likelihood of fraud in the company (Benmelech and Frydman, 2015).

More and more companies invest in corporate social responsibility (CSR) activities nowadays (Dentchev, 2004). The drivers of this behavior have been under investigation for a long time. For example, Petrenko et al. (2016) show that CSR can arise when the CEO of a company wants more attention or when the leader of the company needs to change his image. CEO narcissism is a characteristic that is linked to these earlier mentioned events and CEO narcissism is found to have positive influence on the CSR activities of a company (Petrenko et al., 2016). CSR can have a positive effect on the financial performance of a company (Simanaviciene, Kontautiene and Simanavicius, 2017), but when CEO narcissism is the driver of the CSR activities, this will reduce this positive effect that CSR has on firm performance (Petrenko et al., 2016). Another driver of CSR activities, can be a strategic motivation to gain

5

a competitive advantage or when a company has a political motivation (Campbell, Moore and Metzger, 2002). If not carefully handled, CSR could also harm the competitive advantage of a firm when it is implemented poorly or when there are conflicts of interest between the stakeholders and the managers of the company (Dentchev, 2004).

CEO characteristics are of great importance to a company. They can influence CEO decision making and with that the financial results of an organization (Hambrick and Mason, 1984). Since the military background of a CEO influences their characteristics (Elder, 1968), it is very important to do more research to this military background of CEOs and how it affects decision making. This paper will investigate the relationship between the military background of CEOs and the decisions they make regarding their corporate social responsibility activities. This link is made because the CEO has a significant influence on the decision-making of a company, which also includes the decisions related to corporate social responsibility. If all of this is taken into account, the following research question arises:

RQ: Are CEOs of US companies with a military background associated with more (less) corporate social responsibility activities?

CEO military background is combined with the CSR activities because CSR is a major phenomenon among firms nowadays, and there are still a lot of questions regarding the reasons why companies implement CSR or not. When investigating the CEO characteristics, and especially the military background of the CEO, it is hoped that this influences the kind of CSR decisions that a company makes. The expectation of this research is that there is an influence between these variables. In our main hypothesis, the expectation is set up that there will be a positive influence between CEO military background and CSR investments. This expectation arises from the transformational leader theory that will be further explained in the theoretical framework of this thesis.

This paper will utilize a sample of 15,985 unique observations retrieved from the KLD database and data regarding CEO military background to answer the research question. This data is collected in the period of 2000-2016. The main findings did not include a direct relation between CEO military background and CSR investments. However, it was discovered that CEOs with a military background were less likely to score higher results on the CSR concerns. Which would indicate that military CEOs are not more likely to invest more in CSR but that they are willing to avoid creating CSR concerns.

6

Since CEOs with a military background occur less in the present day, one could be of the opinion that this research question is outdated. However, this paper focusses on CSR, which is a hot topic, and investigates whether military CEOs and their characteristics could influence CSR decision making. Given this, the research question is of high importance for the social dilemmas of today.

Besides this, the research adds to a better understanding of the influence of the characteristics of the CEO and the influence of these characteristics on the company decision making. This research provides a view of the CEO characteristics in the CSR decision making, which gives a broader understanding of this topic. This is crucial for the bases of further research about the CEO characteristics and the influence the CEO has on the company decision making of corporate social responsibility.

This study can be linked to studies about military CEOs and to studies regarding CSR activities. Numerous studies have investigated military CEOs and the influence of these CEOs on the financial performance of the firm. There has also been a lot of research about the characteristics of the military CEO and the organizational outcomes that differ from organizations that have a CEO without a military background. Research about CSR activities mostly investigates the influence of CSR on the financial performance of the firm or the motivation of CEOs to invest in CSR activities. In this study these two variables will be combined to fill in the research gap in this area, which is the combination of US military CEOs and their investments in CSR activities.

Besides this, the literature around military CEOs has not yet been fully developed. For example, Koch-Bayram and Wernicke (2018) indicate in their paper that a lot of aspects of military experience have not yet been investigated properly and that more research should be done on how military CEOs could influence for example governance or the strategy of the company. Koch-Bayram ad Wernicke (2018) thus indicate that the literature of military CEOs and their influence on strategy of the company is not full-grown yet and this paper could contribute to developing the literature on these subjects.

In previous research, the link between military CEOs and corporate philanthropy has been made by Luo, Xiang and Zhu (2017). Luo, Xiang and Zhu (2017), focus on corporate philanthropy and the influence military CEOs have on philanthropy decisions, instead of CSR. In this paper, more components of CSR will be included than just corporate philanthropy, since CSR is so much more. Examples of these components are environment, product diversity, employee relations and human rights.

7

Another way that this paper differs from Luo, Xiang and Zhu (2017) is that this paper studies military CEOs of US companies, while Luo, Xiang and Zhu (2017) studied Chinese organizations. Since Chinese and US military CEOs could have very different characteristics, there could be different outcomes when looking at US companies instead of Chinese companies. Since there hasn’t been a study that investigated the relationship between US military CEOs and CSR activities this gap in the literature needs to be investigated. Luo, Xiang and Zhu (2017) found in their paper that military CEOs from China perform less corporate philanthropy than CEOs that haven’t served in the military. In this paper it will be investigated whether US military CEOs invest more or less in CSR activities, which also includes corporate philanthropy, than non-military CEOs.

8

2. Theoretical framework

Military CEOs have been around for a long time but are starting to disappear from the corporate world. These CEOs have characteristics that could be useful, but on the other hand could harm a company. There has already been done a lot of research regarding characteristics of a military CEO. This research has shown that CEOs that have served in the army, navy, air force, etcetera have good leadership qualities (Waldman, Siegel and Javidan, 2006), but these qualities come from different characteristics. Military CEOs are often found to be overconfident and aggressive (Elder, 1968), which can benefit to their leadership role, but is not necessarily a good quality for the firm’s business results. Since the upper echelons theory by Hambrick and Mason (1984) states that CEO characteristic can influence the organizational outcomes of a company, the claim can be made that leadership characteristics of a CEO influences the outcomes of a company and possibly also other decision-making activities of the CEO.

Another quality of military CEOs that has been investigated before, is their sense of ethics and their ability to remain calm under pressure (Duffy, 2006). This quality of being more honest and ethical is very different from the characteristic of being overconfident and aggressive, but also benefits the leadership role of military CEOs. A more ethical behavior and the ability to remain calm can also influence the organizational performance according to the upper echelons theory (Hambrick and Mason, 1984).

These qualities and characteristics that people gain whilst serving in the military, are qualities that they get when they are still young of age. When they grow older, these characteristics could fade away or stay with them for a longer period of time. Previous research has shown that the longer a person has served in the military, the more these characteristics stay with them when they reach a certain age (Jackson et al., 2012). The study of Jackson et al. (2012) shows that the values that people gain during military training persist in their character after completing this training. They show that for example people that have had military training are more agreeable than people that haven’t had this training, and that this difference in agreeableness persists after they have completed the military training (Jackson et al., 2012). The study of Elder and Clipp (1989) shows the enduring psychological effect of combat and war on people in the military. These effects can for some part be positive qualities that persist for the rest of their lives (Elder and Clipp, 1989). In their study, Elder and Clipp (1989) found that most veterans believe that they gained a lot of desirable experiences from their time in combat like "learning to cope with adversity," "self-discipline,"

9

"greater independence," and "a broader perspective”. These were experiences that Elder and Clipp (1989) found years after the military service of the veterans which were thus experiences that endured for multiple years. These studies are examples that show that the values and qualities of veterans that they gain during their military service won’t fade away over time. This indicates that if a veteran goes into the corporate world and becomes a CEO, these values are still with them and can influence the way they behave.

As mentioned before, the upper echelons theory of Hambrick and Mason (1984) explains that CEO characteristics influence their decision making. With the information stated above, the conclusion can be drawn that there is a possibility that CEOs with a military background make different decisions than CEOs without a military background, since they have different values which they gained during military service (Elder, 1968; Elder and Clipp 1989; Jackson et al., 2012). These decisions for the company are also made regarding CSR activities. In the next paragraphs it will be discussed why the CEO characteristics could influence the CSR decisions.

Corporate Social Responsibility (CSR) can be defined as actions that a firm takes that are not necessarily in the firms’ best interest but can benefit others outside of the company (McWilliams and Siegel, 2001). The framework of CSR is the core concept but can be broken down into multiple different factors according to Carroll (2015). These factors are business ethics (BE), stakeholder management (SM), corporate citizenship (CC) and sustainability (SUS) (Carroll, 2015). When performing CSR activities, organizations can protect the different factors by averting the activities that harm these four factors such as pollution and discrimination and improve the factors by helping society for example with philanthropy, investing in the environment and donating to charities (Carroll, 2015). Carroll (2015) also states that when looking at CSR, there are parts that society require (economic and legal assumptions), expect (ethical behavior) and desire (charity donations).

CSR is a very important aspect in the business world nowadays. Companies need to choose whether their business strategy involves a lot of CSR investments or not and in what way they are going to invest. Still, there are disagreements between studies on whether CSR benefits outweigh the costs of the investments (Margolis, Elfenbein and Walsh, 2007). There has been plenty of research about the drivers behind CSR investment for companies, which has shown that there are multiple reasons why a company would invest in CSR activities, for example to gain a competitive advantage or as a tool for strategic positioning (Dupire and M’Zali, 2018). Kim, Park and Wier (2012) summarize in their paper that there are multiple theories that suggest that there are ethical drivers behind CSR investments and that there is a

10

moral imperative for managers to do the right thing but that there are other studies that argue that managers may engage in CSR to pursuit their self-interest or because they want to conceal the effects of corporate misconduct.

The different reasons for investing in CSR activities or not can be partially explained by the political affiliations of CEOs of the companies (Di Giuli and Kostovetsky, 2014). Di Giuli and Kostovetsky (2014) state in their paper that there is a correlation between the political view of the companies and their CEOs, and the CSR investments that the company does. They state that companies that are situated in a state that has a democratic belief, or if a company has a democratic CEO, the company is more likely to invest in CSR activities (Di Giuli and Kostovetsky, 2014). In this paper it will be tested whether the state that the company is situated in has an influence on the amount of CSR investments. This will be an additional test as a moderator to the main test, which is the influence of military CEOs on the CSR investments. Hence this paper also investigates if the influence that Di Giuli and Kostovetsky (2014) found in their paper adds up for military CEOs.

2.1 Hypothesis development

After defining military background of CEOs and CSR activities, there are numerous expectations that arise. First of all, CEOs with a military background are found to have more leadership characteristics according to Elder (1968). These leadership characteristics were learned on the field and will stay with them when they are CEO of a company. The characteristics of a CEO and thus also their leadership qualities can influence the firms’ decision making. According to the upper echelons theory of Hambrick and Mason (1984) these decisions that CEOs make regarding the firm can influence the firm outcomes.

Leadership is a quality that is often considered to have a positive effect on the CSR activities of a company (Waldman, Siegel and Javidan, 2006). According to the transformational leadership theory of Bass (1985), the CEO with transformational leadership can create a vision that helps employees to understand his ideas of needed change and has therefore the possibility to pursue this change in the organization. According to this theory one would expect a military CEO who has leadership qualities to perform more CSR activities because leadership has a positive effect on the amount of CSR activities according to the review of the transformational leadership theory by Waldman, Siegel and Javidan (2006). In this paper the transformational leadership theory of Bass (1985) is reviewed and combined with CSR engagement.

11

According to Bass (1999), transformational leadership refers to “the leader moving the follower beyond immediate self-interests through idealized influence (charisma), inspiration, intellectual stimulation, or individualized consideration. It elevates the follower’s level of maturity and ideals as well as concerns for achievement, self-actualization, and the well-being of others, the organization, and society” (Bass, 1999, p. 11). When the CEO shows that he is positive towards CSR activities, other employees will also pursue this according to the transformational leadership theory (Bass, 1985).

Bass and Steidlmeier (1999) also argue that transformational leaders are those who sacrifice themselves for the common good. This self-sacrifice gives trust and gains followers for the leader and makes that people accept this leader as a role model. This makes the leader transformational (Bass, 1985). This self-sacrifice and investing in the common goods and society can be seen as CSR in the present time.

In the military, people learn to self-sacrifice, to fight for the common good and to obtain long-term goals (Benmelech and Frydman, 2015). Therefore, military experience can enhance the leadership of the CEO according to the transformational leadership theory of Bass (1985) that claims that self-sacrifice enhances the transformational leadership of the CEO. Obtaining long-term goals, self-sacrifice and investing in the common good are qualities of the kind of leader that would invest in CSR as investing in CSR also requires for the CEO to sacrifice the short-time interests and to obtain long-term goals. This leads to the expectation that military CEOs would invest more in CSR activities than CEOs without a military background.

Another way to consider this expectation is to look at integrity. Duffy (2006) states that military CEOs learned to act with integrity and to be honest during their service. These experiences in their time in the military shape them and have a lot of impact on their time in the corporate world as well as their future character (Duffy, 2006). Since acting with integrity and being honest is a devotion to the public welfare and promotes an image of doing the right thing for one another, it can be expected that military CEOs invest more in CSR activities than CEOs without military experiences. CSR mostly consists of pursuing things that are not necessarily beneficial for the company itself and promotes the idea for employees to take responsibility for society. These theories and statements lead to the following hypothesis:

Hypothesis 1: Military CEOs are more likely to invest in CSR activities than CEOs without a military background.

12

On the other hand, there are papers that show that investing in CSR activities is not always an act of integrity (Luo, Xiang and Zhu, 2017). In fact, Luo, Xiang and Zhu (2017) show that firms engage in CSR activities for various reasons and one should not automatically admire them for their ethical behavior. For example, Koehn and Ueng (2010) found that firms use their charity as a kind of moral window dressing or to “buy” a better reputation. When a firm violates regulations and experiences a worsening of their reputation, CSR activities can help to reduce this (Williams and Barrett, 2000). This also indicates that CSR activities are not always for doing well but can also be performed for selfish reasons. At last, Chen, Patten and Roberts (2008) show that making donations and investing in charity can also be a way to rationalize bad financial performance of a firm. When the financial results are bad and the company is able to blame it on the charity donations, it can be assumed that these donations are not made out of pure good will of the company (Chen, Patten and Roberts, 2008). Luo, Xiang and Zhu show in their study that because of moral window dressing, military CEOs in China perform less corporate philanthropy. These studies would suggest that military CEOs would not necessarily invest more in CSR activities, but because of the selfless characteristics and the integrity of the military CEO, the company would invest even less in CSR activities.

The expected negative relationship between military CEOs and CSR activities is based on the believe that military CEOs are more honest and against moral window-dressing as discussed above. Moral window-dressing is when a company wants to distract the public from their wrongdoings by doing something good like investing in CSR (Chen, Patten and Roberts, 2008). Besides this, CEOs without a military background that experience political pressure from the government, normally invest more in CSR activities (Liao et al., 2019). Military CEOs could experience this pressure from the government as moral window dressing and if that is the case, the negative relationship that is expected because of moral window dressing could be strengthened by political pressures. When political pressure is the reason for CSR investments, military CEOs will view these motives as corrupt and not ethical and according to Luo, Xiang and Zhu, they won’t cooperate with these motives.

Finally, an expectation can be formed from the fact that military CEOs are often more aggressive and masculine than CEOs without a military background (Elder, 1968; Elder and Clipp, 1989). This aggressive form of leadership and masculinity is not something that would be expected when looking at CSR. Studies found that CSR is more correlated with feminine than masculine characteristics (Byron and Post, 2016). Benmelech and Frydman (2015) also found that military CEOs are less aggressive in investing their financial means in R&D. This

13

too, would lead to the expectation that military CEOs are less likely to invest in CSR activities and gives us the following hypothesis:

Hypothesis 2: Military CEOs are less likely to invest in CSR activities than CEOs without a military background.

14

3. Research design

In this paper, CSR data is perceived from the Kinder, Lydenberg, Domini (KLD) database to draw a sample of CSR information. The KLD database provides access to positive as well as negative environmental, social and governance (ESG) performance indicators for publicly traded companies (Godfey, Merrill and Hansen, 2009; Petrenko et al., 2016). For the CEO military background, data from US public firms which have been component firms of the Standard & Poor 1500 index will be used. The sample period of this study is from 2000-2016 because CSR data is mostly available during this period.

3.1 Dependent variable

To test the influence of military background on the amount of CSR activities, it is essential to know how to measure CSR activities. KLD data analyses different groups of CSR which are community, employee relations, environment, human rights, product, diversity and corporate governance. These different groups all get a net score which contains a sum of the companies’ strength and concerns in the particular CSR group. These beneficial issues and harmful issues in each group are taken together to one net score per group and can thus lead to one CSR grade when the groups are added together. In this study the focus lies on the groups: community, employee relations, environment, human rights, product and diversity. Corporate governance is not taken into account in this study because multiple papers suggest that corporate governance is a distinct group that does not fall under CSR and should be left out (Di Giuli and Kostovetsky, 2014; Yuan et al., 2019). In the end, the CSR performance will be rated on one score that is the net sum of the beneficial and harmful issues. This net score variable is called CSR NET SCORE. The total score of the CSR strengths and the total score of the CSR concerns is also taken into account in this study. With these scores the CSR strengths and concerns can be looked at separately which could give more useful insights.

3.2 Independent variable

The central independent variable in this paper is the military background of the CEO. To measure the military background of the CEO, it will be checked whether the CEO has military experience from the army, navy or any other militia. A dummy variable will be used to measure this and will be called MILITARY EXPERIENCE. Besides this dummy variable, other variables will be used to measure the military background of the CEO. These variables will be explained in the next paragraph.

15 3.3 Moderating variables

A variable that measures the political characteristics of the state in which the headquarter of the company is situated is included in the model. This can be either a democratic or a republican political affiliation. Since Di Giuli and Kostovetsky (2014) show that a democratic belief system is more likely to invest in CSR. In our additional test we add a variable that is called DEMOCRATIC, that will be 1 if the headquarters of a company is situated in a Democratic state and 0 if it is situated in a Republican state. With adding this moderating variable, it can be investigated whether the positive influence that Di Guili and Kostovetsky found between democratic views and CSR activities also applies to military CEOs.

Another variable that is added for an additional test is a proxy variable for the time between the retirement from the military of the CEO and the present. This variable will test whether the character traits that the person gains in their time in the military, can fade away during their retirement. This variable is measured in the amount of years that it has been for the person since he left the military and thus how long the CEO has been a veteran. This variable is named: VETERANDURATION. The variable will be 1 when the time that the CEO has been a veteran is 15 years or less and 0 if it has been more than 15 years.

The last moderating variable is the amount of years that the military CEOs served in the army. It is expected that character traits of a person are more influenced when this person has been in the military for a long amount of time rather than a short amount of time (Elder and Clipp (1989). This variable is measured in the amount of years that the person has served in the military and is called MILITARY SERVICE TENURE. This variable will have a value of 1 if the amount of serving time in the military is 5 years or longer and 0 if the serving time was less than 5 years.

3.4 Control variables

Aside from the independent, dependent and moderating variables, control variables will be included. There will be controlled for R&DEX and CASH, because firms with more R&D expenses and more cash are likely to invest more in CSR since there is more money available for this (Lys et al., 2015). Besides this, firm-level control variables will be used. These variables are: FIRMSIZE (defined as total assets), ROA (profits to assets), LEVERAGE (total debt to equity) and BMRATIO (book to market ratio). There will also be controlled for DIVPAY (dividend pay-out ratio) because firms that pay dividends are more likely to be associated with a higher CSR performance (Di Giuli and Kostovetsky, 2014).

16

Besides these firm-level control variables, some CEO-level control variables will be included. These variables are included because the study is focussed on the CEO and should therefore include controls on this level. The controls that are used on CEO-level are: GENDER, AGE and CEOTENURE.

TABLE 1 Variable description

Variable Description Measurement

Dependent variable

CSR NET SCORE Net score of CSR

performance based on CSR strengths and weaknesses

Total grade after subtrecting CSR concerns from CSR strengths

CSR NET STR Net score of total strengths of different CSR factors

Total grade after adding all strength factors

CSR NET CON Net score of total concerns of different CSR factors

Total grade after adding all concern factors

Independent variable

MILITARY EXPERIENCE Whether the CEO has military experience

Value is 1 if the CEO has military experience and 0 if not

Moderating variables

DEMOCRATIC What the political affiliations are in the state that the headquarters of the company is situated in

Value is 1 if the headquarters is situated in a democratic state and 0 in a republican state

VETERAN DURATION How long it has been since the CEO left the military

Value is 1 if the time that has past since the military

service is 15 years or less and 0 if it is more than 15 years

MILITARY SERVICE TENURE

How long the CEO was in the military

Value is 1 if the amount of serving time is 5 years or longer and 0 if it was shorter than 5 years

Control variables

R&DEX Research and development expenses to sales

Value is measured in percentage

CASH Cash to sales Value is measured in percentage

FIRMSIZE Defined as total assets Value is measured in the natural log of the US dollars ROA Return on assets, defined as

profits to assets

Value is measured in percentage

17

percentage

BMRATIO Book to market ratio Value is measured in percentage

DIVPAY Dividend pay-out ratio Value is measured in percentage

GENDER Gender of the CEO

Value is 1 if the CEO is female and 0 if the CEO is male

AGE Age of the CEO Value is measured in the natural log of the years TENURE CEO tenure in company Value is measured in the

natural log of the years

For testing hypothesis 1 and 2 the emperical model (1) below is constructed. This model is used to explain the relationship between the CSR activities and the CEO military background. The data used consists of 15,985 unique observations from 1,885 firms, of which all the data is available. Before performing the analyses, the variables ROA, BMRATIO and DIVPAY were winsorized at the top and bottom 1 percent and the variables R&DEX and LEVERAGE were winsorized at the top and bottom 5 percent to reduce the extreme effect of outliers.

(1) CSR NET SCORE = β0 + β1MILITARY EXPERIENCE + β2DEMOCRATIC + β3VETERAN DURATION + β4MILITARY SERVICE TENURE + β5R&DEX +

β6CASH + β7FIRMSIZE + β8ROA + β9LEVERAGE + β10BMRATIO + β11DIVPAY + β12GENDER + β13AGE + β14TENURE + ε

CSR NET SCORE is used to define the degree to which the company invests in CSR activities. The symbol ε is the error term. All other variables are explained in TABLE 1.

18

4. Emperical results

In this chapter the results of the statistical anlyses will be presented. In Table 2 the descriptive statistics are shown and in Table 3 the Pearson correlation matrix is presented and explained. Furthermore, in Table 4, the results of the regression analyses are analyzed and presented and afterwards a few additional tests will be analyzed to gain more insight in the different CSR areas and the characteristics of the military CEOs.

4.1 Descriptive statistics

Table 2 shows the descriptive statistics for the variables that are used in this research. This table gives an insight in the results of the data collection but does not yet provide an outcome of the research question. The sample data consists of 15,985 observations of 1,885 firms. The dependent variable CSR NET SCORE has an average of 0.0705901 points which is just above 0. The mean of the CSR strengths is 1.718585 and the mean of the CSR concerns is 1.647995. This shows us that on average firms have a little more CSR strengths than they have CSR concerns.

Another fact that the descriptive statistics table shows is that, among the firms that are in the sample, a little above 8.2% of the CEOs have a military background. For these CEOs with a military background, this military experience was on average a little more than 33 years ago, so the CEOs have been veterans for an average of 33 years and they have been in the military for a time period of approximately 5.4 years. From the companies that were investigated, 87.7% had their headquarters in a democratic state.

Table 2 further shows that the research and development expenses are on average 3.967% of the sales and the cash ratio is on average 1.017. The total asests of the firms are the natural logarithm of the amounts in US dollars and are on average 7.664 and the return on assets has an average of 14.729%. The leverage, or debt to equity ratio, is on average 1.568 and the book to market ratio has a mean of 0.533. The divident payout ratio has a mean of 0.289 and from the firms in the sample, 4.194% has a female CEO. The age of the CEO has an average natural logarithm of 3.962 which comes down on approximately 53 years old. The tenure of the CEOs has an average natural logarithm of 1.804.

19 TABLE 2 Descriptive statistics

Variable Minimum Maximum Mean Standard

deviation CSR NET SCORE _{-9 18 0.0705901 2.586218} CSR NET STR _{0 21 1.718585 2.617049} CSR NET CON _{0 15 1.647995 1.822688} MILITARY EXPERIENCE 0 1 0.082125 0.2745636 DEMOCRATIC _{0 1 0.8773125 0.3280884} VETERAN DURATION 2 66 33.11073 10.21074 MILITARY SERVICE TENURE 0 39 5.432331 5.935649 R&DEX 0 0.2154901 0.0396653 0.0651733 CASH 0 35,77039 1.01692 1.643843 FIRMSIZE 2.325422 13.58957 7.663744 1.564025 ROA -0.1291765 0.4556568 0.147294 0.0920247 LEVERAGE 0.1597554 9.81434 1.567864 1.775623 BMRATIO 0.0473581 2.169491 0.5325205 0.366922 DIVPAY 0 3.670816 0.2887276 0.5223145 GENDER 0 1 0.0419375 0.2004526 AGE 3.332205 4.477337 3.96151 0.1356454 TENURE 0 4.110874 1.803793 0.8932301

4.2 Pearson correlation matrix

Table 3 provides the correlations for the variables. This correlation matrix is made to control for possible mulitcollinearity. The matrix does not contain multicollinearity since there are no relations with a correlation of higher than 0.5 which is a signal for multicollinearity. Since there is no correlation with a value higher than 0.5, it can be assumed that there is no evidence for multicollinearity. It is visible that, for example, military experience and CSR net score have a positive significant correlation with each other. A correlation between these variables was expected since these are the dependent and independent variables of this study. Since there are no correlations that exceed the mark for multicolilnearity, there is no need to execute more explicit correlation analyses.

20

1.CSR NET SCORE 2.MILITARY EXPERIENCE 3.DEMOCRATIC 4.VETERAN DURATION 5.MILITARY SERVICE TENURE 6.R&DEX 7.CASH 8.FIRMSIZE 9.ROA 10.LEVERAGE 11.BMRATIO 12.DIVPAY 13.GENDER 14.AGE 15.TENURE

*, **, *** = Coefficients are significant on the p<0.1, p<0.05 and p<0.01 level

TABLE 3

Pearson correlation matrix

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 1. 1 2. .0225** 1 3. .0451*** -.0186** 1 4. -.0537 .0158 .0552 1 5. .0931** . -.0646* .2708*** 1 6. .1419*** -.0412*** .1469*** -.0234 .0938*** 1 7. .0054 -.0500*** .1031*** -.0535 -.0500 .4865*** 1 8. .2478*** .0648*** .0182** -.0616* .0209 -.1905*** -.2802*** 1 9. .0716*** .0348*** -.0375*** .0174 -.0336 -.1776*** -.0623*** -.0479*** 1 10. .0101 .0286*** .0047 .0665** .0424 -.2125*** -.2384*** .3208*** -.1039*** 1 11. -.1375*** -.0112 -.0414*** -.0533 -.1194*** -.1861*** -.0940*** .0767*** -.4344*** -.0726*** 1 12. .0373*** .0111 -.0379*** -.0225 .0251 -.1269*** -.1008*** .1664*** -.1253*** .1727*** .0719*** 1 13. .0986*** -.0240*** -.0054 .0307 .0887** -.0435*** -.0024 -.0033 .0254*** -.0103 -.0152* .0148* 1 14. -.0029 .0690*** -.0033 -.0455 -.1631*** -.0807*** .0416*** .0842*** -.0253*** .0199** .0179** .0523*** -.0321*** 1 15. -.0072 .0290*** -.0218*** -.1133*** -.1962*** -.0179** -.0259*** -.0145* .0265*** .0535*** -.0174** -.0004 -.0323*** .1609*** 1

21 4.3 Regression analysis

Table 4 presents the main results of the fixed effect panel regression analysis with the CSR net score, CSR total strengths and CSR total concerns as the main dependent variables. The first regression (1), with each of these different aspects of CSR, is the CSR variable and all of the control variables. The models (2) show the influence of the military experience on each of these CSR variables. In all of these analyses a fixed-effect panel regression is used for the fixed year effects. The number of observations for this test is 6,951 which is seperated into 1,121 different groups.

In the first analysis it is presented that the control variables have values of approximately .002, -.322, .181, -.041, .048 and .481. As is visible in the model, the values of the R&D expenses, the leverage and the gender of the CEO are significant on respectively the 1 percent, 10 percent and 5 percent level. R&D, as well as leverage and gender, has a positive relation with the CSR net score, which indicates that an increase in these values would increase the CSR net score. The rest of the control variables show no significant relation with the dependent variable. Dividend payout ratio is the only variable that shows a negative influence on the CSR net score, though this relationship is insignificant.

The second analysis shows the relation between the independent variable and the CSR net score. This shows a positive relationship which indicates that a CEO with military experience would have a positive correlation with the CSR net score. However, this relationship is not significant which means that the first hypothesis: Military CEOs are more

likely to invest in CSR activities than CEOs without a military background cannot be

accepted. Since there appears to be no significant relationship at all, the second hypothesis:

Military CEOs are less likely to invest in CSR activities than CEOs without a military background cannot be accepted either.

When splitting up the CSR net score between the total strengths and the total concerns, there can be an alternative fixed-effect panel regression that could present whether only the CSR strengths or concerns can be explained by the variables. When looking at the control variables in relation to the CSR total strengths in Table 4, there is no remarkable difference from the first analysis for the CSR net score. The same can be said about the relationship between the control variables and the CSR total concerns.

Model (2) from the analysis of the CSR total strengths gives an insight of the relationship between military experience and the CSR total strengths. This coefficient implies a negative relationship, however, this relationship is insignificant.

22

The last model from table 4 contains the interaction coefficient of military experience on CSR total concerns. This value indicates a negative relationship (-2.06) between military experience and CSR total concerns. The relationship is significant on the 1 percent level, which indicates that companies with a military CEO score less total concern points on the different areas of CSR.

In conclusion, the results do not support the hypotheses made in this thesis, but there is one interesting finding, namely that there is a negative relationship between the military experience and the CSR total concerns. Nevertheless, the low R-squared could suggest that there are a lot of other variables that are of influence on the CSR scores. This finding will be discussed further in the discussion section of the paper.

*, **, *** = Coefficients are significant on the p<0.1, p<0.05 and p<0.01 level

TABLE 4

Fixed-effects panel regression on CSR score

CSR net score CSR total strengths CSR total concerns Variable (1) (2) (1) (2) (1) (2) MILITARY EXPERIENCE .190 (.132) -.015 (.111) -.206*** (.074) R&DEX -.520 (1.623) -.502 (1.623) -.010 (1.313) -.013 (1.313) .510 (.932) .515 (.932) ROA -.322 (.493) -.337 (.494) .037 (.415) .038 (.415) .359 (.275) .376 (.275) BMRATIO .181 (.134) .183 (.134) .211* (.113) .211* (.113) .030 (.075) .028 (.075) DIVPAY -.041 (.062) -.040 (.062) .046 (.052) .046 (.052) .087** (.035) .086** (.035) CASH .042* (.023) .042* (.023) .033* (.018) .034* (.018) -.008 (.013) -.008 (.013) LEVERAGE .048* (.028) .0475* (.027) .038 (.023) .039 (.024) -.010 (.016) -.009 (.016) GENDER .481** (.198) .478** (.198) .338** (.167) .338* (.167) -.143 (.111) -.140 (.111) Observations 6,951 6,951 6,951 6,951 6,951 6,951 Number of groups 1,121 1,121 1,121 1,121 1,121 1,121 R-squared 0.160 0.160 0.245 0.245 0.015 0.015

23 4.4 Additional regression analyses

One of the variables that has been in the theoretical framework of this paper, is the political blief of the state that the company of the CEO is situated in. In this section an additional test will be presented in which the companies in the democratic states and the republican states are taken seperately into a fixed-effect panel regression. Another additional test that will be discussed in this section is the influence of the amount of time that is has been since the CEO was in the military, so in other words, for how long the CEO has been a veteran. For this test a distinction was made between CEOs that have left the military more than 15 years ago and CEOs that left the military 15 years ago or less. The last additional test is for the CEO tenure and makes a distinction between CEOs that have served in the military for 5 years or more and those that have served for less than 5 years. In Table 5 the results of these additional regression analyses are presented.

In the first model (1), only the firms that have their headquarters in a state with a democratic belief system are taken into account. This test shows us that the military experience of the CEOs of these companies have a positive relation to the CSR net score (.205) which is significant on the 10 percent level. This first model also shows a positive relation between the gender and the CSR net score (.759) which is significant on the 1 percent level. In this model, 6,259 observations were included for 1,050 groups and the model has a R-squared of 0.1692.

In the second model (2), only the firms that are situated in states with a republican belief system are taken into account. This analysis shows that the military experience of the CEOs have a negative influence on the CSR net score. However, this relationship is insignificant. Another finding from this model is that the gender of the CEO has a significant negative influence on the CSR net score. Since there were only 692 observations of companies in a republican state, the R-squared of this model is only .0311 which makes the model less reliable.

In the third and fourth model, a fixed-effect panel regression analyses was run to make a distinction in the amount of years that the military CEO has been a veteran. In other words, with this analyses was checked wether the amount of years that the CEO has been a veteran has a relationship to the CSR net score of the company. In model (3) the regression was run on the non-military CEOs and the military CEOs that are veterans for more than 15 years. This test shows a negative relationship between the military experience (-0.181) which is significant on the 10 percent level. This would indicate that CEOs that have been in the military a long time ago score lower on the CSR net score. In model (4) the same regression

24

was run, but on the non-military CEOs and the military CEOs that are veterans for 15 years or less. For this additional test no significant relation was found between the military experience and the CSR net score. This indicates that the negative relationship between CEO military experience and CSR net score disappears when the time since the CEO was in the military is shorter and the CEO has thus been a veteran for a shorter time period.

In the fifth and sixth model of table 5 the military service tenure is included into the regression model. In model (5) the regression was run on the CEOs without a military background and the CEOs that have been in the military for 5 years or more. A small negative relationship was found between the military experience and CSR net score for this additional test, however this result is insignificant. In model (6) the same regression was run, but on the non-military CEOs and the CEOs that have served in the military for less than 5 years. For this additional test no significant results were find either. Because of these results it cannot be concluded that there are differences between the length of the military service tenure of the CEOs and the CSR net score.

TABLE 5

Additional fixed-effect panel regression analyses

CSR net score Political affiliation test Veteran duration test Military service tenure test Variable (1) (2) (3) (4) (5) (6) MILITARY EXPERIENCE .205* (.139) -.538 (.485) -.181(*) (.123) -.034 (.196) -.117 (.144) -.130 (.145) R&DEX -.762 (1.719) -3.532 (4.954) -.287 (1.749) .305 (1.789) .237 (1.771) -.227 (1.761) ROA -.442 (.523) .305 (1.552) .223 (.440) .107 (.458) .210 (.453) .150 (.444) BMRATIO .147 (.147) .738** (.347) -.126 (.104) -.093 (.107) -.090 (.106) -.108 (.105) CASH .041 (.024) -.026 (.088) .061** (.024) .072*** (.025) .072*** (.025) .053** (.025) DIVPAY -.052 (.066) .014 (.179) -.001 (.048) -.018 (.049) -.008 (.048) -.004 (.025) LEVERAGE .015 (.030) .109 (.102) .016 (.024) .016 (.024) .028 (.024) .004 (.024) AGE -.095 (.238) -.237 (.253) -.182 (.246) -.157 (.244) CEO TENURE -.144*** (.032) -.129*** (.033) -.132*** (.033) -.157*** (.033)

25

(1) = DEMOCRATIC, (2) = REPUBLICAN, (3) = VETERAN DURATION LONG, (4) = VETERAN DURATION SHORT, (5) = MILITARY SERVICE TENURE LONG, (6) = MILITARY SERVICE TENURE SHORT

*, **, *** = Coefficients are significant on the p<0.1, p<0.05 and p<0.01 level

GENDER .759*** (.217) -1.34** (5.828) .320* (.192) .305 (.194) .318* (.191) .326* (.194) Observations 6,259 692 9,736 9,188 9,458 9,508 Number of groups 1,050 144 1,622 1,576 1,599 1,602 R-squared .1629 .0311 0.0368 0.0430 0.0422 0.0363

26

5. Discussion and conclusion

Over a long period of time, investigators have researched the reasons why firms invest in CSR activities. This study adds to the literature by investigating the impact of CEO military backgrounds on the investment in CSR activities. The CEO military background has been discussed by multiple papers as well (Benmelech and Frydman, 2015; Koch-Bayram and Wernicke, 2018; Luo, Xiang and Zhu, 2017), but most research has been done for the influence of the CEO military background on the financial performance of the firm. This study fills the research gap between these two factors by combining CEO military background and CSR activities. In this paragraph the results of the study will be discussed and the limitations of the study and future research will be considered.

First of all, the CEO seems to be an important decision maker regarding the Corporate Social Responsibility activities. The characteristics of the CEO could make a difference in the kind of decision making that the CEO has regarding CSR activities (Hambrick and Mason, 1984). Whether or not the CEO has a military background can influence their characteristics a lot (Elder, 1968). Even though numerous research shows that the CEO has a lot of influence on these decisions, the empirical results of this study show no significant relation between the military background of the CEO and the total score of the CSR activities of the companies. This would indicate that the military background of the CEO would not significantly influence the decisionmaking to invest in more or less CSR activities, since the net total score is not influenced by it. A possible explanation for this finding could be that the CSR net score consists of too many different areas of CSR which conflict with each other. Because of this, there has been made a separate regression for the CSR strengths and the CSR concerns. This separate regression partly shows this to be true by giving a negative, significant, relation between the military background of the CEO and the CSR concerns. For the CSR strengths, still no significant relation was found.

A possible explanation for these findings is that the CSR strengths still consist of too many different factors to gain a significant relation. Since these CSR strengths contain the community, employee relations, environment, human rights, product and diversity strengths of the company. In further research it would be wise to test these CSR strength and concern factors separately to see whether this could lead to a significant relation between military experienced CEOs and the different factors of CSR activities.

As mentioned before, the CSR concerns were significantly negatively influenced by the military experience of the CEOs. This would mean that, when a CEO has a military

27

background, the firm of the CEO would engage less in the activities that would lead to CSR concern points. One way to explain this relationship is to look at the paper of Koehn and Ueng (2010), the article of Duffy (2006) and the paper of Luo, Xiang and Zhu. The paper of Koehn and Ueng (2010) suggests that philanthropy and with that corporate social responsibility is often only used by people to “buy good will” which indicates that people would only invest in CSR for the wrong reasons. This relationship is again mentioned in the paper of Luo, Xiang and Zhu and they claim that military CEOs would not want to participate in this kind of window-dressing which is why there is no positive relationship between the CSR net score and the military CEOs. Duffy (2006) claims that people that have served in the military have certain characteristics that would make them more honest and loyal. Military executives have learned to “do the right thing” from their experiences in the military, which leaves them with a strict moral code (Duffy, 2006). These characteristics could be an indication and explanation why there is a negative relationship between CEOs with a military background and the CSR concerns. These CEOs don’t necessarily invest more in CSR because of their loyalty and honesty, but they do avoid wrongdoings by obeying to their ethical desire to do good.

Another way to explain the negative relationship between the military CEOs and the CSR concerns is to look at the prospect theory of Kahneman and Tversky (1979). This theory suggests that people value losses higher than gains. In this study this could indicate that stakeholders value CSR concerns higher than CSR strengths because for example fines and legal issues increase. CEOs could anticipate this and make sure that the CSR concerns are as low as possible. This however, does not explain why companies with a military CEO score less CSR concerns compared those with a CEO without a military background. Still, this prospect theory of Kahneman and Tversky (1979) can be of important explanatory value when looking at the difference between the CSR strengths and concerns.

Furthermore, this study supports the statement of Di Giuli and Kostovetsky (2014) that firms that are situated in a state with a democratic belief system have a higher CSR net score than firms that have their headquarters in a republican state. In the empirical analysis it was found that CEOs with a military background from a company in a democratic state have a positive influence on the CSR net score. In the first regression analysis there was no significant relationship between the military background of CEOs and the CSR net score for the whole dataset. After separating the firms based on their political affiliations there was a significant positive relationship. It is a possibility that Di Giuli and Kostovetsky (2014) could give the rationale for this relationship that occurred. Di Giuli and Kostovetsky (2014) find in

28

their paper that firms with a democratic leader, or firms that are headquartered in a democratic state, spent more on CSR and score higher on the CSR ratings in the KLD database. In this thesis, no significant evidence was found that the firms with their headquarters in a republican state score higher or lower because of their political affiliations. This may be due to the fact that there were not many observations for firms in a republican state which makes the test less reliable. Another possible explanation for this is of course that the fact that a company and/or it’s CEO has republican beliefs, does not influence their investments regarding CSR activities. Moreover, in this study the relationship between the CEOs with a military background that have been out of the military for a certain time and the CSR net score was investigated. This was measured with the variable VETERAN DURATION, which indicates how long it has been since the CEO has been in the military and thus how long the CEO has been a veteran. The analysis indicated that the longer it has been since the CEO has left the military, the lower their CSR score would be. This would indicate that the values of the military could fade away after a certain time period. Nevertheless, since there were no significant results between the military CEOs and the CSR net score to begin with, there is no indication that this negative relationship would be due to the fact that the values that the CEO learned in the military faded away, but it could indicate that a CEO that has left the military a long time ago slowly loses the ethical values that were gathered by the CEO in the service time. Future research should be done to explain this interesting finding by focusing on the amount of time that a CEO has been a veteran.

Another variable that was investigated in this study is the MILITARY SERVICE TENURE. This was measured in the amount of years that the CEO had served in the military and made a distinction between a long military service tenure (5 years or more) and a short military service tenure (less than 5 years). There was no significant relationship found between the MILITARY SERVICE TENURE and the CSR. This could occur because the CSR net score is broadly defined as mentioned before. Since there are no significant outcomes for this particular variable, there is no indication that the character traits that CEOs gain in the military are better preserved the longer the service tenure was as Elder (1968) suggested. With the absence of significant findings, this theory of Elder (1986) cannot be supported with this research.

5.1 Limitations and future research

This study contains certain limitations. As already mentioned before, CSR consists of multiple factors that together form the corporate social responsibility activities. Companies

29

can score very high on one factor and low on another factor which will lead to a basic CSR net score but still can have interesting results when taken apart. Because of this, there were regressions made with only the CSR strengths and concerns separately, but an analysis with all factors taken apart would be insightful. Future research could take these factors apart and could also include the corporate governance mechanisms as a separate factor to see whether there is a relationship between corporate governance and the military background of CEOs.

This study focusses on military CEOs who occur less and less in the corporate world, since there are fewer people that join the military nowadays. A limitation from this research could be that it might be a little outdated. However, CSR is a topic that is of great importance to investigate more these days. Another problem with the research on military CEOs is that there are not many left anymore. In the dataset that was used for this research only a little over 8 percent of the CEOs had experience in the military.

Besides this, the paper uses the Kinder, Lydenberg and Domini (KLD) database with the CSR net scores to measure CSR performance. A lot of other researchers use this scale in their papers (Dupire and M’Zali, 2018; Di Giuli and Kostovetsky, 2014; Chen, Patten and Roberts, 2008; Kim, Park and Wier, 2012) since it is a very common database to use. However, this score is not a perfect way to measure CSR. Chatterji, Levine and Toffel (2009) discovered that the rates of the KLD database are not always obtaining all the publicly available information and should be interpreted with caution.

Another limitation in this research is the use of the databases, since this study only acquired quantitative data. Because of this there is the possibility for future research to obtain qualitative case analysis and to examine the effect that the military background has on a CEO further by interviewing these CEOs or to make a case study. Finally, future research could expand this research by gaining data from smaller and private firms since this research only used data from public firms from the SP 1500 index. By doing this, future research could improve the understanding of military CEOs and their influence on CSR among more and different environments.

30

6. References

Bass, B.M. (1985). Leadership and performance beyond expectations. Free Press.

Bass, B.M. (1998). Transformational Leadership: Industrial, Military and Educational

Impact. Mahwah, NJ: Lawrence Erlbaum Associates.

Bass, B.M. (1999). Two decades of research and development in transformational leadership. European Journal of Work and Organizational Psychology, 8(1), 9-32.

Benmelech, E., & Frydman, C. (2015). Military CEOs. Journal of Financial Economics,

117(1), 43-59.

Byron, K., & Post, C. (2016). Women on boards of directors and corporate social performance: A meta-analysis. Corporate Governance: An International Review, 24(4), 428-442.

Campbell, D., Moore, G., & Metzger, M. (2002). Corporate philanthropy in the U.K. 1985-2000: Some empirical findings. Journal of Business Ethics, 39(1-2), 29-41.

Carroll, A. (2015). Corporate social responsibility. Organizational Dynamics, 44(2), 87-96. Chatterji, A.K., Levine, D.I., & Toffel, M.W. (2009). How Well Do Social Ratings Actually

Measure Corporate Social Responsibility? Journal of Economics & Management

Strategy, 18(1), 125-169.

Chen, J., Patten, D., & Roberts, R. (2008). Corporate charitable contributions: A corporate social performance or legitimacy strategy? Journal of Business Ethics, 82(1), 131-144. Dentchev, N. (2004). Corporate Social Performance as a business strategy. Journal of

Business Ethics, 55(4), 395-410.

Di Giuli, A., & Kostovetsky, L. (2014). Are red or blue companies more likely to go green? Politics and corporate social responsibility. Journal of Financial Economics, 111(1), 158-180.

Duffy, T. (2006). Military experience and CEOs: is there a link? Korn/Ferry International Report. Retrieved from http://kornferry.com/institute/download/view/id/17026/aid/190. Dupire, M., & M’Zali, B. (2018). CSR strategies in response to competitive

31

Elder, G.H. (1968). Military times and turning points in men’s lives. Developmental

Psychology, 22(2), 233-245.

Elder, G.H., & Clipp, E. (1989). Combat experience and emotional health: Impairment and resilience in later life. Journal of Personality, 57(2), 311-341.

Godfrey, P., Merrill, C., & Hansen, J. (2009). The relationship between corporate social responsibility and shareholder value: An empirical test of the risk management hypothesis. Strategic Management Journal, 30(4), 425-445.

Hambrick, D.C., & Mason, P.A. (1984). Upper echelons: The organization as a reflection of its top managers. Academy of Management Review, 9(2), 193-206.

Jackson, J., Thoemmes, F., Jonkmann, K., Lüdtke, O., & Trautwein, U. (2012). Military training and personality trait development: Does the military make the man, or does the man make the military? Psychological Science, 23(3), 270-277.

Kahneman, D., & Tversky, A. (1979). Prospect theory: an analysis of decision under risk.

Econometrica, 47(2), 263-291.

Kim, Y., Park, M.S., & Wier, B. (2012). Is earnings quality associated with corporate social responsibility? The Accounting Review, 87(3), 761-796.

Koch-Bayram, I., & Wernicke, G. (2018). Drilled to obey? Ex-military CEOs and financial misconduct. Strategic Management Journal, 39(11), 2943-2964.

Koehn, D., & Ueng, J. (2010). Is philanthropy being used by corporate wrongdoers to buy good will? Journal of Management & Governance, 14(1), 1-16.

Liao, Z., Dong, J., Weng, C., & Shen, C. (2019). CEOs' religious beliefs and the environmental innovation of private enterprises: The moderating role of political ties. Corporate Social Responsibility and Environmental Management, 26(4), 972-980. Luo, J., Xiang, Y., & Zhu, R. (2017). Military top executives and corporate philanthropy:

Evidence from China. Asia Pacific Journal of Management, 34(3), 725-755.

Lys, T., Naughton, J., & Wang, C. (2015). Signaling through corporate accountability reporting. Journal of Accounting and Economics, 60(1), 56-72.

32

Margolis, J., Elfenbein, H., & Walsh, J. (2007). Does it pay to be good… and does it matter? A meta-analysis of the relationship between corporate social and financial performance. Unpublished working paper. University of Michigan.

McWilliams, A., & Siegel, D. (2001). Corporate social responsibility: A theory of the firm perspective. The Academy of Management Review, 26(1), 117-127.

Petrenko, O., Aime, F., Ridge, J., & Hill, A. (2016). Corporate social responsibility or CEO narcissism? CSR motivations and organizational performance. Strategic Management

Journal, 37(2), 262-279.

Simanaviciene, Z., Kontautiene, R., & Simanavicius, A. (2017). Assumptions of Corporate Social Responsibility as competitiveness factor. Montenegrin Journal of Economics,

13(3), 149-160.

Waldman, D., Siegel, D., & Javidan, M. (2006). Components of CEO transformational leadership and Corporate Social Responsibility. Journal of Management Studies, 43(8), 1703-1725.

Williams, R., & Barrett, J. (2000). Corporate philanthropy, criminal activity, and firm reputation: Is there a link? Journal of Business Ethics, 26(4), 341-350.

Yuan, Y., Tian, G., Lu, L., & Yu, Y. (2019). CEO ability and Corporate Social Responsibility. Journal of Business Ethics, 157(2), 391-411.