Blockholding shareholders: in control of executive
pay
“Blockholding shareholders and option-based pay in
The Netherlands and the United States”
Faculteit der Economische Wetenschappen
Rijksuniversiteit Groningen
Name: T.A. Naafs
Student number: 1286714
Date: August 2007
Preface and acknowledgement
PREFACE AND ACKNOWLEDGEMENT ... 2
LIJST MET FIGUREN ... 4
1 INTRODUCTION AND PROBLEM STATEMENT ... 5
1.1. INTRODUCTION...5
1.2 PROBLEM STATEMENT...6
1.3 RESEARCH OBJECTIVES...8
1.4 IMPORTANCE OF THE STUDY...10
2. THEORY...11 2.1 INTRODUCTION...11 2.2. REMUNERATION DESIGN...11 2.2.1 Introduction ...11 2.2.2 Governance...12 2.2.3 Developments ...12 2.2.4 Blockholdings...14
2.3 REMUNERATION DESIGN AND AGENCY COST...16
2.3.1 Agency cost and firm size ...16
2.4 VARIETIES OF CAPITALISM...17
2.4.1 Difference between the United States and The Netherlands...17
2.4.2 Remuneration in VoC...19
2.4.3 Varieties of Capitalism, agency cost and remuneration design...20
3. METHODOLOGY ...21
3.1 INTRODUCTION...21
3.2 DATA COLLECTION AND USE...21
3.2.1 Data collection...21 3.2.2 Matching companies ...22 3.2.3 Variables ...22 3.2.4 Statistical techniques...24 4. FINDINGS ...26 4.1 INTRODUCTION...26
4.2 DESCRIPTIVES AND CORRELATIONS...26
4.2.1 descriptives ...26
4.2.2 Transformation...27
4.2.3 Multicollinearity...27
4.3 REGRESSION...29
4.3.1 Introduction ...29
4.3.2 Findings for hypothesis 1 and 2 ...31
5. LIMITATIONS AND S UGGESTIONS...33
6. CONCLUSION AND DISCUSSION...35
6.1 CONCLUSION...35
REFERENCES ...37
APPENDIX 1:RULE 13D-3. DETERMINATION OF BENEFICIAL OWNER...40
APPENDIX 2: MANUAL FOR THE DATA COLLECTION PROJECT ...43
Lijst met figuren
Figures:
Figure 1: Research model ………9
Figure 2: Comparison concentration of blockholders US Demsetz and Lehn (1985) and Becht (1997) ………15
Figure 3: Concentration of blockholders from Becht and Roëll (1999) ……….16
Tables: Table 1: Coding of variables ………..25
Table 2: (In) dependent variables ………..25
Table 3: Descriptives ……….26
Table 4: Distribution of transformed variables ……….27
Table 5: Correlation ………..28
Table 6: Regression analysis……….30
Graphs: Graph 1: % Granted options in 2004 per country ………..32
Graph 2: Blockholders and sales 2004………34
1 Introduction and problem statement
1.1. Introduction
During the past decade the attention towards the growth of executive directors’ salaries has increased. In political, social and scientific circles. The directors were ‘accused’ of self-enrichment, nonetheless more and more companies followed, (Cyert, Kang, Kumar, 2002; Crystal, 1991; Byrne, 1996; Lublin 1996). Obviously this is an interesting subject which raises a lot of questions. For example the Dutch government adopted a Dutch corporate governance code construed by the Tabaksblat committee. This code took effect on the first of January 2004. The code consists of principles of good corporate governance including regulations on remuneration and best practice provisions. It applies to all companies whose registered office is in the Netherlands and whose shares or depositary receipts for shares are officially listed on a government-recognized stock exchange. How countries deal with issues on remuneration is different (Ferrarini and Moloney, 2005; European Corporate Governance Network, country survey 1997) although most continental European countries make regulations on good governance including remuneration regulations. The European commission’s 1988 Transparency Directive requires member states to enact laws directing shareholders of companies listed on a member state exchange to notify the relevant authorities and the company itself within seven days whenever their voting rights cross the stated thresholds. This kind of regulation also exists in the United States (SEC filings) which will be discussed later. Because of these disclosure regulations it is possible to gather specific data on remuneration.
After conducting the data collection project last year I wondered whether the level of salary in the US differs significantly from the levels in the EU, and The Netherlands in particular. Considering the fact that this research would state the obvious (salaries in the US higher than in the Netherlands, Towers Perrin worldwide total remuneration report 2003-20041), I’ve decided to add a more interesting component. During the data collection project it seemed that almost every company in the US awarded its directors in shares or options. I started to wonder whether this would also be the case in the Netherlands considering the difference between the
1
http://www.towersperrin.com/tp/getwebcachedoc?webc=HRS/USA/2003/200311/WWTR_2003_English.pdf
two countries. During my literature review I read a thesis of Ferrarini and Moloney on corporate governance in the EU (2005). Their thesis focuses primarily on the differences in corporate governance in the EU and the disclosure rules but they also suggest a relationship between equity structure and executive compensation (in the form of option grants). They suggest, based on research in some EU countries, that companies with an equity structure consistent of (large) blockholders make less use of option-based remuneration. Because they didn’t investigate this by empirical research but assumed this relationship exists according to data on Spain and Italy (Brunello et al. 2001, Crespi and Gispert, 1999) I decided to use this as subject for my thesis. In my thesis I will investigate whether there is a relationship between equity structure and remuneration structure for companies in The Netherlands and the United States.
The thesis is structured as follows. In the remainder of this chapter the problem statement is stated in section 1.2, the research objectives are explained in section 1.3, the importance of this research is stated in section 1.4. In chapter 2 the theory used for this thesis will be discussed. In chapter 3 the methodology will be explained, in chapter 4 the results will be discussed, chapter 5 is a chapter on the limitations of my research and chapter 6 contains conclusions and suggestions for further research.
1.2 Problem statement
is called dispersed when there are no shareholders who own more than 5% of the total amount of shares issued by that company. Prior research shows that the United States always had a dispersed equity structure compared to a lot of European countries (Demsetz and Lehn, 1985; Roe, 1994; Becht, 1997; Becht and Roëll, 1999) although Marco Becht (1997) states that in 1997 there are more blockholders in the United States than found in research prior to 1997. According to Ferrarini and Moloney (2005) when a company has a ownership structure with (large) blockholders the remuneration is less option-based. This because blockholders monitor the directors so there is less need to align their interests with shareholder interests by share-based pay (see also chapter 2). The purpose of this thesis is to investigate whether the relationship exists between the equity structure (blockholders) and the option-based remuneration.
this project we collected data on all the companies of the Fortune 500 list for 2003 and 2004. Considering the data available from this project I chose to conduct my research also for the year 2004.
1.3 Research objectives
As introduced in the previous section the subject of this thesis is whether there is a relationship between the remuneration structure of the executive directors in the United States and The Netherlands and the equity structure in 2004. The remuneration structure is expressed as the options granted to the executive directors in 2004 divided by the total compensation received in 2004. The total compensation consists of the annual salary in cash plus the annual bonus in cash plus the value of the granted options in 2004. The equity structure is defined as the sum of the total blocks (in percentages) which are reported in the year 2004. This total sum of blockholders is divided in the categories defined by law: 5-10% (hereafter category 1), 10-25% (category 2), 25-50% (category 3), >50% (category 4). If there are no blockholders, or the total ownership of the largest shareholder is smaller than 5%, the company has a dispersed equity structure (category 0).
First there will be investigated whether there is a relationship between the equity structure and the % granted options to the executive directors in 2004. Secondly whether there’s a relationship between the United States and The Netherlands according to the equity structure in 2004 and the % granted options to executive directors in 2004.
The research model is:
Figure 1: research model
Summarizing, this research will try to provide an answer to the following question:
Is there a relationship between the equity structure and the % granted options for the executive directors in The Netherlands and the United States 2004?
In order to answer this question two hypotheses are examined in this study.
Hypothesis 1: Blockholding equity structures are negatively related to the % granted options executive directors received in 2004.
Hypothesis 2: Country (The Netherlands and the United States) is positively related to the % granted options executive directors received in 2004
Remuneration structure (% granted options/total salary)
Industry Equity structure
(In % total blockholders)
1.4 Importance of the study
2. Theory
2.1 Introduction
My hypothesis is whether there is a significant difference between the equity structure and the percentage granted options for the executive directors in The Netherlands and the United States 2004. Because one of the reasons for my hypothesis is the research of Ferrarini and Moloney (2005) the theory review starts with their research and some other literature on governance and blockholdings in section 2.2. Agency cost is also of great importance in relation to remuneration design. In prior studies it proved that performance related pay (in particular option based) is being used to mitigate agency cost, this is explained in section 2.3. Because I use data from the United States and The Netherlands, I discuss the theory of Hall and Soskice on Varieties of Capitalism and the relation to equity structure in section 2.4.
2.2. Remuneration design
2.2.1 Introduction
2.2.2 Governance
Considering this thesis, the distinction between the United States and The Netherlands is made in relation to equity structure and remuneration design. In this respect it has been proven that in the United States dispersed ownership is common (over 50% of the companies have a largest shareholder who holds less than 5% of the shares) and in The Netherlands blockholding ownership dominates (Becht and Röell, 1999; European Corporate Governance Network, 1997; Demsetz and Lehn 1985). In countries where dispersed ownership dominates the reliance on equity-based incentive-driven pay contracts is larger than in countries with blockholding ownership structures (Ferrarini and Moloney, 2005). This is because of the following reasons. Executive pay remedies the agency cost generated by the misalignment of manager and shareholder interests in the dispersed ownership company. Shareholders in dispersed ownership companies have only a fractional interest in firm profits and have limited opportunities to monitor management. Managements unobserved actions can influence shareholder wealth and give rise to agency cost (Jensen and Meckling, 1976). Agency theory suggests that the option-based pay contract can link pay to shareholder wealth and align the interests of shareholders and management. Where a shareholder/owner manages the company there is an alignment of interests and the need for an incentive (option-based) contract recedes. When ownership is concentrated and management is carried out by outsiders, monitoring is carried out by the blockholders. In that case an incentive contract can align managerial interests with those of the company as a whole. Because blockholders are substantial owners of the company they can monitor management and therefore there is less need for option-based remuneration to align director interests with shareholder interests. So theory suggests there is more option-based remuneration in dispersed equity structures.
Hypothesis 1: Blockholding equity structures are negatively related to the % granted options executive directors received in 2004.
2.2.3 Developments
The American labor market used to be more liberal (Hall, Soskice, 2001) than the Dutch labor market so this regulations on maximum period of appointment can be seen as convergence.
2.2.4 Blockholdings
According to Becht and Roëll (1999) an extraordinarily high degree of concentration of shareholder power exists in Continental Europe relative to the US and the UK. Becht examined this concentration in the United States in 1997 and stated that comparisons between the United States and European countries are not entirely suitable because the data available in the US are not available at the same level of detail in Europe. Becht therefore used the SEC filings to make a comparison. In figure 2 a figure shown with the differences Becht found between Demsetz and Lehn (1985) and his own empirical research. Previous research of La Porta, Lopez de Silanes, Shleifer and Vishny is based on the three largest stakes in the ten largest non-financial domestic firms according to Becht (1997). Becht concludes that the percentage beneficial owners in the United States has risen (especially compared to prior research) and that most beneficial owners hold only one block. A block is defined as all securities of the same class beneficially owned by a person, regardless of the form which such beneficial ownership takes2. Data on large shareholders have only recently become widely publicly available throughout the European Union. The reason was given by the European Commission’s 1988 Transparency Directive which requires member states to enact laws directing shareholders of companies listed on a member state exchange to notify the relevant authorities and the company itself within seven days whenever their voting rights cross the stated thresholds. Because each member state makes its own law the data differs from country to country in detail. Becht and Roëll (1999) conclude that as stated above, the blockholdings in Europe are much higher in Europe than in the US. Also that the separation of ownership and control manifests itself in fundamentally different ways. While in the US the main agency problem seem to stem from conflicts of interest between managers and dispersed shareholders, in much of continental Europe there are generally large blockholders present who can and do exercise control over management.
2
These blockholders have the power and information that is necessary to monitor management without the need for incentive contracts. In dispersed ownership companies executive pay is used to resolve the agency problem.
Hypothesis 2: Country (The Netherlands and the United States) is positively related to the % granted options executive directors received in 2004
Figure 3: concentration of blockholders from Becht and Roëll (1999)
2.3 Remuneration design and agency cost
2.3.1 Agency cost and firm size
capital markets) and compensation contracts may also align stockholder and manager interests. Large cash compensation is hypothesized as aligning management interests with debt holders, while large equity based compensation is aligning management interests with equity holders. Debt holders believe that subsequent to signing a debt contract, managers will act opportunistically in the interests of the equity holders. These debt holder concerns are influenced by the remuneration structure of management (Begley and Feltham, 1999). Research of Begley and Feltham (1999) shows that CEO’s cash compensation, the value of the CEO’s equity holdings and the fraction of the firms equity held by the CEO are of influence on debt holder concerns. It shows that debt holders are concerned about manager opportunism because with restrictive debt contracts the managers receive remuneration mostly in cash. With less restrictive debt contracts the equity-based compensation is larger.
Because large firms are more difficult to monitor there also seems a robust relationship between firm size and remuneration. Research showed that large firms pay directors with significantly large relative amounts of stock-based compensation compared to smaller firms (Bryan et al., 2000), (Yermack, 1995), (Gaver, Gaver, 1993). Another reason for large firms to use more option-based compensation is the willingness to incur the fixed administrative costs of implementing sophisticated compensation plans (Yermack, 1995). Because companies in my data set are matched according to size this is not used as a control variable, see section 3.2.2.
2.4 Varieties of capitalism
2.4.1 Difference between the United States and The Netherlands
The United States are from way back more liberal in the matter of free market system than The Netherlands. According to the theory of Hall and Soskice the Unites States can be categorized as a Liberal Market Economy. In liberal market economies, firms coordinate their activities primarily via hierarchies and competitive market arrangements. Market relationships are characterized by the arm’s-length exchange of goods and services in a context of competition and formal contracting. In response to price signals generated by such markets, the actors adjust their willingness to supply and demand goods or services, often on the basis of the marginal calculations stressed by neo-classical economics. In many respects, market institutions provide a highly effective means for coordinating the endeavors of economic actors. The Netherlands on the other hand can be characterized as Coordinated Market Economy. Nations with Coordinated Market Economies tend to rely on non-market modes of coordination in the financial sphere. Therefore they are less dependent on market mechanisms for access to finance. These non-market modes of coordination generally entail more extensive relational or incomplete contracting, network monitoring based on the exchange of private information inside networks, and more reliance on collaborative, as opposed to competitive, relationships to build the competencies of the firm. LMEs tend to rely on markets to coordinate endeavors in both the financial and industrial relation system. LMEs also have a higher developed stock market (Hall and Soskice, 2001).
According the theory of Hall and Soskice: “The financial system or market for corporate governance in coordinated market economies typically provides companies with access to finance that is not entirely dependant on publicly available financial data or current returns. Access to this kind of ‘patient capital’ makes it possible for firms to retain a skilled workforce through economic downturns and to invest in projects generating returns only in the long run” (Hall and Soskice, 2001, p. 22). Because finance is less dependent on balance-sheet criteria, there must be other ways of monitoring the performance for investors. This is possible through close relationships that companies cultivate with suppliers and clients. Also firms share information with third parties who monitor the firm and sanction it if the firm misleads them, such as business associations which have thorough knowledge of the industry.
The internal structure of the firm reinforces the systems of network monitoring in CME’s.
large shareholdings by strategic investors and in LMEs small shareholdings by individual investors. Ownership in CMEs can be characterized by one or more large shareholders with a strategic (rather than purely share value maximization) motivation for ownership. This contributes to my hypothesis because The Netherlands can be characterized as a CME and the United States as a LME. In The Netherlands prior research shows more blockholding shareholders and in the United States more dispersed shareholdings. The theory of Hall and Soskice confirms this.
In addition to the theory of Hall and Soskice (2001)research shows that in highly regulated industries executives will receive lower incentives form compensation or equity ownership, since the reduced range of managerial discretion in these industries diminishes the consequences of decisions. The monitoring of executives in highly regulated industries is larger than in less regulated industries because there is this reduced range of decision making (Yermack, 1995; Demsetz, Lehn, 1985; Smith, Watts, 1992). In these industries option based pay is less effective, therefore the remuneration in these industries will be less option based. This regulation of industries is not controlled by a separate variable but contributes to the theory on difference in monitoring by blockholders in contrast to the dispersed ownership company. Highly regulated so more monitoring and less need to control executive directors actions by option-based remuneration.
2.4.2 Remuneration in VoC
2.4.3 Varieties of Capitalism, agency cost and remuneration design
As indicated in the previous sections there are relationships between agency theory and the varieties of capitalism theory. In the following section the relationship is discussed.
3. Methodology
3.1 Introduction
In this chapter I will explain the design of the research. In section 3.2 the data collection, the data used for the research and the statistic techniques will be discussed.
3.2 Data collection and use
3.2.1 Data collection
reports. For the United States companies it is somewhat more difficult to find out if there are beneficial owners of more than 5%. According to the research of Becht (1997) this information is stated in the SEC filings, forms 13D and 13G. He also mentions form 14D but this form appears to be no longer in use. Considering the fact that these forms are stating the aggregated amount owned (and percentage of total shares) I gathered the information from all the 13G (and a few 13D forms) for the period 2004.
3.2.2 Matching companies
Because the size of companies influences the hypothesis and the existence of blockholders according to Becht (1997) I matched American and Dutch companies per industry so the data set is construed equally. I tried to match four companies per industry, in the data collection project ten different industries were distinguished. Unfortunately for the industry energy (0) only one company could be matched, for the industry health care (5) and utilities (9) there aren’t any Dutch (stock exchange listed) companies. Utilities contains companies as for example energy distributors, these companies are mostly owned by the Dutch government. I matched companies on total sales 2004, number of employees 2004 and industry. If a company has several matches I chose the company that was most similar in business. The total amount of companies equals 54. From all these companies the data are used for all the executive directors. The total amount of directors is 408; 141 Dutch directors and 267 American directors. In appendix 3 a table with all the companies used is added
3.2.3 Variables
The data that are necessary for my thesis consists of the following variables:
Annual cash salary 2004, annual cash bonus 2004, number of options granted in 2004, exercise price of the options granted in 2004. For each company also was registered the industry and the country the company is situated in. For the data collection project we only collected data on total sales for the year 2004. The equity structure is gathered as mentioned before (SEC filings and annual reports). For the blockholders the sum of all blockholders is calculated. Because the companies are matched in both countries on company size (measured in total sales in USD 2004 and number of employees) company size is not stated as a variable.
33
With the collected data the following calculations are made:
Total compensation: annual cash salary 2004+ annual cash bonus 2004+ value of the options granted in 2004. I only use data on remuneration that is granted in 2004. The long-term incentives such as performance related bonuses are not included because the payment of these bonuses differs a lot, sometimes the following year, but sometimes after a few years. This makes it difficult to gather data on for all the directors in the sample. Therefore I chose to use only the data on remuneration granted in 2004.
The value of the options will be expressed as follows:
Number of options granted in 2004 x exercise price options in 2004
In my hypothesis I only use the options granted and not the shares for the following reason. Companies award their directors mostly in options not shares. This because the relationship between management share ownership and performance is not consistent. Shares as incentive proved to be less effective as option-based pay (Morck et al., 1988; Loderer and Sheehan, 1989; McConnell and Servaes, 1990; Core and Larcker, 2002). Therefore the remuneration in shares is of less importance.
If there are more series of options granted the mean of these series will be calculated as in the data collection project. For example if two series of options are granted, one involving 200,000 options at an exercise price of $ 30 and a second of 100,000 options at an exercise price of $ 60, the mean will be ($ 30 x 200,000/300,000 + $ 60 x 100,000/300,000).
Dependent variable:
The percentage granted options 2004 is calculated as follows: Value granted options 2004/ total compensation 2004
Independent variables
Total of blockholders is calculated as follows:
Demsetz and Lehn,1985). This variable is the explaining independent variable for hypothesis 1 and control variable for hypothesis 2.
Country
Country is used as explaining independent variable for hypothesis 2. Industry
The industry the directors are employed in has been used as a control variable for hypothesis 1 and 2.
3.2.4 Statistical techniques
To test my hypotheses I used regression models with the Ordinary Least Squares technique. To define whether a regression model can be used there are a few data considerations.
Data: The dependent and independent variables should be quantitative. Categorical variables, such as religion, major field of study, or region of residence, need to be recoded to binary (dummy) variables or other types of contrast variables. In my research dummy variables have to be used for the variable industry and country as shown in table 1.
Assumptions:
For each value of the independent variable, the distribution of the dependent variable must be normal. The variance of the distribution of the dependent variable should be constant for all values of the independent variable. The relationship between the dependent variable and each independent variable should be linear, and all observations should be independent. As
Variable Code
The Nethe rlands 00
United States 10 Industry: Energy 0000000 Materials 1000000 Industrials 0100000 Consumer discretionary 0010000 Consumer staples 0001000 Health care Financials 0000100 Information technology 0000010 Telecommunication services 0000001 Utilities
Table 1: coding of variables
Independent variables and controls: Dependent variable: Total blockholders % Industry % granted options 2004
The Netherlands/United States
4. Findings
4.1 Introduction
In this chapter the results of the empirical analysis will be discussed. In section 4.2.1 a table is given with the descriptives of the variables used in the model. Section 4.2.2. explains the transformation of the variables. In 4.2.3 multicollinearity and the correlation table are discussed. In section 4.3 the output will be discussed of the statistical results.
4.2 Descriptives and correlations 4.2.1 descriptives Descriptive Statistics 408 ,00 98,30 17,3747 20,72344 2,063 ,121 4,521 ,241 408 ,00 ,98 ,0861 ,21082 2,406 ,121 4,521 ,241 408 % total blockholders % options granted 04 Valid N (listwise)
Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error N Minimum Maximum Mean Std.
Deviation
Skewness Kurtosis
Table 3: descriptives
4.2.2 Transformation
As indicated in the previous section the distribution of the residues is not normal. Because this is one of the data assumptions for regression I transformed the dependent variable % granted options 2004 and the independent variable % total blockholders by a natural logarithm to solve this normality problem. In table 4 a descriptives table is shown with the values for Skewness and Kurtosis for the logarithm transformed variables.
% options granted 2004 Log % options granted 2004 % total blockholders Log % total blockholders Skewness statistic 2.406 0.318 2.063 0.310 Kurtosis Statistic 4.521 -1.644 4.521 -.720
Table 4: Distribution of transformed variables
As shown in table 4 the values for skewness and kurtosis are much closer to zero (normal distribution). Although the value for log transformed % options granted 2004 still has a value of –1.644. When the value for kurtosis is < 2 the variable can be used as if it was normally distributed.
4.2.3 Multicollinearity
Correlations 1 408 -,184** 1 ,000 408 408 -,013 -,136** 1 ,789 ,006 408 408 408 ,000 -,090 -,116* 1 ,993 ,070 ,019 408 408 408 408 -,099* -,116* -,149** -,099* 1 ,045 ,019 ,002 ,047 408 408 408 408 408 ,078 -,196** -,253** -,167** -,215** 1 ,117 ,000 ,000 ,001 ,000 408 408 408 408 408 408 ,086 -,144** -,186** -,123* -,158** -,268** 1 ,084 ,004 ,000 ,013 ,001 ,000 408 408 408 408 408 408 408 ,044 -,101* -,130** -,086 -,111* -,188** -,138** 1 ,371 ,041 ,008 ,083 ,025 ,000 ,005 408 408 408 408 408 408 408 408 -,474** ,179** -,072 ,172** ,304** -,347** -,262** ,249** 1 ,000 ,001 ,204 ,002 ,000 ,000 ,000 ,000 316 316 316 316 316 316 316 316 316 -,960** -,074 ,071 -,088 ,161* -,070 -,098 ,117 ,311** 1 ,000 ,333 ,357 ,249 ,035 ,359 ,202 ,128 ,000 172 172 172 172 172 172 172 172 142 172 Pearson Correlation N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N NL/US Industryx1 Industryx2 Industryx3 Industryx4 Industryx5 Industryx6 Industryx7 log % totalblockholders log %optionsgranted 04 NL/US Indu stryx 1 Industry x2 Industry x3 Industry x4 Industry x5 Industry x6 Industry x7 log% total block holders log %options granted 04
Correlation is significant at the 0.01 level (2-tailed). **.
Correlation is significant at the 0.05 level (2-tailed). *.
Table 5: Correlation
Correlations measure how variables or rank orders are related. Correlation is significant when the p-value (the Sig. 2-tailed value in table 5) is at the 0.01 level. This correlation table shows strong negative correlations between most of the variables except for industry and Log % options granted this correlation is less strong .The only positive correlation is between Log % options granted and log % total blockholders. The high values of correlation indicate there might be multicollinearity. Especially when the value of a correlation coefficient is > 0.9 there could be a multicollinearity problem. In this case the variable country has a value of – 0.960.Therefore a I conducted a test to check whether multicollinearity exists.
part correlations are close to the zero-order coefficient and the Eigenvalues are significantly different from zero. The condition index shows no values above 15 which would indicate multicollinearity. For the variable log % total blockholders there could be a small
multicollinearity problem because the eigenvalue is close to zero and the condition index value is high. For the condition index a value above 15 indicates multicollinearity; the value for log % total blockholders is 14.42 so there is no significant multicollinearity. When looked at the partial and part correlations that are close to the zero- order coefficient this holds. To solve multicollinearity it is best to omit the variable that causes it, in this case it is not a possibility to leave % log total blockholders out so the minimum chance that multicollinearity influences the model is noted.
4.3 Regression
4.3.1 Introduction
Because I want to investigate whether there is a significant relationship between variables I used a regression analysis. As discussed before I transformed the variables that were not normally distributed. First I executed a regression analysis with only the control variable industry and the dependent variable % granted options 2004. Secondly the control variable industry and the explaining independent variable % total blockholders as model 1. Finally the control variables industry and % total blockholders; and country as explaining independent variable in model 24.
4
Variable Model 1 Model 1 Model 2 Model 2 Log % options granted 2004 (control) Log % options granted 2004 Log % options granted 2004 (control) Log % options granted 2004 Constant -3.631 -7.563 -7.563 0.628 Industry X1 X2 X3 X4 X5 X6 X7 -1.126 0.032 -9.13 0.858 -0.725 -0.985 0.729 -1.950 -1.152 -2.619 0.315 0.154 -0.278 -0.303 -1.950 -1.152 -2.619 0.315 0.154 -0.278 -0.303 0.164 -0.385 0.161 0.104 -0.512 0.190 -0.026 % log total blockholders 2004 1.563 1.563 -0.430 Country NL/US -5.563 R-square 0.064 0.235 0.235 0.928 F-value 1.603** 5.113** 5.113** 189.899** Adjusted R-square 0.024 0.189 0.189 0.923 ** P<0.01
4.3.2 Findings for hypothesis 1 and 2
Hypothesis 1: Blockholding equity structures are negatively related to the % granted options executive directors received in 2004.
The R- square statistic measures the proportion of variation explained by the regression model. The value varies between 0 and 1. Values close to 1 indicate a strong relationship and values close to 0 a weak relationship. In table 6 is shown that the R-square value in model 1 for % total blockholders has a value of 0.235. This indicates that some variation is explained by the model (about 24%). The significance value of F statistic is below 0.01 so the model is not due to chance (F(8.133) = 5,113, p= 0.001). Comparing model 1 with the control model the R-square value shows that in the control model R-square has a value of 0.064 and in model 1 0.235. The variable % total blockholders explains a part of the variation and has a significant influence on % granted options 2004. Remarkable is that the sign of the regression coefficient is positive; this contradicts my hypothesis because I expected a negative relationship between % total blockholders and % granted options 2004; more blockholding shareholders, less option based remuneration. So there is a significant relationship but not a negative one. The hypothesis is therefore not true and has to be rejected. Interesting to see is the influence of the industry variable. The coefficients in some industries are positive, other negative which means that some industries contribute to a higher % granted options and other industries diminish the % granted options.
Hypothesis 2: Country (The Netherlands and the United States) is positively related to the % granted options executive directors received in 2004
To see which country granted has the highest % granted options in 2004 I used a graph for more understanding. As shown in graph 1 in The Netherlands the % granted options was much higher than in the United States. This can relate to the developments mentioned in section 2.2.3; the shift in remuneration design from option-based to long term performance based incentives. 1 0 NL/ US 0,25 0,20 0,15 0,10 0,05 0,00
Mean % options granted 2004
5. Limitations and suggestions
Although the initial data set constructed during the data collection project and my own gathering of the Dutch companies is quite large it has its limitations. This especially because I chose to match companies to obtain a balanced data group which causes a loss of data. Because the data on equity structure are obtained for 2004 only also the remuneration data are used only for 2004. This can cause effects which are only true for that year and aren’t true for a longer period of time (although a pilot search on the Dutch annual reports indicates that the size of the blockholders shows not much difference in 2003). To receive a more complete result this research should be conducted for a longer period of time and for all the remuneration parts. As discussed in chapter 2 there are a lot of developments in remuneration practices due to legislation in governance issues and to long term performance related incentives. Differences can be expected in this field so a comparison over several years gives more answers. I only used the % granted options in 2004; for further research the cash bonuses, granted shares and performance linked remuneration (long term incentives) should also be included. For my thesis I only used companies listed on the stock exchanges, therefore I couldn’t investigate all of the industries defined in the American data set simply because there were no companies in The Netherlands in these industries (these were healthcare and utilities).
4 3 2 1 0 Categoryblockholders 50000 40000 30000 20000 10000 0 Mean sales 04
Graph 2: blockholders and sales 2004
6. Conclusion and discussion
6.1 Conclusion
Theory suggests that there is a relationship between the equity structure of companies and option-based remuneration. In blockholding equity structures the % remuneration in granted options should be less than in dispersed equity structures because there is more monitoring by shareholders which causes the interests of executive directors to be more aligned and there is less need for option-based remuneration to align these interests. In my thesis remuneration is measured in % granted options / total compensation to the executive directors in The Netherlands and the United States. The main question whether there is a relationship between the equity structure and the % granted options in The Netherlands and the United States in 2004 has proven to be true. Regression model 1 showed there is a significant influence (R-square value of 0.235) of equity structure (as % total blockholders) on % granted options in 2004. Remarkable is that this relationship is not negative as I expected and therefore hypothesis 1 had to be rejected. Regression model 2 shows a significant regression coefficient of 0.928 for the variable country and explains a very large part of the variation in the model. The hypothesis whether there is a relationship between the % granted options and The Netherlands and the United States in 2004 has therefore proven to be true. It appeared by more research that The Netherlands has a significantly higher % granted options than the United States in 2004. This contradicts the theory and prior research on remuneration design. The possible reason for this difference I expect is the change in remuneration design as described in the section developments in 2.2.3. The theory, which suggests more option-based remuneration in dispersed equity structures is not confirmed at all by my research. My research contradicts this theory through the positive regression coefficient in model 1. The variations in % granted options in The Netherlands and the United States are in this research mostly explained by the variable country.
2 1 NL/US 4 3 2 1 0 Categoryblockholders 4 3 2 1 0 Categoryblockholders 0,400000000000 0,300000000000 0,200000000000 0,100000000000 0,000000000000
Mean %granted options 04
Graph 3: category blockholders and % granted options 2004
References
Bancel, F., Mittoo, U.R., ( 2004), Cross-country determinants of capital structure choice: A
survey of European firms, Financial Management winter 2004 p. 103-132
Barca, F., Becht, M. (eds) (2001), The Control of Corporate Europe, Oxford, Oxford University Press
Bebchuk, L.A., (2003), The case for increasing shareholder power, Harvard Law review Association 2005
Becht, M. (1999), European Corporate Governance: trading off liquidity against control, European Economic Review 43, Elsevier Science BV
Becht, M. (1997), Beneficial ownership of listed companies in the United States, European Corporate governance Network
Becht, M., Roëll, A. (1999), Blockholdings in Europe: An international comparison, European economic review 43, Elsevier Science BV
Begley, J., Feltham, A., (1999), An empirical examination of the relation between debt
contracts and management incentives, Journal of Accounting and Economics 27, Elsevier
Bie de, T., Haan de, L., (2007), Market timing and capital structure: evidence for Dutch
firms, The Economist 155:183-206, Springer
Brailsford, T.J., Oliver, B.R., Pua, S.L.H., (2002), On the relation between ownership
structure and capital structure, Accounting and Finance 42 (2002) p. 1-26, Blackwell
Publishing
Brounen, D., Jong de, A., Koedijk, K., (2006), Capital structure policies in Europe: Survey
evidence, Journal of Banking & Finance 30 (2006) p. 1409-1442, Elsevier
Bryan, S., Nash, R., Patel, A., (2005), Can the agency costs of debt and equity explain the
changes in executive compensation during the 1990s?, Journal of Corporate Finance 12
(2006) 516– 535, Elsevier
Bryan, S., Hwang, L., Lilien, S., (2000), CEO Stock-based compensation: an empirical analysis of incentive-intensity, relative mix and economic determinants, Journal of Business vol.73 no.4, The University of Chicago
Byrne, J., (1996)The best and worst boards. BusinessWeek (November 25)
Cyert R.M., Kang S., Kumar, P., (2002) Corporate Governance, Takeovers, and
Top-Management Compensation: Theory and Evidence, Top-Management Science, Vol. 48, No. 4,
April 2002 pp. 453–469
Commissie corporate governance, (2003), Code Tabaksblat De Nederlandse Corporate governance code, beginselen van deugdelijk ondernemingsbestuur en best practice bepalingen Daily, C.M., Dalton, D.R., (2003), Governance through ownership: centuries of practice,
decades of research, Academy of Management journal vol.46, No. 2, 151-158
Demsetz, H., Lehn. K., (1985), The structure of corporate ownership: causes and
consequences, Journal of Political Economy vol. 93, no. 6, The University of Chigago
European Corporate Governance Network, ( 1997), The Separation of ownership and control:
a survey of 7 European Countries, European commission vol.4
Ferrarini, G., Moloney N., (2005), Executive remuneration in the EU: the context for reform, Oxford Review of Economic Policy vol. 21 no. 2005
Hall, G.C., Hutchinson, P.J., Michaelas, N., ( 2004), Determinants of capital structures of
European SME’s, Journal of Business Finance and Accounting, 31 (5) & (6), Blackwell
Publishing Ltd.
Hall P.A., Soskice D., (2001), Varieties of Capitalism, Oxford University Press
Jensen, M., (2003), Beyond shareholder value, Economist 6/28/2003, Vol. 367 Issue 8330, p. 9-14
John, T.A., John, K., (1993), Top-management compensation and capital structure, The Journal of Finance Vol. XLVIII, No. 3, Blackwell Publishing Ltd.
Johnston, J. (2005), Reward design and CEO succession in the UK, Applied Economics 2005, 37, 1535-1541, Routledge
Lane, C., (2003), Changes in corporate governance of German corporations: convergence to
the Anglo-American model?, ESRC Centre for Business Research, University of Cambridge
working paper No. 259
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., (2000), Investor protection and
corporate governance, Journal of Financial Economics 58, p. 3-27, Elsevier
Lublin, J. S. (1996), Executive pay: The great divide. The Wall Street Journal (April 11).
McClure, K.G., Clayton, R.,Hofler, R.A., (1999), International capital structure differences
among the G7 nations: a current empirical review, The European Journal of Finance 5,
141-164, Routledge
Nanova, T. (2003), The value of corporate voting rights and control: a cross-country
analysis, Journal of Financial Economics 68, Elsevier
O’Reilly, C.A., III, Main, B.G., & Crystal, G.S., (1988), CEO Compensation as tournament
and social comparison: A tale of two theories, Administrative Science Quarterly, 33 (2), p.
257-274
Ortiz-Molina, H., (2007), Executive compensation and capital structure: The effects of
convertible debt and straight debt on CEO pay, Journal of Accounting & Economics 43
(69-93), Elsevier
Smith, C.W., Watts, R.L., (2001), Incentive and tax effects of executive compensation plans, University of Rochester
Smith, C.W., Watts, R.L., (1992), The investment opportunity set and corporate financing,
dividend, and compensation policies, Journal of Financial Economics 32, p. 263-292,
North-Holland
Staat der Nederlanden, (2006), Besluit melding zeggenschap en kapitaalbelang in
effectenuitgevende
Instellingen, Staatsblad 375
Towers-Perrin, (2004), Worldwide total remuneration report 2003-2004, www.towers-perrin.com, accessed April 23, 2007
White, W.L., Fife, R.W., (1993), New challenges for executive compensation in the 1990s, SAGE publications
Yermack, D., (1995), Do corporations award CEO stock options effectively?, Journal of Financial Economics 39, 237–269, Elsevier
Appendix 1: Rule 13d-3. Determination of Beneficial Owner
(a) For the purposes of Sections 13(d) and 13(g) of the Act a beneficial owner of a security includes any person who, directly or indirectly, through any contract, arrangement, understanding, relationship, or otherwise has or shares: (1) Voting power which includes the power to vote, or to direct the voting of, such security; and/or (2) Investment power which includes the power to dispose, or to direct the
disposition of, such security.
(b) Any person who, directly or indirectly, creates or uses a trust, proxy, power of attorney, pooling arrangement or any other contract, arrangement, or device with the purpose or effect of divesting such person of beneficial ownership of a security or preventing the vesting of such beneficial ownership as part of a plan or scheme to evade the reporting requirements of Section 13(d) or 13(g) of the Act shall be deemed for purposes of such sections to be the beneficial owner of such security.
(c) All securities of the same class beneficially owned by a person, regardless of the form which such beneficial ownership takes, shall be aggregated in calculating the number of shares beneficially owned by such person.
(d) Notwithstanding the provisions of paragraphs (a) and (c) of this rule:
(1) (i) A person shall be deemed to be the beneficial owner of a security, subject to the provisions of paragraph (b) of this rule, if that person has the right to acquire beneficial ownership of such security, as defined in Rule 13d-3(a) within sixty days, including but not limited to any right to acquire:
(A) through the exercise of any option, warrant or right; (B) through the conversion of a security;
(C) pursuant to the power to revoke a trust, discretionary account, or similar arrangement; or
(D) pursuant to the automatic termination of a trust, discretionary
account or similar arrangement; provided, however, any person who acquires a security or power
conversion of such security or power. Any securities not outstanding which are subject to such options, warrants, rights or conversion privileges shall be deemed to be outstanding for the purpose of computing the percentage of outstanding securities of the class 28 owned by such person but shall not be deemed to be outstanding for the
purpose of computing the percentage of the class owned(?) by any other person.
(ii) Paragraph (i) remains applicable for the purpose of determining the obligation to file with respect to the underlying security even though the option, warrant, right or convertible security is of a class of equity security, as defined in Rule 13d-1(c), and may therefore give rise to a separate obligation to file.
(2) A member of a national securities exchange shall not be deemed to be a beneficial owner of securities held directly or indirectly by it on behalf of another person solely because such member is the record holder of such securities and, pursuant to the rules of such exchange, may direct the vote of such securities, without instruction, on other than contested matters or matters that may affect substantially the rights or privileges of the holders of the securities to be voted, but is otherwise precluded by the rules of such exchange from voting without instruction. (3) A person who in the ordinary course of business is a pledgee of securities under a written pledge agreement shall not be deemed to be the beneficial owner of such pledged securities until the pledgee has taken all formal steps necessary which are required to declare a default and determines that the power to vote or
to direct the vote or to dispose or to direct the disposition of such pledged securities will be exercised, provided that: (i) The pledgee agreement is bona fide and was not entered into with the purpose nor with the effect of changing or influencing the control of the issuer, nor in connection with any transaction having such purpose or effect, including any transaction subject to Rule 13d-3(b);
(ii) The pledgee is a person specified in Rule 13d-1(b)(1)(ii), including persons meeting the conditions set forth in paragraph (G) thereof; and (iii) The pledgee agreement, prior to default, does not grant to the pledgee: (A) The power to vote or to direct the vote of the pledged securities;
Appendix 2: Manual for the Data Collection Project
Fortune 500:
firms, networks, and directors March 9th 2006
Gerwin van der Laan1 Dirk Akkermans2 Hans van Ees
Arjen van Witteloostuijn
Introduction
Based on the website www.theyrule.net we composed a database containing more than 2,000 directors who together sit on over 5,000 board seats in 493 corporations (the Fortune 500). A group consisting of two researchers and five students will collect information on the directors, the compensation of these directors and the firms they direct. Each student will be given eighty firms and is asked to collect all the information according to this manual. Next, the students are given all data available and are able to write a thesis based on that dataset. For the students, the advantage is that they collect information for 80 firms only, whereas they are given the full dataset to analyse for their thesis. In order to fill out the blank cells, you will be provided with the following (electronic) files.
Firstly, two MS Excel files (Data_network and Data_firmlevel). The variables which are defined on the director level are to be entered in the first file, whereas the second file contains the variables at the firm level. Secondly, a list of countries which is the coding scheme for the nationality variable. Thirdly the data section of the 2003 and 2004 Spencer Stuart Board Index from which you will take non-executive directors’ compensation. Furthermore, we direct you to websites where the other information can be found. The appropriate websites are mentioned in the section on Finding the Data.
Finding the data
corporate websites, documents and filings. If you doubt, contact Gerwin van der Laan. After having dealt with a couple of firms, you will develop a routine of your own, but we suggest that you use the steps provided below especially in the beginning. This schedule has been developed based upon a pilot study and is the one experienced researchers find optimal. If, after a thorough search, you should come to the inconvenient conclusion that the information is not available, leave the cell blank. It is possible that in specific cases you will have to decide how to make the reality match with our data collection instrument. Where this is the case, make a note and discuss it with either Dirk Akkermans or Gerwin van der Laan. Regarding the SEC-filings, two are of particular interest. The DEF form (or DEF/A if the
numbers have been adjusted) provides information to the shareholders and is interesting because of the compensation and board structure information. The 10K (or 10K/A) form provides operational information on the corporation on an annual basis (there are also 10Q forms containing quarterly results). Here, the financial statements and statements on the corporation may be found; it can be compared to the annual report (at corporate websites the forms are often called ‘proxy statements’, though, and can be found as a separate document). Note that the DEF and 10K forms that concern 2004 appear in 2005 and that the search engine records the date of filing.
1. Names
Access the DEF filings at EDGAR company search3. You should select the form for the fiscal year 2004, which appeared in the first half of 2005. In the ‘summary compensation table’ you 3 http://www.sec.gov/edgar/searchedgar/companysearch.html . It is recommended to tick the box “Ownership forms 3, 4, and 5? Exclude”. Furthermore, choose the html filings, because these exclude programming code.3
find the names of the executives that serve on the board of the company. The list of nonexecutives
4 is provided in the first item called ‘Election of Directors’. Expand the list in
you do not have to add the directors which were appointed in 2003, and who left before the fiscal year 2004 started. It is also possible that the database we provided (theyrule) contains directors which are not in the SEC-filings (either as executive or non-executive). Please delete these entries from the file. In short, we take the SEC-filings to be correct and ask you to adjust the database to these filings.
2. Compensation
These can be found in the ‘summary compensation table’. An example is provided below for Alltel. Obtain the variables salary03 and salary04 from Salary in DEF, bonus03 and bonus04 from Bonus in DEF, shares03 and shares04 from Restricted stock award in DEF, and finally
irreg03 and irreg04 from Other Annual Compensation (see picture). The information should
be gathered for all executive directors. Also, you can enter the information for the variable
position using this table. For
executives, there are three possibilities: the executive is either the Chief Executive Officer (1), the Chief Financial Officer (2) or an insider with another job specification. For Alltel, Gardner (not on
the picture) is the CFO, Ford the CEO and the other executives fall in category 3. Just below the ‘summary compensation table’ you find the table ‘option grants in 2004’ which provides the information on stock option plans. For 2003, you have to take the equivalent table from the 2003 DEF form. We need opt03, opt04 (the number of options), optp03, optp04 (the exercise price), and optt03, and optt04 (the exercise date). If there is only one series of options (one line of information per executive in the table) you can easily copy that line (in SEC-terms number of securities underlying options, exercise or base price, and expiration 4 Note that different companies may use different terminology for the same concept. We use non-executive vs. executive directors in this manual, but companies might also call these insiders vs. outsiders or, confusingly, directors vs. executives. If you encounter the word ‘director’ be aware of the fact that this may thus refer to the whole of (non-executive and executive) board positions or to the non-executive directors only. 5 Obviously, this is quite a tedious job and therefore we ask you to regularly send your new entries to
You have been given (part of) the 2003 and 2004 edition of Spencer Stuart Board Index. From this index, we take the non-executive directors’ payment information. Take the column Annual Board Fees from Spencer Stuart and enter it as the salary03 and salary04 values in the file (this variables does not take fees for sitting on a committee into account and assumes full attendance). You do not have to check whether the director served on the board in 2003; just plug in the value for all non-executive directors.
3. Committee membership
The variables auditcom and compcom concern the question whether the director is a member of the audit committee and/or the compensation committee (alternatively: remuneration committee) of the board of directors. In the DEF filing, you find a report of both committees (if present) under the heading ‘Compensation committee report’ and ‘Audit committee report’. At the end of these reports, the names of the members in the committee are mentioned. If the committee does not exist, enter a zero for all directors.
4. Director profiles
options with a price of $40 ($30*200,000/300,000 + $60*100,000/300,000) and the exercise date 2013 (2/3 of 2012 plus 1/3 of 2015). Do not take more specifc expiration dates (months, days) into account.
In some cases names of companies are very similar. Use www.theyrule.net to find the website of the company at hand. At the websites, the ‘investor relations’ or ‘corporate governance’ pages are the most likely to provide the information you are looking for. 8 Note that this website takes quite some time to load (wait for the advertisement to disappear). Furthermore, there is a box which allows you to track the other directors with which the one you searched for sit on the board.
http://www.forbes.com/finance/mktguideapps/personinfo/FromMktGuideIdPersonTearsheet.j
html which may be difficult to find but which is extremely important for the CEO position. This dummy (1 = yes; 0 = no) measures whether the director is recruited from inside the corporation. Thus, did the director have a job with the firm immediately before becoming director. The remaining variables are in the file ‘Data_firmlevel’ which – not so surprisingly – provides information at the firm level.
5. Stock price at the end of the year
Stock prices for companies which still are in the S&P500 index, may be found at the S&P website9. Otherwise, this information is available through the 10K form. Note that you will have to report the closing price, not a high or low price.
6. Duality
Duality occurs if the CEO is not the same person as the chairperson of the board. The easiest
way of finding this piece of information is through searching the DEF-form using the keywords ‘chairman’ or ‘chairperson’. Before starting any search inside these forms, you have to scroll to the top of the form. From the position variable you know who is the CEO. Next, fill out the duality question as follows: if the chairperson and the CEO are the same person, fill out a 1; if not, enter 0.
7. Distribution of sales
If relevant, information on the geographical distribution of sales can be found in the company’s 10K form. Record the absolute value of sales in the regions we distinguish among (see appendix).
8. Credit ratings
website10. Below is an example of what the page looks like. We are interested in the ‘organization rating’, which consists of three components seperated by ‘/’. The first is the rating, the second the expected change, and the third the short-term
rating. Take the first component. The values range from AAA, to A, to CCC, CC, and finally R where R is the lowestrating possible. Refer to the coding scheme in the appendix how to translate these letters into numbers. Secondly, there may be a plus or minus sign after the letter, which indicates the relative
You can change the date at which the closing price is needed into 31 DEC 2003 and 31 DEC 2004
respectively.http://www2.standardandpoors.com/servlet/Satellite?pagename=sp/Page/Indices
Constituhttp://www2.standardandpoors.com. You will have to register before you can log in to obtain credit ratings.This registration is free position inside the class (thus AAA+ are the most creditworthy, while AAA- are worse, but still better than AA+). This relative position is registered through the variable rat_pos, which takes on the values -1 (in case of a minus), 0 (no plus or minus), or 1 (in case of a plus). Next to these variables, we provide you with the variables number of employees, balance
List of Variables data collection
Firm-level variables:
INDUSTRY Industry (filled out) 0 = Energy 1 = Materials 2 = Industrials 3 = Consumer discretionary 4 = Consumer staples 5 = Health care 6 = Financials 7 = Information Technology 8 = Telecommunication serv. 9 = Utilities
STOCKP_3 Stock price at 12/31/2003 x$1 STOCKP_4 Stock price at 12/31/2004 x$1
DUALITY CEO/chairperson duality 0 CEO and board chairperson are not the same person 1 CEO and board chairperson are the same persons
SALES Total sales Millions of dollars SALES_US Sales in the US
SALES_WE Sales in Western Europe SALES_EE Sales in Eastern Europe SALES_AS Sales in Asia
SALES_LA Sales in Latin America
SALES_ROW Sales in the rest of the world (if percentages are given, compute the value yourself)
8 AA 9 AAA
RAT_POS Relative position inside credit rating category
-1 minus sign after rating 0 no sign after rating 1 plus sign after rating Director-level variables:
POSITION Position in the board
1 CEO, 2 CFO, 3 other insider (task within the firm), 4 outsider (no task within the firm, non-executive)
5 undetermined
AUDITCOM Member of audit committee 1 = yes 0 = no
COMPCON Member of compensation committee
1 = yes 0 = no
CHAIRPERSON Chairperson 1 = this director is chairperson of the board 0 = this director is not chairperson of the board
GENDER Gender 0 male 1 female BIRTH_YE Year of birth 4 digits
APPOINTMENT Year of first appointment 4 digits NATIONALITY Nationality See list
EDUCATION Area of education 1 Engineering
2 Business 3 Law 4 Medicine
5 Arts (including languages) 6 Social sciences
7 Sciences (math, physics…) Other (specify)
INTERN_HIRE
Internal hire 0 director is recruited from outside the company, 1 director worked in the company prior to becoming a director
SALARY03 / SALARY04 Fixed salary 2003, 2004 Fixed salary paid to the director (x$1)
BONUS03 / BONUS04 Cash bonus 2003, 2004 Cash bonus paid (non-fixed salary), x$1
NOSERIES03 / NOSERIES04
Number of option series
OPT03 / OPT03 Options 2003, 2004 Number of stock options granted
OPTP03 / OPTP04 Exercise price options Exercise price x$1 OPTT03 / OPTT04 Expiration year Year of expiration (4 digits)
SHARES03 / SHARES04 Shares 2003, 2004 Value of shares granted x$1 IRREG03 / IRREG04 Irregular payments These are all other
Appendix 3: companies included in the data set
Industry company sales 04
0 Royal Dutch Shell 37127,00
0 Ashland Inc. 33437,00
1 Compagnie de Saint-Gobain S.A. 39829,49
1 Corus 17109,87
1 Mittal Steel Company N.V. 27610,14 1 Koninklijke BAM Groep N.V. 9287,95
1 Heijmans N.V. 3226,77
1 Dow Chemical Company 40161,00
1 Georgia-Pacific Corporation 19656,00
1 Alcoa, Inc. 23478,00
1 PPG Industries, Inc. 9513,00
1 Peter Kiewit Sons, Inc 3352,00
2 Akzo Nobel N.V. 15780,07
2 Vedior N.V. 8043,01
2 Koninklijke DSM N.V. 9641,16
2 Imtech N.V. 2570,73
2 Emerson Electric Co. 15615,00
2 Kelly Services, Inc. 4984,00
2 Eaton Corporation 9817,00
2 Washington Group International 2915,40 3 Koninklijke Philips Electronics N.V. 37684,11
3 Draka Holding N.V. 2167,27
3 Best Buy Company, Inc. 24547,00
3 Cooper Tire and Rubber Company 2081,60
4 Unilever N.V. 50203,19
4 Koninklijke Ahold N.V. 64672,40
4 Schuitema N.V. 4133,10 4 Procter & Gamble Company 51407,00
4 Kroger 56434,00
4 Anheuser-Busch Companies, Inc. 14934,20
4 Pathmark Stores, Inc. 3978,50
6 ABN AMRO Holding N.V. 24616,55
6 Allianz 120483,44
6 Binck N.V. 50009,18
6 Fortis N.V. 75514,35
6 ING Groep N.V. 84340,27
6 Morgan Stanley 23765,00
6 American Intl. Group 97987,00
6 Bank of America Corp. 48894,00
6 Berkshire Hathaway 74382,00 6 Citigroup 86190,00 7 Wolters Kluwer N.V. 4055,71 7 Getronics N.V. 2953,79 7 Reed Elsevier N.V. 8822,62 7 Sopheon plc 7926,06 7 Avaya Inc. 4069,00
7 Computer Associates International, Inc.
3299,00
7 Lucent Technologies, Inc. 9045,00
7 EMC Corporation 8229,49
8 Koninklijke KPN Nederland N.V. 14589,84
8 TNT N.V. 15651,96
8 Qwest Communications International, Inc.
13809,00