Executive Compensation in the Netherlands

(1)

Executive Compensation in the Netherlands

(Master Thesis)

By Erik Terpstra

(2)

Executive Compensation in the Netherlands

by Erik Terpstra

Master Thesis

University of Groningen, Faculty of Management and Organization Supervisor 1 : prof. dr. D.M. Swagerman

Supervisor 2: dr. E.P. Jansen

Amsterdam April 24, 2006

‘The author is responsible for the content of this master thesis, the copyright of this thesis rests with the author’

(3)

‘Sed quis custodiet ipsos Custodes’

(4)

Abstract

This paper describes the structures and trends in Dutch executive pay contracts. Based on an empirical study on remuneration data of the executive board members of 71 Dutch listed companies over the period 2002-2004, a number of clear trends are identified. It shows that political interference in compensation practices, for example in the form of Corporate Governance guidelines, has great impact on compensation design, however often at the expense of the economical efficiency of the arrangements. It also shows large increases in equity-based pay and a shift from traditional stock options to new forms of long term incentive plans, of which the performance share is the prime example. While this latter shift can be explained by a number of practical factors, like legitimisation, copycat behaviour, regulatory changes and fundamental flaws in option plans, the increases in equity-based pay are harder to interpret, especially since increased discipline on managers as a result of changes in many governance mechanisms, might decrease the necessity of a strong alignment in executive pay. It is clear that many companies have purposely increased this part of the pay package, however if this is due to a great belief in increased incentive effects, the consequences of external benchmarking, or a high degree of managerial entrenchment, is not clear. It does not seem that the degree of equity-based compensation is contingent on firm-specific, manager-specific or situational factors, however future research might give more insights in possible relationships.

The main relevance of this study is rooted in the awareness that the characteristics of a unique corporate governance regime and environmental context can shape compensation arrangements. Based on a theoretical research, which combines the empirical research results, with a literature study, is argued that an optimal compensation arrangement is a trade-off between complying to corporate governance standards, optimizing potential incentives, avoiding perverse incentives and offering a competitive pay package, which will turn out to be a different function for every company in any moment in time.

(5)

Introduction

The compensation arrangements of top executives have been subject to a tremendous amount of research in the last decades. This research has however mainly focused on data from compensation practices in the United States and the United Kingdom. Naturally, this made sense, since compensation data are widely and on a large scale available, compensation practices are most developed and the economic importance of the US and the UK in the world economy made research in these countries most relevant. Only limited efforts have been made to go beyond Anglo-Saxon compensation data and review executive pay arrangements in, for example, continental Europe (for exceptions, see e.g. Conyon and Schwalbach, 2000 and Bruce et al, 2005). This strong focus on Anglo-Saxon compensation practices has however neglected compensation structures and trends in executive compensation in other countries. In this, a distinction has often been made between the market-based Anglo-Saxon governance system and the bank based Continental-European system (Van Ees et al, 2003), where Anglo-Saxon governance can be characterized as more shareholder-oriented and Continental-European governance as stakeholder-oriented. This is, among others, also reflected in executive compensation arrangements, in the sense that Anglo-Saxon companies have made more extensively use of equity-based compensation to align the interests of managers and shareholders (Abowd and Bognanno, 1995).

The Dutch corporate governance system can be described as a combination of the Anglo-Saxon and Continental-European governance systems and it is therefore interesting to study how executive compensation practices have evolved in such landscape. Dutch executive compensation and corporate governance practices have been subject to heated debate since a couple of years. Large share price increases and subsequent option gains in the late nineties, led former prime-minister of the Netherlands Wim Kok, as one of the first examples, to openly agitate against the ‘exhibitionistic self-enrichment’ of top executives. And since disclosure of the remuneration data of all individual members of the executive board has been made compulsory since 2002, an annual springtime carnival of social outrage and debate is exhibited after, again, large rises in compensation have been revealed.

At the same time, the Dutch corporate governance environment has changed profoundly under influence of the accounting scandals of the beginning of the 21st century. Prime exponent of this changing corporate governance environment has been the implementation of the Dutch Corporate Governance Code in 2003, more commonly referred to as the Tabaksblat Code1. This code has laid down a number of principles concerning the remuneration policy of listed firms and can be considered as one of the most detailed corporate governance codes in Europe. The combination of an unique governance system and quickly changing business environment, makes Dutch executive compensation an interesting topic for research. It can create an understanding of the influence of public pressure and political interference on compensation practices in an environment of increasing shareholder pressure, in which shareholder value creation has become the prime measure of business success. The tension this creates has had, as the results will show, a significant impact on the structure and value of Dutch executive pay packages.

In association with Hewitt Associates2 an empirical study has been undertaken, with the prime objective to present the facts on executive compensation in the Netherlands. The results from this study can help to shape the public debate on the pay of top executives in the

1

Named after Morris Tabaksblat, who chaired the committee, that consisted of several representatives of different stakeholder groups.

2

Hewitt Associates is a global human resources outsourcing and consulting firm delivering a complete range of integrated services to help companies manage their total HR and employee costs, enhance HR services, and improve their workforces.

(7)

Netherlands and foster future research on executive compensation in the specific context of the Dutch governance system. Furthermore, it can help remuneration committees in their tasks in designing optimal compensation packages. Results of this study have been made public in a press release on the 20th of March 2006 and will be presented in an official report of Hewitt Associates. In this paper, the results from this study will be discussed and subsequently be placed in a theoretical framework. Objective of this paper is to provide a theoretical background in which the results of the empirical study can be interpreted. Furthermore, an attempt will be made to assess the effectiveness of Dutch executive pay in a theoretical as well as practical context.

Section I will discuss the basics of agency theory, which serves as the theoretical framework in this research. It will furthermore describe the characteristics of the Dutch corporate governance system in more detail and discuss the recent changes in the main governance mechanisms. The second part of section I will present the research methodology of the empirical study and the literature study.

Section II will present the research findings from the empirical study, which covers the compensation data of all executive board members of 71 Dutch listed companies over the period 2002-2004. The analysis will result in some very clear trends and developments in the structure and value of executive pay packages. Consequently, explanations will be offered, which can explain the trends that were signalled.

Section III will discuss four specific subjects within executive compensation research, which relate to the effectiveness of executive compensation arrangements. Successively, the determinants of executive pay, perverse incentives that can result from incentive plans, the value and valuation of equity-based compensation and the optimal structure of incentive plans will be discussed. An extensive, though not complete, search within existing literature has been conducted and the main views and findings within these subjects will be presented. These theoretical views and empirical findings will be discussed in relation with the research findings on Dutch executive compensation. Finally, conclusions will be drawn on the state of Dutch executive compensation, expected future developments and the effectiveness of Dutch executive compensation arrangements. Furthermore, limitations of this study will be discussed and some guidance for future research will be offered.

(8)

(9)

1. Theoretical Background

Academic research on executive compensation has been greatly influenced by agency theory (Daily et al, 2003). Since the implications of this theory have also been reflected in many compensation arrangements, this is still the necessary point of departion for every study on executive compensation, as it also is in this study. In the next paragraphs the basics of agency theory will be outlined and its consequences on remuneration practices. Subsequently, the discussion on agency theory will be specified to the characteristics of the (quickly changing) Dutch corporate governance system.

The Agency Relationship

Agency theory is based on the separation of ownership and control in the modern firm and the governance issues that arise from this separation (Berle and Means, 1932). In the specific context of executive compensation, agency theory is directed at the agency relationship in which shareholders (principal) delegate work to managers (agents) (Jensen and Meckling, 1976). Shareholders are primarily motivated by value maximisation (dividends and capital gains), they usually hold diversified portfolios of shares and, if ownership is small, have no special interest in monitoring the detailed activities of one firm. The dispersed ownership of shares therefore creates a free rider problem, in which non-monitoring shareholders benefit from the monitoring activities from others, which makes the aggregate expenditure on monitoring decline as the number of shareholders increase (Ang et al, 2000; Shleifer and Vishny, 1986). Agents on the other hand, are motivated by utility (wealth) maximisation, which can consist of notions as self-actualisation, esteem, power and economic interests, which not always fit with the value maximisation goal of shareholders (Rappaport, 1978).

Two problems can occur in this agency relationship. The first is the ‘agency problem’ that arises when the desires and goals of principal and agent conflict and it is difficult for the principal to verify what the agent is actually doing. The second is the ‘problem of risk sharing’, that occurs when principal and agent have different attitudes toward risk (Eisenhardt, 1989). These problems are rooted in assumptions about the behaviour of human beings, of which the most important are that people act in their self-interest, that they are naturally risk-averse and that they are limited by bounded rationality. These characteristics, in combination with a lack of monitoring caused by the diffuse ownership structure, can lead to suboptimal decision-making (resource allocation) from the point of view of the shareholders. Due to a lack of control, managers may put a lack of effort in managing the firm (shirking), they may spend too much on benefits in kind (perks) that enhance their status, reputation and comfort or they may grow the firm beyond the optimal size, which increases their power and, usually, their remuneration (Hall, 1985; Jensen, 1986). Because managers are naturally risk-averse3, they tend to prefer investing in low-risk projects to reduce the risk of business failure and consequently their compensation and (future) employment risk. When risks backfire this usually has significant consequences for the performance and financial health of the company and, indirectly, on the career of the manager. Since investors usually hold highly-diversified portfolios, they however in general prefer high-risk investments. This problem of different attitudes toward risk is called the ‘problem of risk sharing’.

The results of this diversion of interests are widely researched and documented upon in academic literature. One of the major effects of the lack of control (and managerial discretion) is over-diversification. This means that take-overs and mergers have not resulted in synergy

3

The extent of risk aversion can however vary per individual and depends on several factors, among which for example the personal wealth of the individual (Eisenhardt, 1989).

(10)

gains, but merely have increased firm size and have resulted in in-efficiencies, which in turn have decreased shareholder value. For executives however, over-diversification holds several advantages. Diversification has been used to reduce corporate managerial performance risk4 (Amihud and Lev, 1981) and because of the link between firm size and compensation, it provides an avenue for increased compensation (Hoskisson and Turk, 1990). Another form of ineffective behaviour is the creation of managerial slack by managers. This is done by retaining excess internal funds (‘free cash flow’5), which could otherwise be paid out as dividends to shareholders, but now serves as a buffer to decrease risks, to avoid the judgement of the external capital market or is wasted on low-return return projects (Jensen, 1986). The use of take-over defences also serves as an example of misalignment between managers and shareholders. If a take-over bid is designed to replace inefficient management, take-over defences and bid resistance may lower the market valuation of the firm, which is detrimental for the shareholders (Holl and Kyriazis, 1997)6. Finally, the use of ineffective compensation contracts can be seen as a result of the divergence of interests between managing board and shareholders (Bebchuk and Fried, 2004).

The Executive Contract

From an agency perspective, the objective of the executive contract is to align the interests of shareholders and managers in order to maximise the value of the firm (Jensen and Meckling, 1976). This can be done by creating managerial incentives that make executives act in the interests of shareholders. By providing a direct link between realized compensation and company share-price performance (with equity-based compensation7) this can be done effectively. With equity-based compensation executives are encouraged to take more risks, because the upside potential is usually more profitable than the downside risk. Furthermore, every lack of managerial effort or waste because of excessive consumption of perks, will decrease the market value of the firm and subsequently the share holdings of the executive.

Every executive contract should however be reviewed in the context of firm-specific circumstances. The extent of the contract being based on outcome measures (like stock options) or on behaviour monitoring (fixed components) should be determined by factors like task programmability, outcome uncertainty, outcome measurability, length of the agency relationship, the effectiveness and availability of information systems, the extent of risk aversion of the agent and the extent of goal conflict between agent and principal (Eisenhardt, 1989). While task programmability and outcome measurability can generally be described as quite low in the manager-shareholder relationship, hereby favouring outcome-based contracts, (assumptions about) the other factors may vary widely per firm.

By aligning the interests of managers and shareholder with equity-based compensation, managers theoretically will be motivated to act in the interest of shareholders and increase shareholder value. The optimal compensation package from an agency point of view is therefore highly leveraged. That this view goes not without criticism will be discussed in later paragraphs.

4

Business failure in one product unit can be smoothened by the results of other product units.

5

Free cash flow is cash that is available after all positive Net Present Value projects have been realised.

6

This is called the ‘managerial entrenchment hypothesis’. On the other hand scholars also argue that bid resistance might trigger an auctioning process with the purpose to bid up prices, which is in the end favourable for shareholders. This is called the ‘shareholder interest hypothesis’.

7

Equity-based compensation are all pay elements of which the payout is subject to stock price performance. Stock options and (performance) shares are the prime examples of equity-based compensation.

(11)

Governance Mechanisms

Besides the executive contract, there are other (corporate governance) mechanisms that disciplines managers and reduce the agency problem (Hart, 1995). Corporate governance, the whole of mechanisms, can be defined as the formal system of monitoring management performance and ensuring accountability of senior management to shareholders and other stakeholders. In a broader context, it can also be defined as the structures, processes, cultures and systems that engender the successful operation of the organisation (Keasey and Wright, 1993). Four primary governance mechanisms can be identified (Jensen, 1993):

• Legal/political/regulatory system

• Product and factor markets (includes the managerial labour market) • Capital markets (market for corporate control)

• Internal control system

The regulatory system sets legal boundaries on the behaviour of executives. When these boundaries are crossed, executives can face legal action and will consequently be disciplined by law. Corporate governance codes and guidelines also limit the action range and the leeway to shirk of the executive team.

Competition in the product and factor markets imposes discipline on managers, because firms that are managed inefficiently will be unprofitable and will not survive (Jensen, 1986). The managerial labour market is part of this governance mechanism. Managers in the firm face both the discipline and opportunities provided by the market for their services, within as well as outside of the firm (Fama, 1980). Inefficient behaviour and bad performance of managers will be punished in the future wage revision process or in the number of opportunities in the labour market. Bad performing executives will be replaced by others and executives that are replaced often face reputation damage and will have limited opportunities for a new career. They therefore have the incentive to act in the best interest of the firm. Future opportunities for the CEO are even more limited, since internal promotion is impossible. Furthermore, CEO turnover is more sensitive to firm performance than for non-CEOs and few non-CEOs are recruited for a similar position in another firm (Fee and Hadlock, 2004).

The market for corporate control should only operate when internal control efforts have failed. Takeovers can correct managerial failure by displacing bad performing or shirking managers. In cases of bad performance other management teams are likely to offer their services to the shareholders as an alternative for the current team (in the form of a merger or takeover). The market for corporate control is then the competition between these teams for the rights to manage the corporate recourses (Jensen and Ruback, 1983; Walsh and Seward, 1990). In the end, it is the thread of a possible takeover rather than the takeover itself that is seen as the real motivator for the incumbent management (Jensen, 1986).

The internal control system is the primary governance mechanism of the company. In a well functioning market, it should be less costly to implement control changes internally than through external control contests (Walsh and Seward, 1990). The primary internal control mechanism is the supervisory board, which can be described as an information system that shareholders use to monitor the opportunism of executives. The purpose of the board of directors is to ratify and monitor important decisions of the executive board and to choose, dismiss and reward these important decisions agents (Fama and Jensen, 1983). Executive

(12)

remuneration is also an integral part of the internal control system. Many research8 has been devoted to the characteristics of the internal control system, e.g. board size, board composition and ownership concentration, and the correlation with performance measures and (level of) executive compensation, which will also be discussed in chapter 6.

An ineffective internal control system has often been blamed for excessive executive compensation arrangements. Theoretically, the supervisory board, as representative of the shareholders, are assumed to bargain at arm’s length over their pay with the objective to reduce agency costs. Bebchuk and Fried (2004), however argue that boards have not been bargaining at arm’s length, but that their own incentives and preferences dominate this process. These incentives and preferences can be caused by interlocks, board appointment procedures, personal friendships, CEO pressure, cognitive dissonance and other factors. This had led to inefficient pay arrangement and a further misalignment of the interests of managers and shareholders. Bebchuk and Fried (2004) call this the ‘managerial power hypothesis’, but it is also referred to as the ‘managerial entrenchment hypothesis’.

Dutch Governance Issues

The Dutch corporate governance and regulatory system has several distinct features compared to other regimes, which affect the agency relationship and that influence executive pay arrangements. It represents a unique combination of the market-based Anglo-Saxon governance system and the bank-based Continental-European system. Typical of the Dutch corporate governance regime is the two-tier board structure in which the executive board (‘Raad van Bestuur’) is responsible for the daily operations of the firm and the supervisory board (‘Raad van Commissarissen’) has the tasks to appoint, monitor, suspend, and dismiss the members of the executive board, and to monitor and ratify major business decisions. The members of the supervisory board are appointed by co-optation and for a four-year term (Van Ees et al, 2003). Under influence of a number of Anglo-Dutch companies and increased Anglo-Saxon shareholdings, a slow shift in preference for the Anglo-Saxon unitary board structure can be identified.

In the past, this system could be characterised by a large degree of protectionism and managerial entrenchment. By issuing preference shares, tradable depository receipts and priority shares with limited voting rights and with the use of administration offices, ownership was effectively separated from control, thereby building up take-over defences in the case of a hostile take-over bid. This has led to an investor-unfriendly governance climate, leading to the so-called ‘Dutch discount’ on all Dutch share prices9. A market for corporate control hardly existed until a couple of years ago and the Annual Meeting of Shareholders only had limited influence. The co-optation principle and the large influence of the executive board on the appointments of the supervisory board have furthermore led to a large degree of managerial entrenchment (Van der Goot and Van het Kaar, 1997; Van Ees and Postma, 2004). Popular media have often referred to this as the ‘old boys network’, in which the small business community in the Netherlands divided all ‘sideline-jobs’10 among each other. The large increases in executive pay have also been placed in this context as a favour among friends.

However, in the last two decades a series of economic, technical and societal changes have changed the governance structures in the Netherlands significantly. First, an increasingly

8

See for example Beatty and Zajac (1994), Boyd (1994), Conyon and Peck (1998), Core et al (1999), Hoskisson and Turk (1990), Westphal and Zajac (1995) and Yermack (1996).

9

Dutch shares were traded with discounts of up to 20% in comparison with foreign peers, because of these take-over barriers.

10

Ironically, supervisory board memberships were for a long time indeed seen by these directors as just sideline-jobs, with only limited responsibilities and almost unlimited trust in the capabilities of executive directors.

(13)

global business environment has changed Dutch corporate structures. Although already very internationally oriented, a new series of acquisitions have been undertaken by Dutch companies in many parts of the world, making business processes even more globally dispersed. In 2004, on average 58% of the turnover of the largest listed companies and 71% of the turnover of the 25 AEX companies was realized abroad. On top of this, an increasing number of business functions and processes were outsourced or transferred to low-cost countries. The increased competitiveness in the world economy now has a quick impact on business performance when a company is underperforming, which has increased discipline in the product markets.

The labour market for executives has also internationalised quickly. The percentage of foreign CEOs at AEX companies has doubled in only two years from 24% in 2002 to 48% in 2004. These CEOs all bring different attitudes to the boardroom, which subsequently influence corporate governance standards. These attitudes involve, for example, issues like the role of shareholders, the role of the supervisory board and executive remuneration. This globalising managerial labour market has increased competitiveness and brought more dynamism into the managerial labour market.

The increased possibilities of the information age, furthermore, make every move of the CEO visible. Analysts and shareholders have more opportunities now to monitor the performance of managers and a continuous evaluation process has originated, in which stock prices can move quickly when expectations are exceeded or not met. Bad performing executives are more easily singled out by analysts and investors and replaced by others. Research of Booz Allen Hamilton (2003) has indeed showed that CEO turnover has risen quickly in the last decade.

Increased deregulation of many business sectors has led to more competition and dynamism in the Dutch economy. More foreign companies have entered the Dutch market and new competitive business sectors like telecommunications, postal services and energy have been opened. Also, take-over defence possibilities have been restricted, which has led to an increased number of take-overs of Dutch companies by foreign competitors and private equity funds. On top of this, an increased part of the shares of Dutch companies are now in foreign hands, mainly Anglo-Saxon investment funds. Around 2005, approximately 80% of the shares in the main 25 AEX companies were in the hands of foreign investors. These investors are in general more active than the traditionally very passive Dutch investors (mainly large pension funds and banks), which has led to increased pressure and discipline on top executives to perform well. CEOs that do not perform (share price growth) face the possibility of a take-over or increased investor pressure to divest activities or to split up the company and reverse any over-diversification.

Finally, new regulation has been introduced to strengthen the internal control system and increase the rights of shareholders. Also, the social debate is increasingly focused on corporate governance issues. The role and tasks of the supervisory board members have been strengthened and a board membership is no longer seen as a hobby, but instead as a job with important responsibilities. Changes have been so great, that the Netherlands were ranked first, alongside the UK, in the EIRIS corporate governance ranking of 2005. Most important trigger for many regulatory changes has probably been the Ahold scandal in 2003, dubbed by the Economist as the ‘Dutch Enron’. The accounting irregularities that appeared at one of the oldest and largest Dutch companies were for a large part blamed on a failing internal control system (Economist, March 1 2003). This scandal set in motion the realisation of the Tabaksblat Code. In 1997, the Committee Peters had already presented forty recommendations on corporate governance practices, however a review revealed that compliance to these recommendations was in general very limited (De Jong and Roosenboom, 2002). The Tabaksblat Code has replaced the Peters recommendations with the start of 2004.

(14)

It applies to all companies whose registered office is in the Netherlands and whose shares are officially listed on a stock exchange. The Code is for a large part based on the British Combined Code of Good Corporate Governance and presents principles on the role and procedures of the managing board and supervisory board, remuneration of the board members, conflicts of interest and the independence, composition and expertise of the (committees of the) supervisory board. It has established 14 best practice principles concerning the remuneration of top executives, which concern the transparency11 as well as the structure of the compensation package and is in this one of the most detailed codes in Europe. The implementation of the principles as such is not legally binding, but based on a ‘comply or explain’ principle. In the end of 2005 the Monitoring Committee12 concluded that ‘listed companies largely conform to the Corporate Governance Code’. From an international perspective the implementation of the Sarbanes-Oxley act has furthermore led to increased procedures concerning the internal control system of the company and the responsibilities of the members of the board of directors.

All these changes in the field of corporate governance have had a significant impact on the remuneration practices of Dutch companies, which will be outlined in section II and III in the discussion on the research results. In the next chapter, first the research concepts and methods will be discussed.

11

It states that ‘the notes enclosed in the annual accounts have to contain complete and detailed information on

the amount and structure of the remuneration of the individual members of the management board’.

12

The Monitoring Committee was installed to review the compliance to the Code and the topicality and usefulness of the principles.

(15)

2. Concepts & Methods

This paper has two main objectives: the first is to present the facts on executive compensation in the Netherlands and the second is to provide a theoretical framework in which these facts can be interpreted. As part of the second objective, Dutch executive compensation arrangements will be assessed on their effectiveness. With regard to the first objective, an empirical study has been set up in cooperation with Hewitt Associates to analyse Dutch executive pay structures, of which the methodology will be outlined below. With regard to the second objective, an extensive literature study has been undertaken to create a theoretical framework.

Basic Concepts and Definitions

The empirical study focuses on remuneration practices at 71 Dutch listed companies. Since 2002, all Dutch companies that are listed, are required to disclose the remuneration figures of the individual members of the executive board in the annual accounts. The number of companies that are listed on the Dutch Stock Exchange (‘Euronext Amsterdam’) is however rather limited (approximately 150 companies) and the differences in firm size among these companies are considerable. Since remuneration practices are only reasonably developed at large companies, the relevance of the results and the comparability with other studies13 would decrease if all, and consequently many small, companies were included . Therefore a selection of the most prominent and largest companies was made, which has resulted in 75 companies. Two additional criteria were formulated: 1) the companies should have a Dutch origin or substantial managerial activities on corporate level in the Netherlands14 and 2) the company may not have been taken-over by another company or involved in a large restructuring in the period under review, which have caused to change the position, job and responsibilities of the executives. Two companies were left out because of these requirements. After revision of the data, two more companies were left out of the population, because the information provided was incomplete (base salary and bonus figures).

Remuneration data from three different years, 2002, 2003 and 2004, is included. In 2002, for the first time individual remuneration data has been disclosed in the annual accounts and the 2004 data were the most recent available at the time of research. In some cases (qualitative) data on the 2005 remuneration package is taken into account, but only if published in the 2004 annual reports.

Three different positions in the executive board of directors were identified; the Chief Executive Officer (‘CEO’), Chief Financial Officer (‘CFO’) and the other members of the executive board. The roles and responsibilities of the other members of the board are in general too diverse to make a split and representative categorisation15. Only executive directors were included in the study. The remuneration of supervisory board directors (non-executive directors) was not taken into account.

13

Most international studies only contain the largest listed companies of the country in question. In the United States most samples are made of S&P 500 companies and in the United Kingdom most samples are made of FTSE 100 or FTSE 250 companies.

14

A number of foreign companies uses a Dutch listing and holding company primarily for tax purposes, but do not have substantial corporate and operational activities in the Netherlands. Though the principles of the Tabaksblat Code do apply for these companies, it is likely that in these cases remuneration practices most of all reflect those in the country where most corporate activities are seated.

15

The other members have responsibilities varying from product division to geographic or functional

(16)

The definition of compensation, which was used in the study is that of ‘flow compensation’. Executives are given compensation through three primary mechanisms: a) flow compensation, which is the sum of the executive’s annual base salary, annual bonus, LTI grants, benefits and perks, b) value changes of the existing equity portfolio of the executive, and c) the possibility of changes in the market’s assessment of the executive’s human capital (Core et al, 2003; Jensen and Murphy, 1990). The use of flow compensation means that changes in the value of share and stock option holdings, which were built up in previous years, and gains from option exercises and share sales were left out of the analysis.

The executive pay package consists of different components. The fixed part of the compensation package consists of base salary, pension benefits, other benefits, perquisites (perks) and severance arrangements. The short term variable part consists of the annual bonus plan. The long term variable part consists of all equity based compensation (option and share grants and deferred bonus plans) and long term cash bonus plans. All equity-based compensation grants were valued using a Monte Carlo valuation model, resulting in fair value estimates. Pension benefits, other benefits and perks were also valued. In section III and appendix I, a more detailed description of the used valuation methods is given.

Theoretical Concepts

In order to provide a theoretical framework, in which the results of the empirical study can be interpreted, a literature study has been undertaken. The scope of this project and the wide variety of subjects have prevented this overview from being complete, however a reasonable and representative overview is created from all available studies. In every subject, leading articles were sought and if possible, different points of views and research findings were combined to make the conclusions as objective as possible.

The main subjects of study for this literature overview were chosen in relation with the assessment of the effectiveness of Dutch executive pay arrangements. The effectiveness of executive compensation has been defined as to (1) attract, (2) retain, (3) motivate and (4) reward the executive board. These measures of effectiveness are based on the objectives of the remuneration policies of companies as stated by them in their remuneration report. Of the companies that report the objectives of their remuneration policy (n=63), 95% mentions ‘to attract’ as one of their objectives, 94% ‘to retain’, 71% ‘to motivate’ and 21% ‘to reward’. This last percentage is rather small, and one could argue that rewarding an executive is not an actual objective of the remuneration policy, but merely a consequence of the motivation objective. In order to motivate an executive beforehand, one should reward him or her afterwards. This is indeed true and probably explains the rather low percentage of companies that specifically mention ‘to reward’ as an objective16. On the other hand, one can reward an executive without the objective to motivate him. If, for example, it is not communicated beforehand that the possibility of a bonus payout exists, but one is offered anyway (like an additional one-time bonus based on excellent performance), this bonus has not had the objective to motivate the executive, but merely to reward him or her for excellent performance17. Furthermore, if assumptions about the motivational aspects of incentive pay are altered, i.e. that incentive pay does not necessary motivate executives18, one could argue that incentive plans are merely a cost-efficient (or even inefficient) way to reward an

16

It is furthermore observed that many of the companies that mention the reward objective, use it instead of the motivation objective.

17

Of course, one could argue then again, that this one time bonus is offered to motivate the executive in the next performance period, to excel again. However this should not necessary be the objective of the bonus.

18

(17)

executive19. An additional problem with ignoring the reward objective, is that it directly reflects the effectiveness of the motivation objective. If executives are not rewarded adequately or on the wrong measures, they surely will not be motivated. It is therefore necessary to assess the effectiveness of the motivational and reward goal independently, however without ignoring their interrelatedness.

These four measures of effectiveness all have a different influence on the remuneration structure. The first two goals, ‘to attract and retain’ are in general related to offering a ‘competitive’ compensation package, which is primarily related to the level of pay, though the structure of the pay package might also play a minor role20. In chapter 6, therefore the determinants of the level of executive pay will be discussed. The motivation objective is rooted in agency theory and has already been partly discussed in the previous chapter. It will be discussed in greater detail in chapter 8 and 9, in which the determinants of the optimal pay structure will be identified. Chapter 8 will discuss the theoretical incentives resulting from incentive plans and chapter 9 the empirical evidence on the subject.

To adequately reward an executive is interrelated with all other objectives and therefore runs through all other subjects. On the one hand, it is a matter of determining the level of bonus opportunity, which is closely related to the motivation objective and consequently to the discussions in chapter 8 and 9. On the other hand it is also a matter of incentive plan design and a fair performance setting and performance measurement process. Badly designed incentive plans, might reward an executive based on the wrong incentives or on events out of his or her control. Managerial entrenchment may furthermore cause the executive to be rewarded for non-existing performance. Chapter 7 will discuss perverse incentives in incentive plan design and in chapter 6 attention will be paid to the pay-setting process.

Data Description and Characteristics

The companies that form the population of the study are those that are part of one of the two main stock indices, the AEX Stock Index and AMX Stock Index (‘Midkap’), which both consist of 25 companies, and the 25 largest companies ranked by turnover (‘Small Caps’), which are not part of one of these indices. Due to large variations in size and complexity among these companies, these subgroups are also analysed independently. The average turnover of the companies in the population in 2004 was €9.4 billion, the average market value €5.8 billion, and the average number of employees 26,861.

The average age of the CEO in the population was 57.4 years in 2004. He or she was at that time on average 5.2 years in the CEO position, 7.8 years member of the executive board and 14.9 years employed by the company.

The average CFO had an age of 50.2 years, was four years member of the executive board and ten years employed by the company. The other members of the executive board were on average 52.3 years old, were approximately four years member of the board and 14 years employed.

19

For example in case of an annual bonus plan, the company only has to spend cash if performance is adequate and the expense is delayed until the end of the performance period.

20

For example in the sense that more entrepreneurial managers may favour a highly leveraged pay package (large equity incentives) and more risk-averse managers a higher base salary and more benefits. Of course, also other factors determine if a manager will join and stay at the firm, like job pleasure, esteem, power, ambition, reputation of the company, job safety, career perspectives and work environment. Pay however, especially at this hierarchical level, is likely to have a considerable influence.

(18)

Data Collection

No databases exist, which hold Dutch executive compensation data and therefore all data were collected from publicly available recourses, of which the remuneration sections in the annual accounts of companies were the prime source. Furthermore, use was made of all other publicly available information, like the Corporate Governance section on companies’ websites, additional remuneration reports, LTI plan regulations, employment contracts, information concerning annual shareholders meetings and SEC filings. In a few cases a company has been contacted with the request to supply crucial additional information, in some cases with positive response.

Despite regulations and prescriptions concerning transparency and the publication of remuneration data, a lot of data could not be processed without further handling in order to make all different company data comparable. Therefore some rules of thumb were introduced to handle and convert all data into comparable figures. Appendix I will give an overview of these rule of thumbs and will furthermore discuss the valuation methods used. In the next section the research results will be discussed and assessed.

(19)

(20)

3. Structure of Dutch Executive Pay Packages

This chapter will discuss the structure of the different components of the executive pay package, followed in the next chapter by the value of these components and the total compensation package. The structure of the pay package is in most cases similar among the members of the executive board. All executives are usually eligible for the same incentive plans. Some divergence only exists in retirement benefits and other benefits and perks offered, although a recent trend is that executives participate in the same pension plan as the other employees. Long term incentive plans are in many cases broadly implemented throughout the organisation, which makes the employees in other hierarchical levels also eligible for the plan.

Base Salary

Despite an increasing focus on variable pay, base salary is still the prime component of the executive pay package. Base salary is risk-free income and therefore valued high by executives. Furthermore, since the annual bonus opportunity and LTI grant are often set as a target of base pay, increases in base pay automatically lead to increases in these components. Also pension benefits and severance arrangements are in many cases based on the level of base pay. There are examples of executives in the US or UK that do not receive any base salary and instead solely variable pay, however all executives in this study received a fixed base pay. The primary goal of base pay, as part of the total pay package, is to attract and retain executives.

Annual Bonus Plans

Under the annual bonus plan executives are rewarded a sum in cash based on the performance of the firm in the financial year in question. All except one of the companies offered an annual bonus plan to its top executives in the period 2002-2004. Since the introduction of the Tabaksblat Code, a lot more information is published about the structure of annual bonus plans.

A few types of bonus plans can be identified from the available data. The first, and most used type (in 84% of the companies in 2004), is typified by Murphy (1999) as the ‘typical’ bonus plan. This is a bonus plan with an ‘at target’ payout expressed as a percentage of base salary for achieving the performance standard, a ‘minimum bonus’ paid at the threshold performance (usually expressed as a percentage of the target bonus) and a ‘maximum’ which limits the total payout. The range between the threshold and the maximum is called the ‘incentive zone’, in which the bonus payout increases incrementally with increased performance (Murphy, 1999). Other bonus plans that can be identified, are plans that pay a fixed amount for every percentage point that a certain threshold performance is exceeded (8.6% of the companies) and standard profit sharing programs (5.7%). In one example, a discretionary bonus, which is not based on previously-determined and measurable targets, is determined at the end of the year by the members of the supervisory board. In general, the part of the bonus that can be determined discretionary by the supervisory board is limited to an individual or ‘qualitative’ part of the total bonus opportunity. This is mostly due to the Tabaksblat Code which specifies that ‘the variable part shall be linked to

previously-determined, measurable and influenceable targets’.

Murphy (1999) identifies three basic parts of executive bonus plans: performance measures, performance standards and the structure of the pay-performance relationship.

(21)

Performance Measures

All companies that have published information on the performance measures used, have used financial performance measures in their bonus plan. These financial measures make up on average 83% of the potential bonus payout. Of these companies, 31% uses solely financial performance measures. When companies also use non-financial performance measures, the financial part of the bonus makes up for an average 72% of the potential bonus payout.

A clear trend is the increase in the use of individual and qualitative performance goals. Examples of the latter can be the ‘strategic orientation’ of the company, ‘policy progress’ or ‘team performance’. Though these performance measures can be quantified, in practice they usually have a rather discretionary character. In 2002, only 17.9% of the companies used individual measures and 15.4% qualitative measures. In 2004 these figures had increased to 49.2% and 27.7% respectively. If present, these measures make up 28% (average) of the bonus opportunity.

The number of different performance measures used, has increased over the last years. In 2002, still 61% of the companies used only one performance measure, while in 2004 this had decreased to 37%, therefore making the number of companies that use two or more performance measures 63%. These multiple measures are in most cases ‘additive’. This means that performance based on the different measures is separated and therefore each measure can technically be treated as a different plan.

The most popular performance measure is net profit, which is used by approximately 30% of the companies. Earnings per share (EPS) (22.6%), return measures21 (21%), profit measures before taxes and/or depreciation22 (21%) and sales growth (16.1%) are other widely used measures. The only clear trend in the use of specific performance measures is a decrease in the use of EPS, which was still used by 33% of the companies in 2002, but by only 22.6% in 2004. Furthermore, there is an increase in the number of ‘soft’ measures in areas like strategic development and corporate social responsibility.

Performance Standards

Murphy (1999) makes the following categorization in performance standards: ‘budget’ standards, ‘prior-year’ standards, ‘discretionary’ standards, ‘peer group’ standards, ‘timeless’ standards and ‘cost of capital’ standards. There are however too limited data to make any conclusions about the performance standards that are being used in bonus plans. Specific performance goals and targets are hardly published, usually on the grounds that these contain competitive sensitive information. Performance standards will therefore not be discussed in detail.

Pay-Performance Structure

The pay-performance structure usually involves a payout scheme that matches a certain performance with a bonus payout, usually expressed as a percentage of base pay. A typical scheme involves a threshold performance, below which there is no bonus payout, an ‘at target’ performance, which is paid out when the performance goals are met, and a ‘cap’, which limits the maximum bonus payout. The pay-performance relationship between threshold, at target and cap is in most cases linear.

The median at target bonus of the CEO was in 2004 60% of base salary. For the CFO this was 50% of base salary and for the other members of the board 58%. Of all companies, 68%

21

Examples of return measures are ROCE, ROA and ROE

22

(22)

sets the same target bonus level for all members of the executive board, while 32% sets different levels for individual members. This usually means a higher target bonus for the CEO, but also in some cases higher ones for one or more of the other members, usually the non-Dutch members. Bonus opportunities at larger companies are in general greater than those at smaller companies (expressed as a percentage of base pay). The median at target bonus of the CEO at an AEX company was 75% of base pay in 2004, at an AMX company 50% of base pay and at a small cap company 33%. At target bonus opportunities were fairly consistent over the years, however a lack of data in earlier years prevent any hard conclusions. The bonus payout at the threshold performance level is rarely published by companies, however most companies indicate that a threshold level is existent. Furthermore, a couple of companies indicate that in case of clear mismanagement a so-called ‘fairness assessment’ can be applied. This means that the supervisory directors can decide to cut a possible bonus payout because of this mismanagement, despite the performance goals being met. This ‘fairness assessment’ however, will in practice only be applied in exceptional cases (like in fraud situations).

The median maximum bonus that can be earned was 63% of base pay. This figure however cannot be compared with that of the at target bonus, since more and different (more smaller) companies have published the maximum bonus compared to the at target bonus23. CEOs of AEX companies can earn a median maximum bonus of 100%, those of AMX companies one of 60% and those of small cap companies one of 50% of base salary. When looking at the companies that published at target as well as maximum bonus opportunities, it can be seen that CEOs can earn an average 53% on top of their at target bonus (which equals 34% of base salary).

Stock Options

A stock option gives the owner the right to buy or sell an asset at a fixed price on or before a given date. Stock options grew increasingly popular in the eighties of the previous century, first mainly in the United States, but later also in Europe. From the late nineties, stock options constituted the largest part of the US executive compensation package and also regularly formed part of the Dutch executive compensation package (however to a smaller degree)24. Their popularity has been attributed to the absence of a charge against accounting income, favourable tax treatment and the positive incentive effects, which result from the close link between compensation and share price development (Huddart and Lang, 1996). Furthermore, stock options provide an incentive to act in the long term interest of the company, since they usually can only get exercised after the expiration of a vesting period. They can also help to attract highly motivated and entrepreneurial employees, because payout is based on future performance and rewards can be much higher than with normal cash compensation (Hall and Murphy, 2003). When an option ‘vests’ it means that the option can be exercised. The vesting of stock options can be dependent on the attainment of performance conditions, or the expiration of a time period. Options are typically forfeited if the executive leaves the company before vesting, though accelerated vesting is also a common plan characteristic.

Despite their prevalence in the last decades, in recent years new forms of equity-based compensation25 have slowly replaced the stock option in the long term pay package. In the last three years, a clear shift from traditional stock option plans to these other forms of

23

Of all companies, 41% publishes the at target bonus compared to 70% that publishes the maximum bonus.

24

See Jensen and Murphy (2004) for an overview of the rise in use and value of stock option pay in the US.

25

Other long term incentive variants are performance share plans, performance unit plans (a combination of options, shares and cash), deferred bonus plans and long term cash bonuses.

(23)

based compensation, and in particular performance shares, has taken place. In 2002, still 71.8% of the companies granted stock options to the executive board members. In 2004 this number had declined to 66.2%. Based on preliminary data it is to be expected that this will further decrease in 2005 to 45.1% of the companies. Options can be granted within the context of a stock option plan, a performance unit plan (in combination with shares and sometimes cash) or a one-off grant, for example as a sign-on award, though these grants are not included in this analysis. The number of companies with a traditional stock option plan declined from 71.8% in 2002 to 57.8% in 2004 and will further decline in 2005 to an estimated 31%. No new option plans were introduced in 2004 and only one option plan in 2003. Despite a strong shift to performance shares, most AEX companies still hang on to stock options; 70.8% of the AEX companies will still grant stock options in 2005.

Option Plan Structure

The structure of traditional stock option plans has changed significantly in recent years. This is a clear result of the implementation of the Tabaksblat Code and the changed corporate governance climate. The most important trend is the addition of performance conditions on vesting (or grant size) of the grant. In a traditional option plan, options could, if in the money, usually be exercised immediately after the grant. In some cases only time restrictions applied on the vesting of the options, which means that the options could not be exercised until an initial period had expired (usually three years).

In 2002, only 28% of the companies (with option plan) had performance restrictions attached to the vesting of the options. In 2004 this number had increased to 49%. Including the companies that had performance conditions attached to the determination of the grant size, 67% of the companies had in any way performance conditions as part of the option plan. The expectation is that in the next year(s) all other companies that have not attached any performance restrictions to their option plan yet, will implement these or will switch to performance shares or units. Earnings per share is the most popular performance measure that is used in option plans (37%), followed by Total Shareholder Return (33%). The EPS performance condition usually involves a hurdle (e.g. 6% EPS growth) that has to be met to make the options exercisable. TSR is usually used in a relative context; the TSR of the company is measured and compared to those of peer companies. A ranking is determined, which corresponds with a certain vesting percentage, usually defined as a percentage of a target grant (the whole represents the payout scheme).

The average term of an option was 7.26 years in 2004, slightly more than in 2003 (7.08) and 2002 (6.87). Most option grants had a term of 5 years (27% in 2004), followed by a ten-year term (24% in 2004). A small move can be distinguished towards options with longer terms. Of the option grants, 94.3% was granted ‘at the money’ (2004), which means with an exercise price the same as the share price at the date of grant. The rest, 8.7%, was granted at a premium. A couple of companies granted options with different exercise prices to different executives. The percentage of option grants that was granted ‘in the money’ in 2002, was still 16%. No options were granted at a discount (with an exercise price lower than the market price at the date of grant), which is in line with the Tabaksblat Code.

Most options are granted on the basis of a fixed number of options every year (26.8% of the option grants in 2004). This means that every year the same number of options is granted, despite any changes in share price. This can lead to significant changes in grant value every year. Only 7.3% of the option grants is based on a fixed value as a percentage of base pay26.

26

This means that every year an option package is granted, which has the same fair value as a percentage of base pay (e.g. a value of 50% of base pay). The number of options granted therefore fluctuates with share price movements.

(24)

This latter approach leads to a more stable annual grant value. As discussed earlier, 22% of the companies bases the grant on the performance of the executives (over the previous year), though target grants should be set independently and could also be based on a fixed number or fixed value approach. Furthermore, 18.8% of the grants is discretionary determined by the Supervisory Board and 6.3% is based on an individual assessment of the (performance of the) executive. In most cases however, it is not clear on what basis the options are awarded to the executive board members.

Performance Shares and Restricted Shares

Performance shares, also sometimes referred to as Long Term Incentive Plans (which is used here in a broader context), have grown increasingly popular in the Netherlands in recent years. First on a large scale introduced in the UK, and until now not very popular in the US, performance shares have slowly replaced stock options as the preferred form of long term incentive in the Dutch executive pay package. Performance shares are usually granted as part of a performance share plan or performance unit plan (in combination with options and/or cash). Under a typical performance share plan, conditional shares are granted, which vest after a performance period (usually three years). The degree of vesting depends on the attainment of the performance condition. In some case an additional vesting period follows the performance period. Restricted shares are similar to performance shares, but have no performance conditions attached to the vesting of the shares. Only a vesting period restricts the sale of the shares. Restricted shares are not common in the Netherlands and are in most cases only awarded as part of a sign-on bonus.

Of the companies, only 9.9% granted performance shares in 2002. This percentage had increased in 2004 to 35.3%, of which 8.5% were part of a performance unit plan. In 2005 however, the number of companies with a performance share plan will have risen to 42.3% and in total 57.8% of the companies will award performance shares. This means that in 2005 for the first time more companies will grant shares than options. In 2004 was 48% of the newly introduced long term incentive plans a performance share plan and 13% a performance unit plan. In 2003, these percentages were 10% and 30% respectively, however the number of datapoints was much less in 2003. The switch to performance shares was strongest at AEX companies. In 2002 only 17% granted performance shares, in 2005 this will have increased to 88% of AEX companies. Remarkable is that AEX companies also stuck to their stock options. Of the AEX companies, 58% granted options as well as shares to the executive board members. Performance shares will have been granted in 41% of AMX companies and 28% of small cap companies in 2005.

Performance Share Plan Structure

Relative TSR was the most popular performance measure in performance share and performance unit plans (56% in 2004), followed by EPS (16%) and return measures (8%). Half of the in 2004 newly introduced performance share plans used relative TSR as its performance measure, which signals the clear preference towards relative TSR as the performance measure in performance share plans. The use of relative TSR however has significant valuation consequences, which will be further discussed in section III.

More companies base the grant size on a fixed value as a percentage of base salary than in the case of the stock option grant. Of the companies which granted performance shares in 2004, 40% based the grant size on this approach. In 32% of the cases, the basis of grant was unclear and in only 4% of the cases was the grant size based on a fixed number of shares, while this latter approach was most popular with stock options. In 16% of the cases the grant

(25)

size was based on the performance of the company. Technically, after granting the shares these plans qualify as restricted share plans, since only a vesting period restricts the sale of the shares. However, since performance conditions are in some way attached to the plan, these plans are still categorised as performance share plans. For valuation purposes however, these plans are considered to be restricted share plans.

Deferred Bonus Plans

In deferred bonus plans (DBPs) part of the annual bonus payout is deferred into shares of the firm, which consequently have to be retained for a vesting period (usually three years). After this vesting period, bonus or matching shares are awarded, but sometimes only if performance conditions are met. Participation in the plan can be voluntary, but is in most cases compulsory. Another form of encouraging shareholdings by executive board members are so-called shareholding requirements, which oblige the executive to invest in firm shares from its own wealth.

The number of companies with a DBP has increased rapidly in recent years. From 7% in 2002, the percentage of companies with such a plan has increased to 18.1% in 2004 and will have further increased to 25.4% in 2005. Of the newly introduced long term incentive plans in 2004, 22% was a DBP (30% in 2003). DBPs are especially popular among AEX companies, of which 38% offers such a plan (2005). It seems however, that DBPs also become increasingly popular at smaller companies; in 2004, 62% of all DBPs was implemented at an AEX company, in 2005 this had declined to approximately 50% of all plans. In almost all cases, deferred bonus plans are active alongside a performance share or unit plan.

In most cases (36%), 50% of the annual bonus payout is converted into shares. In 27% of the cases this is 25% of the bonus payout. The most used matching ratio was 1:1 (83%), which means that for every share that is deferred into the plan, one bonus (or matching) share is awarded after the vesting period. In 17% of the cases was every share matched by 0.5 bonus share. In 31% of the DBPs, performance conditions are attached to the number of bonus shares granted. The performance measure most used is EPS (60%).

Other Long Term Incentive Plans and Equity Incentives

An alternative for equity based incentives are long term cash bonus plans, which are comparable in structure with annual bonus plans, however with a term of more than one year (usually three). A slight increase in the number of long term cash bonus plans can be seen, from 5.6% of the companies in 2002 to 14.1% in 2004. However in 2005, this will have declined again to 12.7% of the companies. These kind of plans are in particular popular at smaller (AMX) companies. The most used performance measure in long term cash bonus plans is TSR (or stock price development). This performance measure is however often used alongside other measures, quite often qualitative measures focused on the strategic development of the firm. Other used measures are EPS, net profit and the short term bonus payout.

Another form of LTIP are stock appreciation rights, which have the same characteristics as stock options however with a cash payout. These are hardly used by Dutch companies. Furthermore, options or shares can be granted as one-off grant, for example as a sign-on bonus, and therefore not be part of a specific plan. These grants do exist, but not on a regular basis and are not common use.

Executive Compensation in the Netherlands

Executive Compensation in the Netherlands

Executive Compensation in the Netherlands

Abstract

Table of Contents

Introduction

1. Theoretical Background

2. Concepts & Methods

3. Structure of Dutch Executive Pay Packages