MASTER INTERNATIONAL BUSINESS AND MANAGEMENT
EXECUTIVE COMPENSATION,
CORPORATE GOVERNANCE
AND FIRM PERFORMANCE
A CASE STUDY ON THE UNITED STATES
BY:
GAYATRI CHANANA (1412809) UNIVERSITY OF GRONINGEN
FACULTY OF MANAGEMENT AND ORGANIZATION LANDLEVEN 5, 9749 AD
GRONINGEN, THE NETHERLANDS G.Chanana@student.rug.nl SUPERVISORS: Dr. F. A. A. Becker-Ritterspach Dr. A. B Kibriscikli –Özçandarli AUGUST 24, 2007 Abstract
TABLE OF CONTENTS
1. Introduction... 5
2. Literature Review ... 7
2.1 Foundations of Organizational Theory ... 7
2.2 Economic Model of Man versus Self-Actualizing Manager ... 8
2.3 Agency Theory... 9
2.4 Stewardship Theory ... 11
2.5 Comparison of Agency Theory and Stewardship Theory... 13
Table 1: Summary of differences between agency theory and stewardship ... 15
2.6 Manager Power Theory... 15
2.7 Linking Theories ... 16
2.8 Executive Compensation and Firm Performance ... 18
2.9 Factors Affecting Remuneration... 20
2.10 Corporate Governance and Firm Performance ... 22
3. Conceptual Model ... 27
3.1 Remuneration Issues in the Business Cycle... 27
3.2 Keynes Investment Theory ... 28
3.3 Business Investment Theory ... 29
3.4 Merging Keynes Investment Theory with Business Investment Theory... 30
4.3.4 Control Variable... 43
4.4 Data collection ... 43
4.5 Sample... 44
5. Results... 47
5.1 Hypothesis 1a... 47
Table 2: Summary Hypothesis 1a ... 48
5.2 Hypothesis 1b... 50
Table 3: Summary of Hypothesis 1b ... 51
5.3 Hypothesis 2... 53
Table 4: Summary Hypothesis 2... 53
6. Discussion ... 55
7. Limitations ... 61
8. Conclusion ... 65
9. References... 67
1. INTRODUCTION
Executive compensation has generated a spirited debate among academics and practitioners alike for at least 75 years (Gomez-Mejia and Wiseman, 1997). Until the end of the 1990s, little explanative literature has been published regarding executive compensation with the basis of economic efficiency. The focus on compensation has been activated by the principal-agent theory established by Jensen and Meckling (1976), where it is stated that managers (agents) of an enterprise guide their activities in such a way as to maximize the monetary well-being (Lewellen and Huntsman, 1970) of their shareholders (principals). However, the separation of ownership and control in companies created stimulus for executives to follow their own interests and thus, maximize their own utility level instead of their shareholders’. This is commonly expressed as the agency conflict in literature (Alshimmiri, 2004). To reduce this agency conflict, shareholders have two options at hand on how to stimulate their executives and align conflicting interests. One mechanism is of extrinsic nature, financial compensation, and the other mechanism lies within the firm itself, the board of directors, which is a basic element of corporate governance (Agarwal and Knoeber, 1996). Both the extrinsic and the intrinsic factors are believed to impact company performance.
Consequently, this paper presents a framework that combines the predominant managerial models of agency theory, manager power theory, and stewardship theory with the business cycle theory. The objective of this conceptual model is to determine whether the practices in managerial pay setting process are conditioned by business performing cycles.
In addition to the external control mechanism that mitigates agency conflicts, the board of directors is a vital internal instrument that can be used to align the interests of the owner and the agent. Due to the functions that are given to the board of directors, they can actively help in curbing agency costs, which enables the company to reach the company goal, namely maximizing profits. Literature is rich with studies showing the effect of the corporate governance on firm performance through its main dimensions such as structure, composition and process.
2. LITERATURE REVIEW
The literature review will help the reader to understand the main theoretical framework of this research paper. The research presented in this paper is of twofold nature; firstly, the external control mechanism of executive remuneration will be investigated, followed by an exploration of the role of the corporate governance control mechanism.
In order to establish a solid basis for the argumentation in this thesis, the reader will first be introduced to the foundations of organizational theory and the existing managerial theories of agency theory, stewardship theory, and manager power theory. Afterwards, the external control mechanism of executive remuneration is illustrated by reviewing the current academic literature in this field. Here, the important concept of pay for performance comes into play as a method to curb agency conflicts (McConvill, 2006). Following, the internal control instrument represented by the board of directors will be investigated upon, together with its impact on firm performance. The board of directors is in general seen as a vital internal corporate governance structure as it is their role to supervise executives and thus, accomplish the mandate of the company.
2.1 Foundations of Organizational Theory
Due to the increase in the number and diversity of shareholders in ever-larger getting corporations, a shift from owners administering their own business, to managers taking over the operations of the company for the increasing number of stakeholders of the firm, was already detected during the 1920s. This was a conclusion of Berle and Means to the phrase “the separation of ownership and control” in their book: “The modern
Corporation and Private Property” in 1932. Already back then, the authors cautioned
Challenges to develop theories and codes of management are, however, not that recent, with the 19th century industrial revolution giving rise to the need for a systematic approach to management. Henceforth, this gave need for a model that was critical to the design of organizations. Moreover, McGregor (1960) and Argyris (1973) put forward that insights of sociology and psychology - based on the assumptions of Maslow’s (1970) hierarchy of needs – could be used to strengthen the classical organization theory of scientific management. Both academics argued that society has direct implications for managing human behavior in organizations, which consequently led to the creation of the economic model of man and the self-actualizing manager. The next paragraph will deal with these types of managers more carefully.
2.2 Economic Model of Man versus Self-Actualizing Manager
With respect to the economic model of man, a rational self-centered actor will act to maximize his individual economic gain (Donaldson and Davis, 1991). Thereby, he will compute the likely costs and benefits that help him to evade monetary penalty, which in turn enable him to achieve a high remuneration. What is furthermore characteristic about such a rationally minded performer is that he takes an individual decision-making orientation. However, according to McGregor (1960), this individual person shows an inherent dislike towards work; thus, will avoid it, whenever possible. Hence, the self-centered actor must be pressured with punishment in order to put forth adequate effort towards the accomplishment of organizational goals. In all, this kind of manager prefers not to be guided, wishes to avoid responsibility and has relatively little ambitions regarding the organization, and wants security above all, which signals his risk-aversion.
with the corporation, non-financial motivation factors such as self-esteem are most probable to promote the individual’s higher performance. Under proper conditions, the individual accepts and seeks for responsibility, and thus, exercises self-control in the service of objectives to which he is committed concerning the company.
On the basis of the models of man described above, alternative management modes have been specified as scholars aimed at defining principal-agent relationships. As a result, this paper will deal with three dominant management models that are deemed important for the scope of this thesis. These management modes are agency theory, stewardship theory and executive power theory, which try to explain the relationship of a manager with the shareholders more in detail. These theories will be explained in the following section.
2.3 Agency Theory
Wright et al. (2001) affirm that agents are assumed risk averse because agent’s employment security and income are inextricably tied to a single firm; therefore, agents may make decisions in a careful manner so to lower the risk of losing personal wealth. Consequently, agents may show a failing to select projects that generate a positive long-term net present value, hence indicating a loss in value-creation for the principals.
Davis et al. (1997) suggest that in the economic model of man, managers (agents) are appointed to operate as agents of the shareholders (principals), i.e., to carry out business activities for the shareholders that increase their welfare. Since the agent’s activities provide him with value (e.g., remuneration) because of his choice to be of service to this company and not another, he should strive to achieve the highest utility for his shareholders. However, often managers knowingly do not act in the best interests of their firm; this is universally known as the agency problem (Besanko et al., 2004). According to Bebchuk and Fried (2004), executives are most likely to track their own interests at the cost of the company, and thus cannot be trusted. Naturally, when the beneficial terms that can be gained by each party deviate from each other, then agency costs will be part of the corporate governance function.
transferring risk to the agent by basing the contract on the agent’s outcomes, i.e., incentive alignment in terms of compensation design (Beatty and Zajac, 1994, Garen, 1994, Zajac and Westphal, 1997). Agency theory considers how rewards can shift the risk from shareholders to managers in such a way as to align the two parties’ interests (Pennings, 1993). However, the question arises to which extent the external incentive or control may be required and can be put into practice.
Control mechanisms are only helpful when they control the executive behavior in such a way as to link managerial actions with company performance. However, agency theory does not recognize contextual factors, which could result in managerial behavior that aims at goals other than maximizing personal utility level, and thus reduce the degree of control. Hence, this theory needs to be extended in such a way as to enhance the understanding on how the relationship between agents and shareholders can be augmented and thereby lessen the control that is necessary to align the interests of all parties.
2.4 Stewardship Theory
Agency theory provides a useful way of explaining relationships where the parties' interests are at odds and can be brought more into balance through suitable monitoring and a well-planned compensation system. However, stewardship theory can be used to explicate other types of human behavior (Ong et al, 2006) as it attempts to broaden the narrow focus of the self-interested behavior of agents.
of the company’s goal. Being challenged to do so, he not only maximizes his own utility level but also the wealth of his principals as the steward aligns his actions with the objectives and mission of the business. Therefore, control mechanisms may perhaps not be as effective as compared to the agency theory case given that steward’s work is based on trust rather than control. That is why the discrepancy between principals’ interests and stewards in this framework is naturally reduced (Lee and O’Neill, 2003). Preferably, a steward should be given as much autonomy as needed that enables him to self-actualize under the company’s empowering governing structures, while striving for the enterprise’s profits (Davis et al., 1997).
According to Thomas (2002), four intrinsic rewards exist that motivate executives: a sense of meaningfulness, a sense of choice, a sense of competence or quality and a sense of progress. Hence, as a person becomes more senior or has a long tenure in the firm, money as such becomes less of an inducement to work harder. Seligmann (2002) accentuates that an executive’s career entails a deeper personal investment in work by marking the achievement in terms of money and realization of goals. This implies that money is not the sole performance motivator for executives, but is accompanied by the wish to grow professionally and to make a difference within the corporation. Furthermore, the steward recognizes that if he ignores the restrictions on behavior and interests compelled by society, then he is in danger of losing possessions that are more valuable to him than acquired wealth, such as reputation and liberty (Gomez-Mejia et al., 2005).
cooperation are cherished indisputably to improve corporate performance. In turn, it can boost an executive’s well-being when granted with trust, which again may have positive implications on the relationship between the executive and the shareholders, and thus, firm performance. Creating a close and interdependent relationship enables both shareholders and corporate governance to facilitate the well-being of the executives so to align the interests of all parties.
In line with the argumentation above, there might be little confirmation that executives will perform better when their pay is strongly tied to performance or when they are exposed to intense control, especially in cases where intrinsic motivation (e.g. executive’s satisfaction in performing the job) surpasses extrinsic motivation (e.g. receiving a performance related bonus) as stated by McConvill (2006). Furthermore, executives put back their own self-centeredness and make decisions that are affirmative for the interests of the firms’ shareholders in total (Driver and Thompson, 2002). For instance, this can be seen through close cooperation with commissions of enquiry such as the US Securities and Exchange Commission (SEC), whose aim is to protect investors, maintain fair, orderly, and efficient markets, and facilitate capital formation with the executive working with a collaborative and collective interest (Bruce et al., 2005). As a consequence, this might claim more revelation of the executive’s compensation and may even effectively curtail their compensation levels. Thus, these pay for performance sensitivities may not be as applicable as possibly could be assumed in this context. The powerlessness of the pay for performance methodology to amply align the interests of executives and shareholders (McConvill, 2006) can be best explained by the agency theory that neglects on the higher levels of achievement and intrinsic motivation.
2.5 Comparison of Agency Theory and Stewardship Theory
(Wright et al, 2001). In the case of a steward, the executive stands with a cooperative attitude toward the enterprise by interrelating himself with company. Another crucial dissimilarity can be observed in the role of trust. In agency theory, the level of trust is reduced if not eliminated by means of control mechanisms so that the agent avoids opportunistic behavior. On the other hand, stewardship theory assumes that trust is valuable in fostering relationships between the steward and the principals so to align and increase the welfares of each party.
Table 1: Summary of differences between agency theory and stewardship
Theory Agency Stewardship
Managers as Agents Stewards
Approach to governance Economic Sociological
Psychological
Model of man behavior
Individualistic Opportunistic Self-serving Collectivistic Pro-organizational Trustworthy Managers motivated by Own objectives Organizational objectives Manager’s and principal's
Interests Diverge Converge
Structures that Monitor and control Facilitate and empower
Owner's attitude Risk aversion Risk propensity
Principal-Manager
relationship-based on Control Trust
Source: Adapted from Dr. Alfonso Vargas Sanchez (2001)
2.6 Manager Power Theory
There is yet another management mode that has developed over time in academic literature, namely the manager power theory, also known as the executive power theory or rent extraction theory (Murphy, 2002, Aguilera and Jackson, 2003, Granovetter, 1985, Lubatkin at al., 2001). Agency theorists see the managerial power theory as a special case of agency theory in which governance institutions are weak (Bruce et al., 2005). The managerial power theory accepts that corporate governance does not act as rationally as assumed when setting the remuneration package, which stands in contrast to agency theory (Rajagopalan and Finkelstein, 1992).
(Bruce et al., 2005). The ability of executives to exert influence on their own remuneration package is in this context very important, as Bebchuk et al. (2002) empirically display with the existence of ‘rent seeking’ remuneration packages. To be specific, executives often claim to receive performance-related pay, which most of the time legitimates high pay and large pay rises despite attaining the corporate objective. Since these deals yield inefficient pay arrangements and give suboptimal incentives, they consequently impair on shareholder value. When making use of their power at disposal, agents seek not to cross the line of social outrage, otherwise the connection between their rewards and company performance may become too transparent. Therefore, managers need to take care when satisfying their own needs, interests or when enriching them at the costs of the company (Betrand and Mullainathan, 2001). This stands in contrast to stewardship theory where managers do no seek to cross the line of social outrage for another reason, namely reputation. Nonetheless, this argument is preserved in academic literature, which has failed to show an evident positive pay for performance relation that aims at increasing shareholder returns (Bruce et al., 2005). Authors such as Bebchuk and Fried (2004), and Betrand and Mullainathan (2001) stress the use of stock-based compensation, which is considered to be a significant element of the rent-seeking process and which they criticize as lying at the heart of CEO pay.
There are a couple of shortcomings of this theory that need to be explained shortly. First, this theory does not incorporate higher needs as stewardship theory does. Second, with this theory effectuated in management, executives actually detach their interests from the interests of the company to a large extent, as it is assumed in the agency theory.
2.7 Linking Theories
There are several points at hand that need to be discussed more deeply so to understand that these three management modes are in fact closer related to each other then one would initially suspect. Hence, they should be regarded more accurately on a continuum rather than taken on a separate basis.
To start with, there is one important factor that distinguishes agency theory from managerial power theory, namely the assumption of self-interest that is often grouped together with opportunistic behavior in agency theory. To point out, self-interest is not a synonym for opportunism (Donaldson, 1990). According to the former, the agent shows self-interest, which can only manifest as opportunism under solid circumstances (Gomez-Mejia et al., 2005; Jensen and Meckling, 1976). McKechnie (1979) defines opportunism as follows: “opportunism is the adaptation of one’s actions to circumstances in order to further one’s immediate interests, without regard for basic principles or consequences.” Contrary to an executive showing opportunistic behavior, a rational self-interested executive does take into account the consequences of his actions. This is because he understands the importance of long-term cooperation and compromise, which can help him in promoting his interests. Hence, the agent recognizes that he can achieve personal goals more smoothly with the help of cooperation rather than acting on an own account (Olsen, 1971). To recapitulate, the agent-type manager shows characteristics that tend towards a self-interested behavior, whereas the manager of the power theory is inclined to behave more opportunistically since he is heavily occupied with the goal of receiving a high remuneration package without necessarily acting in the interests of the shareholders as described above. This escalates further when information asymmetries apply, ceteris paribus, which complicate the matter for shareholders to determine whether agents bring about opportunistic behavior or not.
To dispute further, also the steward acts to some extent in a self-interested behavior. Gomez-Mejia et al. (2005) state that the steward may pursue self-interests in matters that are more acceptable to society or personal esteem such as independence, prestige and even power. Eventually, these may also create agency costs, albeit to a lesser extent than in the other two management models due to societal pressure.
theory, on the other hand, acknowledges other contextual influences, such that institutional systems may not be as strong as assumed by agency theory. Thus, a balance must be created that recognizes both contextual elements and the theory of principal-agent relationship.
In some contextual settings the executive may be advised to act more as an agent than a steward; naturally, it is also in the interest of the shareholders to control and monitor which management mode is more suitable for their company. For instance, an agent may be required in settings where power structures are more clearly defined and where he can be more risk-averse, for example, in manufacturing settings, whereas a steward may be preferred in more socially embedded settings, where cooperation is asked for, such as in public relations. Hence, it is recommended to the executive to make use of accepted views of maximizing shareholder wealth when making decisions on behalf of them. Moreover, both the executive and the shareholders need to be aware that a mixture of characteristics taken from the management models might enable them to reap the benefits for all parties.
2.8 Executive Compensation and Firm Performance
their own remuneration as the agency theory and manager power theory state. It is obvious that in some situations it is difficult, if not impossible, for the shareholders to specify a quantitative target or adequately monitor performance because of information asymmetry.
Current practice has emphasized on management compensation based on financial targets (Matolcsy, 2000). So far, empirical research has focused on a one-way positive relationship between rising executive pay and rising firm financial performance as a sole determination in the business theories and models of management. Empirically, managerial remuneration is found to have a weak correlation with firm performance. The annual bonus is given in good and in bad times, ceteris paribus. Good performance is supposed to increase the bonus package; however, lowering the bonus does not act as a punishment tool for bad performance, supporting the theoretical background on stewardship. This weak link between managerial remuneration and firm financial performance has been a source of concentrated debate in the scholarly world (Matolcsy, 2000) and financial press (Byrne, 1996). According to Abowd (1990), however, disputes still arise as to whether or not the observed degree of sensitivity of compensation to performance is adequate to solve the principal-agent problem between owners and managers (Abowd, 1990). Firth et al. (1996) for instance, could not prove a relationship between executive remuneration and firm financial performance when taking Norway as their set of sample.
These empirical results have to be interpreted with care as is it is not clear which direction shows more validity. The authors’ use of different methodological approaches and different set of samples as well as variables may be the cause for ambiguous results.
Hitherto, academic literature has often used longitudinal data that cover economic booms and recessions in one study, which are very common in macroeconomics. Consequently, the results of these studies confirm a one-way relationship between compensation and performance. Jensen and Murphy (1990) also dispute that the relation between firm performance and executive remuneration is not “economically significant”. However, their results could underrate the actual relation between company performance and compensation because it includes periods of economic upturns and downturns by using a longitudinal approach (Matolcsy, 2000).
This paper takes another approach in determining the level of compensation in correspondence with firm performance, also to provide another valid set of results that might be less confusing. To be specific, this thesis examines whether remuneration changes during different business performing cycles. It is expected that when firm performance increases, executive compensation package decreases in turn. Vice versa, when firm performance is decreasing, companies will take the bad performance of the firm as a sign to use a higher amount of executive compensation so to stimulate the executive to better performance. To date, these relationships could not be empirically proved with certainty. Hence, this paper attempts to provide evidence on the predicted relationship during two business performing cycles.
2.9 Factors Affecting Remuneration
Much academic dialogue has taken place over the relationship between remuneration and firm performance. Empirical evidence specifies that market performance is an important factor in shaping compensation (Murphy, 1986; Jensen and Murphy, 1992); yet, other authors state the importance of accounting factors such as return on assets (Lambert and Larcker, 1987; Sloan, 1993). Since PRP rewards to executives provide motivation for managers to prompt strategies that boost future stock performance (Coughlan and Schmidt, 1985; Veliyath, 1999), a relation to compensation is expected to exist in general.
There are several firm specific variables that are supposed to have an affect on executive remuneration. In particular, two firm variables will be used for this thesis to check whether they have a positive influence on remuneration, namely the industry the company is active in and the size of the company. With respect to the former control variable, industry, shareholders of the company may wish to make remuneration dependent on performance relative to that of other corporations that are active in the same industry (Holmstrom, 1982; Tirole, 1988; Nicknell, 1995). Simpler expressed, since shareholders want to recompense actions of their executives, but can in fact only observe the output of their company, shareholders can benchmark the profits of their company with the companies in the same industry and thus, control remuneration in order to align manager’s interests with theirs.
The second firm variable that is utilized is firm size. For this measurement, the number of employees will suffice. Rosen’s (1992) argues that the costs of wrong decisions and the benefits of correct managerial behavior are higher the bigger the corporation, and thus, executives require a greater compensation level as they carry more responsibility. Empirically, Firth et al. (1996) find a positive relationship between company size and CEO remuneration.
Furthermore, Mohan and Ruggiero (2007) empirically attest that a lower compensation package attributes to a younger age of the executive.
The last two variables refer to external factors, namely the world and the national economy, which can have an effect on remuneration as well (Matolcsy, 2000).
Considering the first, world economy, for which I took the S & P 1000 Index, it is important to bear in mind that the state of the economy has a high impact on the value of the stock and stock options packages granted to executives. If firm performance is below-average, which is reflected in a relatively low share price, the executive would need to leave the options untouched (Reingold and Jespersen, 2000), which would reduce his wealth.
The national economy is of equal importance since it incorporates national effects, be they political or fiscal. Stock prices of companies that are listed on the NYSE Index can become more sensitive to changes in the NYSE Index (Kaen and Sherman, 1999). As stock prices become transparent by the listing on the Index, companies need to take care more of the volatile behavior of the stock prices and therefore, need to aim for a stock price that is less fluctuating. It is important to understand this relation since stock based compensation represents a significant proportion of executive remuneration in the United States.
2.10 Corporate Governance and Firm Performance
interests. Due to its monitoring function, the board of directors is a vital instrument to reduce agency costs, which is to guarantee that the interests of managers and shareholders are in line.
As put forward by agency theory, the board of directors is regarded as the decisive instrument to control the corporate actions of the managers. The board is assigned to perform the tasks of monitoring and rewarding the managers to ensure that shareholders’ wealth is maximized. Stewardship may be helpful in explaining how the board of directors can ensure that agents are stimulated more to work towards organizational goals. If the board of directors succeeds in facilitating trustworthy relationships with the executives, this might lead to greater cooperative behavior and thus, the executives might be more willing and less risk-averse in realizing organizational objectives besides their own objectives.
Concern over board effectiveness and their weak institutional role, as proclaimed by the manager power theory, has led to changes in board composition and, thus, rising interest in how boards make their decisions. For instance, the Securities Exchange Commission (SEC) and other stock exchanges have deliberately been active in profiling the composition and decision-making process of the board (Bacon, 1979). These calls for reform have resulted in a search for better ways to design effective boards, by looking at different attributes such as the board of directors’ role in structure, the board’s composition, and the process (Zahra and Pearce, 1989).
actions properly. To point out, an effective board structure allows directors to carry out their responsibilities in a timely fashion and to do their job well.
Board composition can contribute (in)directly to company performance. Zahra and Pearce (1989) provide a list of attributes that are based on empirical research. Among these attributes are the size of the board of directors and the mix of director types (dichotomy between inside and outside directors). Size has been often used as a proxy measure for directors’ expertise (Bacon, 1973; Herman, 1981). Outsider’s representation is also increasingly highlighted to have an ambiguous impact on firm performance (Jones and Goldberg, 1982) due to their board independence for management, objectivity on the one hand, and representing multiple perspectives, limited time and general expertise on the other hand. Directors’ background such as age and educational background are other attributes, which are thought to encourage or limit potential effects on firm performance; however, little evidence has been found to be associated with firm performance. Another corporate governance attribute affecting firm performance is considered to be experience of directors, since public legitimacy of the company is expected to rise with the experience of its management (Molz, 1988).
As a final board attribute, the decision-making process of the board affects firm performance. Important indicators of this process are the frequency and length of meetings, level of consensus among directors, and the degree to which the board is involved in evaluating strategies. Kerr and Bettis (1987) also document that the flow of information is an important issue that helps directors contribute meaningfully to the strategic initiates. Furthermore, frequent evaluations result in feedback for corrective moves. Bacon & Brown (1975), for instance, state that ineffective decision making processes weaken boards and limit their contribution to the company performance. As a result, board processes have come under public inspection to guarantee that boards are not nodding through managerial choices (Fleischer et al., 1988; Kohls, 1986).
by many authors (Alshimmiri, 2004). There has been limited empirical research examining the relationship between CEO compensation and board composition (Cahan et al., 2005).
Theoretically, it is expected that the form of board of directors, which has been investigated upon in financial economics, has an impact on firm performance (Zahra and Pearce, 1989; Alshimmiri, 2004). Board size has been found to influence the control role of the board of directors (Kesner et al., 1986). In a small board of directors, coordination and communication are supposed to be exceptionally effective, thereby lessening decision-making problems. This in turn can positively affect firm market performance. Looking from the other angle, large boards may be inclined to include directors with diverse skills and backgrounds. However, efficiency matters in smaller board of directors are expected to supersede the efficiency of the decision making process of larger boards in terms of promoting shareholders interests. Magnan, St-Onge and Calloc’h (1999) show that many corporations reduced the size of the board to make it more effective in terms of coordination and communication. This implies that a small board size enhances the communication efficiency and thus promotes shareholders’ interests. Hackman (1990) finds that smaller groups are easier to coordinate, and free riding becomes less likely in a small size. Yermack (1996) provides proof displaying that firm value is greater when board size declines. He also finds that CEO compensation is more sensitive to performance and that dismissal of poor performing CEO’s is more likely when boards are small.
tend to be more heterogeneous concerning, for instance, their background, value and skills, they may be better able in discerning corporate goals from individual ambitions, and thus present shareholders interests more adequately. Thus, larger boards may be more active in their monitoring role and in evaluating managers.
Some other authors such as Talmor and Wallace (2000) find that board size is not significant at all. Yermack (1996), another scholar, states that there is an inverse relationship, U-shaped, between board size and firm performance. Since the statistical approach for finding empirical proof for the last documented relationship is beyond the scope of this paper, other researchers are invited to examine this relationship more deeply. All the same, there is currently no persuasive argument explaining in which way board size affects firm performance.
3. CONCEPTUAL MODEL
Given the discussion on the management models above, it is obvious that academics are not certain whether agency theory provides the best solution in explaining the relationship between compensation and corporate performance (Gomez-Mejia, Wiseman and Dykes, 2005). In other words, other management modes have to be recognized, which take into account social contexts and thus ask for social solutions that can help to remove agency conflicts, as stewardship theory recommends. Naturally, this does not imply that agency theory becomes valueless, but rather that another model has to emerge that finds a valid and strong foothold for understanding the role of agents in diverse circumstances. Furthermore, since companies are affected by changes in their business performing cycles, shareholders have to find an approach on how to leverage control over their agents in times of changing corporate performance. The suggested management models can help in explaining this pay for performance relationship.
This section presents the conceptual model for this paper, in which the relationship between executive remuneration and corporate performance is examined under the light of Keynesian investment theory.
3.1 Remuneration Issues in the Business Cycle
investment theory incited by Keynes (1936) will be explored in the following paragraph to shed light on the cyclical remuneration pattern.
3.2 Keynes Investment Theory
In macroeconomics, economic fluctuations depict a frequent problem for economists and policymakers. The famous economist John Maynard Keynes (1936) actively promoted an interventionist government policy that could be effective in alleviating market frictions. The government could make use of fiscal and monetary measures to mitigate the adverse effects of economic recessions, depressions and booms. Keynes wanted the government financial plan to be in equilibrium, through either cutting expenditure or raising taxes. Following either policy would enhance the savings and thus lower the demand for both labor and consumption. However, the insufficient buying power in turn would inevitably cause a depression. As a consequence, Keynes (1936) did not see an unbalanced government plan as incorrect; rather, he was in favor of an anti-cyclical fiscal policy that acted against the wave of the business cycle. For instance, raising taxes in order to dampen down the economy and to prevent inflation was considered as one possible solution when growth on the demand-side was overflowing, despite the budget surplus. In contrast, Keynes (1936) supported increased deficits when an economy was suffering from recession or when recovery was obstructed. Examples of this policy are cases where the governments used deficits to finance wars.
enabling a net private income gain. According to Keynes (1936), an increase in individual income will push the private sector to increase spending, thereby enhancing the revitalization of the economy.
In times of economic prosperity, Keynes argued that governments should lessen their expenditures and instead increase their tax revenues. By lowering expenses, the government depresses private excessive spending. Further, through the higher tax expenses, the private income level decreases as a result. When the private income level decreases, individuals in turn have less capital to maintain the high level of expenditures during economic growth. The effect is a slowdown of the disproportionate expenditure and a shift of the economy away from a boom situation towards equilibrium.
Ultimately, the anti-cyclical investment pattern enables governments to fuel economies in times of recession and to control the economic boom by slowing down private expenditure. By means of leveraging incomes, the upswing and downswing economic shocks are recovered. So to say, the government should work out short-term problems rather than waiting for market forces to resolve it. Fiscal stimulus can in fact enhance the market for business output and thus prompt business buoyancy, thereby inducing government and business to act as complements rather than substitutes.
3.3 Business Investment Theory
Analogous to the countrywide economies, businesses in these economies are also faced with fluctuations in their market performance. For example, if one takes the stock market as an indicator for firm performance, one can see that the price of a stock of any particular firm varies over time. However, according to agency theory; the main goal of management is to maximize the value of the shareholders of a firm (Jensen and Meckling, 1976).
remain to deal with changeable performance are positioned in the firm itself. That is why organizations require an internal tool that sways fluctuations in firm market performance. One important device that is available to firm’s lies in the dynamics of executive wage setting.
3.4 Merging Keynes Investment Theory with Business Investment Theory
When fusing business theory with Keynesian theory one can argue that firms should save on expenditure during times of good market performance to create a wage cost buffer, which the company can use in times of economic downturn so to stabilize the compensation package. In other words, shareholders can vote for a low but steady executive remuneration package when firm performance increases. In times of bad performance, the company can decide to use the saved remuneration costs to increase the motivation of its executives and thus, encourage market performance.
As for all economic models, also this model has several drawbacks that need to be taken into account. The Keynesian idea has been heavily criticized by the free market economist Friedrich Hayek (1944). Hayek (1944) argued that despite the utilitarian intentions the theory proposed by Keynes requires central planning, which in turn may lead to totalitarian abuses, as the example in Germany during WW II shows. Another imperative censure is the fact that Keynes studies of economy by relations between aggregates is erroneous, and that recessions are mainly caused by micro-economic factors. Hayek (1944) additionally claims that what starts as a temporary government “repairs”, may turn into an enduring and spreading out of government programs, which sequentially suppresses the private sector and civil society.
paper seeks to investigate whether shareholders use managerial compensation as a tool to induce their agent(s) to work in the interest of the shareholders, and thus, sustain a steady, yet rising market performance of the firm.
Further, due scarcity on the topic of corporate governance and its relation with firm performance, this paper will also focus on this internal tool for shareholders to control executives and help, improving firm performance. Also here, the management modes play a crucial role. In agency theory, harder control mechanisms are required to make sure that the interests of the shareholders and the agent converge. The board of directors must be very active and try to restrict the power given to the agent so that he cannot take advantage of his power. What’s more, the agent has to be induced in such as way as to act pro-organizational. In contrast, when the agent shows a trustworthy behavior and works towards organizational objectives, then the control mechanism, the board of directors, may not be as effective or required to be extremely active in controlling the agent in pursuing shareholders’ interests.
Figure 1: The Impact of the Board of Directors on Firm Performance (internal control)
As already mentioned, shareholders have two options at hand on how to control the manager of their company. One of the options that the shareholders have is the choice to make use of an internal control mechanism, corporate governance or board of directors,
and implementation of strategy of the manager and intervenes in cases where the interests of the managers start to diverge from the interests of their shareholders. Like this, the board of directors can ensure that appropriate steps are taken by the manager so to enhance the firm performance.
Figure 2: The Impact of Executive Remuneration on Firm Performance (external control)
The other option, executive remuneration, enables the shareholders to align their interests with the interests of their manager by means of tying the managers’ remuneration to firm market performance. In this instance, the manager has no other option than giving as much effort as possible and improving the performance of the company so to guarantee his remuneration as agency theory predicts. As Figure 2 exhibits, it is expected that when firm performance increases, executive compensation package decreases in turn. Vice versa, when firm performance is decreasing, companies will take the bad performance of the firm as a sign to use a higher amount of executive compensation so to stimulate the executive to better performance. Since it is in the interest of the shareholders and the manager to level out the market performance of the company, the manager will
Executive compensation decreases Executive compensation increases
Shareholders Firm market performance decreases
understand the motives of the shareholder to reduce compensation in times of economic downturns.
Academically the counter cyclical remuneration model has not been empirically proved on a large scale. However, some evidence of this particular relationship is supposed to be existent and valid, and also documented by both academics and the management press. Matolcsy (2000) mentions an inverse relationship between CEO cash compensation and corporate performance. That is, the poorer the company performance, the higher the cash bonuses received. His results are based on 100 Australian firms for the period 1987-1995. Moreover, in “Management Issues”, Amble (2006) mentions that an inverse relationship between CEO remuneration and performance exists by reviewing a study by pay consultant DolmatConnell & Partners of trends in the TECH 100. The study has found that remuneration increased by almost 16% in 2005 when looking at the relative company performance. This means that remuneration pushes executives in the right direction when company performance is low, for instance.
3.5 Research Questions
Based upon the literature review and the conceptual model, this paper examines whether remuneration changes during different business cycles. More specifically, I will test whether executive remuneration level increases as a result of a decrease in firm performance.
1a)Does the amount of executive compensation decrease when firm market performance increases?
Furthermore, to complete the business cycle, I will investigate whether companies invest more in executive compensation packages in bad times of firm performance so to motivate the executive. The following sub-question will be investigated upon:
Following these specific research questions, the subsequent hypotheses are formulated, which indicate the expected pattern based on the literature review.
Hypothesis 1a: Increases in firm performance leads to decreases in executive remuneration.
Hypothesis 1b: Decreases in firm performance leads to increases in executive remuneration.
Although I am aware that these hypotheses contradict existing literature to some extent, it is expected that testing these hypotheses will disclose information on a highly interesting issue in the financial economics literature. The extent to which remuneration packages align the interests of principals and agents, and are furthermore adjusted during different business cycles, can be examined by analyzing whether there is a relationship between remuneration compensation and firm market performance. If this relationship were found to be of an inverse nature, then this would shed light on a completely new field in academic literature.
Due to empirical scarcity and ambiguous results on the relationship between the size of boards of directors and firm market performance, the following and final question will be examined:
2.) Does the size of the board of directors influence firm market performance?
The hypothesis that explains this pattern is as follows:
Hypothesis 2: The size of the board of directors is negatively related to firm performance.
4. METHDOLOGY
4.1 Methods
The purpose of this paper is to find whether firm market performance affects executive remuneration in an anti-cyclical pattern in the United States and to test in a quantitative way to which extent this is the case. Further intentions of this section are to check whether the size of the board of directors has a direct impact on firm market performance.
In essence, the procedure of deduction might be divided into phases. By testing theory through observation of the empirical world by means of hypotheses, I will be able to either falsificate and discard the Anglo-Saxon laws on principal-agent relationship or create yet unfalsified covering laws that explain and predict causal relationships.
Since this economic model inhibits more than one explanatory variable, a multiple regression model is the best method to test this model. In a general multiple regression model, a dependent variable y is related to a number of explanatory variables x1, x2, …xn through a linear equation that can be formulated as
(1) y = β1* x1 + β2 * x2 + β3 * x3 + …. + βn * xn + ε
The coefficients β1, β2, β3…βn are unknown parameters. The parameter βn measures the effect of a change in the variable xn upon the expected value of y. To allow for differences between the observed and the expected value for y, the random error term ε is added. This random error represents all the factors that cause y to differ from its expected value (Hill, Griffiths and Judge, 2000).
4.2 Model 1
Model 1 will be used to empirically explore hypotheses 1a and 1b. and is defined by the following equation:
EXECUTIVE REMUNERATION =
β1 * FIRM MARKET PERFORMANCE + β2 * INDUSTRY
β3 * NUMBER OF EMPLOYEES + β4 * EXECUTIVE AGE +
β5 * WORLD ECONOMY + β4 * AMERICAN ECONOMY + ε
As concluded in the literature review above, executive remuneration is determined by industry, size of the company in terms of employees, the age of executives, and economic factors, such as the S & P 1000 Index and the NYSE Index. These variables will hence be chosen as control variables. The regression model will be set up with firm market performance as the independent variable next to the five control variables.
When relating the multiple regressions model with the hypothesis stated above, the following relationship between firm performance and executive remuneration is expected based on the anti-cyclical remuneration model:
For hypothesis 1a, a rise in firm performance leads to a decrease in managerial pay. Hence, the beta coefficient for firm performance needs to be negative, so to create a downward pressure on the executive income level.
For hypothesis 1b, a decrease in organizational performance results in a rise in executive remuneration. Thus, the beta coefficient in this scenario must show a negative sign too, so to cause an upward effect on the pay level.
Concluding, in both scenarios, a negative beta coefficient is hypothesized by the anti-cyclical remuneration model.
4.2.1 Operationalization
investigated, resulting in four multiple regressions, there will be also four dependent variables. I want to stress, however, that both the control and independent variables remain equal in every regression.
4.2.2 Dependent Variables
Executive remuneration consists of several components, all of which are taken into account in this study. This enables me to be more specific in judging which parts of income are more closely related to firm market performance:
Executive remuneration:
1. Executive fixed wage plus annual cash bonus. This variable measures the fixed annual compensation and annual cash compensation for each executive over the period 1997-2006.
2. Executive fixed wage plus annual cash bonus plus value of stocks granted. This variable measures the fixed annual compensation, annual cash compensation and stocks granted for each executive over the period 1997-2006.
3. Executive fixed wage plus annual cash bonus plus value of options granted (value of options granted = number of stocks granted * exercise price of option). This variable measures the fixed annual compensation, annual cash compensation and options granted for each executive over the period 1997-2006.
4. Total wealth = Executive fixed wage plus annual cash bonus plus value of stocks granted plus value of options granted. This variable measures the fixed annual compensation, annual cash compensation, stocks granted and options granted for each executive over the period 1997-2006.
4.2.3 Independent Variables
In the United States, market-based performance measures are more suitable as options and stocks represent the largest variable component of executive compensation. Therefore, I will make use of market performance measurements:
Firm Market Performance:
Share Price: monthly share price observations have been transformed into an average annual closing price of the share. Then, the average share prices of all ten years are added and divided by the total number of years in the period, which are ten. Afterwards, each firm’s yearly average share price is expressed as a percentage of the average share price over the total period. Here, a positive value for this indicator implies that the firm is performing above-average, and a negative value implies that the firm is performing below-average.
Market Capitalization: yearly aggregate value of a company, obtained by multiplying the number of shares outstanding by their current price per share. Then, the market capitalization of all ten years are added and divided by the total number of years in the period, which are ten. Afterwards, each firm’s yearly market capitalization is expressed as a percentage of the average market capitalization over the total period. Here, a positive value for this indicator implies that the firm is performing above-average, and a negative value implies that the firm is performing below-average.
4.2.4 Control Variables
Industry:
This variable indicates in which industry each firm in the dataset is active. The firms in the database operate in 15 different industries. Hence, this variable is constructed as a dummy variable, with values ranging from 1 to 15.
Semiconductors; Software & Services; Technology Hardware & Equipment; Telecommunications Services; Transportation.
Number of Employees:
This variable indicates the size of each specific firm in the dataset by looking at the active employees at the end of the company fiscal year.
Executive Age:
Age of executive officer of the registrant.
World Economy:
Monthly S & P 1000 Index observations have been transformed into an annual average Standard & Poor 1000 Index.
American Economy:
Monthly New York Stock Exchange Index observations have been transformed into an annual average NYSE Index.
4.3 Model 2
Model 2, which is devised to test the second hypothesis, has the following equation:
FIRM MARKET PERFORMANCE =
β1 * SIZE OF BOARD OF DIRECTORS +
β2 * RELATIVE INFLUENCE OF SINGLE DIRECTOR + ε
Concerning the second model, the size of the board of directors is the independent variable, and firm market performance the dependent variable. For the control variable, I created a variable that measures the impact of a single director on firm performance
needs to be negative that shows that a smaller board of directors has more impact on firm performance than a larger board of directors. To check for the effect of the influence, the beta coefficient for the control variable needs to be positive, thereby showing that the influence of a single directors is larger in a smaller board.
4.3.1 Operationalization
The following section operationalizes the dependent and independent variables for model 2.
4.3.2 Dependent Variables
Firm Market Performance:
Share Price: monthly share price observations have been transformed into an average annual closing price of the share. Then, the average share prices of all ten years are added and divided by the total number of years in the period, which are ten. Afterwards, each firm’s yearly average share price is expressed as a percentage of the average share price over the total period. Here, a positive value for this indicator implies that the firm is performing above-average, and a negative value implies that the firm is performing below-average.
4.3.3 Independent Variable
Size of Board of Directors:
Number of peoples elected into a board of directors by corporation’s shareholders to oversee the management of the corporation.
4.3.4 Control Variable
Relative Influence of Single Director:
The control variable is created by dividing turnover by board size. With this variable, I expect to differentiate the effect of board size on firm performance more clearly. Here, the relative influence of the board of directors on market performance is measured, not the actual size of the board. When a single director’s influence is minor, given the size of the firm, this implies a relatively large board size. In contrast, when a single director’s influence is extensive, given the size of the firm, this corresponds to a relatively small board size.
4.4 Data collection
Given the fact that this paper uses three units of analysis, the first being the executive, the second unit comprising the firm and the third being economic units, I congregated data from multiple databases.
For information on remuneration, data were drawn from the proxy statements published under the SEC filings for each company for each fiscal year. Hence, I reviewed 330 SEC proxy statements. For the second executive variable, age, data was retrieved from annual yearbooks. The data for both variables ranges from 1997-2006.
firm-level data, share price, needed to be extracted from a different source. DataStream was used to retrieve monthly data, which were then transformed into an average annual share price. DataStream is a database created by Thomson Financial. Access to this database was made feasible through the library of the Faculty of Economics & Business of the University of Groningen. The time frame for all these variables is ten years, 1997-2006.
For the last unit of analysis, economic variables, I obtained data on the S & P 1000 Index and the NYSE Index via DataStream. Again, monthly data was used to take the volatile behavior of the variables into account. The monthly data for both variables was then converted into an average annual index. These variables also have an interval of ten years, dating from 1997 to 2006.
Importantly, only companies that fulfilled the requirement of complete datasets for all ten years were taken. In general, when information could not be retrieved for some elements of the variables used for the research, these values were stated as missing. Furthermore, I controlled for the lagged effect of performance on remuneration by using a one-year time lag. Thus, compensation at time X is regressed to performance at time X-1. For instance, the data for firm market performance and the independent and control variables is taken from the base years 2002 and 2003, and data for executive compensation is taken from the base years 2003 and 2004.
4.5 Sample
out shortly after. Taken as a whole, managers from the United States express a powerful confidence in the motivational usefulness of executive remuneration systems (Pennings, 1993).
Since it is the most important principle behind reliable sampling, only a random sample will guarantee an unbiased result (Hill et al. 2001) . In other words, each member of the population concerned should be given an equal chance of being selected from the sample in order to leave out any bias. Besides this, for reliable results, the sample needs to be representative. Therefore, firms were selected from a listing provided by Forbes on their website www.forbes.com. This list identifies the biggest publicly traded companies in the United States, ranked according to their net income ($ million). They have been selected by Forbes after a thorough review of financial metrics, Wall Street forecasts, corporate governance ratings and other public information. They present a variety of industries in the service sector, such as banking, telecommunication, publishing, as well as the manufacturing sector, such as electronics and food. Furthermore, all these companies are listed on the New York Stock Exchange. Hence, this sample is considered to be a random and representative sample.
The final result of this data collection activity yielded a dataset that compromises 1.623 observations of 438 executives belonging to 30 American stock-exchanged listed firms.
Based on the data collection procedure, I composed six databases. The first two databases consist of data on executive remuneration and above average firm market performance indicators, share price and market capitalization respectively. In the first database, share price, the number of observations is 497. The second database with market capitalization as the performance indicator contains N=553 observations.
The next two databases include data on executive compensation and below average market performance indicators. In the case of share price, the number of observations is 817, in the case of market capitalization, the number of observation counts 901.
5. RESULTS
This section presents the results obtained by the multiple regressions. For reasons of precision, the outcomes are depicted according to the arrangement of the three hypotheses.
As explained in section 4.2.2, the total level of executive remuneration consists of multiple features. Therefore, each hypothesis is tested by means of four regression tests. The first regression takes on the dependent variable, Fixed Wage plus Annual Cash Bonus (EFW+B). The following regression has the dependent variable, EFW plus stocks granted (EFW+B+S). The third dependent variable is EFW plus options granted (EFW+B+O). The final dependent variable measures total executive wealth (EFW+B+S+O). Detailed statistical output can be found in Appendix A.
For matters of clearness, share price will be examined first. Afterwards, market capitalization will be analysed in the sequence of the hypotheses.
5.1 Hypothesis 1a
Table 2: Summary Hypothesis 1a (a) EFW+B (b) EFW+B+S (c) EFW+B+O (d) EFW+B+S+O Hypothesis 1a Coeffici ent Sig. Coeffici ent Sig. Coeffici ent Sig. Coeffici ent Sig. Share Price 0,057 0,231 -0,001 0,991 0,215 0,000 0,205 0,000 Industry -0,188 0,000 -0,142 0,005 0,103 0,040 0,088 0,080 Nr. Total Employees 0,068 0,125 -0,001 0,989 -0,096 0,034 -0,100 0,028 Executive Age 0,158 0,000 0,107 0,021 0,030 0,515 0,038 0,414 S & P 1000 Index 0,416 0,000 0,121 0,217 -0,173 0,072 -0,181 0,061 NYSE Index -0,240 0,011 0,007 0,940 0,086 0,372 0,107 0,270 (a) EFW+B (b) EFW+B+S (c) EFW+B+O (d) EFW+B+S+O Hypothesis 1a Coeffici ent Sig. Coeffici ent Sig. Coeffici ent Sig. Coeffici ent Sig. Market Capitalization 0,089 0,070 0,104 0,033 0,209 0,000 0,215 0,000 Industry -0,163 0,000 -0,179 0,000 0,081 0,077 0,068 0,137 Nr. Total Employees 0,023 0,597 0,001 0,988 -0,054 0,215 -0,055 0,201 Executive Age 0,151 0,000 0,087 0,044 0,048 0,263 0,047 0,278 S & P 1000 Index 0,271 0,009 0,243 0,019 -0,305 0,003 -0,291 0,005 NYSE Index -0,137 0,201 -0,062 0,559 0,140 0,190 0,143 0,183
(a) EFW+B: Executive fixed wage plus annual cash bonus
(b) EFW+B+S: Executive fixed wage plus annual cash bonus plus
stocks granted
(c) EFW+B+O: Executive fixed wage plus annual cash bonus plus
options granted
(d) EFW+B+S+O: Executive fixed wage plus annual cash bonus plus stocks
granted plus options granted
As concerning the first hypothesis, the R square of the model is very low throughout when using the share price as a market performance indicator. The same applies to market capitalization. In general, a low R square implies that the dependent variable executive remuneration can be hardly interpreted by the model. However, in the table of ANOVA, the F-value is of high positive nature, which indicates that the model is statistically significant.
not significant at the confidence level α<0.1 threshold. The model does show positive statistical significance, when options are introduced in the remuneration package and when stocks are granted in addition to options. This means that the when firm market performance is increasing then executive remuneration is also increasing.
In the case of market capitalization, all beta coefficients are positive. The model shows positive statistical significance in all four cases. This implies that executive remuneration increases significantly when market capitalization increases. This outcome stands in contrast to hypothesis 1a, which states that when market performance increases, executive remuneration should decrease as a result. This confusing finding will be explained in the discussion section.
To state upfront, the values for the control variable industry are significant at a confidence level of α<0.1. The findings are nearly the same for the market performance indicator, market capitalization. Only, when total wealth is taken as the dependent variable, then no significant relationship can be detected.
The number of employees does not show any apparent statistical correlation to executive remuneration plus bonuses and when stocks are granted. Yet, EFW+O and EFW+B+S+O reveal negative significant relationships between the control variable and the dependent variable at a confidence level of α<0.05 in the case of share price. In other words, the executive can expect to receive a greater remuneration package the smaller the company is. In the case of market capitalization, no relationship at all can be observed.
When turning to the second last control variable, the world economy, one can distinguish miscellaneous results in the case of share price. There is a positive significant correlation at the confidence level α<0.05 for fixed wage and annual cash bonus, no relationship when stocks are granted, however, a negative statistical relationship when the executive receives options or when considering total wealth.
In the case of market capitalization, the control variable is always statistically significant at a confidence level of α<0.05. In the first two cases, fixed wage plus annual cash bonus and stocks, the relationship is of positive nature. When options are granted or when the total remuneration package is examined, then the correlation gives negative outcomes.
The last control variable, American economy is statistically negatively significant at a confidence level α<0.05 for fixed wage plus annual cash bonus. In all the other cases, there is no evident relationship. With regard to market capitalization, no link is perceived between the dependent variable and the NYSE Index.
It can be concluded that the displayed relationship are of opposite nature, hence, Hypothesis 1a has to be rejected; when market performance increases executive remuneration does not decrease accordingly.
5.2 Hypothesis 1b
