1
The impact of board diversity on CEO compensation
Master Thesis - Finance By DAVIDE CAVALLARO
Abstract
This thesis tests whether board diversity, with respect to gender and nationality, has an effect on CEOs compensation. Over the recent years, CEO compensation has increased substantially, to the point it could be inefficient from the stakeholders perspective. Firm characteristics are to be taken into account when assessing CEOs remuneration. However this thesis finds that another variable, namely board diversity, affects it. The regression model shows how gender diversity is negatively linked to CEO compensation, whereas nationality diversity increases it. Hence the hypothesis of this thesis is supported: CEOs remuneration is indeed inefficient on average, as it should reflect only financial outcomes and firm complexity. On top of that, it is clear how corporate governance is currently failing to reduce CEOs compensation to an efficient level, on behalf of stakeholders. Two possible solutions to this issue can be suggested:
1) increasing the amount of women in boards as they are tougher monitors than men, hence they are more likely to improve monitoring;
2) increasing the rate of attendance to meetings of international board members. With these corrections, the quality of monitoring would be greatly improved, possibly reducing CEOs compensation to efficient levels, as a way to solve the principle-agent conflict.
Author: Davide Cavallaro Date: January 19th 2017 Program: MSc Finance Student number: S3052281 Supervisor: C.Lisciandra E-mail: d.cavallaro@student.rug.nl Word count: 11,037
2 CONTENT
1. INTRODUCTION 3
2. LITERATURE REVIEW 7
2.1 THEORETICAL BACKGROUND 8
2.2 ASPECTS OF DIVERSITY AND HYPOTHESES 13
3. RESEARCH METHODS 16 3.1 DATA GATHERING 16 3.2 DATA ANALYSIS 17 4. DISCUSSION OF RESULTS 19 4.1 DESCRIPTIVE STATISTICS 19 4.2 REGRESSION ANALYSIS 22
5. CONCLUSION AND FINAL REMARKS 27
6. FUTURE RESEARCH 29
7. APPENDIX 31
3 1. INTRODUCTION
This thesis focuses on the possibility that board diversity could affect the remuneration levels of CEOs. Board diversity is measured as the variation in terms of nationality and gender of directors. I explore the way that CEOs remuneration usually works, which is a certain amount decided through negotiations with the board of directors. One should expect compensation to be efficient, in order to align managerial interest and give incentive to exert decisive effort. An efficient compensation is the result of a sound corporate governance process, based on the principal-agent theory: it should be sufficient to align interests of CEOs and stakeholders, while still maximising shareholder value in the long term.
4 independence of directors, which is represented by conflict of interests and collusion, would lead to CEO covering a role of major power; hence the hypothesis of an increase in CEOs compensation. This thesis try to evaluate the role of monitoring and independence, as well as trying to explain their influence over CEO compensation.
Corporate governance finds its roots in a simple but yet efficient model: the stakeholders and shareholders proceed to empower the directors, composing the board, in order to monitor executives. The model is easily explained by a linear relationship, named the principal/agent relationship. The principal is the group of shareholders as owner of the firm equity. The board of directors receives the task of monitoring and assessing managerial activities on behalf of the shareholders. The role of the agent is attributed to managers and executive members.
Boards structure differs based on the structure of each country's economy; it could be formed as a one tier system or two tier system. The dual board is present in countries like Germany, whereas one tier board are more common in the United States. The one tier system boards have both the management and monitoring functions incorporated, the two tier systems tend to separate them. On the one hand, scholars argue that using a two tier system also results in "more monitoring", "less aggressive performance targets", and separation of power by not allowing the CEO to be chairman (Carrasco, 2005). On the other hand it is considered to be "less efficient" from a market point of view. The lack of efficiency arises from the potential lack of communication, and the higher costs of running a Dual Board. In this study, I will provide an overview of both type of boards and the way they influence the remuneration negotiations.
5 CEO dominant control directly reduces directors independence. Together with lack of monitoring, independence could be one of the possible explanations for the recent increase of CEO compensation. A specific board composition and its independence could be affected by diversity forces. Nationality and gender diversity could end up shifting the power relationship among board and CEO. Both internationalization and women quotas are deeply covered in following chapters.
Globalization makes it so that the entire world economy constantly mutates, expanding the possibilities of international relationship and exchanges. At the firm level, it is easy to notice an increasingly multi-nationality engagement, meaning that board of directors experience increased nationality diversity (Masulis, Wang & Xies 2012). Another milestone matured over the last decade is women quotas. These quotas have greatly affected the job market, as they have revolutionized the way board of directors are composed. Not only the composition of boards changes based on mandatory quotas of women, but also its activity of monitor changes drastically. As explained by Ferreira (20015) and Hillman (2015), quotas account for both gender diversity and social-ethnical diversity. Previous research has shown that both internationalization and women quotas influence the quality and the scope of the monitoring efforts exerted by the board of directors. However these two developments affect CEO compensation in different ways, which are explained in detail. Adams & Ferreira (2009) focuses solely on gender diversity an argue that female directors are more independent than male directors. Additionally, women are more likely to attend board meetings, while not being influenced by the executives as much as their male counterparts. Quoting their words, women are “more likely to be tough monitors of CEOs”. The tighter monitoring should lead to a process that reduces CEO power as it rebalances the power struggle between boards and CEOs (Hambrick & Mason 1984).
6 have a lower attendance to board meetings and additional misreporting. The status of foreign director is attributed to directors whose domicile does not correspond to the country of the company Headquarter. Previous research shows that the consequence of internationally diversified board of directors is a lack of monitoring and control possibly leading to a higher CEO compensation (Carter, Simkins & Simmons, 2003). Nonetheless, previous research only pointed out the characteristics of board independence, without exploring the possibility that it could influence CEO remuneration.
Another pressing matter is the lack of research on independence relative to nationality and gender diversity and the way that they could possibly influence remuneration schemes. Past research usually focuses on the effect of diversity over remuneration in terms of monitoring, that way not accounting for the possible influence of independence. Both gender and nationality diversity could have an impact through independence over compensation negotiations. Women and foreign directors on boards should bring more independence as they are less likely to have ties with the management, as well as bringing a new and fresh perspective alongside. Therefore, with this thesis, I aim to answer the following research questions: do gender and nationality diversity have any influence on the remuneration levels of CEOs? Is it possible to capture this effect with a regression analysis? Can I find out whether board diversity affect compensation schemes through tougher monitoring or greater independence?
7 is gathered from the broadly known databases Orbis, Compustat and Boardex; it includes an extensive range of observations across European and American firms, based on the most recent values. The fourth chapter will cover the results emerged from the regression analysis, while the fifth section concludes with a final overview of the topic and the explanation of the results. As far as the univariate analysis is concerned I will provide a descriptive analysis Table. A correlation matrix and various regression models are used for the multivariate analysis. In order to control if the remuneration levels are efficient and deserved, many variables are used as control variables. Size, profitability and riskiness of firms help checking that the compensation exceeds average or expected levels. Additionally, I use the ratio of non-executive directors (NEDs) as a proxy to test the true impact of independence over remuneration levels. NEDs should not have any connections to the firm, that could lead to colliding interests. Therefore, they are less likely to collude with the management, compared to executive directors (Roberts et al., 2005). As a further robustness control, the results are compared with control groups, which purpose is to show the different levels of board diversity based on CEO's compensation. I will divide the dataset into five quintiles sorted by remuneration level, starting from the highest up until the lowest. The goal of this analysis is to check whether the population follows the hypotheses. I expect to see the first quintile, that shows the highest levels of compensation, presents low level of gender diversity and high level of nationality diversity. Another test would be to assess whether diversity variables present endogeneity issues, which means that they could be affected by firms variables. I will additionally test whether the results of the main regression model hold when I divide the dataset in American and European firms.
2. Literature Review
8 2.1 Theoretical background
I start analysing the basics of corporate governance, especially the role of the directors as the control organ of firms. The board of directors act on behalf of the shareholders (principal) and, among other responsibilities, hire/fire and compensate managers and executives (agent). Additionally it tries to solve the pressing matter of conflict of interests among decision makers and residual risk bearers. It strives to reduce the transaction (agency) costs arising from the separation of ownership and control and makes so that the firm, perceived as a complex entity, survives in the long term (Baysinger and Butler, 1985).
The role of the board of directors can be simply defined as follows: the board draws and nominates new executives by means of compensation packages that are found in close negotiations with the potential candidate(s). The remuneration process should be efficient, in the sense that it should be tied to the first and foremost objective, which is increasing shareholder value (Van Veen & Wittek, 2016). The efficiency of remuneration will be defined in this research as a compensation level that is adequate in terms of compensating the CEO based on the firm size, complexity and risk (Boyd, 1994). Remuneration exceeding this defined level, will be treated as being excessive, as it is consequently not explainable by a CEOs remuneration level strictly related to the firm characteristics mentioned previously (Frydman & Jenter, 2010).
9 efficiency or if it is derived by corporate governance failure (Frydman & Jenter, 2010). The former scholars argue that the steep increase of CEOs compensation is a result of an efficient market. Additionally, they state that an increase in compensation surely reflects scarcity on the labour market, leading to an increase in skill demand and complexity of technology markets, as well as stricter monitoring due to higher institutional ownership (Frydman & Jenter, 2010). On the contrary, the opposite view argues that the board is dominated by the CEO and other top executives, who can grant themselves bonuses and pay packages that are beyond efficiency (Bebchuk & Fried 2004, Kuhnen & Zwiebel, 2009). Also Bertrand & Mullanaithan (2001) reject the findings of Frydman & Jenter (2010), as they argue that CEOs are actually not rewarded efficiently. In their research on CEOs compensation level in the oil industry, they find out that on average the remuneration is based on luck, and more specifically on performance beyond the CEO's control; hence confirming the hypothesis of inefficient remuneration.
10 corporate governance literature, as it focuses on agency theory, is that the quality of board control can be directly linked to the independence of the directors (Boyd, 1994).
This thesis will follow the concept of independence and monitoring being close aspects of governance, but still being treated as two separate entities. As stated before, level of independence could predict the quality of monitoring (i.e. scarce independence will likely lead to poor monitoring). However, monitoring usually is a meagre predictor of independence.
Sundaramurthy & Lewis (2003) argue that independent directors are expected to be less inclined to collude with the executives. Independent directors should be free from social cohesion pressures that could possibly connect the board to the CEO (Van Veen & Wittek, 2015). Additionally they should retain their independence within the ratification and monitoring of executive decisions (Roberts, McNulty & Stiles, 2005). Fama and Jensen (1983) argue that the board of directors should avoid collusion with the CEO and dominating power of the CEOs himself, in order to have sufficient monitoring incentives. The same concept could be easily applied to the compensation negotiations as well. A board free of executive collusion is more inclined to set an efficient level of compensation that maximises the value of the firm, keeping in mind the interest of the shareholders. To achieve an efficient level of compensation, the board of directors has to be free of social pressures and colluding ties to the executive (Van Veen & Wittek, 2016).
11 demonstrated that providing alternative options to governance processes, increase the probability of a better monitoring outcome. On the other hand, though, a different perspective could not necessarily result in more effective monitoring in case outsiders may be marginalized and stripped of their power (Carter, Simkins & Simmons 2003).
Talking about board independence, it is important to introduce non-executive directors (NEDs). Stakeholders usually introduce NEDs in boards to enhance the level of independence. NEDs should not have any other connections to the firm, other than their board membership, that could lead to colliding interests. Hence, they are less likely to collude with the managers, compared to executive directors (Roberts et al., 2005).
12 opinion such as the background, beliefs or morality. As I have already made clear before, this thesis will focus solely on two kind of diversity: gender and nationality. The main reason to include them is that they are easily observable, even though there are contrasting theories in the literature about the way they influence CEOs compensation; they are surface level kind of diversity, hence easy to measure and test. However, future research should try to include below-surface type of diversity, to better grasp diversity impact on firm variables.
13 With firms being increasingly multi-nationally engaged, boards are facing increased nationality diversity (Masulis, Wang & Xies, 2011). At the same time women quotas are accounting for social diversity and equality and change in the composition of boards (Ferreira, 2015; Hillman, 2015). Subsequently, boards composition and their relative independence have been subject to diversity pressures due to these distinct developments. Nationality and gender diversity could end up shifting the power relationship between the board and CEO. In the following paragraphs I extensively elaborate on the two developments.
2.2 Aspects of diversity and hypotheses
Gender diversity has been one of the major topics of interest in the literature relative to corporate governance. Research tried to analyse gender diversity features and its effect on board diversity. Female board members have received an increasing attention, especially since the introduction of a quota in many countries, with Norway as the first to implement it (Ferreira, 2015). Although there are very mixed results on the influence of gender diversity and quotas on firm performance (e.g. Erhardt, Werbel & Schrader, 2003; Ahern & Dittmar, 2012; Matsa & Miller, 2013), the direct influence on board independence draws a clearer picture. Adams and Ferreira (2009) argue that female directors are more independent than male directors and have a significant impact on board inputs and firm outcomes. The greater independence is motivated with the fact that women, being a minority, are less likely connected or willing to collude with the CEO. Moreover, the authors show that female directors are more likely to join remuneration and monitoring committees. Remarkably, the attendance rate of male monitors increases as well, as long as female members are represented in the board.
14 reasons could be found in cognitive psychology and decision theory; women bring in fresh perspectives that could reduce or correct informational bias. This can be advantageous as it breaks up group thinking and prevents collusion (Dewatripont, Jewitt & Tirole, 1999; Westpahl & Milton, 2000). However, if it is assumed that gender diversity guarantees independency itself, I would expect to observe the same effect considering internationally diversified boards. Instead, I observe that the benefits introduced by outsiders in the board are offset by an increase of power of national board members, since they tend to be marginalised. In other words, an internationally diversified board is not independent at all. Hence, I argue that a way of coping with excessive CEOs remuneration could be tougher monitoring. Especially important is the presence of women as directors, in order to contrast the lack of independence.
15 interests of stakeholders, contrasting the dominant power of CEOs during compensation negotiations. Therefore, one should expect a lower likelihood of excessive compensation levels, following the managerial power hypothesis. Therefore I take example from the literature on gender diversity in order to model the regression analysis, expecting similar results. The hypothesis is that the ratio of female board directors is negatively related to (possibly excessive) CEO compensation.
The globalization of markets and industries has put diversity pressures on board of directors. The argument that would explain it is that the board of directors has to resemble the multinational firm strategy and employee base of the firm, in order to meet the challenges and complexity of multinational companies (Ruigrok, Pech & Tacheva, 2007). Accordingly, Van Veen & Elbertsen (2008) argue that with increasing development and globalization of capital markets, institutional investors would demand more representation within the board, which increases national diversity (might be ethnicity diversity as well).
16 Circumstantial evidence is given where non-executive directors are visiting plants to better grasp the functioning of the firm. To support this theory, Masulis et al. (2012) report that a higher number of FIDs leads to higher misreporting, which is even more pronounced when they are part of audit committees. As part of the process, the authors claim that the CEO compensation increases as well when more FIDs sit on compensation committees. The reasoning behind this outcome can be found in the research done by John, Knyazeva and Knyazeva (2011). The authors find that the higher the director’s geographic distance from the company headquarters, the lower is their ability and incentives to collect information on executives, hence granting poor management monitoring. Based on the aforementioned literature, it is arguable that a higher number of FIDs leads to more managerial discretion due to the lowered board control and effectiveness. This is what it can be derived from the lower attendance rates and lower ability and incentive to gather information and monitor compensation. Therefore I derive the second hypothesis of the literature, which is: the number of foreign independent directors is positively related to excessive CEO compensation. The regression analysis will find out which of the two basic assumptions holds, namely gender and international diversity influencing CEOs remuneration. Additionally, I will explore whether control or independence are the drivers of such relation.
3. Research Methods 3.1. Data gathering
17 advantage of this selection is that the observed causalities are not attribuTable to either the CEOs or board characteristics linked to a specific industry or country (Boyd, 1994). This means that no specific industry or country is the major driver of the variation in the dependent variable, but the collection of observation in its entirety explains it. The construction ends up leading to a sample size of 1793 firm observations gathered from the most recent corporate analysis, which is 2015. The dataset is formed as a cross section analysis, since the databases allow to gather data for a single period, especially confidential variables (related to compensation or individual board member characteristics).
3.2. Data analysis
In order to analyse the underlying relationship among variables, a quantitative approach has been selected. An ordinary least squares regression (OLS) has been applied to identify the underlying causalities of the diversity variables on the dependent variable of CEO compensation.
CEO compensation, acting as the dependent variable, is measured as total compensation; the variable is constructed as the reported cash compensation of base salary plus bonuses and further deferred income such as corporate shares and options, following Boyd (1994), as well as Lewellen & Huntsman (1970). Gender diversity has been evaluated as the percentage of female directors within the board, measured as the ratio between the number of female directors and the total number of directors. A value of 0 would mean that there is no presence of female directors, whereas a value of 1 would represent a board of solely female directors (certainly unlike to happen). The higher the percentage of women gets, the higher the assumed independence of the board from the prevailing power of CEOs, as well as greater control or monitoring. My hypothesis is that a gender diversified board is less likely to allow excessive compensation of the CEO (Tosi & Gomez-Mejia, 1989; Adams & Ferreira, 2009).
18 a ratio of total number of different nationalities over the total number of directors. A value of 0 would indicate non-diversity with a nationally homogenous board, that is every director’s nationality matches the company’s HQ nationality, whereas a higher ratio should represent a more heterogeneous board. Even though foreign independent directors should not have ties with management and collusion of interests, their presence is usually linked to worse boards performance. Accordingly, a greater ratio of FIDs should be associated to lower board control and quality of monitoring due to lower attendance to meetings and a greater likelihood of intentional financial misreporting (Masulis et al., 2012). My hypothesis is that a nationality diversified board is more likely to allow excessive CEO compensation.
As to account for potential confounding effects, several control variables have been included, namely CEO duality and ratio of non-executive directors.
CEO duality is a dummy variable that indicates that the CEO is chairman of the board of directors, beside his executive responsibilities, corresponding to the value of 1; hence a value of 0 would imply that the CEO is not sitting on the board as the chairman of it. CEO duality is an exceedingly symbolic position that leaves the top executive with more discretion (e.g. Hambrick & Finkelstein, 1987; Patton & Baker, 1987; Harrison, Torres & Kukalis, 1988); one can assume that increasingly managerial discretion leads to an easier avoidance of the board (Hill & Phan, 1991). Unfortunately the databases show little variation in CEOs duality, which means that only few CEOs are chairman of board of directors. Moreover, the database presents a list showing a majority of CEOs who are not chairman of boards or missing observations, which is an unfortunate outcome. For that reason, I end up excluding CEO duality from the model, due to it being insignificant towards modelling the variables relation.
19 as not every non-executive is necessarily independent. However, I ultimately follow the lead of their paper, hence I include the variable to at least control for the impact of such independence.
Lastly, the control variables as implied in the paper by Boyd (1994) have been selected and measured as follows:
Size of companies has been measured as total amount of assets, which should rationally justify an increase of CEO's compensation based on the economic theory of greater complexity.
Profitability of firms has been evaluated as the Return on Equity (ROE), which follows the basic assumption of economic theory, namely the increase of financial performance that should reward CEOs with higher compensation levels.
Riskiness of companies has been measured as the Leverage ratio, namely the ratio of debt over asset, which should be in line with the basic theory that connects higher risk taken from firms to a higher CEO's remuneration.
An additional regression test the difference between American firms and the remaining companies. As a further test, I consider the possibility of endogeneity of the diversity variables; one could argue that diversity is affected by certain firm characteristics like size and performance. Lastly, the dataset is divided in quintiles to show the composition of the variables based on remuneration levels. 4. Discussion of results
4.1. Descriptive Statistics
20 15% is just slightly higher. However, both results are to be considered in line with comparable studies. Masulis et al. (2012) find on average 12.74% foreign directors between 1998 and 2006. Adams and Ferreira (2009) find 2% up to 16% women on boards across industries in 2003. Only 1% of the firms have a CEO that is chairman of the board at the same time. The average firm size in the sample, measured as totality of assets is 4.2 mlln €. As a matter of fact, one could argue that the sample is biased towards large firms, which are more likely to have female directors on the board (Catalyst, 2003; Adams & Ferreira, 2009). Nevertheless, the analysis of the variables (Table 6) present a different picture. On the one hand greater firms, that are part of the first quintile, have the lowest average ratio of gender diversity. On the other hand, the companies that are part of the last quintile show the greatest ratio of women on boards. Hence, the dataset does not show a biased population; it is well diversified and balanced. Lastly, 64% of the board members are NED. The comparably high percentage could be explained by the better understanding of governance by shareholders. The increasingly complex and dynamic corporate environment obliges shareholders to adapt to changes. CEO's compensation has increased over the last decades, as mentioned before, so that boards of directors struggle to cope with it. As shareholders cover the role of principals, they require boards to be more independent and less likely to collude with top executive. That could be the reason behind the vast presence of NEDs, that would act as the monitor entity of companies.
22 4.2. Regression analysis
This section proceeds to present the empirical evidence. In order to establish causality, the influence of the variables composing Table 1 on the total CEO-compensation has been analysed using an OLS approach. The following regression has been applied including controls:
𝐶𝐸𝑂 𝑐𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑖𝑜𝑛 = 𝑔𝑒𝑛𝑑𝑒𝑟𝑖+ 𝑛𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑖𝑡𝑦𝑖 + 𝛴 𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠𝑖+ 𝑢𝑖 (1) Table 3 summarizes the results. If I focus on panel (1), it is clear that without control variables both the independent variables are significant at the 1% level., Gender diversity, though, shows a significant positive impact and nationality diversity a highly significant positive impact. It can be concluded that this outcome is not to be taken seriously, due to the poor explaining power in the form of low R2 (0.03). However, these results change drastically when control variables are included.
a) Gender diversity analysis
23 b) Nationality diversity analysis
The second theoretical statement, which stated that nationality diversity ultimately increases CEO compensation, is supported. With or without controls, nationality diversity is highly significantly and positively correlated to CEO-compensation. Hence, the findings corroborate existing literature, meaning that an increase in nationality diversity does indeed lead to higher CEO-compensation (Masulis et al., 2012). One possible explanation is that the lower attendance rates at board meetings of foreign directors lead to more managerial discretion. Effective board control in the form of monitoring needs presence and involvement of the directors (Roberts et al., 2005). The fact that this result is robust when controlling for firm performance, size and riskiness indicates that higher nationality diversity does indeed lead to excessive compensation.
To substantiate my claim, I need to explore deeply the fact that each of the considered control variables is significantly and positively affecting CEO compensation. To this point, it is clear that the impact of greater nationality diversity is much more relevant than the control variables when I compare the coefficients of the regression analysis. If one would assess the impact of either nationality diversity or control variables, he would notice that CEO's compensation is increased by both. More specifically, however, nationality diversity’s coefficient is as great as the sum of firm size, performance and riskiness. That means that nationality diversity itself increases CEO's remuneration by 50%. Results indicate that larger boards indeed seem to have free-rider problems leading to insufficient board control (Lipton & Lorsch, 1992; Jensen, 1993; Yermack, 1996; Eisenberg, Sundgren & Wells, 1998). Therefore, the regression outcomes offer an explanation for the managerial hypothesis in spite of optimal contracting (Frydman & Jenter, 2010).
CEO-24 compensation. Positive relationship of firm size and remuneration is in line with previous research, as it indicates that CEOs are rewarded for firm complexity (Boyd, 1994). Higher performance leads to higher CEO-compensation, which is in line with the performance based compensation idea of agency theory (Jensen & Meckling, 1979). Lastly, it is worth noting that a higher ratio of NEDs increases CEO-compensation, which contradicts existing literature. Hence, board effectiveness is highly reduced, since outsiders are not executives or not even involved in the daily business. This consequently leads to a higher CEO dominance following the theory of the managerial power hypothesis. Nevertheless, an increase in women's presence as directors is the only way to truly enhance performance of boards as monitoring entity. Due to the fact that they are less likely to collude and meet social expectations of CEOs, as well as being a tougher and objective monitor, they ultimately reduce CEO's compensation to efficient levels.
As an additional robustness test, I check whether the model could be explained only by control variables. In that case, it would mean that CEO's compensation is efficient because affected only by firm specific financial variables. The result of panel 3 (Table 3) shows that financial variables do influence the remuneration. However, if I analyse the R2 (almost twice as low, compared to the full model), it is clear that this analysis is a poor prediction of the model. I would argue that it is required to include the two diversity measures to truly assess the drivers of CEOs compensation.
25 tends to reduce CEO's compensation, while nationality diversity enhances it, whereas every control variable increases remuneration level. Nationality diversity is still the major driver behind inefficient compensation while gender diversity is still the only way to contrast these remuneration schemes. The result of the second subsample (i.e. non USA countries) is slightly different, possibly being related to a different corporate structure. Nationality diversity is the major driver behind CEOs compensation, even though one can see how the coefficient is almost equal to the one of NEDs. This could mean that for the most part, foreign board members are also non-executive. Gender diversity is slightly less significant, but still very important to reduce remuneration. The greatest difference is the performance being insignificant in this analysis; this means that CEOs are not compensated based on firm performance. Overall, both subsamples are able to give a better explanation of the model, due to the fact that both R2 are higher than the first regression.
As far as causality is concerned, I want to assess whether gender and nationality diversity are truly an independent driver of CEOs compensation or whether they are influenced by other firm specific variables. This doubt arises because diversity may be a default value of firms, based on other variables such as size, performance or leverage (Catalyst, 2003; Adams & Ferreira, 2009). To check whether the previous regression analysis is sound, I proceed to run the following regression: gender and nationality diversity are the dependent variables, whereas size, performance and leverage are the independent variables.
26 The next step is to compare the results of the regression analysis with control groups, as presented in Table 6. It needs to be said, that this analysis does not give any insight on causality; the regression output delivers it. However, this test is to assess more specifically the composition of gender and nationality diversity. The data sample is divided in quintiles based on compensation level. The first quintile represents one fifth of the entire population showing the highest level of compensation, up until the last quintile representing the subsample showing the lowest level of remuneration. Few considerations are necessary. First of all, one can notice that both gender and nationality diversity follow a pattern in line with theory and expectation. In fact, I observe an increase in gender diversity alongside a decrease in nationality diversity, as the compensation level decreases. This seems to reflect the effect of women as tough monitors and the poor monitoring arising from a nationally diversified board; both have an influence over remuneration schemes, as mentioned before in the regression analysis. To be precise, the first quintile presents an average of 8% of women and 27% of foreigners in boards. At the other end of the spectrum, the last quintile shows an average of 28% of females and 8% of foreigners in boards.
27 5. Conclusion
The purpose of this thesis was to answer the research question of whether board diversity in terms of gender and nationality has an influence on the remuneration of CEOs. Using the managerial power hypothesis as background, I hypothesized that more diverse boards could have an impact on the power balance between the board and the CEO during the remuneration negotiations. Drawing upon past literature, I expected that higher gender diversity would lead to lower remuneration levels. The reason of such influence is that women have been found to be tougher monitors, breaking up collusion between board and CEO (Adams & Ferreira, 2009). On the contrary, I expected the nationality diversity to positively affect CEO compensation. Research states that the reasons are mainly two, both leading to free riding on CEO control (Masulis et al., 2012): firstly, an increase of managerial discretion and collusion between the two parties as foreign directors are less involved in boards duties and present in the board meeting; secondly, even in the case that foreign directors attend the meetings, cultural and social differences would imply heterogeneity and difficulty of cooperation as monitors.
28 complexity as amount of assets or risk taken in the form of higher level of leverage (Frydman & Jenter, 2010).
However, the results of the regression analysis are robust when controlling for past performance (RoE), firm complexity (total assets) and firm riskiness (leverage ratio). It appears that remuneration level are not efficient, due to the dominant power of CEOs, as well as poor governance and monitoring. Therefore, the research conducted in this thesis adds significant implications to the managerial power hypothesis thread of literature. Never before has a research tried and managed to test different sources of diversity in the same analysis, instead focusing on one type of diversity at a time. If one considers only one aspect of diversity, he might lose sight of the bigger picture, possibly resulting in poorly derived conclusions. Nationality diversity could partially explain how actually boards happen to be dominated by CEOs, instead of monitoring them on behalf of shareholders. The results show that a higher number of foreign directors increases diversity and consequently managerial discretion. I follow the commonly adopted argumentation of Masulis et al. (2012). Foreign directors need to exert more effort in terms of time and money in order to attend board meetings, which ultimately reduces their attendance rate. This consequently leads to free-riding conduct, due to less effective board debates and involvement. Both attendance rate and cooperation during board meetings have been found to be crucial elements of board efficiency and independence (Roberts et al., 2005). Additionally, I find significant and negative impact of gender diversity on CEO compensation. This result is line with previous research, which states that women on boards are tougher monitor, thus reducing the control of CEO over compensation negotiations (Adams & Ferreira, 2009).
29 influenced by firm characteristics, I run another regression (2); both diversity are the dependent variables while the control variables are the independent variables of the model. The results show that in no case firm characteristics predict board diversity, corroborating the goodness of the regression model (1).
As a further robustness test, I compare the results of the regression analysis with quintiles control group, sorted by level of CEOs compensation. The purpose is to check the structure of diversity, as well as the control variables on the different levels of remuneration. The comparison shows results in line with previous research and the hypothesis of this thesis, most importantly the composition of diversity. Gender diversity is inversely related to compensation levels, in in line with the hypothesis that having more women on boards reduces CEO's compensation. On the contrary, nationality diversity is positively related to remuneration level.
The bigger picture shows that really CEO tends to dominate the board of directors over remuneration negotiations. A results that shareholders should take into account whenever they opt to hire or fire directors to act as the monitor organ of the firm. They should strive to compose a balanced board, which means the presence of different members: gender, nationality, tenure, age, education level, ethnicity have a positive impact on the board’s duties. Especially gender diversity, as argued in this thesis, brings a fresh take on board processes and tougher monitoring that could ultimately reduce CEO's compensation to efficient level while maximising shareholder`s value.
6. Future research
30 Another important achievement would be to gather data and observation coming from different periods. Due to lack of time and availability of data, the thesis is structured as a cross section of firms observations. The benefits of that addition would be that longitudinal effects could be measured. Especially, there is the possibility that the observed effects of heterogeneity diminish over time as found by Barkema and Shvyrkov (2007). A potential trend that has been brought to the light is the possibility of gender diversity losing its effectiveness as time passes. One explanation to this hypothesis might be that boards end up falling back into group thinking structures that lead to inertia and inefficient monitoring. Said in other words, women on boards could become too alike to men and the positive effect on board monitoring ultimately diminishes.
Additionally, it would be interesting to test the model based on the implications that different country corporation settings imply. As far as concerns the United States, CEOs pay is expected to be based mostly on equity value and performance. With the presence of more multinational companies, the effect of higher managerial discretion due to higher nationality diversity could possibly be more pronounced. Given how huge is the gap confronting corporate settings of firms belonging to pretty similar countries like Europe and the United States, I can only imagine how many more information it would be possible to gather from the addition of either Asian or African firms.
Further research could even go deeply into analysing the effect of diversity over other important corporate variables. It would be useful to make the most of event studies such as the introduction of the female quota in Norway. In that case it would be possible to test the effect of gender diversity on efficient CEO compensation, but more importantly on firm performance or board independence (e.g. Ahern & Dittmar, 2012).
31 way, has been found to be important for board efficiency (Barkema & Shvyrkov, 2012).However, all these suggestion need to be tested and analysed in future research. Unfortunately such a heterogeneity analysis requires board data on the individual personal member level, which was beyond the scope and intention of this thesis.
7. Appendix
Table 1. Summary statistics. Total compensation and size are presented in ‘000 units and €. Size measured in total assets and performance in Return on equity.
mean standard deviation
32 Table 2. Correlation Matrix. Firm performance is measured by the ROE, firm Size is measured as total amount of assets
CEO Compensation
Gender Diversity
Nationality
33 Table 3. OLS-analysis of the influence on CEO-compensation. Dependent variable: total CEO-compensation. Standard errors are reported in parentheses. Newey-West standard errors have been applied to account for heterogeneity. P-values smaller than 0.01, 0.05 and 0.1 are indicated by ***,** and *, respectively.
Dependent variable: total CEO compensation
34 Table 4. Regression analysis based on country. Dependent variable: total CEO-compensation. Standard errors are reported in parentheses. Newey-West standard errors have been applied to account for heterogeneity. P-values smaller than 0.01, 0.05 and 0.1 are indicated by ***,** and *, respectively.
Dependent variable: total CEO compensation
USA Other Countries
35 Table 5. Causality test. Standard errors are reported in parentheses. Newey-West standard errors have been applied to account for heterogeneity. P-values smaller than 0.01, 0.05 and 0.1 are indicated by ***,** and *, respectively.
Dependent variable
36 Table 6. Control group comparison. Every coefficient is measured as the average of the quintile observation. Values of CEOs compensation and firm size (total asset) are expressed in ‘000 (thousands). ROE measure the profitability, size is the total amount of asset, leverage measure the riskiness.
Compens. Gend.
divers.
Nation.
divers. Size ROE Lever
age NEDs quintile 1 10,420.29 0.08 0.27 14,201.64 12.99 0.71 0.72 quintile 2 2,803.19 0.07 0.19 3,718.48 1.54 0.58 0.69 quintile 3 1,138.72 0.07 0.13 1,965.52 -19.66 1.20 0.66 quintile 4 408.34 0.09 0.10 584.51 -22.29 2.42 0.59 quintile 5 91.77 0.28 0.08 418.39 -37.05 2.91 0.54 Acknowledgment
37 Bibliography
Adams, R., Ferreira, D. 2007. A theory of friendly boards. The Journal of Finance, 62(1), 217-250.
Adams, R., Ferreira, D. 2009. Women in the boardroom and their impact on governance and performance. Journal of financial economics, 94(2), 291-309.
Ahern, K., Dittmar, A. 2012. The changing of the boards: The impact on firm valuation of mandated female board representation. Quarterly Journal of Economics
Bantel, K., Jackson, S. 1989. Top management and innovations in banking: Does the composition of the top team make a difference?. Strategic Management Journal, 10(S1), 107-124.
Barkema, H., Shvyrkov, O. 2007. Does top management team diversity promote or hamper foreign expansion?. Strategic Management Journal, 28(7), 663-680. Baysinger, B., Butler, H. 1985. Corporate governance and the board of directors: Performance effects of changes in board composition. Journal of Law, Economics, & Organization 1.1: 101-124.
Bebchuk, L., Fried, J. 2006. Pay without Performance Overview of the issues. The Academy of Management Perspectives, 20(1), 5-24.
Berle, A., Means, G. 1932. Modern Corporation and Private Property. Mac Millan, New York: NY.
Bertrand, M., Sendhil, M. 2001. Are CEOs rewarded for luck? The ones without principals are. Quarterly Journal of Economics: 901-932.
Boyd, B. 1994. Board control and CEO compensation. Strategic Management Journal, 15(5), 335-344.
Carpenter, M. 2002. The implications of strategy and social context for the relationship between top management team heterogeneity and firm performance. Strategic Management Journal, 23(3), 275-284.
Carrasco, V. 2005. Corporate Board Structure, Managerial Self-Dealing, and Common Agency. No. 04-015.
38
Catalyst. 2003. 2003 Catalyst Census of Women Board Directors of the Fortune 1000. Calayst: New York.
Deckop, J. 1988. Determinants of chief executive officer compensation. Industrial & Labor Relations Review, 41(2), 215-226.
Dess, G., Beard, D. 1984. Dimensions of organizational task environments. Administrative science quarterly, 52-73.
Dewatripont, M., Jewitt, I., Tirole, J. 1999. The economics of career concerns, part II: Application to missions and accountability of government agencies. The Review of Economic Studies, 66(1), 199-217.
Eisenberg, T., Sundgren, S., Wells, M. 1998. Larger board size and decreasing firm value in small firms. Journal of Financial Economics, 48(1), 35-54.
Erhardt, N., Werbel, J., Shrader, C. 2003. Board of director diversity and firm financial performance. Corporate governance: An international review, 11(2), 102-111.
Fama, E. 1980. Agency Problems and the Theory of the Firm. The Journal of Political Economy, 288-307.
Fama, E., Jensen, M. 1983. Separation of ownership and control. The Journal of Law & Economics, 26(2), 301-325.
Ferreira, D. 2010. Board diversity. In R. Anderson & H. K. Baker (Eds.), Corporate governance: A synthesis of theory, research, and practice: 225–242. Hoboken, NJ: John Wiley & Sons.
Ferreira, D. 2015. Board Diversity: Should We Trust Research to Inform Policy?. Corporate Governance: An International Review, 23(2), 108-111.
Finkelstein, S., Hambrick, D. 1989. Chief executive compensation: A study of the intersection of markets and political processes. Strategic Management Journal, 10(2), 121-134.
Francoeur, C., Labelle, R., Sinclair-Desgagne, B. 2008. Gender diversity in corporate governance and top management. Journal of business ethics, 81(1), 83-95.
39
Frydman, C., Jenter, D. 2010. CEO compensation (No. w16585). National Bureau of Economic Research.
Hambrick, D., Finkelstein, S. 1987. Managerial discretion: A bridge between polar views of organizational outcomes. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (Vol. 9, pp. 369–406). Greenwich, CT: JAI Press.
Harrison, D., Klein, K. 2007. What's the difference? Diversity constructs as separation, variety, or disparity in organizations. Academy of management review, 32(4), 1199-1228.
Harrison, J., Torres, D., Kukalis, S. 1988. The changing of the guard: Turnover and structural change in the top-management positions. Administrative Science Quarterly, 211-232.
Herman, E. 1981. Corporate control, corporate power (Vol. 98, p. 1). Cambridge: Cambridge University Press.
Hill, C., Phan, P. 1991. CEO tenure as a determinant of CEO pay. Academy of Management journal, 34(3), 707-717.
Hillman, A., Sabina, N. 2005. Diversity Of What? Multifaceted Boards, Directors, Monitoring And Resource Provision. Academy of Management Proceedings. Vol. 2015. No. 1. Academy of Management.
Jensen, M. 1993. The modern industrial revolution, exit, and the failure of internal control systems. the Journal of Finance, 48(3), 831-880.
Jensen, M., Meckling, W. 1979. Theory of the firm: Managerial behavior, agency costs, and ownership structure. In Economics social institutions (pp. 163-231). Springer Netherlands.
John, K., Knyazeva, A., Knyazeva, D. 2011. Does geography matter? Firm location and corporate payout policy. Journal of Financial Economics, 101(3), 533-551.
Kang, H., Cheng, M., Gray, S. 2007. Corporate governance and board composition: Diversity and independence of Australian boards. Corporate Governance: An International Review, 15(2), 194-207.
40
Kennedy, J. Kray, L. 2014. Who is willing to sacrifice ethical values for money and social status? Gender differences in reactions to ethical compromises. Social Psychology and Personality Science, 5, 52–59.
Kramer, V., Konrad, A., Erkut, S., Hooper, M. 2006. Critical mass on corporate boards: Why three or more women enhance governance (pp. 2-4). Boston: Wellesley Centers for Women.
Kuhnen, C., Zwiebel, J. 2008. Executive pay, hidden compensation and managerial entrenchment. Rock Center for Corporate Governance Working Paper, (16).
Lewellen, W., Huntsman, B. 1970. Managerial pay and corporate performance. The American Economic Review, 60(4), 710-720.
Liao, L., Le, L., Qingliang, T. 2015 Gender diversity, board independence, environmental committee and greenhouse gas disclosure. The British Accounting Review 47.4): 409-424.
Lilliefors, H. 1967. On the Kolmogorov-Smirnov test for normality with mean and variance unknown. Journal of the American Statistical Association, 62, 399-402. Lipton, M., Lorsch, J. 1992. A modest proposal for improved corporate governance. The business lawyer, 59-77.
Lorsch, J. W. 1989. Pawns or potentates. Harvard Business School.
Masulis, R., Cong, W., Fei, X. 2012. Globalizing the boardroom—The effects of foreign directors on corporate governance and firm performance. Journal of Accounting and Economics 53.3: 527-554.
Matsa, D., Miller, A. 2013. A female style in corporate leadership? Evidence from quotas. American Economic Journal: Applied Economics, 5, 136–169.
Miller, M. 2014. Use Cognitive Diversity to Get the Most Out of the Workplace. Emergenetics international.
Mizruchi, M. 1983. Who controls whom? An examination of the relation between management and boards of directors in large American corporations. Academy of management Review, 8(3), 426-435.
Moerland, P. 1995. Alternative disciplinary mechanisms in different corporate systems. Journal of Economic Behavior & Organization, 26(1), 17-34.
41
Roberts, J., McNulty, T., Stiles, P. 2005. Beyond agency conceptions of the work of the non-executive director: Creating accountability in the boardroom. British Journal of Management, 16(s1), S5-S26.
Ruigrok, W., Peck, S., Tacheva, S. 2007. Nationality and gender diversity on Swiss corporate boards. Corporate Governance: An International Review, 15(4), 546-557.
Sundaramurthy, C., Lewis, M. 2003. Control and collaboration: Paradoxes of governance. Academy of Management Review, 28(3), 397-415.
Thomas, A. 2004. Research skills for management studies. Psychology Press. Tosi Jr, H., Gomez-Mejia, L. 1989. The decoupling of CEO pay and performance: An agency theory perspective. Administrative Science Quarterly, 169-189.
Van Veen, K., Elbertsen, J. 2008. Governance regimes and nationality diversity in corporate boards: A comparative study of Germany, the Netherlands and the United Kingdom. Corporate governance: an international review, 16(5), 386-399. Van Veen, K., Wittek, R. 2016. Relational Signalling and the Rise of CEO Compensation:… It is Not Just About Money, It is About What the Money Says…. Long Range Planning.
Vance, S. 1983. Corporate leadership: Boards, directors, and strategy. McGraw-Hill Companies.
Walsh, J., Seward, J. 1990. On the efficiency of internal and external corporate control mechanisms. Academy of management review, 15(3), 421-458.
Westphal, J., Milton, L. 2000. How experience and network ties affect the influence of demographic minorities on corporate boards. Administrative Science Quarterly, 45(2), 366-398.
Westphal, J., Zajac, E. 1995. Who shall govern? CEO/board power, demographic similarity, and new director selection. Administrative science quarterly, 60-83. Wiersema, M., Bantel, K. 1992. Top management team demography and corporate strategic change. Academy of Management Journal, 35(1), 91-121.
