Management Phd Thesis

8/3/2019 Management Phd Thesis

1/230

Explaining executive pay:

The roles of managerial power and complexity

D I S S E R T A T I O N

of the University of St. Gallen,

Graduate School of Business Administration,

Economics, Law and Social Sciences (HSG)

to obtain the title of

Doctor of Business Administration

submitted by

Lukas Hengartner

from

Waldkirch (St. Gallen)

Approved on the application of

Prof. Winfried Ruigrok, PhD

and

Prof. Simon Peck, PhD

Dissertation no. 3217

Publisher: Deutscher Universitts-Verlag, Wiesbaden 2006


2/230

The University of St. Gallen, Graduate School of Business Administration, Economics, Law and

Social Sciences (HSG) hereby consents to the printing of the present dissertation, without hereby

expressing any opinion on the views herein expressed.

St. Gallen, June 12, 2006

The President:

Prof. Ernst Mohr, PhD


3/230

PREFACE V

Preface

The idea of studying executive pay was born in a lecture on option valuation in spring

2002 at the Norwegian School of Economics and Business Administration NHH. At that

time, I started wondering how companies that had granted large amounts of stock

options to their executives would react to the sharp decreases in share prices that could

be observed since 2000. Based on this interest, I wrote my NHH master thesis about

Employee stock options and falling share prices investigating a sample of Norwegian

IT companies. These companies were especially likely to both have granted substantial

amounts of options and suffered from severe declines in share prices. The topic of

executive pay continued to attract my interest, mainly because I felt that it offered

opportunities to make a real contribution to academic researchers and policy makers

alike. In summer 2002, I wrote my master thesis at the University of St. Gallen HSG. I

investigated the determinants of executive stock option plan adoptions of Swiss stock-

listed companies. Major results were that both large outside shareholders and CEO

duality influenced whether companies would adopt stock option plans. At that time, very

little information about executive pay was disclosed in Switzerland. However, the SwissStock Exchange SWX issued new listing guidelines that required companies to disclose

detailed information on Corporate Governance starting from the annual report 2002 to be

published in 2003. Part of this information was the total compensation of the highest-

paid member of the board of directors and the total compensation for the top

management team. With these guidelines, the way was open to empirical studies on

executive pay in Switzerland, a topic that had received much attention in the US for a

long time.

The goal of this dissertation was twofold. On one hand, I wanted to make a contribution

to the international research community interested in top executive pay. By developing

and testing an innovative and broad measure of the link between managerial power and

pay as well as complexity and pay, I am confident to have made such a contribution. On

the other hand, this work should provide a detailed and comprehensive overview of the

executive pay in Switzerland. By presenting an extensive and updated review of the

literature and detailed statistical data, I hope to provide valuable input for compensation

consultants, HR managers and policy makers on this highly contentious issue.


4/230

PREFACEVI

I am indebted to my interview partners and all participants of the survey used in this

work. Interview partners were Gary Steel, Head of HR and secretary to the compensation

committee ofABB, David Tankel, Managing Director ofNew Bridge Street Consultants,

Hanspeter Fssler, member of the compensation committee ofDaetwyler, and Prof. Dr.

Rolf Dubs, chairman of the compensation committee ofSchindler.

Further, I would like to thank all persons who supported me in writing this study. My

special thanks go to my academic advisors, Prof. Dr. Winfried Ruigrok and Prof. Dr.

Simon Peck, for their valuable guidance and suggestions as well as for the academic

freedom I was granted during the preparation of this paper. I also appreciated the

instructive comments of Elin Therese Dahl and Peder Greve, who reviewed drafts of the

manuscript.

Zurich, June 15, 2006 Lukas Hengartner


5/230

OVERVIEW VII

Overview

Preface ..............................................................................................................................V

Overview.........................................................................................................................VII

Table of Contents............................................................................................................ IX

List of tables................................................................................................................ XVII

List of figures................................................................................................................XIX

List of abbreviations .....................................................................................................XXI

Summary....................................................................................................................XXIII

1 Introduction...............................................................................................................1

2 The compensation setting process ............................................................................9

3 Research review on executive pay...........................................................................13

4 Research review on director pay.............................................................................48

5 Are executives paid for the complexity of the job they have?................................53

6 Can powerful managers extract rents? ..................................................................66

7 Research methods..................................................................................................102

8 Research results.....................................................................................................119

9 Discussion..............................................................................................................159

10 Conclusions ...........................................................................................................169

Appendix - questionnaire ..............................................................................................179

References......................................................................................................................183


6/230

VIII


7/230

TABLE OF CONTENTS IX

Table of Contents

Preface ..............................................................................................................................V

Overview.........................................................................................................................VII

Table of Contents............................................................................................................ IX

List of tables................................................................................................................ XVII

List of figures................................................................................................................XIX

List of abbreviations .....................................................................................................XXI

Summary....................................................................................................................XXIII

1 Introduction...............................................................................................................1

1.1 Why study executive pay?...............................................................................1

1.2 The evolution of executive compensation research.......................................2

1.3 Literature gap and research questions ..........................................................4

1.4 Overview of the paper .....................................................................................6

2 The compensation setting process ............................................................................9

2.1 Purpose of compensation policies ...................................................................9

2.2 Roles and responsibilities ................................................................................9

2.3 Organization of the compensation committee .............................................10

2.4 Elements and allocation of executive pay ....................................................10

2.5 The influence of executives on their compensation.....................................11

3 Research review on executive pay...........................................................................13

3.1 The relationship between pay and performance.........................................14

3.1.1 The debate over size versus profits..............................................................14

3.1.2 Agency theory and optimal contracting.......................................................15

3.1.3 The pay-for-performance sensitivity ...........................................................16

3.1.4 Relative performance evaluation .................................................................21

3.1.5 Non-financial firm performance and pay ....................................................22


8/230

TABLE OF CONTENTSX

3.1.6 Pay structure and subsequent performance..................................................23

3.1.7 The certainty-equivalent approach ..............................................................26

3.2 Other determinants of executive pay ...........................................................273.2.1 Economic explanations................................................................................27

Complexity...........................................................................................................27

The influence of risk in optimal contracting........................................................27

Managerial labor market ......................................................................................29

Investment opportunities......................................................................................30

Managerial discretion...........................................................................................31

Human capital ......................................................................................................33

Market forces .......................................................................................................34

Executives personal characteristics ....................................................................34

Stewardship theory...............................................................................................35

3.2.2 Political explanations...................................................................................35

Managerial power ................................................................................................35

Impression management ......................................................................................35

3.2.3 Social explanations ......................................................................................36

Social comparison................................................................................................36Isomorphism ........................................................................................................36

3.3 Other effects of executive pay .......................................................................37

3.3.1 Stock options, dividend payments and share repurchases ...........................37

3.3.2 Unvested stock options, restricted stock and manager retention.................37

3.3.3 Equity incentives and risk............................................................................37

3.3.4 Compensation, M&A and divestitures ........................................................39

3.3.5 Long-term compensation and capital investments ......................................40

3.3.6 Contingent pay and expectations management............................................40

3.3.7 Incentive compensation and earnings management.....................................41

3.3.8 Stock options and litigation .........................................................................41

3.3.9 Crowding out ...............................................................................................41

3.4 Determinants and consequences of pay differentials..................................42

3.4.1 Explaining pay differentials tournament theory .......................................42

3.4.2 Pay differentials and performance...............................................................43

3.4.3 Other determinants and consequences of pay differentials .........................44


9/230

TABLE OF CONTENTS XI

3.5 Stock option repricing ...................................................................................44

3.6 Stock options versus stock grants.................................................................46

3.7 International comparison of pay levels and mix .........................................47

4 Research review on director pay.............................................................................48

4.1 Directors as the shareholders agents ..........................................................48

4.2 Overview of outside director compensation ................................................49

4.3 Determinants and consequences of director compensation .......................49

4.4 Summary of director pay studies .................................................................51

5 Are executives paid for the complexity of the job they have?................................53

5.1 Towards a framework of complexity and pay.............................................53

5.1.1 Defining complexity ....................................................................................53

5.1.2 Complexity as a determinant of pay level and structure..............................53

5.1.3 Prior operationalizations of complexity.......................................................54

5.1.4 Conceptual framework of complexity and executive pay ...........................55

5.2 Firm internationalization and executive pay...............................................56

5.3 Firm diversification and executive pay ........................................................58

5.4 Firm size and executive pay ..........................................................................59

5.5 Market uncertainty and executive pay ........................................................61

5.6 Politicized environment and executive pay..................................................62

5.7 Summary of the literature on complexity-pay relation..............................63

5.8 Literature gap and complexity hypotheses..................................................656 Can powerful managers extract rents? ..................................................................66

6.1 Towards a framework of power and pay.....................................................67

6.1.1 Defining executive power............................................................................67

6.1.2 Executive pay theories and executive power...............................................69

6.1.3 Prior studies on the link between executive power and pay........................70

6.1.4 Related studies on CEO power ....................................................................72

6.1.5 Conceptual framework of executive power and compensation...................74


10/230

TABLE OF CONTENTSXII

6.2 Ownership power and executive compensation ..........................................76

6.2.1 Executive ownership....................................................................................76

6.2.2 Non-executive directors ownership............................................................78

6.2.3 Shareholder concentration ...........................................................................79

6.3 Structural power and compensation ............................................................81

6.3.1 Non-executive and independent directors ...................................................81

6.3.2 CEO duality .................................................................................................83

6.3.3 Presence of a compensation committee .......................................................84

6.3.4 Composition of the compensation committee .............................................86

6.3.5 Board size ....................................................................................................87

6.4 Tenure power and executive compensation.................................................88

6.4.1 CEO tenure ..................................................................................................88

6.4.2 Interdependent directors ..............................................................................90

6.5 Network power and executive compensation ..............................................91

6.5.1 Executive board memberships.....................................................................92

6.5.2 Reciprocal and social interlocks ..................................................................93

6.5.3 Non-executive board memberships .............................................................94

6.6 Credibility power and executive compensation ..........................................94

6.6.1 CEO celebrity status ....................................................................................94

6.6.2 Prior firm performance ................................................................................95

6.6.3 Expertise ......................................................................................................97

6.7 Summary of the literature on power pay relation ...................................97

6.8 Literature gap and power hypotheses........................................................100

7 Research methods..................................................................................................102

7.1 Development of hypotheses.........................................................................102

7.2 Sample and data...........................................................................................102

7.2.1 Archival data..............................................................................................102

7.2.2 Survey........................................................................................................104

7.3 Model specification ......................................................................................105

7.3.1 Regression analysis....................................................................................105

7.3.2 Unit of analysis..........................................................................................106


11/230

TABLE OF CONTENTS XIII

7.4 Measuring executive compensation............................................................107

7.4.1 Cash compensation ....................................................................................108

7.4.2 Share awards..............................................................................................108

7.4.3 Stock options .............................................................................................108

7.5 Measuring complexity .................................................................................109

7.5.1 Archival data..............................................................................................109

7.5.2 Survey data ................................................................................................110

7.6 Measuring executive power.........................................................................111

7.6.1 Ownership power.......................................................................................111

7.6.2 Structural power.........................................................................................112

7.6.3 Tenure power.............................................................................................113

7.6.4 Network power ..........................................................................................113

7.6.5 Credibility power .......................................................................................114

7.6.6 Survey data ................................................................................................114

7.7 Other variables.............................................................................................115

7.7.1 Firm financial performance .......................................................................115

7.7.2 Firm risk ....................................................................................................116

7.7.3 CEO age.....................................................................................................116

7.7.4 Anglo-American board members and top managers .................................116

7.7.5 Disclosure effects.......................................................................................117

7.7.6 Industry effects ..........................................................................................117

7.7.7 Consultants ................................................................................................118

8 Research results.....................................................................................................119

8.1 Descriptive statistics ....................................................................................119

8.1.1 Overview of the Swiss pay scene ..............................................................119

CEO compensation ............................................................................................119

Top management compensation.........................................................................120

Non-executive compensation.............................................................................122

8.1.2 Evidence on complexity variables .............................................................123

8.1.3 Evidence on managerial power variables ..................................................124

8.1.4 Other variables...........................................................................................126

8.2 Basic determinants of executive pay ..........................................................127


12/230

TABLE OF CONTENTSXIV

8.2.1 The determinants of the level of executive pay.........................................127

8.2.2 The determinants of equity-compensation mix .........................................130

8.3 The influence of complexity on executive pay ...........................................1318.3.1 The influence of individual complexity variables on pay levels ...............131

8.3.2 The influence of individual complexity variables on pay mix ..................134

8.3.3 Survey measures of complexity and pay levels and mix...........................135

8.3.4 The complexity construct and executive pay.............................................137

8.4 The influence of power on executive pay ...................................................140

8.4.1 The influence of individual power variables on pay levels .......................140

8.4.2 The influence of individual power variables on pay mix ..........................143

8.4.3 Survey measures of power.........................................................................145

8.4.4 The power construct and executive pay.....................................................147

8.5 The overall influence of power and complexity on executive pay ...........150

8.6 Sensitivity analysis .......................................................................................155

8.6.1 Does the composition of the compensation committee matter? ................155

8.6.2 Alternative measures for complexity variables .........................................156

8.6.3 Alternative measures for power variables .................................................157

8.6.4 Alternative measure of firm performance..................................................157

9 Discussion..............................................................................................................159

9.1 Complexity and executive pay ....................................................................159

9.2 The incremental contribution of firm size .................................................160

9.3 Technological complexity and executive pay.............................................160

9.4

The components of complexity ...................................................................161

9.5 The components of managerial power .......................................................162

9.6 Interdependent directors and executive pay .............................................162

9.7 Compensation committee existence and executive pay ............................163

9.8 Managerial power and pay structure.........................................................164

9.9 CEO duality as a self-serving opportunity? ..............................................166

9.10 Board size and executive pay ......................................................................167


13/230

TABLE OF CONTENTS XV

9.11 The influence of Anglo-Americans.............................................................167

10 Conclusions ...........................................................................................................169

10.1 Contributions to the literature....................................................................169

10.2 Summary of results......................................................................................170

10.2.1 Overview of executive and director compensation ...............................170

10.2.2 Basic determinants of executive pay .....................................................171

10.2.3 Complexity and executive pay ..............................................................172

10.2.4 Power and executive pay .......................................................................173

10.3 Implications for policy makers and practitioners .....................................175

10.4 Limitations....................................................................................................176

10.5 Future research............................................................................................177

Appendix - questionnaire ..............................................................................................179

References......................................................................................................................183


14/230

XVI


15/230

LIST OF TABLES XVII

List of tables

Table 1: Empirical studies on the pay-performance relation. ................................................ 20

Table 2: Overview of studies on directors pay....................................................................... 52

Table 3: Empirical studies on complexity-CEO pay level relation........................................ 64

Table 4: Empirical studies on complexity executive pay mix relation. .............................. 65

Table 5: Major empirical studies on the CEO/board power variables and CEO pay level.... 97

Table 6: Empirical studies on CEO power pay mix relation............................................... 99

Table 7: Descriptive statistics on CEO compensation. ........................................................ 120

Table 8: Descriptive statistics on top management compensation....................................... 121

Table 9: Descriptive statistics on non-executive director compensation. ............................ 122

Table 10: Descriptive statistics on complexity variables. .................................................... 124

Table 11: Descriptive statistics of executive power variables. ............................................ 125

Table 12: Summary statistics of other variables. ................................................................. 127

Table 13: Correlation matrix of basic determinants............................................................. 128

Table 14: OLS estimation of common determinants on total executive compensation....... 129

Table 15: Tobit model estimates of common determinants on equity-pay mix................... 130

Table 16: Correlation matrix based on individual complexity variables. ............................ 131

Table 17: OLS estimation of individual complexity variables on compensation levels. ..... 133

Table 18: Tobit estimates of individual complexity variables on compensation mix.......... 135

Table 19: Descriptive statistics on survey complexity measures......................................... 136

Table 20: Univariate estimates of surveyed complexity on CEO compensation. ................ 136

Table 21: Univariate Tobit estimates of CEO equity-based compensation on survey

complexity variables. ........................................................................................................... 137

Table 22: Correlation matrix of archival complexity variables. .......................................... 138

Table 23: Principal component analysis and scoring coefficients of archival

complexity construct............................................................................................................ 138

Table 24: Principal component analysis and scoring coefficients of survey

complexity construct............................................................................................................ 139

Table 25: Relationship between complexity constructs and CEO compensation. ............... 140

Table 26: OLS estimation of individual power variables on compensation level................ 142


16/230

LIST OF TABLESXVIII

Table 27: Tobit estimates of individual power variables on executive compensation mix.. 144

Table 28: Descriptive statistics on survey power measures................................................. 146

Table 29: Univariate estimates of survey power variables on CEO compensation. ............ 147

Table 30: Principal component analysis of archival power construct.................................. 148

Table 31: Scoring coefficients of archival power construct................................................. 149

Table 32: Principal component analysis on survey power measures. .................................. 150

Table 33: Scoring coefficients of survey power construct................................................... 150

Table 34: Regressions of the archival complexity and power constructs on executive pay. 152

Table 35: Tobit regressions of the complexity and power constructs on executive pay

structure................................................................................................................................ 154

Table 36: Regressions of the survey complexity and power constructs on CEO

compensation. ...................................................................................................................... 155

Table 37: OLS estimation of the composition of the compensation committee on TMT

compensation. ...................................................................................................................... 156


17/230

LIST OF FIGURES XIX

List of figures

Figure 1: Dissertation overview............................................................................................... 7

Figure 2: Determinants and consequences of executive compensation. ................................ 13

Figure 3: Conceptual framework of complexity and executive compensation. ..................... 55

Figure 4: Conceptual framework of executive power and compensation. ............................. 75


18/230

XX


19/230

LIST OF ABBREVIATIONS XXI

List of abbreviations

CEO Chief Executive Officer

DOI Degree of internationalization

EJPD Eidgenssisches Justiz- und Polizeidepartement

FASB Financial Accounting Standards Board

FETE Foreign employees total employees

FSTS Foreign sales total sales

GBP British Pound

Ln Natural logarithm

LTIP Long-term incentive plan

M & A Mergers & acquisitions

MCHF Million Swiss Francs

MUSD Million US Dollar

R & D Research & Development

SIC Standard Industry Classification

UK United Kingdom

US United States

USD US Dollar

S & P 500 Standard & Poors 500

SWX Swiss Stock Exchange

TMT Top management team


20/230

XXII


21/230

SUMMARY XXIII

Summary

This paper examines the roles of complexity and power as explanations for executive

pay level and mix. It is hypothesized that complexity and power are positively associated

with the level of compensation. Further, I hypothesize that complexity is positively and

executive power negatively related to the proportion of executive compensation that is

based on stock options and restricted stock. Despite numerous research articles

investigating variables associated with complexity and executive power and their effect

on executive pay, a broad conceptualization of both complexity and power with respect

to the setting of executive pay is missing. This dissertation attempts to close this gap. For

a sample of 199 Swiss stock-listed companies in 2002-2003, I collect a large number of

archival variables to empirically test the effects of complexity and power constructs on

executive compensation. I augment these results with a sample of 47 survey-based

responses to obtain a richer conceptualization of executive power and firm complexity.

Multiple regressions are run on an individual variables basis as well as on factor-

analyzed categories. Results are broadly consistent with hypotheses. Complexity is

positively associated with CEO and top management team compensation levels andequity-pay mix. Managerial power is generally positively associated with pay levels.

Contrary to my hypothesis, I also find evidence that some dimensions of executive

power are positively related to the proportion of equity in total compensation.


22/230

XXIV


23/230

INTRODUCTION 1

1 Introduction

1.1 Why study executive pay?

Executive compensation has attracted widespread attention in recent years and has

become one of the focus topics in corporate governance (Felton, 2004). At least three

reasons have contributed to this increased attention on executive compensation for

academics and practitioners alike: A fascination with and sometimes lack of

understanding for the high levels of CEO compensation, the importance of the CEO as

the main strategic decision maker and the increased transparency of executive

compensation data in many countries.

First, the pay of corporate leaders has undisputedly escalated in the last few decades,

while workers pay has stagnated. The average real pay for chief executive officers of

S&P 500 firms increased significantly during the 1990s, growing from 3.5 MUSD in

1992 to 14.7 MUSD in 2000. Most of this increase reflects the escalation in stock

options valued at the time of grant, which now constitute the single largest component of

executive pay (Hall and Murphy, 2003). Average real CEO pay in the S&P 500

decreased again to 9.4 MUSD in 2002.1 Even for large stock-listed companies, these

amounts are substantial. Average CEO compensation now constitutes around 8% of

corporate profits for US firms (Balsam, 2002: 262; Bebchuk and Fried, 2003). Over the

same period, the pay of most employees has increased only marginally, increasing the

ratio of CEO pay to the average factory workers pay from 1:42 in 1980 to 1:531 in 2000

(Felton, 2004) and 1:431 in 2004 (Economist, 2005a). In Switzerland, some CEOs seem

to have multiplied their total compensation in recent years, in spite of modest stock

market developments (Schtz, 2005). At UBS, a large bank, top executives now earn 230times the average pay of low-paid employees (Wittwer, 2005a: 27). The rapid increase in

CEO compensation is difficult to comprehend for many people, given simultaneous

corporate downsizing, employee layoffs, plant closings (Abowd and Kaplan, 1999; Dial

and Murphy, 1995; Murphy, 1999; Murphy, 1997) and poor financial performance (e.g.

1 At the same time as average pay decreased, median pay rose, indicating that the declining averagewas the result of a significant fall in the earnings of a few extremely highly compensated executives(Useem, 2003).


24/230

CHAPTER 12

Krauer, 2004). In short, there is a widely held belief that top executives are overpaid

(Gomez-Mejia, 1994). Anecdotal evidence indicates that CEOs of large Swiss

corporations are the best paid in Europe (Schtz, 2005: 73).

A second reason for the high attention paid to CEO compensation by practitioners and

academics is the CEOs position at the apex of the firm. As the main strategist for the

firm, his or her compensation may have a direct bearing on business decisions affecting

the future of the entire organization. How a firms top managers are compensated and

how much equity of their firm they hold has a significant impact on corporate strategic

decisions (Sanders, 2001a; Datta et al., 2001; Bliss and Rosen, 2001), the value

development of the firm (Morck et al., 1988; McConnell and Servaes, 1990) or the

compensation policies for middle and lower-level managers (Gomez-Mejia, 1994).

Furthermore, the structure of executive compensation mainly the large option grants

has been linked to recent corporate collapses in the US. With large stock options payoffs

in view, executives manipulated their books and inflated stated earnings in the search for

higher share prices (Hall and Murphy, 2003; Bebchuk et al., 2001; Bolton et al., 2002;

Felton, 2004; Frey, 2004; Hambrick et al., 2005). This has led experts to conclude that

much may be wrong with executive compensation (Elson, 2003) and stock options in

particular (Hall and Murphy, 2003) today.Third, executive pay is increasingly transparent, as companies are required to publish

increasing details of remuneration data in various countries (e.g. Schildknecht, 2004:

140). The Swiss Stock Exchange SWX required its listed companies to disclose the

compensation of the top management team and the board of directors in 2002.

Subsequently, interest in top management pay rapidly increased and has been a regular

topic in news publications. An amendment of the law to make more detailed disclosure

mandatory is under way (Gerny, 2004; EJPD, 2003).

1.2 The evolution of executive compensation research

Along with the increase in executive compensation levels, the volume of related research

papers has soared. Over the last several decades, hundreds of studies have been

conducted on the determinants and to a lesser degree the consequences of executive

compensation. Already more than a decade ago, Gomez-Mejia (1994) stated that

probably no other single variable had received as much empirical attention across

different business fields and related social sciences as executive compensation.


25/230

INTRODUCTION 3

Spurred by the separation of ownership and control (Berle and Means, 1932; Fama and

Jensen, 1983) and agency theory (Jensen and Meckling, 1976), research has mainly

focused on the strength of the relationship between pay and firm performance (e.g.

Lewellen and Huntsman, 1970; Murphy, 1985). Conversely, managerialists (Tosi et

al., 2000) were interested in whether pay was related to firm size and sales growth rather

than financial performance (Baumol, 1959; Ciscel, 1974). This started the sales versus

profit debate originally tested by McGuire et al. (1962), which returned inconsistent

results.

In a seminal study, Jensen and Murphy (1990a) showed that a change in total

shareholder wealth of 1,000 dollars was associated with a change of 3.25 dollars of total

CEO wealth, and concluded that the size of the relationship is too weak to provide useful

incentives. Since then, numerous studies have replicated such investigations of the pay-

performance relationship (e.g. Lippert and More, 1994; Hall and Liebman, 1998;

Murphy, 1999; Brunello et al., 2001). However, little consistency has been found for the

relationship between pay and performance, partly due to differences in sample periods,

methods applied, different performance measures and control variables included (Joskow

and Rose, 1994). In a meta-analysis of 137 CEO pay studies, Tosi et al. (2000)

documented that firm performance accounts for less than 5% of the variance, while firmsize accounts for more than 40% of the variance in total CEO pay. The same authors

noted that the findings of studies on executive pay as a control mechanism are

remarkably inconsistent not only with theory but with each other (Tosi et al., 2000:

305).

Due to these discouraging results, recent work has begun to move beyond the pay-

performance framework to suggest other potential factors influencing executive

compensation (Barkema and Gomez-Mejia, 1998). These factors include the role of risk

(Gray and Cannella, 1997; Carpenter, 2000; Aggrawal and Samwick, 1999; Miller et al.,

2002), human capital (Harris and Helfat, 1997; Combs and Skill, 2003; Sanders et al.,

2001), skill (Kornhauser et al., 2005), social similarity (Belliveau et al., 1996), power

(Sridharan, 1996; Barkema and Pennings, 1998), CEO reputation (Milbourn, 2003),

symbolism and impression management (Zajac and Westphal, 1995; Siegel and

Brockner, 2005), the role of managerial discretion (Finkelstein and Boyd, 1998) and

stakeholder management (Coombs and Gilley, 2005). Overall, these advances have


26/230

CHAPTER 14

shown that executive pay levels and structure are the consequence of the interplay

between various firm- and CEO-specific factors.

Economists and financial scholars have turned their focus towards design specificationsof stock option plans (Brenner et al., 2000; Hall and Murphy, 2000; Johnson and Tian,

2000a; Archarya et al., 2000; Bettis et al., 2005), stock option repricing (Chen, 2004;

Carter and Lynch, 2001; Chance et al., 2000; Challaghan et al., 2004; Chidambaran and

Prabhala, 2003) and the valuation of executive stock options (Carpenter, 1998; Hall and

Murphy, 2002; Johnson and Tian, 2000b; Cuny and Jorion, 1995).

Furthermore, researchers have increasingly become interested in the consequences of

compensation. They have studied how compensation policy affects managerial decisions

such as investment and debt policy (Coles et al., 2002), risk taking (Carpenter, 2000) or

corporate acquisition decisions (Datta et al., 2001) and how compensation is related to

future firm performance (Sanders, 2001b; Hanlon et al., 2003).

1.3 Literature gap and research questions

This work moves beyond the basic agency-theoretic hypothesis of linking pay to firm

performance. It extends the existing literature by investigating the concepts of

complexity and power in the design of executive compensation contracts.

Some CEOs have more complex and demanding jobs than others, and these CEOs may

command a pay premium in the managerial labor market. Although complexity was

early on suggested as an important determinant of CEO pay (Finkelstein and Hambrick,

1989), it has received very little theoretical and empirical attention in prior research on

executive compensation. Few studies have explicitly invoked a complexity explanation

for CEO pay. Moreover, operationalization of complexity has by and large been limited

to firm diversification (Finkelstein and Hambrick, 1989) or a firms degree ofinternationalization (Sanders and Carpenter, 1998). Much remains unclear on how the

notion of complexity affects executive compensation. Are executives really paid for the

complexity they manage? What kind of complexity are executives paid for? Does

complexity affect the structure of executive pay contracts?

This dissertation attempts to shed more light on how complexity influences executive

pay:


27/230

INTRODUCTION 5

What is the role of complexity in the determination of executive compensation level and

structure?

To answer this research question, I developed an integrative concept of complexity. Thisconcept models complexity along different dimensions based on a large set of variables.

The variables were collected through archival and survey data collection methods to

capture both objective and subjective notions of complexity. Applying this thorough

method, I have attempted to give a more comprehensive account of how complexity

affects compensation level and structure.

Power stands at the heart of agency theory (Fama, 1980; Jensen and Meckling, 1976).

According to the agency perspective, the evolution of the public corporation has

dispersed corporate ownership. This has created a separation between ownership and

control and an opportunity for CEOs to maximize their personal wealth at the expense of

shareholders (Berle and Means, 1932; Jensen and Meckling, 1976). To empirically test

this basic agency assumption, researchers focused almost exclusively on the magnitude

of interest alignment in terms of the pay-performance sensitivity. A significant and high

pay-performance sensitivity would indicate that agency problems are effectively

mitigated. The level of compensation, arguably an important indication of a CEOs

pursuit of self-interest, has traditionally played a minor role, as agency theorists advocatethat it is not how much you pay but how you pay (Jensen and Murphy, 1990b). The

relatively scarce empirical literature on power has yielded ambiguous results. This is

mainly due to insufficient operationalization of power variables (Grabke-Rundell and

Gomez-Mejia, 2002) rather than poor theoretical ground. Better constructs and

operationalization of variables are necessary to capture the full picture of how more

powerful CEOs, as opposed to less powerful ones, manage to extract rents. Bebchuk and

Fried (2003; 2004) argue that managerial power is the reason for the extraction of rent in

recent years and decades. However, a valid construct measuring cross-sectional

differences in managerial power and its influence on compensation is still missing. A

call by Hambrick and Finkelstein (1996: 281) that a broader conceptualization of power

is needed in the context of executive compensation is still unanswered.

This dissertation attempts to close this gap by answering the following question:

What is the role of executive power in the determination of executive compensation?


28/230

CHAPTER 16

I developed a conceptualization of board-CEO power grounded in theory and earlier

empirical results and created a measurement method that adequately captures the

multiple dimensions of executive power and its relation to compensation level and mix. I

also augmented archival data with survey data, as archival data cannot adequately

capture the nuances underlying executive compensation decisions (Gomez-Mejia, 1994).

1.4 Overview of the paper

The paper is structured as follows. In chapter 2, I describe the process of executive pay

setting in Switzerland. Today, many Swiss companies have established a compensation

committee. This committee is the body of the board mainly responsible for setting

executive compensation. The influence of executive management on pay determinationmay vary substantially across companies.

The vast literature on top management compensation is reviewed in chapter 3. Major

streams of research are discussed. The most influential concept in studying executive

compensation is agency theory. Due to the non-alignment of interests between

shareholders and managers, performance-based compensation helps reduce agency costs.

Researchers have focused on measuring how executive pay and wealth responds to

concurrent changes in firm performance in determining the extent of reductions inagency costs. While significant relations between pay and performance have been

detected, results were generally disappointingly low. As a result, other concepts have

been studied to increase the understanding of executive pay. Human capital, risk,

discretion, or managerial power have all been invoked and partially confirmed as

determinants of executive pay. Academic research also investigated the effects of

compensation structure and level on a firms strategic behavior.

Chapter 4 contains a review of the literature on director pay. Closely linked to executivepay, non-executive director pay has only recently received more attention by academics.

One reason for this lack of interest may be the implicit assumption of most studies on

executive pay that the board of directors acts in the interests of shareholders. Empirical

results show that the determinants of executive and director pay are remarkably similar.

In Chapter 5, the notion of complexity and its application to executive compensation are

discussed. Research hypotheses are formulated. While complexity has been examined as

a determinant of CEO pay in earlier studies, operationalization has generally remained


29/230

INTRODUCTION 7

weak. Prior studies used proxies such as firm size and firm diversification to gauge the

extent of complexity. This chapter develops a broad concept of complexity related to a

CEOs job. In this respect, complexity is multidimensional. Firm size, diversification,

internationalization, a politicized environment and market uncertainty are identified as

different dimensions of complexity. It is hypothesized that complexity is associated with

higher levels of pay and a higher proportion of equity-based pay out of total pay.

Figure 1: Dissertation overview

9. Discussion

10. Conclusions

1. Introduction

2. Compensationsetting process

3. Literature reviewon executive pay

5. Are executives paid for thecomplexity of the job they have?

7. Research methods

8. Research results

6. Can more powerful managersextract rents?

4. Literature reviewon director pay

Chapter 6 contains the literature review and the development of the research gap and

hypotheses with respect to managerial power. Notions of power have been frequently

studied in executive compensation research. Agency theory, the dominant theoreticalapproach, argues that the separation of ownership and control in a large organization

creates a power base for executive management. However, most prior studies have

applied single measures of power, and results are often inconsistent and ambiguous. A

more complete concept of power is developed.

Research methods used to test the hypotheses developed in chapters 5 and 6 are

discussed in chapter 7. The hypotheses are tested applying both archival data and survey

data to capture a broad concept of power and complexity. I collected compensation data


30/230

CHAPTER 18

for 199 Swiss stock-listed companies in 2002 and 2003, resulting in a dataset of 398

firm-year observations. Both CEO compensation and top management team

compensation are investigated. A large number of complexity and power variables were

also collected for these years. In addition, I conducted a survey among compensation

committee chairmen to capture more subtle and subjective elements of power and

complexity. This second data set consists of 47 observations in 2004. To reduce the

number of variables in regression analysis, I apply principal components factor analysis.

Research results are provided in chapter 8. In general, results confirm that several

dimensions of complexity are significantly associated with executive compensation

levels and mix. The individual complexity dimensions size, internationalization, and

politicized environment are all related to executive pay, while diversification and market

uncertainty show no significant association with pay. Further, the subjective measure

perceived complexity is significantly related to CEO pay. Results based on variables

transformed by factor analysis show that these dimensions all load on one principal

component and confirm a strong association between complexity and compensation. The

fifteen power variables sort into six principal components. Several of these components,

such as outside ownership power and structural power, are associated with executive

compensation levels.

I discuss the results in chapter 9. For instance, several dimensions of managerial power

are significantly associated with compensation mix. However, the sign of the results is

not consistent with the idea that more powerful managers prefer cash over equity-based

compensation. One explanation may be that internal and external political constraints

restrict the amount of cash compensation, and that stock options and share awards

provide better opportunities for managers to extract rents. Conclusions are provided in

chapter 10.


31/230

THE COMPENSATION SETTING PROCESS 9

2 The compensation setting process

The general roles of the board of directors are to set the companys strategic direction, to

advise and monitor top management and to otherwise protect the interests of

shareholders (e.g. Styles and Taylor, 2003; Hilb, 2005). Determining executive pay is an

important task in fulfilling the boards duties (Finkelstein and Hambrick, 1996). This

chapter describes institutional details of the executive pay setting process.

2.1 Purpose of compensation policies

Many companies have a charter for the board of directors and their committees. Thesecharters outline in detail the purpose, tasks and responsibilities of the board and its

committees. The purpose of compensation policies as frequently stated is that executive

compensation is set such as to retain and attract executives who are needed to ensure the

competitiveness and long-term success of the business. Most companies publishing a

charter for the compensation committee further state the goal of establishing a link

between pay and performance. For instance, the charter for the compensation committee

of the UBS Board of Directors states that the compensation committee will support

policies and practices to attract, motivate and retain executives, ensure competitiveness,

long-term business success, shareholder interests and a strong pay-performance link.

While other companies use a different wording, the contents largely remain the same.

Companies that do not have or do not disclose a compensation committee charter often

state the same purpose in their annual reports. This would imply that compensation

committees only pay as much as needed to attract and retain the necessary individuals at

the top and that they aim at establishing a strong link between pay and performance.

Some companies further mention that the value of the position or individual qualificationrequired for that position play a role in determining compensation levels. This indicates

that complexity influences executive pay.

2.2 Roles and responsibilities

The compensation committees role is to either determine executive compensation or to

make recommendations to the full board of directors. The board expects to adopt the

compensation committees recommendations with possible modest modifications


32/230

CHAPTER 210

(Newman and Mozes, 1999). The compensation committee typically surveys market

compensation levels, establishes performance benchmarks and salary policies, and

evaluates managements performance against financial and non-financial goals. These

activities, however, often depend on managements cooperation in providing background

information and advice (Crystal, 1991). Even though the CEO usually is not a member

of the compensation committee, he may sometimes attend its meetings. In companies

that do not have a compensation committee, the full board assumes the role of

determining executive pay. This is especially the case in small boards where all tasks are

fulfilled by the combined board. Some Swiss companies further combine the specific

responsibilities of nomination and compensation in a nomination & compensation

committee. Many compensation committees use external compensation consultants toestablish market conditions. These consultants typically compare the pay levels and

structure of firms in the same industry and of similar size and make recommendations to

the board of directors on how to compensate the CEO and his top management team. In

some companies, the internal audit function regularly reviews the compensation setting

process and submits a report to the board of directors.

2.3 Organization of the compensation committee

The compensation committee typically consists of three to four members of the board of

directors, who are appointed for one year. A majority and sometimes all of its

members are non-executive and independent. This should ensure that the objectivity

required to perform their duties is not impaired. However, it is not uncommon that the

CEO and other executives attend the meeting of the compensation committee without

formal voting power. The head of HR occasionally functions as the secretary to the

compensation committee. Furthermore, internal or external specialists may be invited to

hold presentations if deemed necessary. Meeting minutes are usually made available toall members of the board of directors. Compensation committees meet less frequently

than the full board, typically twice or three times a year, if circumstances do not require

otherwise.

2.4 Elements and allocation of executive pay

Compensation is largely made up of five components: Salary, bonus, option grants, share

grants and benefits. Salary and bonus are usually paid in cash. Bonuses are typically

awarded for achievement of accounting-based performance targets of the prior year and


33/230

THE COMPENSATION SETTING PROCESS 11

are awarded once a year. Benefits include additional cash contributions to companies

pension funds, company cars or other perquisites. Not all companies make use of equity-

based option or shares grants. Similar to bonus payments, option and restricted stock

grants are usually made once a year. Only a small minority of companies has an award-

frequency that is higher than that. The adjustment of the strike price of options, called

repricing, is possible. However, little is known about repricing practice in Switzerland.

Interestingly, a few companies explicitly mention that the board considers the awards

given to the CEO and other top executives in past years when it determines current-year

compensation. While it remains unclear what elements of compensation that would

include, it indicates a certain slack in compensation levels. This suggests that once a

certain level of CEO compensation has been established, for instance after a period of

strong growth or profitability, a later significant downward adjustment seems unlikely.

In Switzerland, employee options of stock-listed companies are taxed at grant. The

taxable income is determined by the fair value of options at the time of grant adjusted for

the vesting period. The fair market value is usually determined by the Black-Scholes

option-pricing formula (Black and Scholes, 1973). For each year that the options are

blocked for exercise, the tax value is reduced by roughly 6%. Taxation at grant means

that employees are taxed for a potential benefit they may never actually be able to cashin.

2.5 The influence of executives on their compensation

Sections on executive compensation in annual reports revealed numerous instances

where CEOs appeared to have direct or indirect influence upon the contracting process

for their own pay. Some companies openly acknowledge in annual reports that CEOs

and other corporate executives such as the head of HR helped structure their own

compensation with the role of board committees potentially limited to ratifying

management proposals. For example, Tamedias 2004 annual report states that the

remuneration of the top management team is determined based on the proposal of the

CEO to the Board of Directors. Similarly, Kaba described the role of its board of

directors by writing in its 2004 annual report, it is the task of the compensation

committee todetermine the compensation policies for the top management team upon

proposal of the delegate to the board of directors and to approve the emoluments of the

top management team... Even when CEOs cannot directly influence their own pay, they


34/230

CHAPTER 212

often propose the pay of their immediate subordinates, which in turn may influence their

own pay, because paying regular members of the top management team more necessarily

reduces the gap between the CEO and the average TMT member. This, in turn, provides

a legitimate reason to increase CEO compensation to ensure internal fairness. For

instance, SIGs 2004 annual report states that the level of compensation for the other

members of the Group Executive Committee is setby the Nomination &

Compensation Committee, based on the proposal of the CEO.

In addition to these acknowledgments of direct management participation in setting the

terms of compensation, many annual reports suggested the presence of conflicts of

interests in the contracting process. As discussed above, a few Swiss stock-listed

companies reported having CEOs or other executives who served as members of their

own compensation committee. Other companies reported that non-executive directors

benefited from personal consulting contracts or from diversion of company business to

their principal employers. In addition to these channels for CEOs to grant favors to the

directors who set their compensation, the process for recruiting and reappointing

members of the board itself had long been understood to be influenced by the CEO in

most companies. A central tenet of this paper is that CEOs exert influence over their

board of directors and compensation committees in these and other ways, and that theyexploit this power to increase the value and lower the riskiness of their compensation.


35/230

RESEARCH REVIEW ON EXECUTIVE PAY 13

3 Research review on executive pay

The number of research papers on executive pay has increased dramatically over the last

few decades. Academics of such diverse fields as accounting (e.g. Antle and Smith,

1986), economics (e.g. Jensen and Murphy, 1990a), finance (e.g. Baker et al., 1988;

Carpenter, 2000), human resources (e.g. Kostiuk, 1990), management (e.g. Barkema and

Gomez-Mejia, 1998), industrial relations (e.g. Agarwal, 1981) and sociology (e.g. Allen,

1981a) have published studies on the determinants and effects of executive

compensation.

Figure 2: Determinants and consequences of executive compensation.

Economic

explanations Performance

Complexity

Risk

Managerial labor market

Investment opportunities

Managerial discretion

Human capital

Market forces

Executives personal

characteristics

Stewardship theory

Executivecompensation

Cash compensation

Equity-based

compensation

Total compensationPolitical

explanations Managerial power

Impression management

Social explanations

Social comparison Isomorphism

Performance

Dividend payments

Stock repurchases

Managerial retention Risk

Mergers, acquisitions and

divestitures

Capital investments

Expectations

management

Earnings management

Litigation

This vast body of literature has generated not only useful insights, but also many

contradictory findings. Any review on executive pay will therefore have to remain

incomplete. Good general reviews are provided by Gomez-Mejia (1994), Finkelstein and

Hambrick (1996), Gomez-Mejia and Wiseman (1997) or Murphy (1999). Core and Guay

(2003) review the literature on equity-based executive pay. Balsam (2002) provides a

good introduction into and overview of executive pay. Figure 2 presents an overview of


36/230

CHAPTER 314

the determinants and consequences of executive compensation level and structure, as

discussed in this and subsequent chapters.

3.1 The relationship between pay and performance

3.1.1 The debate over size versus profits

Prior to 1980, only a handful of studies of executive compensation were published. Most

of these studies focused on whether pay was more closely tied to company size or

company profits. Economists were interested in examining hypotheses derived from the

traditional theory of the firm that top managers operate to maximize profits. It was

contended that executive compensation (i.e. salary and bonus2) would be closely linked

with profitability (e.g. Lewellen and Huntsman, 1970). According to this view,

competition in the managerial labor market (Fama, 1980) and the structure of the

managers compensation contracts unite managerial interests with shareholder interests.

Consequently, managers act to maximize profits and shareholder wealth.

To recognize elements of oligopolistic competition, an alternative managerialist

hypothesis was introduced. Due to the separation of ownership in large firms, managers

seek their own personal goals, such as maximization of perquisites, power and control,

and they achieve these goals by maximizing sales and not profits or shareholder wealth.The managerialist hypothesis therefore stated that compensation would be more closely

related to sales revenues subject to a minimum profit constraint (Baumol, 1959). This

alternative view was also termed sales-maximization hypothesis.

This started the sales versus profit debate. Researchers examined whether changes in

executive pay were more closely related to changes in sales revenues or changes in

profits. For instance, McGuire et al. (1962), Ciscel (1974) and Schmidt and Fowler

(1990) among others concluded that top executive compensation appears to be drivenmore by organization size than performance. At the other extreme, Deckop (1988),

Lewellen and Huntsman (1970) and Masson (1971) accorded firm performance a

stronger role than size. These early studies suffered from severe methodological

problems. For example, collinearity can make a difference in how one interprets

regression coefficients, if both firm size and profits are correlated (Ciscel and Carroll,

2 At that time, the award of equity-based pay was still rare.


37/230


1980). Also the type of firm performance measure used affects the results (Masson,

1971).

3.1.2

Agency theory and optimal contractingSince the late 1970s, agency theory has emerged as the major theoretical approach to

studying executive pay. The original foundation of agency theory, however, goes back to

the early 1930s. At that time, Berle and Means (1932) argued that the owners of a

company usually do not have the power to place important pressure upon management,

since individual ownership interests are very small. They built their arguments on three

propositions: First, economic power is concentrated with a few large corporations in

each industry. Second, stock ownership of these corporations is dispersed. And third,

executives manage companies without owning them.

The relationship between shareholders of a publicly owned corporation and the

corporations executives is a classic example of a principal-agent problem. Jensen and

Meckling (1976: 308) define an agency relationship as a contract under which one or

more persons (the principal(s)) engage another person (the agent) to perform some

service on their behalf, which involves delegating some decision-making authority to the

agent. Agency theory assumes that the agent is a self-interested rational individual and

seeks to pursue personal value maximization, which may not necessarily be aligned with

shareholder value maximization. There is goal incongruence between managers and

shareholders. For instance, building a diversified and large company may reduce the

managers risk of unemployment and increase his compensation (Murphy, 1999).

However, shareholder value may not necessarily be maximized with such a strategy.

Another example of agency cost is the purchase of a corporate jet or a luxury office

building. Both of these actions are unlikely to result in a net increase in shareholder

wealth.A further building block of agency theory is information asymmetry. If shareholders had

complete information regarding the CEOs activities and the firms investment

opportunities, they could design a contract specifying and enforcing the managerial

action to be taken in each state of the world. In most cases, however, shareholders do not

know what actions the CEO can take and they cannot perfectly observe managerial

actions and investment opportunities. Therefore, information asymmetry excludes the

design of a detailed action plan to be implemented by the CEO.


38/230

CHAPTER 316

Agency theorists argue that a compensation contract can be written that alleviates

problems associated with goal incongruence and information asymmetry between

managers and shareholders. The traditional agency-framework defines an optimal or

efficient contract as one that maximizes the net expected economic value to shareholders

after transaction costs (such as contracting costs) and payments to employees (Core et

al., 2003). In this view, the board of directors sets executive pay to minimize agency

costs. Compensation policy will be designed to give the manager incentives to select and

implement actions that increase shareholder wealth. As actions cannot be directly

specified in the compensation contract, the principal will negotiate outcome-based

elements of compensation in the form of bonus payments, stock options and restricted

stock awards (Eisenhardt, 1989). Equity-based compensation is a particularly effectivemeans to solve the agency problem (Haugen and Senbet, 1981), especially as it is hard in

practice to use salary and bonus to reward and penalize CEOs (Hall and Liebman, 1998).

Agency theory further predicts that variable pay and equity holdings are used as

complementary sources of pecuniary incentives, where their relative importance varies

across firms and industries, depending on the relative costs and gains (Lambert and

Larcker, 1987).

3.1.3

The pay-for-performance sensitivityOptimal contracting suggests the alignment of shareholders and managers interests

through the provision of salary revisions, bonus compensation, and equity-based

compensation.3 As a measure of agency costs reduction, the sensitivity between pay and

performance has developed as the most studied phenomenon in executive compensation

research. Pay-performance sensitivity is defined as the dollar change in the CEOs pay or

wealth associated with a dollar change in the wealth of shareholders. The underlying

idea is that the existence of competition in capital markets makes the survival of the firm

depend on the provision of incentive compensation, which encourages the CEO to act in

the shareholders interest. Firms which fail to compensate managers in this way will face

higher costs and thus will not compete successfully with firms whose managers act in the

shareholders interest. Pay performance sensitivity is typically measured by tracking

3 Supporting incentive effects of equity compensation, the announcements of proposed changes inequity-based managerial compensation packages have been met with positive stock market reactions(Brickley et al., 1985; Larcker, 1983).


39/230


current changes of firm financial performance (typically total shareholder return) to

changes in CEO compensation or wealth.

Murphy (1985) and Coughlan and Schmidt (1985) document that changes in executivecash compensation are positively related to current year stock price changes. While the

relation between changes in executive cash pay and changes in shareholder wealth is

statistically significant, the effect is small and explains little of the variance in executive

compensation (Jensen and Zimmerman, 1985). Bentson (1985) establishes the

importance shares and option holdings have in calculating the appropriate pay-

performance sensitivity. He investigates CEO wealth effects of 29 large conglomerates

between 1970 and 1975 and finds that managers tend to gain and lose along with

shareholders, if the changes in value in CEO share- and option holdings are included in

the calculations.

In a seminal article, Jensen and Murphy (1990a) provide estimates for the total wealth-

performance sensitivity for a large sample of US companies between 1974 and 1986.

Jensen and Murphy estimate the value-enhancing magnitude of different compensation

elements such as salary revisions, cash bonus, stock options, stock ownership, and

dismissal decisions. Both cash compensation and total compensation (including stock

option grants and gains from exercising stock options) are positively related to firmperformance. However, the authors argue that the economic significance of these

compensation and wealth changes is low. The CEO receives an additional 1.35 cents of

cash and 3.3 cents of total compensation for each 1,000 USD increase in shareholder

wealth respectively. Their estimates of the total CEO pay-performance relation including

pay, options, stockholdings and dismissal indicates that CEO wealth changes 3.25 USD

for every 1,000 USD change in shareholder wealth. The results further demonstrate that

the incentives generated by stock ownership are large relative to direct pay incentives.

Showing a decline in real executive compensation since the 1930s and a decreased

sensitivity in the pay-performance relation since then, Jensen and Murphy (1990a) use

their findings to challenge the principal-agent paradigm. They conclude that these values

are too low, and that boards of directors should seek to increase equity-based

compensation and CEO shareholdings. In a related article, Jensen and Murphy (1990b)

argue that, from a shareholder perspective, the level of compensation is of secondary

importance to the appropriate mix of compensation. Compensation would be too low to

attract the highly-talented individuals necessary to run big companies: The very best


40/230

CHAPTER 318

lawyers or investment bankers can earn substantially more than the very best corporate

executives. Highly talented people who would succeed in any field are likely to shun the

corporate sector, where pay and performance are weakly related, in favor of

organizations where pay is more strongly related to performance and the prospect of

big financial rewards more favorable (Jensen and Murphy, 1990b: 149). They further

write that the resulting general absence of management incentives in public

corporations presents a challenge for social scientists and compensation practitioners

(Jensen and Murphy, 1990a: 262). By proposing incentive compensation, these

contributions greatly influence companies adopting equity-based compensation plans,

which consequently increased total executive pay.4

In the academic field, the contribution of Jensen and Murphy has spurred a vast amount

of literature investigating the size of the pay-performance sensitivity. Lippert and More

(1994) confirm Jensen and Murphys (1990a) results for a sample of 310 US firms

between 1974 and 1988. Their total pay-performance sensitivity is 0.0035 or 3.5 USD

CEO pay-change per 1,000 USD change in shareholder wealth. Haubrich (1994) studies

whether principal-agent and optimal contracting theory is indeed inconsistent with the

low value of pay-performance sensitivity reported by Jensen and Murphy (1990a) by

calculating numerical solutions to agency models developed by Grossman and Hart(1983) and Holmstrom and Milgrom (1987). He concludes using reasonable

assumptions about executive risk aversion, CEO effort and other variables, principal-

agent theory can yield quantitative solutions in line with the empirical results of Jensen

and Murphy (Haubrich, 1994: 259). Hall and Liebman (1998) study 478 US large

companies between 1980 and 1994. Investigating changes in total CEO wealth

including holdings of stock and stock options they argue that there is a strong link

between the fortunes of CEOs and the fortunes of the companies they manage (Hall and

Liebman, 1998: 654). As changes in firm market value are often extremely large in

absolute terms, even relatively low wealth-performance sensitivities may lead to very

large dollar rewards and punishments for CEOs. Applying the Jensen/Murphy (1990a)

statistic of CEO wealth changes, they estimate a change of 6.00 USD for every 1,000

USD change in firm value for 1994. Adjusting for increasing firm size over the

4 A further reason for the explosion of option grants is the 1994 reform that limited tax deductibility ofexecutive cash pay to 1 MUSD. Options turned into an alternative form of tax-favored compensation.


41/230


investigation period and the corresponding natural tendency for the Jensen/Murphy

statistic to fall over time, they find that sensitivities increased fourfold between 1980 and

1994 to 12 USD. The inclusion of stock and stock option revaluations in addition to

changes in direct pay increases compensation sensitivities by a factor of about 30

compared with direct pay-performance sensitivity. Murphy (1999) tests both cash pay-

performance sensitivity and elasticity for a sample of 500 US firms between 1990 and

1996 and reports an average sensitivity of 14 USD and elasticity of 0.263.

In the UK, pay-performance sensitivities are lower than in the US. Conyon and Murphy

(2000) report an average pay-performance sensitivity of 2.33 GBP per 1,000 GBP in the

largest 510 UK companies. Buck et al. (2003) report a sensitivity of 1.55 GBP of CEO

rewards per 1,000 GBP increase in shareholder value.

Very few studies on the pay-performance relationship exist outside the US and the UK.

For a sample of 48 German firms between 1968 and 1994, Conyon and Schwalbach

(2000) find a cash pay-performance elasticity of 0.071. Their pay measure is per capita

income of the top management team. Brunello et al. (2001) use survey data from 106

Italian companies and find a significant pay-performance sensitivity for upper- and

middle-level managers. However, their estimate is considerably lower than the ones

based on US data. In addition to cross-national differences, the lower sensitivity may bea result of including middle-level managers. The authors further report that the pay-

performance sensitivity is higher in foreign-owned firms, in listed firms, and in firms

affiliated with a multinational group. Table 1 summarizes a selection of studies

estimating cross-sectional differences in the pay-performance relation.

These studies of the pay-performance link suffer from two major methodological

weaknesses. First, there is no ex ante certainty on the time-lag of the pay-performance

relationship. The arbitrary research design of regressing current-year changes inexecutive pay on current year shareholder wealth need not necessarily be the appropriate

agency-theoretic proposition. Boschen and Smith (1995) show that CEO compensation

responds to changes in firm performance over the next 4-5 years, with the cumulative

response of pay to performanceroughly 10 times that of the contemporaneous

response (p. 577). Second, relying on shareholder return or accounting performance

measures provides an incomplete picture of how the board of directors evaluates CEO

performance. Hayes and Schaefer (2000) investigate the proposition that employment

contracts may be based on performance measures that are observable only to the parties


42/230

CHAPTER 320

of the contract (but not to the researcher). They show that unexplained variation in

current compensation is positively related to future performance and that this

relationship is stronger when there is more noise in the firms market and accounting

returns. These results suggest that compensation is informative regarding future

performance and that the board of directors takes into account subjective performance

measures when setting executive pay. Together, both articles demonstrate that examining

the contemporaneous relationship between pay and observable financial performance

underestimates the strength of the pay-performance relationship. In short, the total pay-

performance relationship may be high enough to elicit suitable incentives for executives,

especially when variations in equity holdings are considered.

Table 1: Empirical studies on the pay-performance relation.

STUDYSAMPLE(FIRMS)

COUNTRY TIME PERIODCOMPENSATION

VARIABLESP-PSENSITIVITY/

ELASTICITYA

Murphy (1985) 73 US 1964-1981 CEO totalcompensation

= 0.12

Coughlan andSchmidt (1985)

149 US 1977-1980CEO cash

compensation= 0.14

Jensen and Murphy(1990a)

430 US 1974-1986CEO total

compensation= 0.0033

Lippert and More

(1994) 310 US 1974-1988

CEO total

compensation = 0.0035

Hall and Liebman(1998)

478 US 1980-1994CEO total


Murphy (1999) 500 US 1990-1996CEO cash


= 0.263

Conyon andSchwalbach (2000)

102

48

UK

Germany

1969-1995

1968-1994

HPD* cashcompensation

TMT cashcompensation

= 0.067

= 0.071

Brunello et al(2001)

106 Italy 1993-1996Managerial cashcompensation

= 0.00024

Buck et al. (2003) 287 UK 1996-1997CEO total


AMeasuring pay and performance in absolute values leads to interpreting the regression coefficients as

sensitivities, while measuring pay in logarithms and performance in rates of return leads to interpreting the

regression coefficients as elasticities. An advantage of the elasticity approach is that it is relatively invariant to

firm size (Gibbons and Murphy, 1992). The primary advantage of the sensitivity approach is that sensitivities

have a more natural economic interpretation. The pay-performance sensitivity represents the executives share of

value creation. This sharing rate seems a natural m

Documents

Management Phd Thesis