Master Thesis for IB&M University of Groningen Supervisor: Dr. C.H. Slager June 18th, 2018
Determinants of CEO-Employee Pay
Ratios: New Evidence After the
Implementation of the Dodd-Frank Act
Huisman, Jaap J., s2549700
Abstract
The top 1% earners have increasingly been able to expand their gains disproportionately compared to the average bottom 99%. This study explores this phenomenon by examining the drivers of income of the top 1% in the corporate setting. More specifically, this paper investigates the determinants of CEO-employee pay ratio at the corporate governance level. Additionally, it provides brand new data on US CEO-employee pay ratios after the implementation of the Dodd-Frank Act of 2010 requiring companies to disclose CEO to median employee pay ratio from 2017 onwards. This paper tries to answer the following question: what are the determinants of CEO-employee pay ratios within publicly listed firms? The study draws on a hierarchical multiple regression analysis to examine the determinants of pay ratios of firms included in the S&P 500. The novelty of this study lies in extending the limited pay ratio literature with new US data and the link with income inequality. The research leads to the following findings: CEO as a (co-)founder and unionization of employees reduces pay ratios, while director compensation increases pay ratios.
Table of Contents
1. Introduction 3
2. Literature review 5
2.1. Executive compensation 5
2.2. Pay ratio literature 8
2.3. SEC: the new rule 10
2.4. Hypothesis development 10 2.5. Conceptual model 14 3. Methodology 16 4. Results 20 4.1. Characteristics 20 4.2. Procedures 22 4.3. Regression results 23 4.4. Additional testing 26 5. Discussion 28 5.1. The results 28
5.2. The measure of pay ratios 31
5.3. Pay ratios and the link with income inequality 32
6. Conclusion 34
6.1. Summary 34
6.2. Theoretical contributions 34
6.3. Practical implications 35
6.4. Limitations 35
6.5. Recommendations for future research 37
7. References 40
8. Appendix 45
1. INTRODUCTION
Income inequality within countries has been rapidly increasing since the late 1970s and after the recent global crisis of 2008 hefty bonuses for top executives were severely criticized. According to a report by Oxfam (2018) 82% of the wealth created in 2017 went to the top 1% wealthiest of the world, while the poorest 50% did not increase their wealth. Top executives and CEOs are often considered to be part of the top 1%, whereas the average employee will be in the other 99%. Executive compensation in the S&P 500 has seen a six fold increase since 1980 (Edmans, Gabaix and Jenter, 2017). The increasing income inequality has become a major concern for the political and economic sphere, but has been somewhat neglected in the corporate sphere. While executive compensation has been vastly researched, the CEO-employee pay ratio been insufficiently examined despite recent interest.
Furthermore, governments all across the globe have increasingly started to regulate transparency regarding executive pay and pay ratios. In the US, the SEC, or Securities and Exchange Commission, as of 2017 requires companies to disclose the CEO pay, the median employee pay and the CEO-median employee pay ratio. This is likely to cause a wave of controversy as CEO pay ratio is expected to far exceed the 95:1 found by Faleye et al. (2013) due to self-selection bias. More accurate might be the findings of the American Federation of Labour and Congress of Industrial Organizations, who state a ratio of 347:1 in S&P 500 firms in 20161. The China Securities Regulations Commission issued several revised regulations requiring more transparency in top pay disclosures in 2001, 2005 and 2007. Korean regulators have already required listed firms to publish pay ratios since 2000. These regulatory differences over time and across countries are determined by the political and institutional environment. In response to the rapid rise of executive pay and growing income inequality the political and institutional environment across the globe appears to increasingly require more disclosure of executive compensation by publicly listed companies. Adding more pressure on the compensation levels of executives in companies. A prime example is the reform of the Dodd-Frank Act by the Securities and Exchange Commission (SEC) in the US in 2010 requiring firms to disclose CEO-employee pay ratios from 2017 onwards. These will be published from 2018 onwards by the publicly listed firms in the US. This has spurred
the UK government to begin talks on regulation requiring pay ratio disclosure.
When examining income inequality and executive compensation levels, it is important to place these levels in the right contextual and historical perspective. The economists Piketty and Saez (2003) examined the levels of income inequality over the course of the 20th century and found that income inequality has risen dramatically since the 1970s. Additionally, they compared CEO compensation to the average income of workers and found that since the 1970s and 1980s CEOs compensation has risen disproportionately compared to the average income. This reflects the same pattern observed for the top incomes in the economy. Frydman and Saks (2010) examined the development of executive compensation and found similar results as Piketty and Saez (2003). The patterns observed for both executive compensation and income inequality seem to reflect a homogenous development over time. Piketty and Saez (2003) note that income structures for the very top earners have increasingly been shifting away from wage incomes and towards capital gains. This shift seems to more or less coincide with changes in the structuring of executive compensation towards more pay for performance compensation following the implementation and anticipation of SEC 162(m) US regulation in 1993 (Goolsbee 2000; Perry and Zenner, 2001).
various factors including firm size and industry. Firm performance was found to decrease pay ratios, although this finding proved insignificant after controlling for industry effects. This paper proceeds with a review of the executive compensation literature, followed by a review of the pay ratio literature. Thereafter, the hypothesis and conceptual model are proposed, then procedures are discussed in the methodology and analysed in the results section. This paper closes with a discussion and conclusion that includes the limitations of this study and possibilities for future research.
2. LITERATURE REVIEW
2.1. Executive compensation
Executive compensation is the total pay that members of the top management team receive. That includes the main executives, such as the CEO. Executive compensation has gathered increasing interest from academics, since compensation levels have skyrocketed over the last few decades. There are two main theoretical views in the literature of executive compensation (Bebchuk and Fried, 2003). The first is optimal contracting reflective of agency theory, where executive pay is structured through incentives in order to maximize shareholder value. Edmans, Gabaix and Jenter (2017) dub it as the shareholder value view. In this view agency problems might occur as managers may be inclined to act according to their own interest rather than to the interests of shareholders. To minimize the agency costs the shareholders structure pay to increase their profits, and incentivize managers to do so by tying their pay to performance, or through stock options. However, conflicts of interests also create problems as CEOs often appoint board members, who in turn determine the compensation of the CEO. Thus, the board members have little incentive to hassle over the pay of the CEO.
in simpler terms, to use his/her position to increase his/her own compensation. The ability of CEOs to use their managerial power to increase its own pay is influenced by several factors. Firstly, it is dependent on the ownership structure of the firm. It is argued that in firms with large block holders, the compensation of the CEO is reduced (Dyl, 1988; Hambrick and Finkelstein, 1995; David, Kochhar and Levitas, 1998). Shareholders are then more likely to actively engage in monitoring behaviour and can exert more pressure on CEOs.
In short, the main difference between the two views is that the shareholder value view argues that executive compensation is decided by shareholders, while the managerial power view argues that executive compensation is decided by executives themselves (Bebchuk and Fried, 2003; Edmans et al., 2017). Edmans, Gabaix and Jenter (2017) argue that these two views are drivers of executive compensation, but they add a third driver namely an institutional perspective that includes drivers arising from practices in the institutional environment. This includes taxation, regulation and accounting policy. This research focuses on managerial power of executives as a driver of executive compensation, yet it does not dismiss the other two as drivers of executive pay.
Another explanation might be the change in the structure of executive compensation from cash towards more equity based compensation from the 1970s onwards. The majority of compensation in the US in the 1970s was cash, whereas it was options in the 1980-1990s and performance stocks in the 2000s (Edmans, Gabaix and Jenter, 2017). Currently in the US, the majority of firms’ executive compensation structures have been tied to performance through equity based and non-equity based compensation, apart from the base salary. The base salary currently often consists of less than 20% of the total compensation. According to Edmans et al. (2017) the causes of this structural change in compensation have not yet been completely grasped. Another more obscure form of executive compensation is ‘stealth pay’, where executives are compensated through more invisible structures such as retirement benefits, severance pay, perks and deferrals (Bebchuk and Fried, 2004b). There are even comprehensive compensation structures in case the CEO is terminated from his position, including life insurance policies. Therefore, full transparency in the structure and total compensation is necessary for shareholders to assess whether their executives are acting in their interest or in the interest of themselves. Since 2006 the SEC requires companies to disclose compensation in the form of pensions. Pension related bonuses are found to jump 17-33% prior to pension freezes and retirements, but not in firms with strong corporate governance (Stefanescu, Wang, Xie and Yang, 2018) indicating that pension benefits are prone to exploitation by managerial power.
the rising pay of CEOs in a few companies can trigger an increase in pay of CEOs of other corporations due to the benchmarking process of determining CEO pay. In an increasingly globalized world, benchmarking is likely to lead to a convergence towards the highest international standard of CEO pay, which are seen in the US (Schmid, Altfeld and Dauth, 2018). In order to avoid CEO and executive hijacking, corporations will increase compensation towards American levels when they are more involved with American CEOs, corporations and industry peers. (Schmid et al., 2018). Thus, benchmarking leads to a rise in executive compensation at the top end of the pay ratio distribution.
2.2. Pay Ratio Literature
The literature on executive compensation is vast. One underdeveloped theme within the executive compensation is the pay ratio literature. However, there are various pay ratios. The existing literature might confuse the quick reader by the different terminology used to examine various ratios regarding pay of executives. Therefore, this study makes a clear distinction in the various forms of pay ratios. To be clear, this study revolves around the CEO-employee pay ratio. Other studies have used the terminology ‘pay ratio’ for examining an executive-employee pay ratio and the pay ratio between the CEO and other executives associated with the CEO pay premium (Henderson and Fredrickson, 2001; Bebchuk, Cremers, and Peyer, 2011). The latter is sometimes called the CEO pay slice, although some still use the terminology ‘pay ratio’ for this concept. For accuracy and consistency the term ‘pay ratio’ should only be associated with one pay ratio, namely the CEO-employee pay ratio. Other ratios should be treated differently and therefore have different and consistent terminology to prevent confusion. The CEO-employee pay ratio branch of the executive compensation literature is somewhat underdeveloped and further understanding is needed. This paper will try to broaden the theoretical scope of executive compensation literature by examining the determinants of CEO to employee pay ratios in the US, by examining pay ratio measures and providing new pay ratio data after the implementation of the SEC Dodd-Frank Act in 2017.
rank is necessary for productivity and even incentivizes it (Lazear and Rosen, 1981; Rosen, 1986). In tournament theory small differences between individuals can lead to large discrepancies in reward (Connelly, Tihanyi, Crook and Gangloff, 2014). This win-or-lose structure - hence tournament theory - is argued to function as a motivating system in order to boost productivity. However, critics have attempted to debunk the theory. Dai, Kong and Xu (2017) have suggested that a pay gap is beneficial for productivity, but an exorbitant gap is harmful. Additionally, the success of an individual in tournament theory is not only dependent on the individuals’ efforts, but also on random factors that are uncontrollable by the individual and on sheer dumb luck (Connelly et al., 2014). A CEO, the winner, has no other incentive left, yet he or she still has a job to fulfil. Therefore, exuberant rewards are necessary to satisfy the needs of the winner, demystifying the argument of productivity as the major driver of pay gaps. Furthermore, tournament theory neglects other important factors of motivation and job satisfaction (Connelly et al., 2014). This approach is somewhat in line with the shareholder value view in that it attempts to motivate through compensation which is set by shareholders. Both theories view compensation as a remedy for agency problems.
participation of employees leads to lower pay ratios.
2.3. SEC: The new rule
The Securities and Exchange Commission issued the Dodd-Frank Wall Street Reform and Consumer Protection Act in 2010 requiring firms to disclose pay ratios in their annual reports or proxy statements. The law was a response to the global financial crisis of 2008. The Dodd-Frank Act came into effect in 2017, thus corporations shall be publishing their pay ratios in 2018. The Act eliminates the self-selection bias as reported in Faleye, Reis, and Venkateswaran (2013) who found a CEO-employee pay ratio of 95:1 in US firms that voluntarily disclosed pay ratio information. Previously, US firms were not required to disclose pay ratios by law and therefore the companies that disclose the pay ratios were likely to be in the lower bound of the pay ratio distribution. They argue that since the data suffers from self-selection bias since the actual number is deemed far higher than the 95:1 ratio found (Faleye et al. 2013). This study will provide new and more accurate data on the actual pay ratio within the US that are listed on the S&P 500. The pay ratio is based on the calculation of CEO pay to median worker pay, as opposed to an average employee calculation method. For any ratio the robustness of the calculation is crucial. The pay ratio is no exception. The SEC requires companies to publish pay ratios calculated as the ratio of CEO total compensation to the total compensation of the median worker. They include all the temporary workers (including part-timer workers, seasonal workers, and ad hoc workers) and fully employed workers of the companies’ global workforce. Corporations are however, allowed to exclude less than 5% of foreign employees as under SEC ruling.
2.4. Hypothesis development
issues associated with it. The compensation the CEO receives is seen in the shareholder value view as a measure to control these agency problems, while the managerial power approach views the compensation as a source of agency problems (Bebchuk and Fried, 2004). In order to reduce agency problems shareholders have a monitoring role towards the managers that are in the executive branch of management of the firm. It is argued that in firms with large blockholders, the compensation of the CEO is reduced (Dyl, 1988; Hambrick and Finkelstein, 1995; David, Kochhar and Levitas, 1998). Shareholders are then more likely to actively engage in monitoring behaviour and can exert more pressure on CEOs. Hartzell and Starks (2003) also find that institutional ownership concentration lowers executive pay and is associated with more performance enhancing structures of pay. Concentration of ownership implies an increased ability to monitor the principal-agent issues associated with the CEO-owner relationship, by the shareholders. Subsequently, it is expected that:
Hypothesis 1a: The presence of blockholders is negatively associated with pay ratios
A specific case is the CEO as a founder, or a family member of the founders. This puts CEOs in a unique position since they act as the executive branch of the firm, but are also considered as a shareholder, who normally have a monitoring role of the executive branch. This constitutes in greater managerial power for the CEO, which is associated with higher compensation levels (Bebchuk et al., 2002; Bebchuk and Fried, 2004). The CEO then has long-term ties with the board and can use that relationship in order to influence board decisions and obtain higher compensation levels. However, Gomez-Mejia, Lazzara-Kintara and Makri (2003) find that CEOs related to the owners receive less pay than CEOs who are not related to the owners. They attribute this to the observation that related CEOs trade a lower compensation level for having a higher job certainty. Consequently, this study expects the following:
Hypothesis 1b: CEO as a (co-)founder is positively associated with pay ratios
theory. They find a positive relation to pay ratios for firm size, performance and, CEO-chairman duality. However, a negative relation for employee unionization and physical capital intensity is observed. Their findings suggest that indicators reflective of larger CEO bargaining power lead to higher pay ratios, whereas indicators reflecting larger employee bargaining power leads to lower pay ratios (Faleye et al., 2013). The bargaining power of the CEO is an outcome of managerial power. The managerial power of the CEO is the CEO’s power relative to the board. Therefore, board composition is important for the levels of pay. As noted by Finkelstein (1992) part of the CEO’s managerial power is determined by the structural power given to the CEO by the position it holds within the firm and board. According to agency theory, CEO-Chairman duality enhances the CEOs ability to grow his/her influence fostering entrenchment of the CEO in the company and the board. Thereby abating the effective ability of the board to monitor the CEO. Fernandes et al. (2012) argue that in the US, the corporate governance system is different from other countries. The US has specific conditions that influence the pay of the CEO, that are not found in other countries. In the US, CEO-Chairman duality is found to increase CEO pay (Fernandes, et al., 2012). Thus, CEO-Chairman duality in the managerial power approach improves the influence a CEO has and thereby increases the CEO’s ability to increase his total compensation. Therefore, this study anticipates the following hypothesis:
Hypothesis 2: CEO-Chairman duality is positively associated with pay ratios
the CEO has influence over the compensation board members receive, they are in turn more likely to accept higher compensation of the CEO. This phenomenon is dubbed as ‘mutual back scratching’ or as Brick, Palmon and Wald (2006) call it ‘cronyism’. They find a strong relationship between the compensation of board directors and that of the CEO. Their evidence is backed by Li and Roberts (2017). In the US, a high ratio of independent directors on the board is found to increase CEO pay (Fernandes et al., 2012). Following the above discussion, this paper expects the following:
Hypothesis 3a: Board independence is positively associated with pay ratios
Hypothesis 3b: Director compensation is positively associated with pay ratios
the board is less likely to approve the proposed level of pay (Bebchuk and Fried, 2004a). Outrage costs are a feature of the pressure that stakeholders can exert on corporations and executives, creating a downward pressure which limits executive pay. Therefore, this paper proposes the following hypothesis:
Hypothesis 4: Outrage costs are negatively associated with pay ratio
At the other end of the pay ratio scale are the employees. The relative power of employees vis-à-vis management is best represented by labour unions. Labour unions are found to be a likely force constrain the bargaining power of CEOs, thereby limiting their ability to increase their own compensation, while increasing both bargaining power and compensation of employees. The decreasing trend of labour union strength over the last few decades is likely to cause an increase in income concentration at the top (Piketty and Saez, 2006; Western and Rosenfeld, 2011; Lin and Tomaskovic-Devey, 2013). The strength of labour unions varies across countries since different legal systems and institutional structures determine the bargaining power of labour unions. A common example is the high labour representation in Germany compared to the low representation of labour in the US. Labour unions are found to have negative impact on CEO compensation (Huang, Jiang, Lie and Que, 2017). Power structures determine how much managerial power and bargaining power CEOs and employees have. Shin (2014) found that bargaining power of labour reduces the pay dispersion between executive and employee. The bargaining power of employees is mainly determined by the presence and strength of labour unions. The bargaining power of CEOs grows when labour unions are absent or severely limited in their strength. Faleye et al. (2013) find that the pay gap between CEO and employee is largely dependent on the balance of bargaining power of the CEO versus the bargaining power of the employee. They also find that when employees are unionized their bargaining power is larger thus resulting in a lower pay gap. Therefore, this paper argues the following:
2.5. Conceptual model
3. METHODOLOGY
This study employs a quantitative approach using a multiple regression analysis to examine the determinants of CEO-employee pay ratios in the US. The sample consists of publicly listed active companies on the S&P 500. These are the 500 large companies in the US based on market capitalization. Data is collected on the S&P 500 from the ExecuComp database. Data on CEO compensation and median compensation for the pay ratios are collected manually from DEF14A proxy statements available as of May 22nd 2018 for the fiscal year 2017. Data for the control variables is gathered from the Orbis database. Data on CEO as a (co-)founder and CEO-Chairman duality is gathered from ExecuComp and based on the executive’s title stated in ‘Annual title’. Both are included as a dummy variable. The CEO is determined as ‘Annual CEO Flag’ from the ExecuComp database. The number of board members and independent board members, along with the compensation per non-employee director, have been manually collected via the Spencer Stuart US Board Index 2017 from company DEF14A proxy statements. Data on unionization of employees was collected from the US Bureau of Labour Statistics. The LexisNexis Academic database provided the data regarding outrage costs. Table 1 provides a clear overview of the data, its measurement and of the data sources.
The independent variables are ownership, CEO-Chairman duality, board composition,
outrage costs, and unionization of employees. Ownership is measured as two independent
variables: blockholders measured as the number of shareholders that own a minimum of 5% shares and CEO is )founder measured as a dummy variable where 0 = CEO is not a (co-)founder and 1 = CEO is a (co-(co-)founder of the firm. CEO-Chairman duality is measured as a dummy variable, where 0 = CEO is not also Chairman of the board, while 1 = CEO is also Chairman of the board. Board composition is also measured twofold: board independence, which is determined as the ratio of the number of independent board members to the total number of board members, and director compensation measured as the average compensation received by the independent, or non-employee directors. These independent variables reflect the managerial power of the CEO.
The following independent variables are reflective of external power. Unionization of
employees is measured as the sectoral percentage of employees that are members of labour
Company etc.) and the names were double-checked to prevent errors. The name of the corporation is a more reliable method of searching than the name of the CEO, since the CEO changes more frequently than the corporate name. Therefore, this research has focused on the name of the corporation rather than the name of the CEO. However, in cases of mergers and joint ventures name changes might have occurred. Additional checks have been conducted to secure more robustness and prevent biases in the data collection. Furthermore, in-depth examination within the articles were conducted when the name of the company is also a common word in our everyday vocabulary (f.e. Gap resulted in 48 hits, but only 1 involved the company). When no articles were found in relation to the firm name and executive compensation a zero was included in the dataset, because it does not reflect a missing value. It merely implies that there has been no or limited outrage costs.
A number of variables have been included as control variables. Industry type to control for industry effects computed as a dummy variable on the classification provided by the Orbis database. The variable firm size controls for firm size effects (Gabaix and Landier, 2008), which is measured here as the total number of employees. CEO age will also be included as a control variable since Zhang, Guo and Hu (2017) found a positive relationship between CEO age and executive-employee pay ratios. In India CEO age was also found to positively influence executive compensation (Ramaswamy, Veliyath and Gomes, 2000; Edmans et al., 2017). Firm performance also has been found to influence executive compensation in
multiple countries (Conyon and He, 2011). In this study firm performance is measured as Net Income. Additionally, this study controls for firm leverage following Faleye et al. (2013) and Shin et al. (2015), which is calculated as the total long-term debt divided by the total assets.
Tobin’s q is often used as a control variable in the executive compensation literature and is
outliers in the analysis. At the lower bound of the pay ratio CEOs can voluntarily decide to reduce their pay, thereby reducing the firm’s pay ratio significantly. Additional tests for robustness purposes regarding outliers have been conducted and are discussed later on. Table 1 below provides a clear overview of the variables and their measurements.
Table 1
Overview Variables
Variables Measurement Log Inclusion Data Source
Pay Ratio CEO total pay divided by median
employee total pay
Yes Proxy statement
Blockholders Number of shareholders with >5% shares
No Orbis
(co-)Founder Dummy No ExecuComp
Chairman Duality Dummy No ExecuComp
Board Independence Ratio of independent board members
No Proxy statement*
Director Compensation Average pay of non-employee directors
Yes Proxy statement*
Outrage Costs Number of articles with firm name and executive compensation
No LexisNexis Academic
Unionization Percentage of sectoral union membership
No US BLS**
CEO Age Age of CEO No ExecuComp
Firm Size Number of employees Yes Orbis
Firm Performance Net income Yes Orbis
Firm Leverage Long-term debt divided by total assets
No ExecuComp
Tobin’s q Market value divided by total assets
No ExecuComp
4. RESULTS
4.1. Characteristics
Table 2
Paired samples t-test results
Variable and group Mean Std.
Deviation
Std. Error Mean
Number of observations
Pair 1 CEO Total Compensation 2017 – Dummy
13,785,008 9,192,478 414,850 491
Pair 2 CEO Total Compensation 2017 - Pay Ratio
13,531,696 7,731,754 388,535 396
Pair 3 Tobin’s q – Dummy 1.471 1.378 .062 490
Pair 4 Tobin’s q – Pay ratio 1.386 1.409 .071 395
Pair 5 (co-)Founder - Dummy .04 .203 .009 491
Pair 6 (co-)Founder – Pay ratio .05 .214 .011 396
Table 3
Descriptive of variables
Note: *The descriptive of these variables are taken as the natural log of those variables.
**This column shows the skewness of the variables after natural log transformation, if applicable.
4.2. Procedures
Due to skewness issues, log transformation has been conducted on most variables with skewness issues. CEO is (co-)Founder has plausible skewness problems. However, since CEO is (co-)founder is measured as a dummy variable, a log transformation was not constructed. Outrage costs was not included as a natural log of outrage costs, because the sample size would have been reduced to proportions that are close to violating the minimum rule of thumb for a multiple regression analysis. Skewness in the measurement of outrage costs are also to be expected considering the nature of the data. The possibility of no outrage costs allows for skewness in the distribution. Table 4 below depicts the correlation matrix of the variables included. This allows for an overview of possible interrelationships between variables, and thus to identify any plausible multi-collinearity problems.
Variables N Minimum Maximum Mean Std. Deviation Skewness**
Table 4 Correlation matrix 1 2 3 4 5 6 7 8 9 10 11 12 13 1.Pay ratio 1 2.Blockholders 0.056 1 3.(co-)Founder -0.076 0.059 1 4.Chairman duality 0.075 -0.022 .103* 1 5.Board independence 0.054 -0.050 -.121** .214** 1 6.Director compensation .270** 0.078 .137** -0.028 .093* 1 7.Outrage costs .173** -0.050 -0.040 .119** 0.086 .126** 1 8.Unionization -.202** -.092* -.118** 0.028 .111* 0.020 -0.067 1 9.CEO age 0.035 -0.051 0.045 .341** 0.032 -.148** 0.049 -0.073 1 10.Firm size .536** -0.075 -0.010 .146** 0.069 .105* .367** 0.016 0.051 1 11.Firm performance .136* -.153** -0.045 .204** 0.087 .123* .508** 0.012 0.066 .536** 1 12.Firm leverage 0.074 0.057 -0.042 -0.046 0.025 -0.078 -0.045 .138** 0.004 -0.083 -.108* 1 13.Tobin's q 0.084 0.010 .236** -0.036 -.168** .184** -0.089 -0.026 -0.082 -0.012 -.103* 0.007 1 Note: *Significant at the 5% level (2-tailed)
**Significant at the 1% level (2-tailed)
4.3. Regression results
Table 5
Regression coefficients on CEO-employee pay ratio
Note: The dependent variable Pay Ratio is measured as the natural log of the Pay Ratio. Blockholders is measured as the number of shareholders that have a minimum of 5% total shares. (co-)Founder and Chairman duality and are both measured as a dummy variable. Board independence is calculated as the ratio of independent board members. Director compensation is measured as the natural log of the average compensation received per non-employee director. Outrage costs are measured as the number of articles in which the firm name is mentioned in combination with executive compensation. Unionization of
Independent Variables Model 1 Model 2 Model 3
(Constant) -.326 -1.002 -1.601 (1.061) (.914) (.943)* Blockholders .012 .009 .006 (.011) (.010) (.010) (co-)Founder -.289 -.275 -.289 (.094)*** (.080)*** (.080)*** Chairman Duality .089 .048 .040 (.043)** (.038) (.039) Board independence .000 -.000 .001 (.003) (.002) (.002) Director compensation .492 .448 .568 (.193)** (.167)*** (.171)*** Outrage costs .005 -.003 -.001 (.003)* (.003) (.003) Unionization of workers -.020 -.022 -.020 (.004)*** (.003)*** (.004)*** CEO age .004 .004 (.003) (.003) Firm size .342 .253 (.030)*** (.036)*** Firm performance -.094 -.066 (.045)** (.045) Firm leverage .306 .193 (.091)*** (.097)** Tobin’s q .009 -.011 (.012) (.013)
Industry dummy No No Yes
R-Squared .141*** .432*** .504***
Adjusted R-Squared .120*** .409*** .456***
F-Statistic 6.944*** 18.496*** 10.440***
workers is measured as the percentage of workers in the industry who are members of a labour union. Firm size is measured as the natural log of the number of employees. Firm performance is measured as the natural log of net income of the firm. Firm leverage is calculated as total long term debt divided by total assets. Tobin’s q is calculated as the market value of the firm divided by total assets. The unstandardized coefficients are reported in the results of this table. The numbers within the parentheses are the standard errors of the coefficients.
*Significance level at 10% **Significance level at 5% ***Significance level at 1%
The hierarchical multiple regression analysis has led to the following results, as can be seen in table 5. Firstly, model 1 shows the results of the independent variables, without the inclusion of the control variables, regressed on pay ratios. In this scenario, there is a significant negative relationship between (co-)founder and pay ratios (B = -.289, p < .001). This entails that when the CEO is also the (co-)founder of the firm, pay ratios are 2,89% lower than when the CEO is not a (co-)founder of the firm. There is also a significant positive relationship between chairman duality and pay ratios (B = .089, p < .05). This means that when the CEO of the firm is also the chairman on the board of directors, pay ratios are higher than when the CEO is not also chairman. The results suggest a significant positive relationship between director compensation and pay ratios (B = .492, p < .05). This indicates that an increase in non-employee director compensation leads to an increase in pay ratios. Outrage costs were found to be positively related with pay ratios at the 10% significance level (B = .005, p=.071). This seems to suggest that an increase in outrage costs leads to an increase in pay ratios. Unionization is found to have a significant negative relationship with pay ratios (B = -.020, p < .001), thereby implying that an increase in labour unions membership by employees is accompanied by decreasing pay ratios. Blockholders and board independence were positively but insignificantly related to pay ratios.
(B = 0.003, p = .204). The control variables themselves have led to a number of relationships. The relationship between CEO age and pay ratios is positive but insignificant (B= .004, p= .204). Firm size is found to be significantly positive in relation to pay ratios (B = .342, p < .001). This means that an increase in firm size is associated with an increase in pay ratios. The relationship between firm performance and pay ratios is found to be negative and significant (B = -.094, p < .05). Therefore, an increase in firm performance leads to a decrease in pay ratios. Firm leverage is found to be positively and significantly related to pay ratios (B = .306, p < .001). Thus, an increase in firm leverage is associated with an increase in pay ratios. Tobin’s q is positively, but insignificantly, related to pay ratios (B = .009).
Thirdly, model 3 shows the results of the independent and control variables, including a control for industry regressed on the dependent variable pay ratios. Once more, only the major changes will be discussed as many results remain similar to the previous model. The relationship between director compensation and pay ratios has remained positive and significant, yet the coefficient has increased after the addition of the industry control (B = .568) in model 3 as opposed to (B = .448) in model 2. The coefficient for firm size dropped from (B = .342) in model 2 to (B = .253) in model 3, yet the significance of the positive relationship between firm size remains unchanged (p < .001). The relationship between firm performance and pay ratios remains negative, but it is no longer significant. Firm leverage remains positively related to pay ratios, but it has dropped in coefficient and significance (B = .193, p < .05). The relationship between Tobin’s q and pay ratios switched from positive to negative after the addition of industry controls, yet it remains insignificant.
4.4. Additional testing
although this is to be expected. Further investigation through the variance inflation factor has not delivered any problematic cases. All variables had values below 2.5, which is significantly below the threshold of 5. Therefore, no multi-collinearity issues have arisen.
In order to verify whether the assumption of independent errors holds, a Durbin-Watson test was conducted. This test examines the serial correlations among errors with a value of 2, which would mean that there is no correlation among the errors. This test provided very decisive results (t=1.952). The assumption of independent errors in the multiple regression analysis therefore holds. Multiple tests were conducted to determine whether there were significant outliers or influential cases. A robustness check was run by re-running the multiple regression analysis after deleting possible outliers. The outliers were identified by investigating the studentized deleted residuals that were 3 times or more greater than standard deviations. The results of the re-run multiple regression analysis led to the same results as the multiple regression that included all cases (R2 = .549, F(27,273) = 12.328, p < .001). A test on leverage points was also run and all cases with a leverage value of 4 or higher were excluded in the following multiple regression. Across all three models the results remained statistically significant and highly similar to the results from the regression included in this paper (R2 = .501, F(25,275) = 11.037, p < .001). To examine whether there are influential points a Cook’s distance test was run. No cases were found to be above the influential threshold of 1. Re-running the regression was superfluous.
majority of the corporations that had outrage costs in the previous method only had a larger number of mentions in world newspapers. For corporations that were not mentioned in US newspapers, but were mentioned in world newspapers, the number of articles in which they were mentioned remained low. Separate multiple regression analyses were conducted on both methods of data collection for outrage costs. The results were highly similar for all variables and model fit with the exception of outrage costs in the second method (R2 = .504, F(27,277) = 10.436). This proved not significant at the 10% level in model 1 without the control variables. Due to the unchanged results regarding the analysis and the fact that this study focuses on the corporations listed in the US-based S&P 500, the first method was chosen for this regression, i.e. data on outrage costs was collected from major US newspapers. For the second method of outrage costs in model 1 the coefficient was also positive (B = .003), followed by a negative coefficient in model 2 (B = -.002) and in model 3 (B = -.001), yet the relationship between outrage costs and pay ratios were insignificant in all three models.
5. DISCUSSION
5.1. The results
lower pay ratios for (co-)founder CEOs as opposed to non-founder CEOs is likely to be attributable to a higher sense of responsibility for employees and more outside pressure, but above all, the corresponding wealth associated with founding a successful firm. The sense of responsibility arises from the (co-)founder maintaining tighter relationships with employees through a shared history. When the CEO is a (co-)founder the reputation and name of the CEO are often associated with the name and reputation of large firms in the public eye. Subsequently, the CEO is often under increased scrutiny creating outside pressures from the public and shareholders. ‘It’s personal’ could be a plain way of explaining these pressures. Most important, however, is the associated wealth that founders often have when founding a large successful corporation. The names of these CEOs are frequently included in lists of the richest individuals in the US or the world. These CEOs have sometimes amassed billions of US dollars in wealth over time, thereby reducing the need for higher compensation levels. Thus, resulting in voluntary pay cuts leading to lower pay ratios.
The structural power associated with the CEO as described by Finkelstein (1991) through CEO-Chairman duality has not been fully supported in this study. The hypothesis that chairman duality is associated with higher pay ratios has found some support. However, the findings failed to hold after controlling for firm size, performance, leverage and industry. This seems to contradict the findings of Fernandes et al. (2012) that chairman duality increases CEO pay. An explanation could be that chairman duality is somewhat relevant for CEO pay, but not for pay ratios.
with CEO compensation and pay ratios in the S&P 500 corporations, since all are required to have a majority of independent board members, thus limiting the differences within board compositions across the S&P 500. The other board composition hypothesis, stating that director compensation is associated with higher pay ratios, has been supported by the findings in this study. The results provide evidence confirming the earlier findings of ‘mutual back scratching’ by Brick et al. (2006). In this study, an increase of 1% in director compensation is found to increase pay ratios by 0.568%. Brick et al. (2006) explained this relationship by arguing that CEOs exert their managerial power to influence the compensation of board members, who in turn respond favourably to propositions of higher CEO compensation, thus also resulting in higher pay ratios. This study provides additional backing to their findings. Employees do not normally hold an influence over the board and cannot try to increase their compensation in similar ways as the CEO. Employees merely have the means of uniting through labour unions in an attempt to persuade board members to increase their compensation. This relationship is discussed later on.
The external power hypothesis regarding outrage costs, stating that outrage costs are associated with lower pay ratios, is not supported by this study. There was some evidence that outrage costs are associated with pay ratios. However, this relationship was found to be insignificant after adding the control variables. This study could not find supporting evidence of the influence of outrage costs on pay ratios as suggested by some (Johnson et al., 1997; Thomas and Martin, 1999; Bebchuk and Fried, 2004a). The findings in this study could suggest that Wall Street has spun out of the control of the public, failing to influence CEO pay and pay ratios. However, outrage costs as measured in this study seem to be largely the result of firm size and some element of reputation. Large well-known firms appear to be mentioned more often in major newspapers, regardless of the compensation levels of their CEOs and of pay ratios. Other less familiar and/or smaller firms did not have outrage costs or had less outrage costs despite larger pay ratios. Outrage costs however, could still influence pay ratios as anecdotal evidence points out. A different measurement for outrage costs could possibly lead to significant findings.
from the results of this study. This finding confirms the findings of Huang et al. (2017). Employees can exercise influence over the firm by collective bargaining power through labour union participation. Labour unions can either exert pressure on the firm to limit CEO compensation levels or to increase employee compensation. The latter is likely to be the most successful option, which in turn reduces pay ratios. Evidence seems to suggest, however, that the influence of unions is fairly limited, highlighting the decline of recent decades in labour union participation and strength (Piketty and Saez, 2006; Western and Rosenfeld, 2011; Lin and Tomaskovic-Devey, 2013).
5.2 The measure of pay ratios
Since the top 1% have disproportionately benefited from economic and capital gains compared to the other 99% of the population, it is important to understand how the top 1% have been able to increase their income. This paper examines a subset of the top 1% income earners namely CEOs. CEO total compensation as opposed to rank-and-file employees in the US has diverged in the benefit of CEOs. This research examined the CEO-employee pay ratios of the S&P 500 as reported in proxy statements following the implementation of the Dodd-Frank Wall Street Reform and Consumer Protection Act which requires the disclosure of CEO-employee pay ratios. Subsequently, this research has aimed to explore the determinants of CEO-employee pay ratios. The CEO still reaps the rewards through the company’s profits as pay is largely related to firm performance, while the employees merely receive compensation for their labour. The degree of foreign employees is therefore irrelevant. The Dodd-Frank Act was a major step towards improved transparency and regulation regarding CEO compensation, despite large corporation protest, but it still lacks a clear and consistent measure to significantly compare CEO compensation of all major firms. Another measure of CEO-employee pay ratios sometimes conducted is measuring employee compensation as GDP per capita. This ensures easier comparability across countries as GDP data is more available and less costly to determine than median firm employee data. However, this measure is based on gross domestic product rather than compensation levels. US GDP per capita in 2016 was $57,638 according to the World Bank database2. Median employee pay does seem to be a more accurate measure of employee compensation at the firm level, reflected in the higher mean median employee pay of $78,790 of the S&P 500 firms. This sample consists of the largest firms in the US. The inclusion of smaller firms could change the median employee levels significantly. Therefore, depending on whether countries require firm level pay ratio disclosure through regulations a CEO to GDP per capita ratio might be more appropriate to analyse pay ratios at the national and global level.
5.3. Pay ratios and the link with income inequality
Income inequality is an major economic phenomenon and measuring its drivers is both complex and time-consuming. Due to time constraints this study merely explores one of its drivers namely the ability of the top 1% to increase their monetary gains. More explicitly, the
focus lies on the ability of CEOs to extract increasingly higher levels of compensation compared to the average employee. Bivens and Mishel (2013) support the rent extraction or managerial power approach by stating that the top 1% has increasingly improved their income by extracting or creating rents from the markets. They argue that the increasing rent extraction is facilitated by increased opportunities and/or incentives for rent-shifting behaviour. They continue by arguing that the rent extraction from the top 1% thwarts the ability of low and middle incomes to benefit from economic growth. Two studies have examined trends of income inequality and trends of executive compensation over the course of the majority of the 20th century. Piketty and Saez (2003) analysed income inequality from 1913-1998 and found a U-shaped curve, while Frydman and Saks (2010) examined executive compensation from 1936-2005 and found a similar pattern of executive compensation forming a J-shaped curve. This could reflect the right hand side of the U-shaped curve found by Piketty and Saez (2003), as Frydman and Saks’ (2010) data does not incorporate the first 20 years which would likely form the left-sided part of the U-shape from Piketty and Saez (2003). Piketty and Saez (2006) state that the income concentration for the top 0.1% earners is not homogenous across countries. They found a more L-shaped curve in France and Japan compared to the U-shaped curve in Anglo-Saxon countries such as the United States, United Kingdom and Canada. This has spurred debate among scholars as to why income inequality has increased more in Anglo-Saxon countries compared to other countries. Lin and Tomaskovic-Devy (2013) argue that the financialization of the US economy has changed income dynamics and disproportionately benefited the top earners at the expense of the general employees. The changing income dynamics could be spurred on by the changes in the structure of executive compensation. This created increases in pay dispersion and income inequality. Literature on how corporations impact the level of income inequality is currently severely under examined. Cobb (2016) examines firm-level factors that influence income inequality at the societal level. He proposes a model in which executive compensation increases the level of income inequality through performance based pay, external hiring of executives and external benchmarking. The research described above leads me to believe the following proposition for future research:
6. CONCLUSION
6.1. Summary
The average US pay ratio over the S&P 500 firms was 273:1 in 2017. Thus, the average CEO in the US earns 273 times more than the average employee. CEOs of the S&P 500 combined earned in total $6,75 billion over 2017, while the median employees combined earned in total $38,7 million over 2017. Combined these employees would have to work 175 years to earn the same amount as CEOs earned combined in just 1 year. Pay ratios in firms are thus likely to increase income inequality. The main focus of this study however, were the determinants of CEO-employee pay ratios. This paper addresses the identified gap in the executive compensation literature namely pay ratios. A hierarchical multiple regression analysis on pay ratios was run to examine the determinants. Results showed that unionization of employees reduces pay ratios, although this effect is low compared to other factors. This illustrates the declining power of labour unions of recent times. The results also showed that pay ratios are lower when the firm’s CEO is also the (co-)founder. Director compensation was found to increase pay ratios. Additionally, some evidence was found pointing towards outrage costs and chairman duality influencing pay ratios, although these results did not hold after adding control variables. Firm size is a major factor driving pay ratios. The results seem to suggest that external power through unionization and outrage costs is rather unsuccessful or restricted in limiting pay ratios, whereas managerial power of the CEO is more influential in determining pay ratios. CEOs can exercise their managerial power to influence pay ratios through ‘mutual back-scratching’ with directors. These findings are applicable to the US. Whether they are applicable to other countries could be examined in future research.
6.2. Theoretical contributions
level of CEO compensation and/or increase employee compensation. The findings of this study confirmed and complemented the managerial power theory and external power forces. It also contradicted some earlier findings that influence executive compensation. Determinants of executive compensation are, therefore, not necessarily the same as determinants of pay ratios. Mutual back-scratching is found as a tool for CEOs to use their managerial power to influence their pay by increase pay of directors. Structural CEO power through Chairman duality was not influential found to be influential. External forces, such as union membership of employees, lead to lower pay ratios. The finding that (co-)founder CEOs lead to lower pay ratios could imply that the wealth of CEOs should also be taken into account when examining pay ratios. This paper also contributes by providing a guide for future research.
6.3. Practical implications
This paper provides a theoretical and empirical analysis of pay ratios allowing useful insight for both regulators and firms. The year 2018 marks the birth of the pay ratio disclosures by US corporations. The average US pay ratio across the S&P 500 was 273:1 and could be higher still if some CEOs had not chosen to voluntarily reduce their compensation. When employees have to work 2-3 lifetimes in order to earn the same as CEOs in one year, questions should arise on the sustainability and usefulness of the current executive compensation practices. The US government could decide to tackle income inequality by limiting pay ratios through regulation or through taxation. The measures regulators could embark upon could tackle the pay ratios directly through a maximum pay ratio rule or indirectly by limiting executive compensation or increasing employee pay. The latter, could be achieved through increasing the minimum wage or opting for a minimum living wage. Whether the US will tackle pay ratios, in an effort to curb income inequality levels, is yet to be seen. Regulators from other countries could use this study as a guide to decide to implement similar disclosure regulations.
6.4. Limitations
generalizability across firm sizes and countries. However, most executive compensation studies have constructed their sample to the S&P 500, or the S&P 1500 including Midcap, and Smallcap companies, since US regulations provide researchers with detailed information regarding executive pay. Additionally, manual data collection might have led to recording errors, although this effect was minimized by double-checking. The number of articles in which the firm is mentioned in combination with executive compensation is a rather rough measurement of outrage costs. It might be more reflective of firm size and reputation than of actual outrage costs. Additionally, the pressure on firms arising from the outrage following a newspaper article is unlikely to be homogenous across articles.
Sample and data limitations have prevented this study to conduct an additional robustness measure of calculating the CEO-employee pay ratio, and in particular employee pay. Following other studies (Faleye et al., 2013; Connelly et al., 2013; Xu et al., 2017) the alternative measure of employee compensation is calculated as: (total labour costs - CEO total compensation) / number of employees. Total labour costs are not easily accessible in most databases, thus creating a small sample size that reduces accuracy and generalizability of the findings. It is also likely to violate the minimum number of observations required for a multiple regression analysis.
limitations are beyond the control of this study and could be addressed by the SEC in the future.
Lastly, this research has focused on pay ratios of US firms only. As highlighted by Fernandes et al. (2012) determinants for executive compensation in the US are often different to other countries. Therefore, the results of this study should be considered in the US context only. Future research could examine whether these findings are also applicable to other countries. This study is bound by the nature of the research setting, a Master Thesis, which reduces the ability employing a study of large scale and scope. Therefore, this paper does not address the first and last concern of the criticism highlighted in the following section, but instead aims to provide recommendations for future research to address these concerns appropriately.
6.5. Recommendations for future research
(Fernandes et al., 2012; Schmid et al., 2018).
Secondly, antecedents of executive compensation and pay ratios, and the actual pay ratios themselves differ across countries. Fernandes et al. (2012) found significant differences in the determinants of CEO pay across countries. They determined that the relation between CEO pay and firm size and firm leverage is stronger in the US. Tobin's q and CEO pay, on the other hand, were found to be lower in the US, while they found a positive relation between CEO pay and board size outside the US, but not in the US. Furthermore, they found that when the CEO is also the chairman of the board and when there are more independent board members, the compensation of the CEO is higher in the US, but not outside the US. This study, however did not find evidence of chairman duality and board independence in the US. In addition, the CEO-employee pay ratio in the US is likely to be (one of) the highest in the world, as we found a pay ratio of 273:1. In context, Shin, Kang, Hyun and Kim (2015) found an average executive-employee pay ratio of 7.6:1 for Korean listed firms from 2000-2009. This ratio has increased from around 6:1 to 8.5:1 over the same period, showing that the pay ratio has also increased in South Korea. The time frames are not similar, yet it is highly unlikely that the Korean ratio in 2016 has converged towards similar levels as the US. The Korean pay ratio is still very low compared to the US. Therefore, the US could act as an outlier in comparison to countries around the world, a position not entirely unfamiliar to the US in international business research. The US should therefore not be taken as a norm for global standards, nor be looked at by boards of other countries in order to determine whether the compensation their CEO receives is fair. International benchmarking based on the US would only allow executives to increase their ability and justification for rent seeking behaviour. Executive compensation and pay ratio studies based in the US should therefore be wary of generalizing their results to other countries.
often measured through Hofstede’s cultural dimensions. The cultural dimension of power distance has been found to be a determinant of CEO compensation (Tosi and Greckhamer, 2004). The power structures within countries are likely to be reflected in the pay structures of CEOs, resulting in higher CEO-employee pay ratios. Strong power structures in countries also reflect that, to a certain extent, the members of a culture accept or expect a level of inequality in power and income. In such cultures a larger discrepancy between the income levels of a CEO and the average employee are accepted or even expected. Additionally, Tosi and Greckhamer (2004) find that the cultural dimension of individualism also determines total pay and the structural composition of pay. Companies in highly individualistic countries are more likely to have higher pay ratios than less individualistic countries. Individualistic CEOs are more likely to engage in rent extractive behaviour by using their managerial power. CEOs, thereby, increasing their own pay rather than employee pay or rather than rewarding shareholders which results in higher CEO-employee pay ratios.
disclosure.
7. REFERENCES
Bebchuk, L. A., Fried, J. M., & Walker, D. I. (2002). Managerial power and rent extraction
in the design of executive compensation (No. w9068). National bureau of economic
research.
Bebchuk, L., & Fried, J. (2004a). Pay without performance (Vol. 29). Cambridge, MA: Harvard University Press.
Bebchuk, L., & Fried, J. (2004b). Stealth compensation via retirement benefits, NBER
Working Paper Series No. w10742.
Bebchuk, L., Cremers, K., & Peyer, U. (2011). The CEO pay slice. Journal Of Financial
Economics, 102(1): 199-221.
Bivens, J., & Mishel, L. (2013). The pay of corporate executives and financial professionals as evidence of rents in top 1 percent incomes. Journal of Economic Perspectives, 27(3): 57-78.
Boyd, B. K., Santos, M. F., & Shen, W. (2012). International Developments in Executive Compensation. Corporate Governance: An International Review, 20(6): 511-518.
Brick, I. E., Palmon, O., & Wald, J. K. (2006). CEO compensation, director compensation, and firm performance: Evidence of cronyism?. Journal of Corporate Finance, 12(3): 403-423.
Connelly, B., Haynes, K., Tihanyi, L., Gamache, D., & Devers, C. (2013). Minding the Gap.
Journal Of Management, 42(4): 862-885.
Connelly, B. L., Tihanyi, L., Crook, T. R., & Gangloff, K. A. (2014). Tournament theory: Thirty years of contests and competitions. Journal of Management, 40(1): 16-47.
Conyon, M., & He, L. (2011). Executive compensation and corporate governance in China.
Journal of Corporate Finance, 17(4): 1158-1175.
Dai, Y., Kong, D., & Xu, J. (2017). Does fairness breed efficiency? Pay gap and firm productivity in China. International Review of Economics & Finance, 48(1): 406-422.
David, P., Kochhar, R., & Levitas, E. (1998). The effect of institutional investors on the level and mix of CEO compensation. Academy of Management Journal, 41(2): 200-208.
DiPrete, T. A., Eirich, G. M., & Pittinsky, M. (2010). Compensation benchmarking, leapfrogs, and the surge in executive pay. American Journal of Sociology, 115(6): 1671-1712.
Dyl, E. A. (1988). Corporate control and management compensation: Evidence on the agency problem. Managerial and Decision Economics, 9(1): 21-25.
Van Essen, M., Otten, J., & Carberry, E. J. (2015). Assessing managerial power theory: A meta-analytic approach to understanding the determinants of CEO compensation. Journal of
Management, 41(1): 164-202.
Faleye, O. Reis, E. & Venkateswaran, A. (2013). The determinants and effects of CEO-employee pay ratios. Journal of Banking and Finance, 37(8): 3258-3272.
Finkelstein, S. (1992). Power in top management teams: Dimensions, measurement, and validation. Academy of Management journal, 35(3): 505-538.
Frydman, C., & Saks, R. E. (2010). Executive Compensation: A New View from a Long-Term Perspective, 1936–2005. Review of Financial Studies, 23(5): 2099–2138.
Gabaix, X., & Landier, A. (2008). Why Has CEO Pay Increased So Much?. Quarterly
Journal of Economics, 123(1): 49-100.
Gomez-Mejia, L. R., Larraza-Kintana, M., & Makri, M. (2003). The determinants of executive compensation in family-controlled public corporations. Academy of management
journal, 46(2): 226-237.
Goolsbee, A. (2000). What Happens When You Tax the Rich? Evidence from Executive Compensation. Journal of Political Economy, 108(2): 352-378.
Hambrick, D. C., & Finkelstein, S. (1995). The effects of ownership structure on conditions at the top: The case of CEO pay raises. Strategic Management Journal, 16(3): 175-193.
Henderson, A. D., & Fredrickson, J. W. (2001). Top Management Team Coordination Needs And The Ceo Pay Gap: A Competitive Test Of Economic And Behavioural Views. Academy
of Management Journal, 44(1): 96–117.
Huang, Q., Jiang, F., Lie, E., & Que, T. (2017). The Effect of Labour Unions on CEO Compensation. Journal of Financial and Quantitative Analysis, 52(2): 553–582.
Johnson, M. F., Porter, S. & Shackell. M. B. (1997). Stakeholder Pressure and the Structure of Executive Compensation. Working paper Michigan State University, University of Massachusetts at Amherst, Notre Dame.
incentives. Administrative Science Quarterly, 38(3): 438-461.
Lazear, E. P., & Rosen, S. (1981). Rank-order tournaments as optimum labour contracts.
Journal of Political Economy, 89(5): 841-865.
Li, M., & Roberts, H. (2017). Director and CEO pay reciprocity and CEO board membership.
Journal of Economics and Business, 94: 54-65.
Lin, K.-H., & Tomaskovic-Devey, D. (2013). Financialization and U.S. Income Inequality, 1970–2008. American Journal of Sociology, 118(5): 1284–1329.
Main, B. G., O'Reilly, C. A., & Wade, J. (1995). The CEO, the board of directors and executive compensation: Economic and psychological perspectives. Industrial and Corporate
Change, 4(2): 293-332.
Oxfam, (2018). Reward work, not wealth: To end the inequality crisis, we must build an economy for ordinary working people, not the rich and powerful. Oxfam Briefing Paper,
January.
Perry, T., & Zenner, M. (2001). Pay for performance? Government regulation and the structure of compensation contracts. Journal of Financial Economics, 62(3): 453–488.
Piketty, T., & Saez, E. (2003). Income Inequality in the United States, 1913-1998. The
Quarterly Journal of Economics, 118(1): 1–41.
Piketty, T., & Saez, E. (2006). "The Evolution of Top Incomes: A Historical and International Perspective."American Economic Review, 96(2): 200-205.
Randall, T. S. & Martin K. J. (1999). The Effect of Shareholder Proposals on Executive Compensation.” University of Cincinnati Law Review, 67(4): 1021–1081.
Compensation in India. MIR: Management International Review, 40(2): 167-191.
Rosen, S. (1986). Prizes and incentives in elimination tournaments. American Economic
Review, 76: 701-715.
Schmid, S., Altfeld, F., & Dauth, T. (2018). Americanization as a driver of CEO pay in Europe: The moderating role of CEO power. Journal of World Business, 53(4).
Shin, T. (2014). Explaining Pay Disparities between Top Executives and Nonexecutive Employees: A Relative Bargaining Power Approach, Social Forces, 92(4): 1339–1372.
Shin, J. Y., Kang, S. C., Hyun, J. H., & Kim, B. J. (2015). Determinants and performance effects of executive pay multiples: Evidence from Korea. ILR Review, 68(1): 53-78.
Stefanescu, I., Wang, Y., Xie, K., & Yang, J. (2018). Pay me now (and later): Pension benefit manipulation before plan freezes and executive retirement. Journal of Financial Economics, 127(1): 152-173.
Tosi, H. L., Werner, S., Katz, J. P., & Gomez-Mejia, L. R. (2000). How much does performance matter? A meta-analysis of CEO pay studies. Journal of Management, 26(2): 301-339.
Tosi, H.L & Greckhamer, T. (2004). Culture and CEO Compensation. Organization Science, 15(6): 657-670.
Weisbach, M. S. (2007). Optimal executive compensation versus managerial power: A review of Lucian Bebchuk and Jesse Fried's pay without performance: The unfulfilled promise of executive compensation. Journal of Economic Literature, 45(2): 419-428.
Western, B., & Rosenfeld, J. (2011). Unions, norms, and the rise in U.S. wage inequality.
