MSc Thesis 2018
University of Groningen, Faculty of Economics and Business
THE RELATIONSHIP BETWEEN CEO COMPENSATION AND CORPORATE
RISK-TAKING. AN EXAMINATION WITH THE MODERATING EFFECTS OF
BOARD SIZE AND UNCERTAINTY AVOIDANCE.
Abstract:
This thesis investigates the relationship between executive compensation and corporate risk-taking by including US and European firms in a sample over the period of 2011 to 2015. The effects of executive compensation are analyzed with moderating factors of the firm board size and country-level uncertainty avoidance scores. This study finds a positive linkage between executive compensation and corporate risk-taking. Both the firm board size and uncertainty avoidance scores have a significant direct impact on corporate risk-taking, but are found to have no significant influence on the executive compensation – risk-taking relationship.
Keywords: Corporate risk-taking, CEO compensation, US firms, European firms, board size, uncertainty avoidance.
Student Number: S2541815
Name: Allard Joosten
Supervisor: Dr. W. Westerman
Co-Assessor: Dr. N. Selmane
Study Program: MSc International Financial Management
2
Preface
Ever since I have been studying in the worldwide International Business field I find myself being highly intrigued by the differences in business perspective on a global scale. I find it interesting to discover and investigate the varying ways in preferences of doing business. Hence, I have written my Bachelor thesis in the cross-cultural domain on the varying levels of job satisfaction between the Netherlands and Japan. I was born and raised in the Netherlands and have been fortunate to have travelled to Asian countries on several occasions. In addition, during my bachelor International Business at the University of Groningen, I have had the opportunity to study for one semester at the Macquarie University in Sydney, Australia. The experience of studying abroad and participating in courses such as ‘Asian Business
Environment’ and ‘Japanese Economy’ has amplified my interest in differing business
environments. Through these encounters and having first-hand experience with these business perspectives my fascination towards these countries and their way of doing business has grown. These experiences have increased my interest in business perspectives and accompanying factors which are aligned with these perspectives.
3
1. Introduction
Research has boomed over corporate risk-taking practices in the recent International Financial Management literature; financial crises have often been linked to excessive corporate risk-taking, which indicates that corporate risk-taking can have a significant influence on both the individual firm as the economy as a whole. Infamous scandals such as the Enron scandal of 2002 affirm the serious impact excessive corporate risk-taking has on firm operations. The concept is therefore extensively researched to get a better understanding of what exactly can be defined as main determinants of corporate risk-taking.
One of these determinants worth mentioning is the compensation level of the executive director of the firm. A possible rationale behind the relationship between compensation levels and corporate risk-taking is the notion that managers pursue their own interests. Enhancing manager compensation may increase the executive’s willingness to decrease risk aversion in future firm operations as this could pose beneficial outcomes for the executive manager.
The relationship between the Chief Executive Officer compensation levels and corporate risk-taking demands further investigation. Compensation levels and corporate risk-risk-taking share ground in decision making, however, it is vital to introduce new perspectives to this
relationship to find out what exactly influences this relationship and how these main variables are effected by each other. Firm-level and country-level characteristics are therefore added in this research to exhibit the relationship from multiple perspectives.
This paper elaborates on an international sample including the USA and several European countries to have the opportunity to compare these different business perspectives on
4 Netherlands, Switzerland, Spain, Belgium and Sweden. The compensation levels as well as corporate risk-taking levels might show dissimilar patterns between these countries, which can potentially be explained by diverse risk avoidance levels embedded in the culture of each of the countries.
Besides the influence of varying cultural preferences, more factors could pose an impact on the relationship between CEO compensation and corporate risk-taking. One of these factors, which is elaborated upon this study is the board size of the firm. The size of the board, as a firm-level characteristic, could influence this relationship as well, since the number of active people on the executive board may determine the level of autonomy that the CEO possesses. The aim of this paper is to gain a deeper understanding on the relationship and influencing factors between CEO compensation and corporate risk-taking. Hence, the following research question has been constructed:
To what extent does CEO compensation impact corporate risk-taking of firms?
The following sub-research questions have been composed to investigate the moderating effects of country-level and firm-level characteristics;
How does firm-level board size affect the relationship between CEO compensation and corporate risk-taking?
5 The findings of this study indicate a positive relationship between CEO compensation
incentives and corporate risk-taking. Increasing the compensation packages for executives thus leads to higher risk-taking activities of the firm, which in this study is examined by looking at the research and development intensity of this firm. As this indicates that firm risk-taking is influenced by the degree of CEO cash compensation, additional variables have been investigated to further examine the factors that contribute to this finding. Consequentially, this study examines the influence of the firm board size on the relationship between executive compensation and risk-taking. In addition, this study incorporates the effects of country-level uncertainty avoidance scores as well. While both variables are found to be of significant influence as a stand-alone variable, I do not find statistically significant effects for these factors as a moderating effect on the main relationship.
This paper is structured as follows; section 2 presents a review of relevant literature on the relationship between CEO compensation and corporate risk-taking. Within this section the hypotheses are constructed. Section 3 presents the methodological segment of this study by introducing the data collection process and the regression models. Section 4 exhibits the regression results. Lastly, section 5 presents the discussion, implications and limitations of the study.
2. Literature Review
6
3.1 Chief Executive Officer compensation
The topic of Chief Executive Officer compensation has been extensively researched throughout recent years (e.g. Brick et al., 2006; Firth et al., 2006; Finkelstein et al., 2009). The amount of research has risen since the acceleration of these compensation levels during the 80s and the 90s. Evidently, the debate on the determinants and effects of these managerial pay levels has been stimulated simultaneously (Hall and Murphy, 2003). However, many forms of executive compensation levels exist, therefore it makes sense to distinguish between various forms of compensation. Jensen and Murphy (1990) argue that shareholder wealth should be taken as the basis for executive pay and that compensation incentives, such as stock options, bonuses and salary revisions should be based on performance. From this perspective, a high corporate performance would be positively related to the compensation level of the executive director.
According to Firth et al. (1999), compensation packages for executives are important tools in attracting, motivating, retaining and awarding these managers. Although heterogeneity is present in payment packages across firms, Frydman and Jenter (2010) find that most CEO compensation packages contain the following basic components: salary, payouts from long-term incentive plans, annual bonus, restricted option grants and restricted stock grants. However, the relative importance between these components within the compensation packages has shifted throughout recent years. Frydman and Jenter (2010) describe different patterns in US CEO compensation; a time period between 1936 to the 1950s where salaries and annual bonuses were the main drivers of CEO compensation packages. These bonuses were tied to annual accounting measurements, which led to bonus payouts in either cash or stock. After this time period, bonuses were constructed on multi-year performance plans. Hence payments from long-term incentive plans became more important in CEO
7 within the CEO compensation levels. This type of compensation can be seen as one of the key drivers of the growth in CEO pay experienced over the last years (Frydman & Jenter, 2010). The increase in stock compensation compared to cash compensation is suggested to be assigned to accounting regulations which led to a relative decrease in costs of option compensation for organizations (Murphy, 2002). However, as Frydman & Jenter (2010) explain, cash compensation incentives allow a straightforward measurement of executive compensation. This study will emphasize the impact of cash executive compensation and the impact it has on corporate risk-taking.
3.2 CEO compensation and firm performance
Many authors have found linkages between CEO remuneration and corporate performance and risk-taking (Coles et al., 2006; Finkelstein et al., 2009; Fernandes et al., 2011; Ayadi et al., 2012). Ayadi et al. (2012) point out that the relationship between CEO compensation and firm risk-taking is one of the underlying causes of the 2008 financial crisis. The trend of CEOs taking high risks in return for potentially high rewards, which could be achieved via high bonuses in the financial sector, caused excessive risk-taking in the financial sector. Ayadi et al. (2012) stress the positive, causal effects of equity-based compensation for executives on corporate risk-taking.
8 in the organization if the executive compensation packages are structured in optimizing
corporate risk-taking (Coles et al., 2006). Coles et al. (2006) argue that CEOs are influenced by these remuneration incentives and are encouraged via these packages to increase overall risk-taking of the firm.
However, not all scholars find this relationship between CEO pay and firm performance to be positive. Basu et al. (2007) amplify a negative relationship in Japan in investigating these variables. They argue that relatively weak Japanese governance structures are the root cause for a higher degree of agency problems. Basu et al. (2007) argue that these agency problems lead towards overcompensation of the executive, which in turn causes negative firm
performance.
3.3 Agency theory and CEO compensation
9 investments (Boyd, Haynes & Zona, 2010). The mechanism of structuring executive
remuneration is helpful in this trade-off as components of the compensation package can help reach consensus between shareholders and managers. For instance, agency problems can be mitigated by equity-based compensation incentives for CEOs as these incentives induce risk-averse managers to invest in value-enhancing higher risk investments (Haugen & Senbet, 1981; Smith & Stulz, 1985). As Holmström (1982) mentions, the compensation package of the executive is an efficient tool that can be used as an indicator whether the executive has performed according to shareholder preferences.
Shareholders could vary in their risk preferences depending on the originating country of the company and its shareholders. Therefore, besides investigating the effects of CEO
compensation levels on corporate risk-taking levels, it is interesting to consider cultural preferences underlying the relationship as well.
3.4 Uncertainty Avoidance level and corporate risk-taking
Scholars mention that managerial biases may affect corporate investment decisions. Yu and Thuan (2014) argue that CEO attributes have an impact on the relationship between CEO compensation and risk-taking. Attributes mentioned which influence this relationship are dominance, age and overconfidence. Yu and Thuan (2014) further mention that these
10 Previous studies have addressed national culture as an impact on corporate governance
(Hofstede, 1991; Chow et al., 1999; Jansen et al., 2009; Greckhamer, 2011). An important cultural preference dimension within the concept of corporate risk-taking is the risk aversion level embedded in the national culture. An adequate tool that can be used for measuring risk aversion is the Cultural Dimensions model by Hofstede (Hofstede, 1991). Hofstede has conducted research on national cultures and has defined six dimensions in his Cultural Dimensions model. Of these six dimensions, the cultural dimension which will be addressed in this study is the Uncertainty Avoidance dimension. Hofstede (1991) defines this dimension as the degree to which members of a society feel uncomfortable with uncertainty and
ambiguity. This dimension addresses the national preference for having rigid codes for
(managerial) behavior or having a rather relaxed attitude which puts practice above principles. This tool is valuable for this research as we can derive the underlying desire of the culture for an increased or decreased preference in risk-taking. By including the values of this dimension in the model, we note whether the cultural aspect of risk aversion plays a role in the influence of CEO compensation on corporate risk-taking. Slocum and Lei (1993) state that higher corporate risk-taking is observed at US CEOs as uncertainty avoidance levels of the US are relatively low. As Slocum and Lei (1993) mention, these embedded cultural values may impact the risk aversion level of the CEO. Since CEO preferences are impacted, these cultural influences might influence the relationship between CEO compensation and corporate risk-taking.
3.5 Country selection rationale
11 that US CEOs implement their own pay levels. Taking an international perspective by
including European countries in the analysis allows the opportunity to analyze whether indeed the CEO compensation structures are higher in the US and what possible effects this has on firm operations via corporate risk-taking. Weber and Hsee (1998) elaborate on the existence of differing country-level risk preferences. Having different risk aversion levels among countries and investigating the relationship between executive compensation and corporate risk-taking on an international scale could offer interesting insights on the subjects.
The countries included in the research sample are the US, the UK, Germany, France, Italy, Finland, Ireland, the Netherlands, Switzerland, Spain, Belgium and Sweden. One of the main reasons for selecting these countries is their geographic diversity as it allows a comparison on these components between Europe and the United States. The geographic distance between these countries is obviously large and fulfils the desire to investigate the research subject through business perspectives active in different continents. In addition, these countries show appropriate levels of cultural diversity through which cultural aspects can be investigated in the analysis. USA has been included in the sample as it is a world leader within global
economy. Large economic players in the Europe such as the UK and Germany are included in the sample to allow investigation of countries who play a larger role in the European
economy. This makes a comparison with the US more realistic. Furthermore, European countries such as the Netherlands, France, Finland and Ireland are added, which could highlight potential cultural differences in explaining the relationship between executive compensation and corporate risk-taking.
12 countries vary significantly in their levels; we find that the US has a slightly below average score on UA (score of 46). The European countries, however, show divergent scores on uncertainty avoidance; the UK has a relatively low score on UA (score of 35), which would imply that people in the UK are willing to take risks. France scores relatively high on UA (score of 86). The score on UA for Germany is 65 and Netherlands show a score of 53 on Uncertainty Avoidance. The highest score output for UA is for Belgium (score of 94) and the lowest score output for UA is for Sweden (score of 29). Table 1 (Uncertainty Avoidance Score by Country) shows the country-level uncertainty avoidance scores for all countries included in the research sample. With the varying distribution of the uncertainty avoidance levels, we can investigate whether the UA level influences the relationship between CEO compensation and the corporate risk-taking level within the country.
3.6 Board size
13 executive power, Finkelstein (1992) argues that CEOs experience relatively high structural power as a result of the embedded authority of their organizational position. Keltner et al. (2003) holds that powerful managers are the individuals who have few constraints and a great amount of resources. This level of constraints is partially due to the board composition of the firm. Researchers have found differing mechanisms that alter the degree of power that a manager perceives (Daily et al., 2003). Elaborating on this via the agency theory, larger director boards are seen as more powerful in the sense of fulfilling the monitoring role and in assuring effective alignment between the interests of the shareholders and the decision making process of the executive (Jensen & Meckling, 1976; Finkelstein et al., 2009). This implicates that having a larger board of directors potentially decreases the managerial power that the CEO possesses, which influences the decision making of risk investments of the firm. Adding to this notion, Brick et al. (2006) mention that an inefficient CEO monitoring of the board may lead to overcompensation of the executive, which potentially results in an ineffective firm performance. In this view, it is of significant importance to address board structures in terms of total directors present on the board and to investigate the impact it has on the relationship between CEO compensation and corporate risk-taking.
3.7 Hypothesis development
This section of the paper highlights the hypotheses constructed for the relationships described in the literature review of the previous sections. The main focus of this study is to emphasize the impact of CEO compensation on corporate risk-taking. As many authors find this
14 To further explain the main relationship of this study and to stress potential influencing
factors on this relationship, the board size of the firm is included in the investigation to encompass the effects this has on the main relationship. However, no clear consensus has been reached whether the impact of board size is set to be positive or negative. Hence, the second hypothesis is divided into hypothesis 2a and hypothesis 2b, which are structured as follows;
H2a: A larger firm-level board size strengthens the relationship between CEO compensation and corporate risk-taking
H2b: A larger firm-level board size weakens the relationship between CEO compensation and corporate risk-taking
Besides the firm-level factor of board size, this study investigates the international influence of country-level uncertainty avoidance scores on the relationship between executive
compensation and corporate risk-taking. As this impact has not yet been stressed on an extensive manner, no clear consensus has been set out on this moderating impact. Therefore, hypothesis 3 is divided into hypothesis 3a and hypothesis 3b as well, which are formulated as follows;
H3a: A high country-level uncertainty avoidance score strengthens the relationship between CEO compensation and corporate risk-taking
15 The hypotheses are structured based upon the conceptual model of this study, which is shown in figure 1.
3. Methodology
The focus of this study is to investigate the effects of executive manager compensation levels on corporate risk-taking. All data will be gathered for firms active in the United States and Europe. Looking at the conceptual model, the variables are constructed as follows.
4.1 Dependent variable
16 be performed by scaling the ‘Research & Development Expenses’ variable over the ‘Total Assets’ variable found in Compustat Global and Execucomp databases. In addition, as some companies lack data for the ‘Research & Development Expenses’ variable found in
Compustat Global database, data on R&D expenses is retrieved by the ‘Research and Development’ variable in the Asset4 ESG database via Datastream.
Research & development intensity is an adequate proxy for corporate risk-taking levels as expenses on research & development is commonly used as a measurement for risky corporate policies (Li et al., 2013). These R&D investments have a higher rate of risk as beneficial outcomes are uncertain and distant. Hence, the proxy for corporate risk-taking is structured as shown in the following formula:
𝑅𝐷 = 𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑙𝑜𝑔𝑎𝑟𝑖𝑡ℎ𝑚 𝑜𝑓 ( 𝑅&𝐷 𝐸𝑥𝑝𝑒𝑛𝑑𝑖𝑡𝑢𝑟𝑒𝑠 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 )
4.2 Independent variable
17 ExecuComp database as well as BoardEx, which are accessed via Wharton Research Data Services. BoardEx records data on executive compensation via the ‘Total Compensation’ variable. These variables are significantly large compared to the values derived from the corporate risk-taking proxy (RD), which is why we derive the natural logarithm of the found values. Hence, the formula for executive compensation is structured as follows:
𝐸𝑋𝐶𝑂 = 𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑙𝑜𝑔𝑎𝑟𝑖𝑡ℎ𝑚 𝑜𝑓 𝑇𝑜𝑡𝑎𝑙 𝑆𝑒𝑛𝑖𝑜𝑟 𝐸𝑥𝑒𝑐𝑢𝑡𝑖𝑣𝑒𝑠 𝐶𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑖𝑜𝑛
4.3 Moderating variable: Uncertainty Avoidance
The research includes a country-level variable, which is the ‘Uncertainty Avoidance’ (UA) level derived from Hofstede’s Cultural Dimensions (Hofstede et al., 1991). The country scores for UA reflect the risk aversion levels of the country. These scores can be drawn from the Hofstede Insights website1. This research will investigate the country-level moderating
effect of Uncertainty Avoidance scores by using the following formula:
𝑈𝐴 = 𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑙𝑜𝑔𝑎𝑟𝑖𝑡ℎ𝑚 𝑜𝑓 𝑈𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦 𝐴𝑣𝑜𝑖𝑑𝑎𝑛𝑐𝑒 𝑠𝑐𝑜𝑟𝑒
4.4 Moderating variable: Board Size
In addition to the country-level moderating variable Uncertainty Avoidance, the research includes the firm-level variable ‘board size’, which is measured by the number of executive directors as well as the number of non-executive directors within the board. The
18 corresponding variable extracted from the Execucomp database is ‘Board Size’. The formula for the board size moderating variable in this study is structured as follows:
𝐵𝑆𝐼𝑍𝐸 = 𝐵𝑜𝑎𝑟𝑑 𝑆𝑖𝑧𝑒 (= 𝑡𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑏𝑜𝑎𝑟𝑑 𝑚𝑒𝑚𝑏𝑒𝑟𝑠)
4.5 Control variables
Control variables are included in this research to strive towards a more efficient regression analysis and to aim at isolating the effects of our independent variables on the dependent variable. Following previous studies such as Fernandes et al. (2011) and Hayes et al. (2012), this study includes control variables of firm size and leverage. Firm size is included in this study as a control variable as larger firms have different perspectives on firm performance and its risk-taking behavior (Chaudhuri, Khumbakar & Sundaram, 2016). In addition, Tosi et al. (2000) mention that firm size accounts for 40% in total variance of executive compensation. The size of the firm is measured as the natural logarithm of the company’s total assets.
19 𝑆𝐼𝑍𝐸 = 𝐹𝑖𝑟𝑚 𝑠𝑖𝑧𝑒 = 𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑙𝑜𝑔𝑎𝑟𝑖𝑡ℎ𝑚 𝑜𝑓 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠
𝐿𝐸𝑉 = 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 = 𝑇𝑜𝑡𝑎𝑙 𝐷𝑒𝑏𝑡 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠
4.6 Data Collection
The data on the variables mentioned above are retrieved via the Asset 4 ESG database in Datastream, Compustat Global, Execucomp and BoardEx databases. The data on Hofstede’s Uncertainty Avoidance country level is extracted from the website of Hofstede, which indicates individual scores on all cultural dimensions of Hofstede’s model.
The sample period reaches from the year 2011 to 2015. Data is collected for random firms active in the aforementioned countries. This has led to a sample of 600 firm-year
observations. Since not all companies within these databases offer data on CEO compensation levels, the database has been screened to have a clear, valid sample. Multiple databases have been assessed for companies lacking data on either executive compensation levels or
20
4.7 Model Estimation
We distinguish between multiple models in this research in explaining the dependent variable of corporate risk-taking. Model 1 is constructed to test the effect of CEO compensation levels on corporate risk-taking using the dependent variable proxy of research & development intensity. The second model includes the firm-level factor of board size and the third model incorporates the country-level uncertainty avoidance scores. Model 4 includes the interaction effect of executive compensation and board size whereas model 5 incorporates the interaction effect of executive compensation and country-level uncertainty avoidance scores.
(1) RDi,t= β0+ β1∗ EXCOi,t−1+ β2∗ SIZEi,t−1+ β3∗ LEVi,t−1+ εi,t−1
(2) RDi,t= β0+ β1∗ EXCOi,t−1+ β2∗ BSIZEi,t−1+ β3∗ SIZEi,t−1+ β4∗ LEVi,t−1 + εi,t−1
(3) RDi,t= β0+ β1∗ EXCOi,t−1+ β2∗ BSIZEi,t−1+ 𝛽3∗ UAi,t−1+ 𝛽4∗ BSIZEi,t−1∗ UAi,t−1
+ β5∗ SIZEi,t−1+ β6∗ LEVi,t−1 + εi,t−1
(4) RDi,t= β0+ β1∗ EXCOi,t−1+ β2∗ BSIZEi,t−1+ 𝛽3∗ EXCOi,t−1∗ BSIZEi,t−1+ 𝛽4∗ UAi,t−1
+ 𝛽5∗ BSIZEi,t−1∗ UAi,t−1+ β6∗ SIZEi,t−1+ β7∗ LEVi,t−1+ εi,t−1
(5) RDi,t= β0+ β1∗ EXCOi,t−1+ β2∗ BSIZEi,t−1+ 𝛽3∗ EXCOi,t−1∗ BSIZEi,t−1+ 𝛽4∗ UAi,t−1
+ 𝛽5∗ EXCOi,t−1∗ UAi,t−1+ 𝛽6∗ BSIZEi,t−1∗ UAi,t−1+ β7∗ SIZEi,t−1+ β8
21
4. Empirical results
This section of the paper will introduce the descriptive statistics and regression results of the regression analysis. Several regressions will be run of which the results will be discussed in the section after the current one. Many authors, such as Conyon & He (2011), investigating the influences of CEO compensation make use of linear regression methods such as Ordinary Least Squares. This study will perform a linear regression method on the above mentioned variables to estimate the unbalanced panel dataset.
5.1 Descriptive statistics
23
5.2 Correlations
Table 4 (Correlation Matrix) represents the covariance correlation between all variables used in the regression models. The purpose of the correlation matrix is to describe relationships between explanatory variables. The correlation matrix offers insights whether
24
26 The models (1-3) represented in table 5 (Regression Analysis) and models (4-5) shown in table 6 (Regression Analysis) show the coefficients for the regressions. Model 1 analyzes the sole effect of executive director compensation on corporate risk-taking via research &
development intensity. This relationship is the basis for hypothesis 1. Models 2-5 are added to study the impact of including board size and country uncertainty avoidance scores in the analysis. Models 2 and 4 are added to investigate the effects of board size on this relationship, which is practiced to test hypothesis 2. Models 3 and 5 include the impact of country-level uncertainty avoidance scores on the relationship between CEO compensation and corporate risk-taking, which is addressed in hypothesis 3.
Model 1 documents the coefficient of executive compensation on research intensity without the influences of board size and uncertainty avoidance scores. In this model, we notice a positive coefficient of 0.686 for executive compensation, which is significant at the 1% level. The accompanying standard error and p-value are 0.115 and 0.000, respectively. As we observe a positive, significant influence of executive compensation on research &
development intensity, we may assume that hypothesis 1 is supported. Conclusively, we draw upon the finding that CEO compensation is positively related to research & development intensity.
27 In Model 2, the influencing factor of board size is added to the model. Hypotheses 2a and 2b are constructed upon the positive and negative influencing effect of board size on the main relationship, respectively. Model 2 shows that a positive coefficient of 0.684 is again found for the main relationship of executive compensation and R&D intensity, which is significant at the 1% level.
Looking at the board size variable, the coefficient of board size (0.045) is positive and
significant at the 1% level. The accompanying standard error and p-value are 0.017 and 0.000, respectively. This indicates that board size as a stand-alone variable has a positive, significant influence on corporate risk-taking. Besides, it can be noted that the coefficient of executive compensation has decreased from 0.686 (model 1) to 0.684 (model 2), which might indicate that the influence of board size weakens the relationship between executive compensation and research & development intensity. This suggestion would be in line with hypothesis 2b. However, to investigate this moderating effect of board size on the compensation – risk-taking relationship this study incorporates an interaction effect between executive
compensation and board size. This interaction is shown in model 4, which can be found in table 6 (Regression Analysis). The interaction effect encompasses the combined coefficient of executive compensation and board size on R&D intensity. Thus, this effect exhibits the moderating effect of board size on the main relationship.
28 and 2b, which indicates that board size does not have a significant moderating effect on the relationship between executive compensation and corporate risk-taking.
The control variables in model 2 still pose negative coefficients, which are significant at the 1% level. Conclusively, the adjusted R-squared for model 2 is 0.149, which indicates that including the board size variable increases the overall explanatory power of the model
compared to model 1. Model 4 show negative coefficients for the control variables, positive at the 1% level as well. The adjusted R-squared for model 4 is 0.152.
Model 3 includes the country level variable for uncertainty avoidance in the analysis. Hypotheses 3a and 3b are constructed upon the influencing factor of uncertainty avoidance scores on the relationship between executive compensation and research & development intensity. In this model, we find a positive coefficient for executive compensation of 0.641, which is significant at the 1% level. Standard robust errors and the corresponding p-value for this variable are 0.118 and 0.000, respectively.
29 Furthermore, an interaction effect of board size and uncertainty avoidance show a negative coefficient of 0.162, which is significant at the 5% level with a corresponding standard robust error of 0.079. The interaction effect entails the combined influence of board size and
uncertainty avoidance scores on research & development intensity. While both board size and UA variables show positive coefficients in model 2 and 3, this interaction effect shows a negative coefficient on RD. This can be interpreted as that the effect of the combined action of board size and UA is less than the sum of the individual effects of board size and UA. However, model 3 does not provide a rationale to state that uncertainty avoidance has a significant moderating impact on the main relationship. To incorporate the moderating effect of UA on the compensation – risk-taking relationship, an interaction effect of executive compensation and uncertainty avoidance is included in model 5. This interaction effect shows a coefficient of 0.571 with a corresponding standard robust error of 0.451. The interaction of compensation and UA does not show a statistically significant effect. Thus, this study does not find support for hypotheses 3a and 3b, which indicates that uncertainty avoidance does not have a significant moderating effect on the relationship between executive compensation and corporate risk-taking.
The control variables of firm size and leverage again pose to have negative coefficients both found to be significant at the 1% level in models 3 and 5. The found adjusted R-squared of models 3 and 5 are 0.150 and 0.153, respectively.
5. Discussion & limitations
This section of the paper elaborates on the implications of the results of the regression
30
6.1 Discussion
31 deliver adequate provisions on executive compensation governance (Barney, 2009).
Regardless, the implementation of the Sarbanes-Oxley Act and the specific sections mentioned, evoked the debate on executive compensation policies and strategies. These developments led to an increased attention on the relative level of CEO remuneration.
Next to the rising debate on compensation incentives, corporate risk-taking procedures have been in the public’s attention for a long period of time. Particularly, events which involved excessive corporate risk-taking such as the aforementioned scandals increased the relevance of this topic in current International Financial Management. As modern global economics is heavily dependent on large firms and their long-lasting existence it is vital that the risk-taking practices of these firms remain effective. Historical events, such as the Lehman brothers debacle which led to the 2008 financial crisis, have pointed out that global economics is dependent on responsible risk-taking of worldwide companies. Within the risk-taking domain, lawmakers and company shareholders differ in their risk preferences. Lawmakers aim at stabilizing the well-being of the economy by striving to avoid excessive risk-taking whereas firm shareholders aim at realizing high stock returns via risk-taking of the executive
managers. Executive managers are seen as more risk averse than the shareholders of the company. One important tool for shareholders to overcome risk aversion at the managerial level is to encourage these executives to participate in increased risk-taking via compensation packages. Thus, these remuneration incentives influence risk-taking practices of the
companies.
32 compensation and how these incentives impact firm performance. Previous researchers have investigated the influences of corporate governance scores on corporate risk-taking (Cyert et al., 2002; Coles et al., 2006; John et al., 2008; Larcker et al., 2012). This study incorporates board size and Uncertainty Avoidance scores to the analysis to further encompass the factors that contribute to this relationship. The findings present a positive, significant relationship between executive compensation and corporate risk-taking. An increased compensation package can be linked towards higher research and development activity relative to the assets of the firm. This finding indicates that executives are indeed influenced in their risk aversion levels by the remuneration packages set by the company. As previous studies such as Coles et al. (2006) and Boyd et al. (2010) found as well, shareholders can use executive compensation as a tool to drive managers towards higher risk-taking activities. The results show that an increased executive pay is related to a higher ratio of the firm’s research and development expenses scaled over firm assets since R&D investments are seen as higher risk projects. Implications hereof are that the board of the firm should be aware of this mechanism, which provides the board with reasoning to closely monitor executive compensation and to assess whether these executives strive towards effectively aligning their own goals with company objectives.
33 processes on firm risk by the amount of directors on the board. The combined effect of
executive compensation and board size does not significantly influence corporate risk-taking. Country uncertainty avoidance scores have been incorporated in this study’s models to investigate whether cultural aspects play a moderating role in the compensation – risk-taking relation. To comprise risk aversion preferences embedded in national cultures, the Uncertainty Avoidance dimension as defined by Hofstede (1991) is included in my model. My findings imply that country uncertainty avoidance scores do not moderate the main relationship. This indicates that cultural risk preferences have no combined effect with compensation packages in explaining corporate risk-taking.
Overall, my findings imply that shareholders should take into consideration that remuneration incentives designed for increased corporate risk-taking will indeed motivate executives to increase their risk activities if it generates beneficial outcomes for the executives.
Shareholders and company policy-makers should act in a responsible manner in designing executive compensation packages, since we have seen in the past that excessive corporate risk-taking has a substantial impact on global economics.
6.2 Limitations
Evidently, this study is subject to limitations. One of the limitations of this study is driven by data availability. It is difficult to retrieve a significant amount of data on executive
compensation levels as this data is relatively scarce. Databases such as Datastream,
34 Furthermore, availability of data on Research & Development expenses is limited as well. To ensure an accurate sample, I choose not to assume undisclosed values of R&D to be zero for the companies included in this study’s sample. However, since both R&D expenses and executive compensation data is limited, it is difficult to acquire a decent sample size on the variables included in the analysis. Evidently, the sample size of this study needs to be
mentioned as a limitation of this study as it reduces the generalizability of its findings. Future research could benefit from incorporating multiple countries originating from various
continents in the world to derive a more globally scalable model. Since this study solely considers European and US firms, it could be interesting to conduct research on the compensation – risk-taking relation in Asian countries for example.
Another limitation of this study is that this study uses solely research & development intensity as a proxy for corporate risk-taking. The proxy of volatility of corporate earnings has been used as a proxy within this research as well, however, this proxy did not pose any significant results. Therefore, the dependent variable of volatility of corporate earnings has been left out of this research. It could well be that the sample of this study has proved to be inadequate to explain the effects of CEO compensation on corporate risk-taking via the volatility of
corporate earnings. Future research could certainly benefit from including a secondary proxy for corporate risk-taking as this increases the explanatory power of the independent variables and thus the modelling of corporate risk-taking as whole.
35 emphasized differently in a culture which values Individualism enormously. Thus,
incorporating multiple cultural aspects in future research could enrich the understanding of cultural influences on the executive compensation – corporate risk-taking relationship. Besides further elaborating on the cultural aspects and the impact it has on the main
36
6. References
Adkins, L.C. and Hill, C.R. (2011). Using Stata for principles of econometrics, fourth edition. 4th edn. New York: John Wiley & Sons.
An, Z., Huang, G., Li, D., & Xiao, S. (2014). Foreign institutional ownership, risk-taking, and crash risk around the world. Working Paper, available at
http://ssrn.com/abstract=2406743.
Ayadi, R., Arbak, E., & Pieter De Groen, W. (2012). Executive compensation and risk taking in European banking. Research Handbook on International Banking and Governance, JR Barth, C. Lin, C. Wihlborg, eds., Edward Elgar.
Barney, J., L. (2009). Corporate Scandals, Executive Compensation, and International Corporate Governance Convergence: A U.S.-Australia Case Study, Temple International and Comparative Law Journal, Vol 23(2): 231 – 267.
Basu, S., Hwang, L. S., Mitsudome, T., & Weintrop, J. (2007). Corporate governance, top executive compensation and firm performance in Japan. Pacific-Basin Finance Journal, Vol 15, 56- 79.
Berle, A.A., Means G.C. (1932). The Modern Corporation and Private Property. New York: Macmillan
Boyd, B., Haynes, K., & Zona, F. (2010). Dimensions of CEO–board relations. Journal of Management Studies, Vol 48: 1892–1923.
Brick, I. E., Palmon, O., & Wald, J. K. (2006). CEO compensation, director compensation, and firm performance: Evidence of cronyism?. Journal of Corporate Finance, Vol 12(3), 403-423.
37 Chow, C. W., Harrison, G. L., McKinnon, J. L., & Wu, A. (1999). Cultural influences on
informal information sharing in Chinese and Anglo-American organizations: an exploratory study. Accounting, Organizations and Society, Vol 24(7), 561-582. Coles, J. L., Daniel, N. D., & Naveen, L. (2006). Managerial incentives and risk-taking.
Journal of Financial Economics, 79(2), 431-468.
Conyon, M. J. & He, L. (2011). Executive compensation and corporate governance in China., Journal of Corporate Finance, Vol 17, 1158-1175.
Croci, E., Gonenc, H., Ozkan, N. (2012). CEO compensation, family control, and institutional investors in Continental Europe., Journal of Banking & Finance, Vol 36, 3318–3335. Cyert, R. M., Kang, S. H., & Kumar, P. (2002). Corporate governance, takeovers, and top
management compensation: Theory and evidence. Management Science, Vol 48(4), 453-469.
Daily, C., Dalton, D., & Cannella, J. A. (2003). Corporate governance: Decades of dialogue and data. Academy of Management Review, Vol 28: 371–382.
De Cesari, A., Gonenc, H., & Ozkan, N. (2016). The effects of corporate acquisitions on CEO compensation and CEO turnover of family firms. Journal of Corporate Finance, Vol 38, 294-317.
Dee, C. C., Lulseged, A., & Nowlin, T. S. (2005). Executive compensation and risk: The case of internet firms. Journal of Corporate Finance, Vol 12(1), 80-96.
Eisenhardt, K. (1989). Agency theory: An assessment and review, Academy of Management Review, Vol 14: 57–74.
38 Fernandes, N., Ferreira, M., Matos, P., Murphy, K. (2013). Are US CEOs paid more? New
international evidence. Rev. Financ. Stud. 26, 323–367.
Ferris, S.P., Javakhadze, D., Rajkovic, T. (2017). CEO social capital, risk-taking and corporate policies. Journal of Corporate, Vol 47, 46-71.
Finkelstein, S. (1992). Power in top management teams: Dimensions, measurement, and validation. Academy of Management Journal, Vol 35: 505–538.
Finkelstein, S., Hambrick, D. C., & Cannella, A. A. (2009). Strategic leadership: Theory and research on executives, top management teams, and boards. Oxford University Press, USA.
Firth, M., Fung, P. M., & Rui, O. M. (2006). Corporate performance and CEO compensation in China. Journal of Corporate Finance, Vol 12(4), 693-714.
Firth, M., Tam, M., Tang, M. (1999). The Determinants of Top Management Pay. The International Journal of Management Science, Vol 27, 617 – 635.
Frydman, C., & Jenter, D. (2010). CEO Compensation. CESifo Working Paper, No. 3277. Gande, A. & Kalpaty, S. (2017). CEO compensation and risk-taking at financial firms:
Evidence from U.S. federal loan assistance. Journal of Corporate Finance, Vol 47, 131-150.
Gordon, J., N. (2002). What Enron Means for the Management and Control of the Modern Business Corporation: Some Initial Reflections, The University of Chicago Law Review, Vol 69, 1233, 1234.
39 Greckhamer, T. (2011). Cross-cultural differences in compensation level and inequality across
occupations: A set-theoretic analysis. Organization Studies, Vol 32(1), 85-115. Hall, B. J. & Murphy, K. J. (2003). The Trouble with Stock Options, Journal of Economic
Perspectives, Vol 17, 49-70.
Haugen, R., Senbet, L. (1981). Resolving the agency problems of external capital through options. Journal of Finance, Vol 36, 629–647.
Hayes, R.M., Lemmon, M., Qiu, M., (2012). Stock options and managerial incentives for risk taking: Evidence from FAS 123R. Journal of Financial Economics, Vol 105, 174-190.
Hofstede, G., Hofstede, G. J., & Minkov, M. (1991). Cultures and organizations: Software of the mind (Vol. 2). London: McGraw-Hill.
Holmström, B. (1982). Moral hazard in teams, The Bell Journal of Economics,. Vol 13(2):324—40
Jansen, E. P., Merchant, K. A., & Van der Stede, W. A. (2009). National differences in incentive compensation practices: The differing roles of financial performance measurement in the United States and the Netherlands. Accounting, Organizations and Society, Vol 34(1), 58-84.
Jensen, M. C., Meckling, W. H. (1976). Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Finance and Economics, Vol 3(4):, 305—60 Jensen, M.C., & Murphy, K. J. (1990). Performance pay and top-management incentives.
Journal of Political Economy, Vol 98 (2), 225-264.
40 Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition.
Psychological Review, Vol 110: 265– 284.
Kuo, C. S., Li, M. L., & Yu, S. E. (2013). Non-uniform effects of CEO equity-based compensation on firm performance – An application of a panel threshold regression model. The British Accounting Review, Vol 45, 203-214.
Larcker, D. F., McCall, A. L., Ormazabal, G., & Tayan, B. (2012). Ten Myths of' Say on Pay'. Rock Center for Corporate Governance at Stanford University Closer Look Series: Topics, Issues and Controversies in Corporate Governance, No. CGRP-26. Li, K., Griffin, D., Yue, H., Zhao, L. (2013). How does culture influence corporate
risk-taking?, Journal of Corporate Finance, Vol 23, 1-22.
Murphy, K., J. (2002). Explaining executive compensation: managerial power versus the perceived cost of stock options. The University of Chicago Law Review, Vol 69(3), 847-869
Osei-Bonsu, N., & Lutta , J., G. (2016) CEO Cash Compensation and Firm Performance: An Empirical Study from Emerging Markets. Business and Economic Research, Vol 6 (2), 79-99.
Sarbanes-Oxley Act. (2002) §§ 304(a)(1)-(2),15 U.S.C. §§ 7243 (a)(1)-(2) (2008). Shapiro, S. P. (2005). Agency Theory. Annual Review of Sociology, Vol 31, 263-284. Shaw, K. W., & Zhang, M. H. (2010). Is CEO Cash Compensation Punished for Poor Firm
Performance? The Accounting Review, Vol 85(3), 1065-1093.
Slocum, J. W., & Lei, D. (1993). Designing global strategic alliances: Integrating cultural and economic factors. Organizational change and redesign: Ideas and insights for
41 Smith, C., Stulz, R. (1985). The determinants of firms' hedging policies. Journal of Financial
Quantitative Analysis, Vol 20, 391–405.
Tosi, H., Werner, S., Katz, J., Gomez-Meija, L., (2000). How much does performance matter? A meta-analysis of CEO pay studies. Journal of Management, Vol 26 (2)
Weber, E. U. & Hsee, C. K. (1998). Cultural differences in risk perception, but cross-cultural similarities in attitudes towards perceived risk. Management Science, 44(9), 1205-1217.
Yu, H. & Thuan, L., T. (2014). CEO compensation, CEO attributes and corporate risk taking evidence from US listed corporations. Banks and Bank Systems, Vol 9(4)
