CEO Remuneration
MASTER THESIS WITHIN: Business Administration
NUMBER OF CREDITS: 30hp
PROGRAMME OF STUDY: Civilekonom
AUTHOR: Anisa Odzak
Amanda Skog
JÖNKÖPING May, 2017
A study on the impact of the Capital Requirement Directive IV’s
bonus cap on the variable remuneration
Acknowledgement
Throughout the process of completing the thesis, several individuals have been involved that we would like to express our sincere gratitude to.
Firstly, we would like to express our thankfulness to our supervisor for providing us with valuable and relevant guidance and support in the development of the thesis.
Secondly, we wish to show our appreciation to the individuals in our seminar group that have provided us with constructive criticism in order to help us present a thesis of higher research
standards.
Jönköping International Business School May 2017
Civilekonom thesis within Business Administration
Title: CEO Remuneration: A study on the impact of the Capital Requirement Directive
IV’s bonus cap on the variable remuneration
Authors: Anisa Odzak & Amanda Skog
Date: 2017-05-22
Key terms: Chief Executive Officer, Remuneration, Bonus Cap, EU Banking Sector, CRD IV, Managerial Power Theory
Abstract
Background: Over the past decades, there has been a massive increase in the level of CEO
remuneration, which has not been in line with the growth in company earnings and firm size. This massive increase in particularly variable remuneration was one of the main causes to the financial crisis during 2007-2008. As a result of the financial crisis, the issue of excessive CEO remuneration has become increasingly highlighted. Thus, the regulators have been striving to come up with new regulations to align the long-term interest of the banks with the variable remuneration. A regulation that has been implemented by the EU countries to combat this issue is the bonus cap that is an amendment to CRD IV. The bonus cap limits the variable remuneration to 100% of the fixed remuneration, with the exception that it can be raised to 200% if approved by shareholders. However, the bonus cap has not been as efficient as expected.
Purpose: The purpose of this thesis is to examine how the bonus cap has affected the variable
CEO remuneration. This will be done by analysing and comparing the total remuneration within the 20 largest banks in EU. The investigation is concentrated on the time period 2012-2015.
Method: A sample of 20 banks has been chosen based on three criterions, namely the banks with the largest amount of assets, banks with EU headquarters and banks that provide the information necessary for the research. When collecting the data needed to fulfil the purpose, quantitative data in form of numbers from the banks financial reports was retrieved from their respective website.
Conclusion: The results show that the variable remuneration has increased since the
implementation of the bonus cap. This increase is likely driven by the increase in fixed remuneration. Another consequence of the bonus cap is that the variable remuneration has become
correlation between performance and pay, meaning that the variable remuneration has had a weakened dependence on the performance of the bank. The results of the development in variable remuneration provide an insight about the effectiveness of the bonus cap, what impact it has had and that an improvement of the bonus cap is desirable.
Abbreviations
CEO Chief Executive Officer
CRD IV Capital Requirement Directive IV
DKK Danish Krone
EU European Union
EUR Euro
GBP Great British Pound
SEK Swedish Krona
Table of Contents
1.
Introduction ... 1
1.1 Background ... 1 1.2 Problem ... 2 1.3 Purpose ... 3 1.4 Disposition of thesis ... 42.
Literature Review ... 6
2.1 Managerial power theory ... 6
2.2 Fixed and variable remuneration... 8
2.2.1 Fixed components ... 8
2.2.2 Variable components ... 8
2.2.2.1 Stock option ... 9
2.2.2.2 Bonus ... 9
2.3 Bonus cap ... 11
2.3.1 The impacts of the bonus cap ... 12
2.4 Optimal design of remuneration package ... 14
3.
Method ... 16
3.1 Selected banks ... 16
3.2 Data collection ... 17
3.3 Data analysis ... 18
3.4 Strengths and limitations of the research ... 19
4.
Empirical findings ... 21
4.1 Evolvement of CEO remuneration ... 21
4.2 Evolvement of variable-fixed remuneration ... 23
4.3 Evolvement of correlation between pay and performance ... 24
5.
Analysis ... 26
5.1 Evolvement of CEO remuneration ... 26
5.2 Evolvement of variable-fixed remuneration ... 29
5.3 Evolvement of correlation between pay and performance ... 30
6.
Conclusions ... 33
7.
Discussion... 35
Figures
Figure 1 Components of a "typical" annual incentive plan ... 10
Tables
Table 3.1 Total assets, December 31, 2015 ... 16Table 4.1 Evolvement of CEO remuneration ... 21
Table 4.2 Variable-to-fixed remuneration ... 23
Table 4.3 Correlation between variable pay and profit before tax ... 24
Appendix
Appendix 1 - Currency Conversion ... 40Appendix 2 - Remuneration components ... 41
Appendix 2.1 - Remuneration components 2012 ... 41
Appendix 2.2 - Remuneration components 2013 ... 43
Appendix 2.3 - Remuneration components 2014 ... 45
Appendix 2.4 - Remuneration components 2015 ... 47
1. Introduction
1.1 Background
The issue of Chief Executive Officer (CEO) remuneration has been widely discussed by market participants over the past decades. This was mainly due to a massive increase in the level of remuneration that CEOs are receiving, which was not in line with the growth in company earnings and firm size (Bebchuk & Grinstein, 2005). The enormous growth in CEO compensation in combination with recent corporate scandals, mainly during the financial crisis of 2007-2008, has contributed to highlight the issue of CEO remuneration and how it could be improved (Angeli & Gitay, 2015).
The financial crisis caused many near-collapses of several European banks. A key cause of the financial crisis was the incentivisation of CEOs and employees to take risks in order to obtain short-term gains, without considering the negative consequences that might arise because of their actions. These large amounts of remuneration, especially to the CEOs, contributed to extreme losses at major banks and are one of the reasons to why the crisis lasted as long as it did (Nguyen & Shlomo B, 2013).
In the aftermath of the financial crisis, it was necessary to enable better regulations of the remuneration practices to avoid future crises and achieve financial stability. For that reason, the European Commission has issued several recommendations dealing with the issue of executive remuneration within the banking sector (Avgouleas & Cullen, 2014). The focus of the recommendations issued is that the remuneration should be distributed with the aim to align the remuneration policies with the long-term interests of the banks (European Commission, 2016; Avgouleas & Cullen, 2014; Nguyen & Shlomo B, 2013).
A continuously evolving regulation in Europe has had a core focus on limiting the variable remuneration (European Parliament, 2013a). In order to limit the variable remuneration, important amendments to the current Capital Requirements Directive (CRD IV), primarily aimed at preventing excessive risk-taking, were implemented by the countries in EU from the beginning of 2014. The additional amendment to CRD IV, called the bonus cap, implies that the variable remuneration of CEOs and other employees will be limited to 100% of the fixed remuneration. This cap can in turn be increased to 200% with the approval of the shareholders (Directive 2013/36/EU).
The limit implies that the CEOs variable remuneration will increase based on performance until it reaches its maximum amount. Accordingly, excessive risk-taking will not lead to excessive bonuses like it did previously (Murphy, 2013). A previous study on banks’ remuneration revealed that there exists a striking difference between the ratio of variable and fixed remuneration when comparing the pay practices in the EU before and during the financial crisis (Murphy, 2013). The implementation of the bonus cap, and how banks respond to a regulation like this, creates an interesting topic concerning how the variable and fixed remuneration has evolved.
1.2 Problem
The improper usage of the variable remuneration by banks has been a widely discussed topic by regulators. As mentioned in the background, the high proportion of variable pay has encouraged the CEOs to increase their financial performance in the short-term, with the consequence of excessive risk-taking by CEOs (Murphy, 2013).
During the years before the financial crisis, the variable remuneration doubled in the finance industry. The banks’ remuneration policies were one of the factors contributing to the recent financial crisis (Angeli & Gitay, 2015). For this reason, the remuneration of bankers has been a hot topic for politicians on EU level since then (Partington & Patrick, 2013). In addition, the banking sector is one of the most questioned and criticized industries when it comes to excessive remuneration (High-level Expert Group, 2012).
The regulations regarding remuneration in banks have been modified and changed to minimize a risk-taking approach. The bonus cap is supposed to be one solution for the excessive variable remuneration (European Commission, 2013). However, the bonus cap has not been as effective as expected. Several banks have tried to circumvent the limited variable remuneration by increasing the fixed remuneration to provide the CEOs with a higher variable pay (Angeli & Gitay, 2015). This implies that the bonus cap has not completely worked as a solution for the excessive variable remuneration since banks have come up with ways to continuing providing their CEOs with high remuneration. Therefore, this study will investigate the effectiveness of the bonus cap, to see if further developments are necessary to ensure that the history of excessive variable remuneration paid to CEOs does not repeat itself.
1.3 Purpose
The purpose of this thesis is to examine how the bonus cap has affected the variable CEO remuneration. This will be done by analysing and comparing the total remuneration within the 20 largest banks in the EU. The banks are ranked in accordance with their total assets in 2015 and the investigation is concentrated on the time period 2012-2015.
The bonus cap, which was implemented by the EU countries in 2014, aims at preventing excessive amounts of variable remuneration and the corresponding risk-taking by CEOs and other employees (European Commission, 2013). In order to compensate for the limited variable remuneration some banks have increased the fixed remuneration by adding new components. One way of increasing the fixed remuneration is by paying monthly allowances to the CEOs that are based on their responsibilities and position and cannot be linked to performance. These allowances can be adjusted downwards and upwards over the years. By increasing the fixed remuneration through the use of added components, the maximum limit of variable remuneration will in turn increase (Partington & Patrick, 2013). To examine if the bonus cap has fulfilled its aim, the following research question has been stated:
How has the CEO remuneration evolved after the implementation of the bonus cap?
The quantities of variable remuneration increased greatly during the years before the financial crisis. As a response to the introduction of the bonus cap, the fixed remuneration has instead increased more than the variable remuneration in proportion to the total remuneration throughout the EU (Angeli & Gitay, 2015). A study made on banks in Poland showed that the structure of the fixed and variable remuneration had changed for CEOs after the financial crisis. In 2010, basic salary accruing to the fixed remuneration constituted of a larger percentage of the banks total remuneration when compared to before the financial crisis (Slomka-Golebiowska, 2013). The implementation of the bonus cap, and its subsequent limit of the variable remuneration, might have caused the proportion of the fixed and variable remuneration to change when compared to before the implementation, leading up to the following research question:
How has the variable remuneration evolved in relation to the fixed remuneration?
A previous study showed that high amounts of remuneration were awarded to key personnel even though the same personnel did not perform well, leading to a lack of correlation
between pay and performance (Bebchuk & Fried, 2009). In addition, another previous study revealed that there exists a strong correlation between pay and performance when the organization is performing well. Meanwhile, a weak correlation exists between pay and underperformance of an organization (Gregg, Jewell, & Tonks, 2010), meaning that CEOs receive a great remuneration that does not always reflect the performance of the bank. In addition, the provided remuneration to CEOs should according to CRD IV be aligned with the long-term performance of the bank, implying that a correlation between performance and pay should exist (Directive 2013/36/EU). Therefore, it would be interesting to see how the bonus cap has affected this correlation, which leads to the following research question:
How has the correlation between pay and performance evolved?
1.4 Disposition of thesis
The thesis is outlined as follows:
Chapter 2 – Literature Review: The chapter begins with the managerial power theory that is a theory of high importance for the topic of matter. The chapter continues with an outlining of the difference between the fixed and the variable remuneration of the CEO together with their components. The chapter sheds an extra focus on variable remuneration before heading over to an extensive explanation of the bonus cap and the impacts that it is predicted to have. The chapter closes with a model about the optimal design of a remuneration package. Chapter 3 – Method: The chapter starts with displaying an overview of the selection of banks used in the study. The procedure for data collection is then extensively explained and followed by an explanation of how the data has been analysed. The chapter is completed with a discussion of the strengths and limitations of the chosen method.
Chapter 4 – Empirical findings: The chapter covers the findings from the different banks financial reports during the chosen years. The three headlines included are covering the data collected and portrayed in tables in order to answer the three research questions.
Chapter 5 – Analysis: The chapter analyses the empirical findings in relation to the literature and the managerial power theory. This section is analysed in the same arrangement as it is
presented in the empirical findings, implying that the structure of the analysis aims towards answering the three research questions.
Chapter 6 – Conclusion: The chapter concludes and fulfils the purpose. The conclusion is based on the empirical findings and analysis.
Chapter 7 – Discussion: The chapter reflects on the major problems with the investigation and ethical effects of the study. The discussion also proposes future research of the topic.
2. Literature Review
2.1 Managerial power theory
The managerial power theory can be used to explain the issue of excessive CEO remuneration. The theory is dealing with the same agency problem as the classical agency theory where there are conflicts of interests between shareholders (principals) anddirectors (agents). In this study, the CEO will represent the agent. This principal-agent problem arises due to the separation of ownership and control. As a result, the owners of the banks do not run the bank and it is the CEO who has significant power. Since the shareholders own the bank, they are more likely to think about the long-term interests of the bank to increase their return on their shares. Whereas the CEOs might be more interested in short-term profits to receive high bonuses and they will use their power to benefit their wants (Bebchuk & Fried, 2003).
There exist two different views on how the CEO remuneration and the agency problem are connected: the optimal contracting approach and the managerial power theory. The optimal contracting approach is the most dominant view and states that the CEO remuneration will work as a remedy to the agency problem as it will provide the CEOs with incentives to work towards maximizing shareholder value, instead of benefiting themselves. In contrast to the optimal contracting approach, the managerial power theory emphasizes that the CEO remuneration is a part of the problem itself and that the remuneration does not always work as a remedy to the agency problem (Bebchuk & Fried, 2003).
The managerial power provides the CEOs with the power to influence their own remuneration. As a consequence, the CEOs can misuse their power to extract rents, which consist of a higher amount of remuneration than the CEOs would normally have received if their obtained remuneration would have the aim to maximize shareholder value. Thus, the remuneration of CEOs will not mitigate the agency problem between the shareholders and CEOs since the rents are not aligning the interests of the CEOs with the interests of the shareholders (Bebchuk, Fried, & Walker, 2002).
The managerial power theory proposes that managerial power and rents are correlated, which means that the more power a CEO has, the greater the rents will be. The CEO’s power is influenced by several factors. A significant factor that influence the power of the CEOs is
how many independent directors there are within the bank as well as how many directors there exists that the CEO can easily influence (Bebchuk, Fried, & Walker, 2002). According to Schneider (2013) there are several incentives, both economic and non-economic, that encourage independent directors to support the CEO remuneration. One non-economic incentive is that they want to remain in their position and be re-elected. When the independent directors want to be re-elected, it would benefit them if they would get the support from the CEO. This is important, as the CEO has a great influence over the nomination of directors that will be placed on the company’s slate of directors. If the directors want to get the CEO’s support, they should favour the proposed CEO remuneration. Another reason for why the directors would like to please the CEO is because the CEO can influence the directors’ compensation and reward directors that have shown their support of the CEO (Schneider, 2013).
Even though the managerial power theory states that the CEO’s remuneration is correlated with the power of the CEO, it also implies that the CEOs cannot use their power to obtain unreasonable amount of remuneration to such a degree that would create outrage within the society. The larger amount of outrage that the CEO remuneration would contribute to within the society, the harder it would be for CEO’s remuneration packages to be approved by the directors. This is because outrage can lead to bad reputation for the directors, which is something preferably to avoid. If the society does not approve the CEO remuneration, that is suboptimal for shareholders, the directors are more likely to disapprove of the remuneration. If the CEO remuneration, and rent extraction, generates much outrage within the society, the CEO can try to minimize this outrage by camouflaging their extraction of rents to prevent it from being easily identifiable by outsiders. Consequently, insufficient remuneration structures might be adopted that will lead to suboptimal incentives that can hurt the shareholder value (Bebchuk, Fried, & Walker, 2002; Bebchuk & Fried, 2003).
2.2 Fixed and variable remuneration
Remuneration is a form of compensation consisting of monetary and non-monetary rewards, given in exchange for the performance of a specific task. Monetary rewards consist of a fixed pay and a variable pay. Non-monetary rewards are benefits that in some cases are legally required such as health insurance (Madhani, 2010).
Fixed pay is a non-contingent payment that is independent on the performance. It is rather guaranteed in return for participation. Fixed pay is based upon skills, previous experience and the difficulty of performing the task. Variable pay is participation based but essentially dependent on performance of the bank and the employee. It is not guaranteed but has a moderately uncertain nature. The uncertainty stems from the dependence that variable pay has on the quality of the performance and achievements, which are the key indicators determining the amount of the reward. Variable pay is more volatile than fixed pay and fluctuates annually (Madhani, 2010).
2.2.1 Fixed components
According to CRD IV, fixed remuneration comprises of base salary, regular pension and benefits (Directive 2013/36/EU). Base salary is paid to the CEOs based on the terms stated in the contract. The base salary is influenced by the company’s size, the industry in which they are operating, the experience of the CEO and how similar companies are setting the base salary of the CEOs. A characterization of the base salary is that it is not related to the performance of the CEO or how well the company is currently performing (Mallin, 2016). Pensions provide CEOs with the right to receive a part of their salary after reaching a specific age (Cadman & Vincent, 2015). The CEO can also receive several benefits including for example healthcare and a company car (Mallin, 2016).
2.2.2 Variable components
According to CRD IV, variable remuneration comprises of bonuses and stock options, which are contingent on performance (Directive 2013/36/EU). The use of variable remuneration in the rewarding of CEOs has been a widely-discussed topic since many market participants agrees that a large variable remuneration can lead to an increased risk-taking by the CEOs (Angeli & Gitay, 2015)
2.2.2.1 Stock option
The first variable component is the stock option, which allows the CEOs to buy shares at a stated exercise price. The CEOs are able to purchase these shares during a specified period of time and are being rewarded with stock option when certain performance objectives have been fulfilled (Aboody & Kasznik, 2000). Stock options can also work as a strong contribution to the excessive risk taking by CEOs. The options that the CEOs are receiving will provide the CEO with an advantage compared to shareholders who own common shares. This is because the CEO is provided with the right to acquire shares for a stated exercise price, during a specified time period. Consequently, the CEO will be protected from the losses that might occur for the common shareholders. When the value of the shares declines for common shareholders, the value of the stated exercise price for stock options will not be affected. Hence, if the share price is below the stated exercise price, it will not have any negative impacts on the CEOs option pay. As a result, the CEOs will be more willing to engage in excessive risk taking since the CEO will not experience any losses when the share price is below the stated exercise price (Bebchuk & Spamann, 2010).
2.2.2.2 Bonus
The second variable component is the bonus. Bonuses are paid as a result of good performance, based on both the performance of the CEO and the bank, and are normally rewarded once a year (Mallin, 2016). The bonus is also encouraging CEOs to engage in excessive risk-taking. This is a consequence of the linkage between variable remuneration and performance as the CEOs will take risks in order to obtain short-term gains and maximize their yearly bonus plans (Angeli & Gitay, 2015). Murphy (2001) has created a categorization of the bonus plan that exemplifies how several companies set their internal bonus plan. The bonus plan is divided into three basic components. The first component is performance measures, as the majority of companies rely on some sort of measurement of their performance. These measurements are mainly measures of accounting performance where the net income, operating profits and revenues are some of the existing measurements. The second component is performance standards where each of the measurements are provided with a standard and a target for how the CEO and the company should perform. An example of this is when the performance is measured against the annual budget goals to ensure that the performance has met or exceeded the standards. The final component is the pay-performance relations that displays how the CEO remuneration is influenced by the
performance of the CEO and the bank. Based on the three components of the bonus plan, Murphy (2001) has come up with a model (see Fig.1), which is explaining the formation of bonuses. In line with the bonus plan, the bonus will usually not be paid to the CEOs until a threshold performance has been attained. This threshold bonus will equal the minimum bonus that will be rewarded to the CEOs when the performance target has been achieved. In contrast to the threshold bonus, there is usually also existing an internal bonus cap which will limit the bonus to a fixed amount. Consequently, an “incentive zone” is created which corresponds to the range between the cap and the threshold. Within this range, the yearly CEO bonus will be set which will be influenced by the measured performance of the CEO and the company. This implies that performance improvements will result in increased bonuses and CEOs will be encouraged to take risks to perform better and accordingly receive a higher bonus (Murphy, 2001).
2.3 Bonus cap
The bonus cap is an amendment to CRD IV. CRD IV, also known as Directive 2013/36/EU, is aimed at preventing excessive risk-taking by highlighting that the remuneration policy should not encourage risk-taking that is beyond the level of tolerated risk for each organization. The remuneration policy should be in accordance with the strategy and objectives of the bank. In addition, the remuneration policy should have a focus on the long-term interest of the bank (Directive 2013/36/EU). The objective with CRD IV is accomplished by confirming that the remuneration policy of every credit institutions and investment firm reflects the performance and risk management of the association. One way of confirming this is by letting a competent authority review the risk management of these associations each year. Besides this, banks must provide calculation showing their risk exposure (European Parliament, 2013b). The rules concerning excessive risk-taking will be applicable on personnel whose task has a substantial impact on risk, personnel with control functions and others that are paid in the same manner. The directive applies to all banks located in the area of the EU, regardless of where their headquarters are located, and it also applies for banks located anywhere in the world with an EU headquarter (European Parliament, 2013b; Directive 2013/36/EU).
In order to restraint excessive risk-taking, the bonus cap has been imposed. The ratio of the bonus cap is regulated to 1:1 of the fixed remuneration, but it can be offset to 1:2 if shareholders approve. Shareholder approval corresponds to the votes of 66% of the shareholder that own half of the shares represented when the decision is made. It could also correspond to a majority of 75% of the ownership rights represented. Another limit of the bonus cap concerning the variable remuneration, states that 25% of the variable remuneration should constitute of long-term deferred assets, implying that the remuneration will be provided to the CEOs over a certain amount of years. By long-term it is suggested to be a time period of at least 5 years. In addition, at least 40% of the variable remuneration should be deferred over a time-period of three to five years in line with the nature of the business. A substantial portion of the variable remuneration, implying 50% or more, should be balanced between ownership interest in terms of shares or in comparable instruments in an unlisted organization and equivalent ownership interests. This portion should be designed to align the long-term interest of the institution with the incentives given. Malus or clawback
arrangements are covering the total variable remuneration, implying that the variable remuneration can be taken back after it has been distributed (Directive 2013/36/EU). 2.3.1 The impacts of the bonus cap
The introduction of the bonus cap has received a lot of criticism (Murphy, 2013). There exist two articulated objectives with the bonus cap; these are to reduce excessive risk-taking and to decrease the levels of remuneration (European Parliament, 2013b). According to Murphy (2013), the cap will have an opposite effect than it is actually striving for, as it will result in five different shortcomings heading in the contradictory direction.
Firstly, the excessive risk-taking will increase rather than decrease as a consequence of the next shortcoming. The second shortcoming will be an increase in fixed remuneration, since banks will need to increase the fixed proportion to keep giving bonuses as previously. The increase in fixed remuneration will be a way of circumventing the bonus cap, that restraints the variable remuneration in proportion to the fixed remuneration, as an increase in the fixed proportion will lead to an equivalent increase in the variable proportion. Labour cost in the banking sector were previously given as variable pay, making it easier to lower the remuneration for bankers during crises and hard times. This flexibility helped banks overcome times with less profits, without endangering bankruptcy. If the fixed remuneration grows it will make banks more vulnerable in crises since the flexibility of the remuneration will decrease, resulting in an additional risk of failure. With a higher fixed remuneration, bankers will be more likely to take risks since their total remuneration is not affected to a larger extent if the investment fails. This implies that one of the problems with the bonus cap is that traders do not share the downside of the risks. Previously, bankers that took large risks that resulted in unfavourable outcomes did not get a bonus, and with a fixed remuneration being set below the market levels, the low salary as such represented a penalty. By this, a system of high base salaries and lower variable incentives will result in more risk-taking, which is contrary to the purpose of the bonus cap (Murphy, 2013).
Thirdly and fourthly, the bonus cap will reduce the competitiveness of the EU banking sector and the incentives to create value (Murphy, 2013). Murphy (2013) argues that the competitiveness will be affected by the decreased flexibility of the remuneration. A large fixed remuneration will make it harder to align pay and performance, as variable remuneration is used to incentivize hard work. A mitigated connection between pay and
performance will lead to lower profitability because traders will be less eager to perform and make profits. It will also reduce incentives to create value since the variable pay that is used to connect pay and performance will be limited in a way that prevents it from serving its purpose. This will result in fewer incentives to take action in the interest of shareholders and society. Lastly, Murphy (2013) claims that the talented EU bankers will leave for better-paid jobs. The reason for this is that if the EU does not offer these bankers competitive market remuneration these workers will change to another employer outside the EU where a better pay is offered (Murphy, 2013).
In addition, the variable remuneration will stop to fluctuate to a great extent, instead it will rather become constant to the fixed remuneration. The assumption behind this prediction lays in the mitigated connection between pay and performance that results in that the function of the variable remuneration to incentivise performance will not be used as before and therefore diminish. A consequence of the prediction that variable remuneration will become more constant to the fixed remuneration will be that new different components of fixed pay will evolve, since the variable remuneration will lose its function (Murphy, 2013). One example of this is the use of allowances, which can be adjusted downwards and upwards over the years and are predicted to be used by an increased number of banks after the implementation of the bonus cap (Partington & Patrick, 2013). According to Murphy (2013) there cannot be one solution that fits every bank, instead of using the approach one size fits all, the directive should have used another approach that focused on giving bonuses that are based on quality of the transaction instead of volume.
2.4 Optimal design of remuneration package
According to Matousek and Tzeremes (2016) it is important to balance the fixed and variable pay in order to have a well-designed remuneration package. Higher levels of CEO variable pay lead to more risk-taking investments, which tamper the overall performance of the organization (Matousek & Tzeremes, 2016).
The significance of having a well-designed remuneration package is also emphasized in another study. A reason why the remuneration package should be well designed is in order to accomplish three objectives. These objectives are, in this case; in order to attract the right CEOs to the lowest cost, to retain the right CEOs to the lowest cost, and to encourage CEOs to work towards creating long-term shareholder value and circumventing actions that could harm the value created (Jensen, Murphy, & Wruck, 2004)
These objectives can only be accomplished by designing three different dimensions. The first dimension states that a CEO will only stay at a workplace if the expected benefits at that place are the highest compared to other potential employers. Benefits refer to the expected pay, the risks associated with the benefits and the cost of switching employer. Continuing, the second dimension distributes the remuneration components in a way were the benefits to the CEO are maximized or in another way where the total cost for the bank is minimized. The last dimension outlines the relationship between pay and performance. It states which actions and outcomes are rewarded but also which are punished. By stating this it is expected to show how hard a CEO should work towards the rewards and the importance of being productive (Jensen et al., 2004).
Well-designed remuneration packages are important in diverse ways to manage different CEOs (Jensen et al., 2004). According to Madhani (2010) there are different objectives for the use of a remuneration package consisting of variable remuneration. An objective is the use of the variable remuneration as a motivation to facilitate the CEO to put in a greater effort into their working tasks but also to encourage the CEOs to take risks that might result in a positive outcome for the bank. If CEOs only obtain fixed remuneration, they will be more likely to only put in as much effort as needed in order to remain at their position as CEO. This implies that they will be less willing to take risks even though it might lead to future gains for the bank. The use of variable remuneration can function as a motivation for CEOs to work harder due to its connection to performance (Madhani, 2010). Jensen et al.
(2004) confirm this statement by stating that a remuneration package consisting of a high fixed pay and a low variable pay will be less motivating when it comes to performance than a remuneration package with a low fixed pay and high variable pay. Another advantage of variable remuneration is that it can attract new employees to work for the bank as well as retain valuable current employees (Hermalin & Wallace, 2001). Meanwhile, a package with high retirement benefits will attract CEOs that plan to be employed by a firm for a longer period of time (Jensen et al., 2004).
3. Method
3.1 Selected banks
At the commencement of the sample selection, three criterions were used as a guide to find a suitable sample that will answer the purpose on how the variable CEO remuneration has been affected by the bonus cap. Firstly, the sample selection was centered on leading banks in EU according to their assets in 2015, providing the study with banks that have substantial impact on the financial stability. Secondly, the bonus cap applies to all banks located anywhere in the world with headquarters in EU, therefore only banks with EU headquarters were included to enable a clear impact on the consolidated financial reports. Finally, only banks that provided usable information regarding the CEO remuneration were included. The sample has been derived from Relbanks statistics that is based on the assets of the banks portrayed in the balance sheets found in their respective annual report (Relbanks, 2016). Table 3.1 gives an overview of the size of the assets in the different banks examined and the country of origin for each.
Total assets in Included in Number Name of the bank Country US $ billion the final sample:
1. HSBC Holdings Plc UK 2 410 Yes 2. BNP Paribas SA France 2 180 Yes 3. Credit Agricole SA France 1 858 Yes 4. Deutche Bank AG Germany 1 781 Yes
5. Barclays Plc UK 1 660 Yes
6. Banco Santander SA Spain 1 465 No 7. Societe Generale SA France 1 459 Yes 8. Group BPCE France 1 275 Yes 9. Royal Bank of Scotland Group Plc UK 1 208 Yes 10. Lloyds Banking Group Plc UK 1 195 Yes 11. UniCredit SpA Italy 941 Yes 12. ING Group NV Netherlands 920 No 13. Banco Bilbao Vizcaya Argentaria SA Spain 820 No 14. Banque Federative du Credit Mutuel SA France 809 Yes 15. Intesa Sanpaolo SpA Italy 740 Yes 16. Rabobank Group Netherlands 733 No 17. Nordea Bank AB Sweden 707 Yes 18. Standard Chartered Plc UK 640 Yes 19. Commerzbank AG Germany 582 Yes
20. KfW Group Germany 550 Yes
21. Danske Bank A/S Denmark 483 Yes
22. DZ Bank AG Germany 446 No
23. Cassa Depositi e Prestiti SpA Italy 435 No 24. ABN AMRO Group NV Netherlands 427 Yes
25. CaixaBank SA Spain 376 No
26. Nationwide Building Society UK 304 Yes 27. Svenska Handelsbanken AB Sweden 300 No 28. Skandinaviska Enskilda Banken AB Sweden 297 Yes Table 3.1 Total assets, December 31, 2015.
At the starting point of this research, the selected banks were the 25 largest banks within the EU. During the process of collecting the data, it was acknowledged that several of the banks could not be included in this study. To arrive at a final sample of 20 banks, eight banks were excluded and three new banks were added in order to have an even number of banks. CaixaBank SA and Rabobank Group were excluded as they were missing a clear outline of the remuneration accruing to the CEO for the year 2012. ING Group NV, Banco Bilbao Vizcaya Argentaria SA, Svenska Handelsbanken AB and Banco Santander SA paid their CEOs with stock options and shares that did not have a specified value. Without a specified monetary value of the shares that the CEO received each year, it was not valid to include them since it would have given a misleading result. Cassa Depositi e Prestiti SpA and DZ Bank AG were also excluded as they were missing an outlay of the remuneration accruing to the CEO concerning all sampled years.
3.2 Data collection
Quantitative data in form of numbers from the banks’ financial reports found on their respective website was retrieved to fulfil the purpose of this study. This already published information is classified as secondary data (Bryman, 2016). To ensure that the collected data would provide the study with valid results, the data was firstly collected individually by the authors of this study. Furthermore the data was revised collectively to confirm that the collected data represents the values found in the financial reports. In the examination of the data found in the financial reports, the data on the different CEO remuneration components was transferred to excel sheets and lastly assembled in Appendix 2. To enable an analysis of the change in variable remuneration, the collected components were further grouped into the category of fixed and variable remuneration as can be seen in table 4.1.
In addition to the assembled data regarding remuneration, data on the profit before tax has also been collected from the banks’ financial reports to see the relationship between performance and pay, as stated in the third research question. Profit before tax has been used as a parameter for performance since it provides a more valid parameter than if the profit after tax would be used, as profit after tax would be biased by the different tax rates in the different countries. The parameter for pay comprises of variable remuneration as the variable remuneration, in contrast to the fixed remuneration, is based on the performance of the CEOs and the banks.
To facilitate a clear comparison of the assembled data, the values provided from each bank has been converted into the currency of euro (EUR) by using the European Central Bank’s currency rates. The different currency rates used are each based on the last day of the calendar years included, and are all provided in Appendix 1.
3.3 Data analysis
To answer the three research questions of this thesis, three types of data have been analysed. The analysis has been subdivided into three parts corresponding to each research question. Regarding the first research question ‘How has the CEO remuneration evolved after the implementation of the bonus cap?’ an overview of the fixed, variable and total remuneration for each bank concerning every sampled year has been carefully studied. The reason why the fixed remuneration has been included is to enable a comparison of the development of the variable remuneration against the fixed and the total remuneration. This comparison has been done in order to get a greater understanding of the impact of the bonus cap on the variable remuneration. The values for the average, median and the increase/decrease in percentage between each sampled year concerning the fixed, variable and total remuneration, has been identified for all banks collectively. These sets were analysed by the two authors independently. Each resulting analysis was then compared and reconciled.
The second research question ‘How has the variable remuneration evolved in relation to the fixed remuneration?’ has been analysed by putting the already gathered data on variable remuneration in relation to the fixed remuneration for each sampled bank. The analysis of the second research question was greatly influenced by the findings and analysis concerning the first research question.
To answer the third research question ‘How has the correlation between pay and performance evolved?’ the statistical program Statistical Package for the Social Sciences has been used. To enable a correlation to be found, a Pearson correlation between the profit before tax and variable remuneration during the sampled years for each bank has been performed. The Pearson correlation provided the study with a general correlation for all of the banks and a level of significance that was tested against the stated hypothesis below: H0=0 - There exists no significant correlation between pay and performance
To determine if the null hypothesis should be rejected or not, the received p-values were tested against the chosen significance level of 10%. A level of 10% has been selected to ensure that only a difference that really exists has been detected. The correlation and p-values have been examined jointly by the two authors.
3.4 Strengths and limitations of the research
The sample of 20 banks used in this research cannot fully represent the impact of the bonus cap on all banks affected by it. If all banks in the EU affected by the bonus cap would have been examined it would be too time consuming and therefore this limitation of the sample can be seen as a weakness of the chosen method. Even if the extension of the sample is not broad, a pattern on the impact of the bonus cap has been discovered. Due to the fact that all sampled banks were selected based on the value of their assets, a sample bias may also occur since some banks were not given a chance to be included in this study.
By using the financial reports and mostly the annual reports of the sampled banks, the presumption is that the numbers provided are correct and reliable. This presumption is strengthened by the circumstance that the reports that have been used have been audited before publication. Since the focus in this report will be on the CEO remuneration found in the financial reports, which is mainly numerical, the quantitative research strategy is the most appropriate research method to use. A major advantage with using the quantitative research method when collecting the data is that it will provide the study with easily comparable data, both among the banks and among the years. A negative aspect with using a quantitative approach is that the data will not be as descriptive and it is harder to look deeper into the data collected in order to make a detailed analysis of the data (Bryman, 2016).
At the commencement of this research the idea was to examine the different components accruing to the variable remuneration. When starting the data collection, it was clear that this approach needed to be changed due to the number of different components. Before a change was made, an attempt to categories the different variable remuneration components was performed that made the authors realize that this categorization could be subject for misinterpretations. Therefore, a categorization of fixed and variable remuneration has been used to minimize the risk of delusions. The way of grouping the different components belonging to the fixed and variable remuneration are all based on the guidelines provided in the amendment to CRD IV by the European Parliament, which strengthens the validity of
the research. There is a possibility of subjectivity from the researchers, due to any misinterpretation of a specific remuneration component that has resulted in that it is grouped in the wrong category.
As some of the banks included in this sampled have used a currency other than the EUR in their financial reports, their currency has been converted to EUR to enable a comparison. This implies that the numbers used may be influenced by fluctuations in the exchange rates. During the collected time period, several of the banks have experienced a change of CEO, making the remuneration in most cases rise that particular year. In some cases, were there was a change of CEO, no variable remuneration was paid out that specific year. Meanwhile Deutsche Bank AG has two co-CEOs, increasing the remuneration as pay is given to two persons. Since the other sampled banks do not have any co-CEOs, an average value has been derived for Deutsche Bank AG to make the remuneration between the sampled banks consistent. It is important to consider that these different factors could have an impact on the findings. In several cases where a CEO has been changed, a termination payment has been paid out. These termination payments usually aggregate to large amounts and have not been included in our study. The reason for this is because they would influence the findings negatively and also since they were not mentioned in the guideline of the amendment to CRD IV (Directive 2013/36/EU).
4. Empirical findings
4.1 Evolvement of CEO remuneration
T ab le 4 .1 : E vo lv em en t o f C E O re m un er at io n B an ks 2012 2012 2012 2 01 3 2 01 3 2 01 3 2 01 4 2 01 4 2 01 4 2 01 5 2 01 5 2 01 5 V ar ia bl e F ix ed T ot al V ar ia bl e F ix ed T ot al V ar ia bl e F ix ed T ot al V ar ia bl e F ix ed T ot al H SB C H ol di ng s Pl c 6 0 65 4 33 3 0 42 5 19 9 1 07 9 52 6 5 97 0 97 3 0 38 2 63 9 6 35 3 60 4 3 67 6 98 5 4 14 0 45 9 7 81 7 43 4 1 43 3 34 5 8 45 0 85 9 9 88 4 19 B N P P ar ib as S A 1 6 80 0 00 1 1 53 1 08 2 8 33 1 08 2 1 40 1 12 1 2 53 1 08 3 3 93 2 20 1 5 31 2 00 1 2 54 5 68 2 7 85 7 68 2 2 89 8 85 1 2 54 5 68 3 5 44 4 53 C re di t A gr ic ol e S A 45 0 0 00 1 0 08 8 95 1 4 58 8 95 1 0 70 0 00 1 0 13 2 89 2 0 83 2 89 98 7 0 00 1 0 01 9 55 1 9 88 9 55 93 8 0 00 96 4 6 28 1 9 02 6 28 D eu tc he B an k A G 3 0 57 8 05 2 6 18 0 93 5 6 75 8 98 5 1 74 5 97 3 3 11 7 56 8 4 86 3 53 2 8 61 9 58 4 9 71 5 87 7 8 33 5 45 0 3 5 61 9 64 3 5 61 9 64 B ar cl ay s Pl c 1 7 97 5 74 2 2 11 7 39 4 0 09 3 13 0 1 9 21 5 55 1 9 21 5 55 3 7 92 5 28 3 2 26 3 45 7 0 18 8 73 2 7 23 6 19 2 2 84 8 97 5 0 08 5 16 So ci et e G en er al e S A 0 1 3 05 9 25 1 3 05 9 25 1 4 06 0 70 1 3 05 9 25 2 7 11 9 95 94 8 7 67 1 3 05 9 25 2 2 54 6 92 1 4 74 2 00 1 3 05 9 25 2 7 80 1 25 G ro up B PC E 56 2 5 69 55 5 2 56 1 1 17 8 25 89 0 9 94 55 5 2 92 1 4 46 2 86 85 1 8 58 55 5 2 92 1 4 07 1 50 82 7 4 57 55 5 2 88 1 3 82 7 45 Ro ya l B an k o f S co tla nd G ro up Pl c 0 2 0 16 9 10 2 0 16 9 10 0 1 9 34 7 49 1 9 34 7 49 45 9 6 23 1 9 16 8 06 2 3 76 4 28 1 8 35 2 75 3 3 21 7 52 5 1 57 0 27 Ll oy ds B an ki ng G ro up Pl c 1 8 19 6 30 2 3 20 7 94 4 1 40 4 24 5 7 91 0 52 3 1 75 0 03 8 9 66 0 55 10 5 05 8 42 4 3 15 0 60 14 8 20 9 01 8 3 13 9 18 3 6 39 2 12 11 9 53 1 30 U ni C re di t S pA 2 2 39 6 01 1 7 64 9 23 4 0 04 5 24 1 1 76 3 49 1 7 38 8 05 2 9 15 1 54 1 3 78 4 86 2 2 39 9 08 3 6 18 3 94 2 6 88 7 72 2 4 29 2 20 5 1 17 9 92 B an qu e Fe de ra tiv e du C re di t M ut ue l S A 0 80 5 8 27 80 5 8 27 0 80 5 7 25 80 5 7 25 0 1 5 97 1 97 1 5 97 1 97 0 81 3 4 04 81 3 4 04 In te sa S an pa ol o S pA 90 0 0 00 35 0 0 00 1 2 50 0 00 0 34 9 0 00 34 9 0 00 47 5 0 00 35 0 0 00 82 5 0 00 61 2 0 00 35 0 0 00 96 2 0 00 N or de a B an k A B 66 0 5 54 1 2 93 9 83 1 9 54 5 37 83 1 2 80 1 3 56 6 15 2 1 87 8 95 1 0 84 2 67 1 3 14 6 79 2 3 98 9 46 1 2 40 3 71 1 3 39 5 30 2 5 79 9 01 St an da rd C ha rt er ed P lc 4 2 48 9 01 1 5 82 5 38 5 8 31 4 39 2 9 44 6 74 2 0 15 0 82 4 9 59 7 56 32 0 4 02 3 8 86 8 30 4 2 07 2 32 0 4 4 75 0 62 4 4 75 0 62 C om m er zb an k A G 0 1 3 92 0 00 1 3 92 0 00 0 1 3 81 0 00 1 3 81 0 00 52 1 0 00 1 3 86 0 00 1 9 07 0 00 48 5 0 00 1 4 25 0 00 1 9 10 0 00 K fW G ro up 25 0 0 00 1 5 72 4 00 1 8 22 4 00 26 0 0 00 1 2 02 2 00 1 4 62 2 00 26 9 0 00 2 2 42 7 00 2 5 11 7 00 25 0 0 00 1 5 36 6 00 1 7 86 6 00 D an sk e B an k A /S 21 4 4 48 1 5 81 5 57 1 7 96 0 06 67 0 30 3 1 37 0 24 3 2 04 0 54 37 6 0 76 1 7 99 7 93 2 1 75 8 69 33 5 0 04 1 8 35 8 21 2 1 70 8 25 A B N A M RO G ro up N V 0 97 5 0 00 97 5 0 00 0 1 0 69 0 00 1 0 69 0 00 0 1 0 62 0 00 1 0 62 0 00 0 1 0 50 0 00 1 0 50 0 00 N at io nw ide B ui ldi ng S oc ie ty 1 1 95 9 32 1 5 62 3 09 2 7 58 2 40 1 1 70 6 85 1 5 37 7 23 2 7 08 4 08 1 6 58 7 50 1 6 42 0 59 3 3 00 8 09 2 6 29 6 07 1 8 36 6 37 4 4 66 2 44 Sk an di na vi sk a E ns ki lda B an ke n A B 52 3 3 05 1 0 76 5 56 1 5 99 8 60 47 4 6 53 1 1 13 6 57 1 5 88 3 10 51 4 1 06 1 1 26 3 71 1 6 40 4 77 63 0 2 85 1 2 79 9 39 1 9 10 2 24 M ea n 1 2 83 2 88 1 5 09 5 17 2 7 92 8 04 1 4 99 7 30 1 6 60 7 39 3 1 60 4 68 1 6 45 1 78 2 1 30 4 56 3 7 75 6 34 1 5 70 8 36 2 0 55 2 27 3 6 26 0 63 M ed ia n 61 1 5 62 1 4 77 1 54 1 8 88 4 69 86 1 1 37 1 3 68 8 08 2 1 35 5 92 90 0 3 13 1 6 19 6 28 2 3 87 6 87 88 2 7 29 1 5 36 6 00 2 7 80 1 25 In cr ea se /d ec re as e p er y ea r ( m ea n) 0, 00 % 0, 00 % 0, 00 % 16 ,8 7% 10 ,0 2% 13 ,1 6% 9, 70 % 28 ,2 8% 19 ,4 6% -4 ,5 2% -3 ,5 3% -3 ,9 6% In cr ea se /d ec re as e p er y ea r ( m ed ia n) 0, 00 % 0, 00 % 0, 00 % 40 ,8 1% -7 ,3 3% 13 ,0 9% 4, 55 % 18 ,3 2% 11 ,8 0% -1 ,9 5% -5 ,1 3% 16 ,4 4% N ot e: Re ga rdi ng D eu ts ch e B an k, a n a ve ra ge v al ue h as b ee n u se d b as ed o n t he tw o c o-C E O s r em ue nr at io n.
Table 4.1 displays the total CEO remuneration together with the categorization of fixed and variable remuneration. The remuneration is shown for each bank concerning all the sampled years. A mean and a median value has also been calculated collectively for all banks in each sampled year. To see how the CEO remuneration has changed when compared to the previous year, two columns showing that increase or decrease in percentage for the median and mean values have been included.
When looking at table 4.1, it clearly demonstrates that the average variable remuneration has increased from the year 2012 to 2014. Meanwhile, during 2015 the average variable remuneration has decreased compared to the previous year. It is important to have in mind that when comparing the average variable remuneration in 2015 with the corresponding remuneration in 2012, there is still an increase with almost €300 000 even though the variable remuneration has experienced a decrease in 2015. Similar findings are noticed regarding the median, as the median variable remuneration has experienced an increase of nearly €270 000 from the year 2012 to 2015.
Correspondingly, the fixed remuneration demonstrates an equivalent development when it comes to the average fixed remuneration as it has increased throughout the years 2012 to 2014 with a small decline in 2015. Even though the fixed remuneration has experienced a decline in the year 2015, it has still experienced an increase of approximately €500 000 over the sampled period. In contrast to the average fixed remuneration, the median fixed remuneration has fluctuated over all sampled years but there has still been an increase of almost €60 000 during the time period of 2012 to 2015. In the observation of the variable and fixed remuneration, it is apparent that both the variable and fixed remuneration has increased since the implementation of the bonus cap.
To examine what the increase in both the variable and fixed remuneration comprises of, Appendix 2 has been observed. When looking at the components regarding the variable CEO remuneration found in Appendix 2, it is hard to see a clear pattern. Some of the banks have experienced changes regarding the variable remuneration. These changes comprise of both increases and decreases in different directions for the various sampled banks during the sampled period. Meanwhile, the components accruing to the variable remuneration in a few banks, have not experienced major changes or no changes at all.
In opposite to the variable remuneration, it is easier to see what the increase in fixed remuneration comprises of. When looking at Appendix 2, it is clear that additional components have been added to the fixed remuneration from 2014 such as allowances, role based pay, fixed share awards, but also that current components have risen. These changes have in turn raised the fixed remuneration in several of the banks after the implementation of the bonus cap.
4.2 Evolvement of variable-fixed remuneration
Table 4.2 displays the variable remuneration for each bank in relation to the corresponding fixed remuneration during the sampled years. This table is provided to examine how the ratios have changed since the implementation of the bonus cap in order to see if the variable remuneration has become more constant to the fixed remuneration. When examining the retrieved ratios and their evolvement, several banks have had a change within the range of ±0,1 after the implementation of the bonus cap. Whereas the change was significantly larger before the implementation of the bonus cap. In order to find out if the banks have experienced constant changes in their variable remuneration, a change of ±0,1 has therefore been regarded as a constant change. When looking at changes up to a maximum of ±0,1
Table 4.2: Variable-to-fixed remuneration
Name of the bank 2012 2 013 2 014 2 015
HSBC Holdings Plc 1,99 2,17 0,81 0,71 BNP Paribas SA 1,46 1,71 1,22 1,83 Credit Agricole SA 0,45 1,06 0,99 0,97 Deutche Bank AG 1,17 1,56 0,58 0,00 Barclays Plc 0,81 0,00 1,18 1,19 Societe Generale SA 0,00 1,08 0,73 1,13 Group BPCE 1,01 1,60 1,53 1,49
Royal Bank of Scotland Group Plc 0,00 0,00 0,24 0,55
Lloyds Banking Group Plc 0,78 1,82 2,43 2,28
UniCredit SpA 1,27 0,68 0,62 1,11
Banque Federative du Credit Mutuel SA 0,00 0,00 0,00 0,00
Intesa Sanpaolo SpA 2,57 0,00 1,36 1,75
Nordea Bank AB 0,51 0,61 0,82 0,93
Standard Chartered Plc 2,68 1,46 0,08 0,00
Commerzbank AG 0,00 0,00 0,38 0,34
KfW Group 0,16 0,22 0,12 0,16
Danske Bank A/S 0,14 0,02 0,21 0,18
ABN AMRO Group NV 0,00 0,00 0,00 0,00
Nationwide Building Society 0,77 0,76 1,01 1,43
Skandinaviska Enskilda Banken AB 0,49 0,43 0,46 0,49 Note: Regarding Deutsche Bank, an average value has been used based on
between the different sampled years, the most constant changes occur between the years 2014 and 2015 as eleven of the sampled banks are experiencing a change within this range. Meanwhile, only seven banks experienced a similar change between the years 2013 and 2014 compared to eight banks between the years 2012 and 2013.
4.3 Evolvement of correlation between pay and performance
Table 4.3 demonstrates a Pearson correlation, enabling the authors to reject or accept the stated null hypothesis in section 3.3. The aim of a Pearson correlation in this study is to show the strength of linkage between variable pay and profit before tax for the sampled years. The different rows that are used are representing the variable pay for the chosen years. These rows are in turn divided into three different components. The first component, the Pearson correlation, is showing the correlation between the two different variables. If the correlation is strong, the value found should in this case approach a value of 1. The significance, referring to the second component, is displaying a p-value. A p-value is showing the probability of improperness regarding the obtained correlation. If the probability of improperness is high, it is difficult to conclude that a correlation exists between the two variables. A level of 0,10 (10%) is used in this study when determining if the p-value is significant or not, meaning that if the p-value is below 10% a correlation exists between variable pay and profit before tax. N, the third row, equals the sample of 20 banks and will remain the same for all sampled years.
Table 4.3: Correlation between variable pay and profit before tax
Profit before tax Profit before tax Profit before tax Profit before tax
2012 2013 2014 2015
Variable pay 2012 Pearson Correlation 0,324 Significance 0,163
N 20
Variable pay 2013 Pearson Correlation 0,501*
Significance 0,025
N 20
Variable pay 2014 Pearson Correlation 0,182
Significance 0,442
N 20
Variable pay 2015 Pearson Correlation 0,025
Significance 0,918
N 20
As seen in table 4.3, the Pearson correlation between the variable pay for 2012 and profit before tax for 2012 equals 0,324. In order for a perfect correlation to be present this number should be 1. This implies that it exists a correlation but it is not strong since the value of the correlation of 0,324 is far from 1. However, the p-value equals 0,163 meaning 16,3% probability of improperness regarding the correlation. Since 16,3% is fairly close to 10% it means that it is not far from being significant. Continuing, the Pearson correlation between the two variables during 2013 is 0,501, which is a value that is relatively close to 1. The fact that a correlation probably exists is confirmed by the p-value that is 0,025 or 2,5% showing that there is a minor probability of improperness. Since the value of 2,5% is below 10% it is significant, which implies that a correlation exists between the variable pay and profit before tax in 2013. In contrast, the Pearson correlation in 2014 is equivalent to 0,182, which is distant from 1. The p-value is 0,442, which implies that the likelihood of improperness is great, disenabling the existence of a correlation. Lastly, the Pearson correlation in 2015 is at the low level of 0,025. Meanwhile, the p-value is 0,918 or 91,8% indicating that the probability for improperness is substantial and that no or an insignificant correlation is present.
5. Analysis
5.1 Evolvement of CEO remuneration
The two articulated objectives with the bonus cap are as mentioned to decrease the amount of remuneration and to reduce excessive risk-taking (European Parliament, 2013b). Risk-taking has been a widely-discussed problem for financial institutions specifically since the financial crisis. This is a consequence of the excessive risk-taking by CEOs in their aim to obtain short-term gains. These short-term gains were given as high bonuses that did not reflect the quality of their investments in terms of the banks’ performance (Nguyen & Shlomo B, 2013; Angeli & Gitay, 2015).
When looking at the findings in table 4.1 it is apparent that the total level of remuneration has increased from the year 2012 to 2015, even though it has experienced a small decline in 2015. The rise in the CEO remuneration includes both an increase in the fixed and variable remuneration, where the increase in total amount is higher regarding the fixed remuneration. This implies that one of the objectives with the bonus cap, which strives towards decreasing the level of remuneration, has not been fulfilled. An increase in remuneration is in line with one of the predictions by Murphy (2013) who argues that the variable remuneration will increase as a consequence of an increased fixed remuneration. Consequently, Murphy claims that the excessive risk-taking will have the opposite outcome compared to what the bonus cap is striving for, as the risk taking by CEOs will experience an increase rather than a decline. The increase in excessive risk-taking will be a result of a higher fixed remuneration as CEOs will be able to engage more in excessive risk-taking since their total remuneration is not impacted to a larger extend if the investment does not succeed (Murphy, 2013). Based on the findings, this study cannot answer if Murphy’s predictions about the excessive risk taking is true. Nevertheless, it is safe to assume that the risk-taking might experience an increase as a consequence of the increased fixed remuneration.
Banks have come up with several ways to circumvent the bonus cap by increasing the fixed remuneration in order to camouflage the excessive variable remuneration (for a detailed overview see Appendix 2). One example found is that the use of allowance pay has escalated, mainly by the UK banks, which seems to be an attempt to increase the fixed remuneration. Allowance pay in 2012 was used by two banks in the research sample, meanwhile in 2015 the use of allowance pay had increased to six banks. An advantage with using allowances is that this component can be adjusted over the year. Allowances can enable banks to pay their CEO a high total remuneration as previously, since the variable component can be increased in relation to the increased fixed remuneration. Other banks have tried to circumvent the bonus cap by adding another additional component, or by increasing a current component after the implementation of the bonus cap, in order to increase the fixed remuneration. These findings are in line with another of Murphy’s (2013) predictions, where he claims that new components will be added to the fixed remuneration.
Another assumption made by Murphy (2013), is that the variable remuneration will not continue to fluctuate to a great extent over the years, as a result of the bonus cap. The findings in table 4.1 displays that this assumption may not be incorrect as the increase/decrease in percentage per year for the mean variable remuneration from the year 2014 to 2015 has not fluctuated as much as the previous years in the sampled period. A lack of great fluctuations implies that the change in variable remuneration is becoming more stable. When looking at table 4.1, it is clear that the years before the implementation of the bonus cap experienced a greater fluctuation than the years after the implementation of the bonus cap. Since the sampled period only includes two years that can be analysed after the implementation of the bonus cap, the study is not provided with sufficient data to produce a valid conclusion whether Murphy’s assumption is correct or not in the long-term. Also, when using the mean value for the sampled banks to analyse the fluctuation, the value analysed will not be representative for all of the banks as the amount of variable remuneration differ greatly between the chosen banks. A possible explanation for why the values will not be representative could be that some banks provide their CEOs with a large variable remuneration whereas the CEOs at other banks do not receive an equally large amount of variable remuneration. The mean value will also be affected by the banks that are not providing their CEOs with a variable remuneration for some years. This is mainly a result of the changing of CEOs and it is the reason why the median value has also been used in the
upcoming paragraph since it will exclude the many zero values which are impacting the mean value.
When analysing the increase/decrease in percentage per year for the median value in table 4.1, it is clear that the prediction made by Murphy (2013) is further supported, as the change in the median value for the years 2014 to 2015 is almost the same. This implies that the fluctuation between those years is not great. As the bonus cap was implemented by the countries in the EU in 2014, Murphy’s prediction about the fluctuation of the variable remuneration can be assumed to be true for this short time frame. Similarly to using the mean value, the time frame must be taken into consideration, which prevents a valid conclusion to be drawn regarding Murphy’s assumption about the fluctuation of the variable remuneration in the long-term. Another aspect to consider while analysing the findings is that the changing of CEOs, that some banks are experiencing during the sampled period, may affect the variable remuneration provided by some banks during those years. This can affect the validity of both the mean and median value of the variable remuneration as certain banks are not providing their CEOs with a variable remuneration during some years. The importance of having a remuneration consisting of an appropriate level of variable and fixed remuneration has been highlighted by Jensen et al. (2004), as it will have positive consequences for the banks. Some of the positive consequences are to attract and retain a valuable CEO as well as encouraging CEOs to strive towards great performance. When looking at table 4.1 it is clear to see that the majority of the chosen banks have tried to circumvent the bonus cap by increasing the fixed remuneration, which will allow them to increase the level of variable remuneration. A possible explanation for this increase in fixed pay is that the banks have realized the importance of being able to provide CEOs with an attractive total remuneration that will enable the banks to retain their CEOs. According to Murphy (2013), CEOs that do not receive a competitive market remuneration will switch employer. By increasing the total remuneration, it can be seen as a solution for banks to retain their CEO.
The managerial power theory can provide a possible explanation for why the total remuneration has increased since the implementation of the bonus cap. The theory proposes that it is the level of power that the CEO has that influences the remuneration (Bebchuk & Fried, 2003). This indicates that it is the power of the CEO that can influence the
