essays in executive compensation · 2. executive compensation trends after 1930 in the usa 7 3....
TRANSCRIPT
ESSAYS IN EXECUTIVE COMPENSATION
By
João Paulo Torre Vieito
A DISSERTATION
Presented to the Faculty of Economics at the University of Porto in the partial fulfilment of requirements for the degree of Doctor
of Business Sciences
Advisor: António Melo Cerqueira
Coadvisor: Elísio Fernando Moreira Brandão
University of Porto Faculty of Economics and Management
Porto, Portugal
May, 2008
I
THE AUTHOR
João Paulo Torre Vieito has a Masters in Finance and a MBA in Management of
Commercial Operations from the Catholic University of Porto. His honours degree is in
Organisation and Business Management from ISCTE. He has been professor of finance
since 1993 at the Polytechnic Institute of Viana do Castelo and VicePresident of the
School of Business Sciences since 2005.
II
ACKNOWLEDGEMENTS
I am particularly indebted to my dissertation advisor Prof. António Melo
Cerqueira, and my coadvisor Prof. Elísio Fernando Moreira Brandão, from the
University of Porto Faculty of Economics and Management, for their valuable
guidance in developing this research. In addition, I would like to thank Prof. Walayet
Khan, from Evansville University, Prof. Kevin Murphy from the University of Southern
of California, Prof. Amir Licht, from the Interdisciplinary Center Herzliya, Prof. Sheng
Huang from Washington University, Prof. Sara Robicheaux, from Birmingham
Southern College Alabama and Prof. Mário Augusto, from the University of Coimbra,
for their helpful comments on this work.
I would like also to thank my wife, Sandra Manso, for her extraordinary support
during this project with the family, and my children Patricia Vieito and Gonçalo Vieito,
for the moments that I have not spent with them watching them grow up.
Finally, I would also like to dedicate this dissertation to my parents, Filipe Vieito
and Maria José Vieito and to my uncles and aunt Luciano Vieito, Maria de Fátima
Vieito and João Vieito. They were willing to sacrifice so much throughout their lives so
that I could succeed. I feel fortunate to have been blessed with such wonderful parents
and aunts and uncles.
III
ABSTRACT
ESSAYS IN EXECUTIVE COMPENSATION
Advisor: António Melo Cerqueira; Coadvisor: Elísio Fernando Moreira Brandão
This dissertation consists of four parts. In the first part, we review the literature on
executive compensation in the last 13 years and in the other three parts, we investigate a
few critical questions in the executive compensation area as described below.
We analyze whether or not the determinants, the form and total executive compensation
are the same for the new versus old economy firms (first essay), NYSE and NASDAQ
listed firms (second essay) and S&P500, S&PMidCap and S&PSmallCap firms (third
essay). For all these three groups of firms we also analyze whether or not the NASDAQ
crash in 2000 and the SarbanesOxley (SO) Act in 2002 changed the structure of
executive compensation.
In all the cases we find that the factors that explain executive compensation are
generally different. However, in a few instances of common factors the intensity of
these factors (coefficients) is different and this difference is generally statistically
significant. Our results also reveal that the determinants and forms of executive
compensation did indeed change after the NASDAQ crash and the SarbanesOxley Act.
IV
TABLE OF CONTENTS
Page List of tables VII Introduction 1 CHAPTER ONE: Executive Compensation: The Finance Perspective Perspective 5
1. Introduction 6 2. Executive Compensation Trends after 1930 in the USA 7 3. Executive Compensation, Firm Performance and Relative Performances 9 4. Executive Compensation and Agency Theory 18
5. Executive Compensation and Mergers and Acquisitions 23
6. Executive Compensation and Dividend Policy 27 7. Executive Compensation and Capital Structure 28
8. Risk and Executive Compensation 29
9. References 32 CHAPTER TWO: Essay 1 Trends and Determinants of Executive Compensation in the New and Old Economy 44
1. Introduction 45 2. Literature Review and Research Questions 47 3. Data, Sample Selection and Statistics 49 3.1. Data and Sample Selection 49 3.2. Statistics 50 4. Research Design 56 4.1. Dependent Variables 56 4.2. Independent Variables 58 4.2.1. Financial Variables 58 4.2.2. Governance Variables 62 5. Empirical Results 64 5.1. Summary Statistics 64 5.2. Determinants of Executive Compensation in the New and Old Economy 65 6. Conclusions 73 7. References 74 8. Appendix 77
CHAPTER THREE: Essay 2 Executive Compensation: NYSE and NASDAQ listed firms 84
1. Introduction 85 2. Literature Review and Research Questions 86 3. Data, Sample Selection and Statistics 88
V
3.1. Data and Sample Selection 88 3.2. Statistics 89 4. Research Design 95 4.1. Dependent Variables 95 4.2. Independent Variables 96 4.2.1. Financial Variables 96 4.2.2. Governance Variables 100 5. Empirical Results 102 5.1. Summary Statistics 102 5.2. Determinants of Executive Compensation for NYSE and NASDAQ
listed firms 103
5.3. Analysis of the Results 110 6. Conclusions 114 7. References 115 8. Appendix 118
CHAPTER FOUR: Essay 3 Executive Compensation of S&P Listed Firms 125
1. Introduction 126 2. Literature Review and Research Questions 128 3. Data, Sample Selection and Statistics 130 3.1. Data and Sample Selection 130 3.2. Statistics 131 4. Research Design 133 4.1. Dependent Variables 133 4.2. Independent Variables 138 4.2.1. Financial Variables 139 4.2.2. Governance Variables 140 5. Empirical Results 143 5.1. Summary Statistics 143 5.2. Determinants of Executive Compensation in S&P500, S&PMidCap and
S&PSmallCap listed firms 134
5.3. Impact of NASDAQ crash on the determinants of executive compensation 151
5.4. Analysis of the Results 154 6. Conclusions 157 7. References 158 8. Appendix 161
Conclusions and Possible Extensions 168
VI
List of the Tables Chapter 2: Page 1 Number of Items of Compensation by SIC Code 50
2 Mean Total Executive Compensation for New and Old Economy Firms (1992 2004)
51
3 Executive Compensation as a Percentage of Total Compensation in New and Old Economy Firms (19922004)
53
4 Statistics From Principal Component Analyses 59 5 Summary Statistics 64
6 Fixed Effect Regression: Least of Square Dummy Variables CEOs New Economy Executives 66
7 Fixed Effect Regression: Least of Square Dummy Variables Directors New Economy Executives
67
8 Fixed Effect Regression: Least of Square Dummy Variables CEOs Old Economy Executives
68
9 Fixed Effect Regression: Least of Square Dummy variables Directors – Old Economy Executives 69
10 T Test of Equality of Fixed Effect Regressions Coefficients CEOs 77
11 T Test of Equality of Fixed Effect Regressions Coefficients Directors 80 12 Pearson Correlation of Independent Variables 83 Chapter 3:
1 Average Total Compensation between 1992 and 2004 Adjusted to Inflation 90
2 Yearly Inflation Adjusted Total Compensation Trends of NYSE and NASDAQ Listed Firms Between 1992 and 2004 91
3 Yearly Percentage of Each Compensation Component of NYSE and NASDAQ Listed Firms ( 19922004)
92
4 Statistics from Principal Component Analyses 97 5 Statistics from Regression Variables 102 6 Fixed Effect Regression Analysis of Compensation Determinants 105
7 Fixed Effect Regression Analysis of Compensation Determinants of CEO Compensation for NYSE Listed Firms 106
8 Fixed Effect Regression Analysis of Compensation Determinants of CEO Compensation for NASDAQ Listed Firms 107
9 Fixed Effect Regression Analysis of Compensation Determinants of Director Compensation for NYSE Listed Firms 108
10 Fixed Effect Regression Analysis of Compensation Determinants of Director Compensation for NASDAQ Listed Firms 109
11 T Test of Equality of Fixed Effect Regressions Coefficients CEOs (NYSE vs. NASDAQ) 118
VII
12 T Test of Equality of Fixed Effect Regressions Coefficients Directors (NYSE vs. NASDAQ) 121
13 Pearson Correlation 124 Chapter 4:
1 Mean Total Compensation Levels Adjusted for Inflation by Years (1992 2004)Top Five 132
2 Executive Compensation Components as a Percentage of Total Compensation for S&P500, S&PMidCap, S&PSmallCap Listed Firms (19922004) 134
3 Statistics from S&P500, S&PMidCap and S&PSmallCap Listed Firms 143
4 Fixed Effect Regression Analysis of Determinants of CEO Compensation for S&P500 Listed Firms
145
5 Fixed Effect Regression Analysis of Determinants of Director Compensation for S&P500 Listed Firms 146
6 Fixed Effect Regression Analysis of Determinants of CEO Compensation for S&PMidCap Listed Firms 147
7 Fixed Effect Regression Analysis of Determinants of Director Compensation for S&PMidCap Listed Firms
148
8 Fixed Effect Regression Analysis of Determinants of CEO Compensation for S&PSmallCap Listed Firms 149
9 Fixed Effect Regression Analysis of Determinants of Director Compensation for S&PSmallCap Listed Firms
150
10 Fixed Effect Regression Analysis of CEO Total Compensation Determinants before and after NASDAQ Crash in 2000 for S&P Listed Firms 152
11 Fixed Effect Regression Analysis of Director Total Compensation Determinants before and after NASDAQ Crash in 2000 for S&P Listed Firms 153
12 T Test of Equality of Fixed Effect Regressions Coefficients CEOs 161
13 T Test of Equality of Fixed Effect Regressions Coefficients Directors 164
14 Pearson Correlation of the Independent Variables 167
1
Introduction
There was little research interest in the area of executive compensation until the
end of 1980s – on average two articles per year before 1985, according to Murphy
(1999). However, in recent years the research interest is growing exponentially. How to
motivate executives to increase shareholders wealth without performing fraudulent acts
is now one of the most interesting and challenging topics of research in the executive
compensation area. Executive compensation has become so important that it is now
investigated in vast areas such as mergers and acquisitions, firm performance, capital
structure and dividend policy.
Several factors have contributed to this phenomenon: a significant number of
privatizations of state controlled enterprises, the reduction in a significant number of
international trade barriers and the free flow of information associated with the
development of new technologies have started to create a small "integrated global
village." The combination of all these factors has led to an unprecedented worldwide
economic growth, and this increase in global competition has intensified the search for
executive talent across the world. Firms now compete for highly qualified executives
globally, hoping that their knowledge will be instrumental in increasing the share value
of the firms that they will manage. The search for good managers has made the average
compensation for Chief Executive Officers (CEOs) in S&P500 listed firms more than
double since 1970, with their compensation in the last few years generally tied to firm
performance, essentially based on stock options.
In this dissertation we first review the empirical and theoretical research in
executive compensation area and than analyze whether or not the factors that explain
executive compensation are the same in three different groups: New versus Old
Economy 1 , NYSE versus NASDAQ listed firms and S&P500, S&P Mid Cap and S&P
Small Cap listed firms. We also analyze whether or not the forms and total value of
executive compensation are the same in these groups and whether or not they changed
after the NASDAQ Crash in 2000 and the SarbanesOxley Act in 2002. We use fixed
1 New and Old economy firms are defined by authors like Anderson, Banker, and Ravindran (2000) and Murphy (2003) as firms competing in the computer, software, Internet, telecommunications, or networking fields.
2
effect regression analyses and collect data from the ExecuComp database from Standard
and Poor’s, which has collected information, since 1992, on the five best paid
executives from firms listed in the S&P1500 indexes.
We receive theoretical as well as empirical motivation to undertake the current
research. Scholars such as Ittner et al. (2003), Anderson et al. (2000), Murphy (2003),
Stathopoulos et al.(2004) and Chen and Hung (2006), document that new economy
firms are fundamentally different in terms of many characteristics they possess
compared to the old economy firms, and if these differences exist, we believe that total
value, the form, and the factors that explain executive compensation can also be
different between these two groups. There is only one study, developed by
Stathopoulos, Espenlaub and Walker (2004), which analyzes the executive
compensation for both new and old economy firms. But this study is confined to the
English market only. Most studies focus on the preponderance of stock options as a
form of executive compensation, without analyzing the remaining components of the
compensation. With the exception of the study developed by Murphy (2003), all other
studies on executive compensation analyze small periods of time; therefore, the
conclusions achieved by these studies must be validated for a longer period of time.
As far as literature review in terms of market structure and executive
compensation is concerned, we find only one research, developed by Firth, Lohen et al.
(1996), that analyzes the factors that explain CEO compensation in Norwegian Stock
Exchange listed firms. As far as we know, no research has yet analyzed whether or not
the factors that explain executive compensation for firms listed on the NYSE versus
NASDAQ are the same. We believe that NASDAQ listed firms are different than the
NYSE listed firms in particular in terms of liquidity and level of cash flows. Sapp and
Yan (2000) document that some small companies listed on NASDAQ changed to
AMEX exhibited improved liquidity. The same results are achieved by Chung et al.
(1999, 2001) and Bacidore and Lipson (2001), who document that NASDAQ listed
companies, which changed to the NYSE, experienced reduced transaction costs. If
NASDAQ and NYSE have different transaction costs, the liquidity of these firms will
be different. Murphy (2003) also reports that NASDAQ listed firms are essentially
technological firms with low levels of cash flows. If the levels of cash flows of these
3
two groups of firms are different, we then expect that factors that explain executive
compensation of these two groups of firms will be different.
Lambert et al. (1990) analyze the relationship between the changes in executive
compensation and the firm size and find that this relationship is positive and significant.
Most of the research on executive compensation inserts a variable of size (which can be
Assets, Market Value or Sales, or even the natural logarithm of this variable) in the
regression analysis to control for firm sizes, but we go further in our study. Instead of
inserting one additional size variable in multiple regressions, we focus on three different
size groups of large firms (S&P 500), medium size firms (S&P Mid Cap), and small
firms (S&P Small Cap) individually.
Given the limitations of the current studies we extend the executive
compensation research by analyzing what factors explain executive compensation in the
three different groups (New and old economy; NYSE and NASDAQ, and S&P500,
S&P Mid Cap and S&P Small Cap). We also analyze whether or not the form and total
values of executive compensation are similar in the above three groups and whether or
not these values changed after the NASDAQ Crash and SarbanesOxley Act.
Essentially, we believe that controlling for firm size and industry effect, the factors that
explain executive compensation may be different because each S&P index represents
different groups of firms with different characteristics, such as net income, sales and
dividends.
Effectively, after the NASDAQ crash, we saw a series of financial scandals
associated with the bankruptcy of some of the large American companies based on
fraudulent accounting practices and executive selfdealing. The SarbanesOxley Act of
2002 was established on July 30th to solve this problem. It introduced sweeping
changes in the governance, reporting, and disclosure requirements of public firms with
the aim of improving accuracy, reliability, and timeliness of the information provided to
investors. This Act contains provisions which have a significant impact on the benefits
and compensation of public company executives. The major changes in this area include
the following provisions: to prohibit publiclytraded companies from making or
arranging loans for their directors and executive officers; to expedite Securities and
Exchange Commission (SEC) reporting to insider traders; to prohibit corporate directors
and executive officers from trading employers` securities during planned blackout
4
periods with respect to those securities and to require an employee retirement Income
Security Act to cover individual account plans to provide a 30day notice of blackout
periods. So far little is known about the real impact of this legislation in terms of
executive compensation. Narayanan and Seyhun (2005) document that the 2002
SarbanesOxley Act has curtailed, but not eliminated, managerial influencing of the
grant day stock price and Murphy (2003) also reports that some changes have occurred
in terms of the number of stock options granted to executives. Based on these findings,
we expect that the NASDAQ Crash and SarbanesOxley Act have changed the form and
total values of executive compensation.
This dissertation is organized in the following way: Chapter I presents the
review of literature on executive compensation since 1995 based on the best
publications in the area. Chapter II (first essay) analyzes the trends and determinants of
executive compensation in new and old economy firms; Chapter III (essay 2) analyzes
the executive compensation in NYSE and NASDAQ listed firms; Chapter IV (essay 3)
analyzes the executive compensation of S&P listed firms. Finally, we describe the final
conclusions and possible extensions.
5
CHAPTER I
Executive Compensation: The Finance Perspective
6
1. Introduction
Executive compensation is presently one of the most interesting and innovative
fields of research in the finance area. It was only in the 1990s, with the growth of the
world economy, that shareholders felt the need to contract executives and give them
incentives to make firms’ stock market growth increasingly faster each year. Academics
and researchers started searching for the best form of compensation to motivate these
executives. It was not only the values that mattered but also the way in which
executives were paid: with more short term compensation (salary or bonus) or more
long term compensation (stock options, restricted stocks, longterm incentives plans) or
even with other forms of compensation like perks, and the impact of these
compensation policies on all the fields of finance.
In this paper we describe the literature review on executive compensation from
the period between 1995 and 2007. The reason for choosing this period of time is that
we believe that 13 years is enough time to cover a set of research studies that are
representative of executive compensation research lines.
To better understand the state of the art in terms of executive compensation we
aggregate this research by topic. First, we describe the evolution of executive
compensation in America since 1936. After that, we describe the different relationship
between executive compensation, firm performance and relative performance, agency
theory (asymmetric information: adverse selection, moral hazard and double moral
hazard), mergers and acquisitions, dividend policy, capital structure and risk.
7
2. Executive Compensation Trends After 1930 in the USA
The number of empirical studies on executive compensation has increased
exponentially since the beginning of the nineties. Before this period, there is little
knowledge about executive remuneration in America and worldwide.
The most important study that collected information on executive compensation
evolution in America for a longterm period was undertaken by Frydman and Sacks
(2005) based on data since 1936. The authors divide and analyse the period into three
important groups: after World War II, executive compensation decreases slightly; from
the 1940s until the 1970s executive compensation grows slowly, and after the 1990s we
see a fast growth in executive compensation. To these findings we can also add the
information that the NASDAQ crash in 2000 stopped this growth.
There is a generally accepted idea that until the end of the 1980s, probably the
beginning of the 1990s, executives were paid only with salary and bonus and that other
compensation components, like executive stock options or restricted stocks, were then
created in this period to satisfy an increasing desire for pay by firms’ performance,
essentially because shareholders wanted executives to maximise the firms’ values, and
indirectly maximise their wealth. But this is not true. Frydman and Sacks (2005)
documented that between 1934 and the 1950s executives were paid mainly with cash
compensation and in some cases with bonuses. Stock options appeared in 1951 and 18%
of the analysed executives received this kind of compensation component 2 . Due to
advantages in terms of taxes, restricted stock was also introduced at this time in the
USA as a compensation component. Until the 1970s, this kind of compensation
component was used by shareholders to motivate executives to expand the firm’s
production capacity. This was the shareholders’ main goal: to have big companies with
high production capacity. But with the significant economic changes that occurred in
the USA at this time, large organisations found themselves with production excess,
which indirectly led to significant falls in the firms’ market value. With these losses in
terms of the firm’s value, shareholders felt the need to contract executives and motivate
2 The authors report that a small number of companies give stock options to their executives but in the
analysed sample, no executive receives these.
8
them to increase shareholders’ wealth again. If shareholders motivate executives based
on firm performance, by giving part of their compensation in stock options, they give
them the possibility of owning a small part of the firm’s capital in the future. In this
way, it is understood that executives will make greater efforts to increase the firm’s
stock market value to exercise their stock options, and at the same time increase
shareholders’ wealth.
It was only in the nineties, with a significant growth in the worldwide economy,
and with the sprouting of companies associated with new technologies, also called “new
economy” firms, that the academic community, at worldwide level, focused its interest
on the executive compensation problem, first essentially for CEOs and then for the
remaining employees.
With the market in expansion, and great worldwide expectation created for the
stock return of the companies associated with new technologies, the market value of
these companies continued to rise. Financial analysts found strong incongruence
between the market value of these firms and traditional model evaluations. It became
urgent to develop new evaluation models that would determine the real value of the
stocks of this new concept of organisation, which works at worldwide level and
generally does not have a significant sales force. Some authors, like Sanders and Boivie
(2004), also defend that in the case of US Internet firms, market valuation was strongly
associated with the level of executive stockbased incentives compensation.
The new economy firms’ stock prices grew so much that, predictably, in the year
2000, the NASDAQ crash occurred.
Most of the compensation plans based on stock options are now outofthe
money and the reason for granting them to executives – to motivate them – has
disappeared.
In an attempt to retain good executives and other employees, many shareholders
reformulated the stock option plans, readjusting the exercise price to an attractive value.
Others opted to attribute new stock option plans.
The first few years of this new century were characterised by a significant number
of fraudulent firm bankruptcy cases and this bankruptcy was, in most cases, related to
compensation policies based on stock options. Executives wanted to increase the firm’s
value so much, in order to exercise their options, that they sometimes lied to the market
9
and manipulated the firm’s accounting data to achieve their goals. To solve these
problems, in 2002 the SarbanesOxley act was created in the USA, which introduced a
significant number of changes in terms of corporate governance.
The real impact of these new corporate governance rules in terms of executive
compensation is still not clear and investigating this is one of the goals of this study.
After this brief description of the evolution of executive compensation since
1936 in the USA, we next describe a group of research studies that relate executive
compensation to different fields in the finance area.
The majority of the studies on executive compensation to date have been based
on data from only one country, and do not compare executive compensation in different
parts of the world; see for example, Lowe et al. (2002). One of the reasons for this is
that, in most countries, information on executive compensation is scarce, disperse and is
aggregated to all the boards and not described for each top executive.
3. Executive Compensation, Firm Performance and Relative Performances
The relationship between executive compensation and firm performance has
been widely analysed in academic publications in the area of accounting, essentially
comparing executive compensation methods with vast groups of accounting items, but it
has not been examined so much in the financial area that focuses on firms’ stock market
price. Some of the findings in this area are described by Devers et al (2007), Bebchuck
and Fried (2004) and Murphy (1999).
Firms that implement executive compensation plans based on performance
generally create more ambitious and difficult strategies (Dow and Raposo, 2005) than
companies that do not give this kind of compensation to their executives, and when the
adoption of these compensation plans for CEOs is announced to the market,
shareholders’ wealth generally increases (Morgan and Poulsen, 2001). In most cases,
the market will respond positively because it believes that the CEO will develop efforts
to increment the firm’s stock market value to the level that will guarantee that stock
options will be exercised.
Short term executive compensation components have been reported as being
negatively related to firms’ corporate social performances and positively related to long
10
term executive compensation items (Deckop et al. 2006). Conversely, authors like
Sanders (2001a), found better firm performances when firms adopt yearend
readjustments of executive compensation than when firms adopt executive
compensation contracts with predominantly longterm compensation components, such
as stocks and restricted stocks. In other words, executives, essentially CEOs, will
secure better performances for the firm if they know that the company will readjust their
compensation in positive terms at the end of the year, and performances will not be as
good if this compensation is based on longterm compensation components.
Better performances are also achieved in US multinational firms when they
contract CEOs with international experience (Carpenter et al. 2000 and 2001). It has
also been reported that when a CEO is at the same time the Chairman of the Board, this
situation influences, if only in small terms, the firm’s long term performance (Baliga et
al. 1996).
Another interesting finding about firm performances and executive compensation
is that in firms that are controlled by the owners, or in firms where the owner is also the
manager, the relationship between pay and firm performance for all the employees is
higher than in other cases (Wener et al. 2005). But if the CEOs are family CEOs, in
familycontrolled firms or the firms have stakeholder management, they generally
receive less total compensation than outsider CEOs. The compensation increases when
family ownership also increases (GomezMejia et al., 2003) but the gap exists even
when firms’ performances increase (Coombs and Gilley, 2005). Performances, in the
case of banks, have been reported by Magnan and StOnge (1997) as being associated
more with executive compensation in a high managerial discretion context than in a low
managerial one.
The relationship between the change in CEO compensation and firm strategy
has been described as positive when firm performances are low, and negative when firm
performances are higher (Carpenter, 2000). Firms that use defensive strategic
orientations have better firm performances when they pay their executives in cash and
bonuses, and when they evaluate the firm performances based on accounting items.
Firms that adopt prospector strategic orientations will have better performances if they
pay their executives in stock, or stock options, and when they use market measures to
evaluate managerial performance. Furthermore, firms that adopt governance reforms
11
have been reported as achieving higher levels of market performances than firms that do
not adopt governance reforms (Tuschke and Sanders, 2003).
Generally, in firms where agency problems are higher, the performance levels
are lower and these firms pay more to their CEOs than companies with low agency
problems (Core et al, 1999 and Adams et al., 2005). If CEOs’ actions are not monitored,
they will try to extract higher compensation from shareholders than in companies where
the monitoring process is efficient. Pay to board members based on firm performance
has also been described as reducing the likelihood of “WhiteCollar” crimes being
committed (Schnatterly, 2003).
At present, it is generally accepted that compensation based on performance is
more common for CEOs than for other executives (Ang et al., 2002 and Aggarwall and
Samwick, 2003), because they have more power than other low level executives to
influence a firm’s stock price. The amount that is paid to executives to motivate them is
important, but more important than that is to choose what percentage total compensation
stock options should represent (Kole, 1997 and Mehran, 1995). Some executives may
develop maximum efforts just with a small number of stock options, but to motivate
others, it is necessary to give a significant amount of this compensation component. But
future firm performance is not only dependent on the total compensation, or the
percentage of stock options received, but also on how much is paid to the other
members of the top management team (Carpenter and Sander, 2002, 2004)). In other
words, if the CEO receives a lot more than other top management executives, the latter
will probably not be actively involved in the goal of increasing firm performance and
will pass this responsibility to the CEO. The behaviour and economic view of the gap in
terms of executive compensation between CEOs and other top executives is described
as balanced as predictors of firm performance by Henderson and Fredrickson (2001).
When a firm makes changes in the top management team, the gap between the CEO and
other top management team members’ compensation is positively related to the number
of participants in the firm tournament but the changes in terms of how much is paid to
the executive team has only a small influence on determining company performance
(Conyon et al., 2001). Essentially, this gap must exist but the differences should not be
very significant in order to guarantee that all the members of the management team will
be motivated to increase firm performance. It is logical that hierarchical firm position
12
influences not only the design of the executive compensation plan, but it is most
important that the hierarchical position is related to the level of interaction of each
member with the strategic firm orientation (Boyd and Salamin, 2001). Generally, top
executives will receive more total compensation and have more influence over strategic
firm decisions than low level executives. Total compensation, and longterm
compensation of top executives that are not CEOs, have been reported as positively
related to firm performances (Carpenter and Sanders, 2004) Generally, top executives that have a strong professional reputation are paid
more in terms of performance (Milbourn, 2003) because shareholders know that their
decisions and actions will directly influence a firm’s stock market price. They also
receive more total compensation than noncertified CEOs in cases where firm
performance was high, but less remuneration when performance was poor (Wade et al.,
2006). CEOs who are considered celebrities will internalise such celebrity and tend to
be overconfident about the efficacy of their past actions and their future ability to
increase firm performances (Hayward et al., 2004).
Granting stock options to executives is a way for shareholders to guarantee that
if they lose wealth, so will the executive. Narayanan (1996) also points out that when
executives are remunerated essentially with cash, in the long term they will make fewer
firm investments than optimal, but if they are essentially remunerated with stock
options, the firm’s investment will be higher than optimal value in the long run.
Because of this, the authors defend that the best way to guarantee that executives will
make the optimal firm investment is to create compensation contracts with both
restricted stocks and cash.
As previously said, the shareholders’ main intention, in giving their executives
these kinds of compensation components, is to motivate them to make the firm’s stock
price grow to the desirable level where they will be able to exercise the stock options
and, at the same time, increase shareholders’ wealth. But this is not always the only
reason why stock options are given to executives. Authors like Core and Guay (2001)
also argue that companies give stock options to their executives when they are facing
capital requirements and financing constraints, because this is an expert way to pay
executives without cash. On the other hand, Kato et al. (2005) documented that
companies with more future growth perspectives also give stock options to their
13
executives. The reason for this is not only associated with the fact that they might have
to face small cash flows in order to pay in cash or bonus, but also because pay based on
the firm’s stock price will motivate executives to increase the size of the firm.
If companies give stock options, or restricted stocks 3 , to their executives to
increase firm performance, what happens if these executives cannot make the firm’s
stock grow to the desirable value that will give the capacity to exercise the options?
Generally, in order not to lose their executives, firms with a low level of performance
use repricing, also called resetting, of stock option plans (Chen, 2004 and Brenner et al.,
2000). In other words, companies will reduce the actual exercise price to a more
attractive value where the probability of stock options being exercised is higher in the
future. What shareholders want is to motivate executives again to increase the firm’s
stock market value and, indirectly, their wealth.
Stock option repricing is more common not only in firms with low performance
levels (Chen, 2004 and Brenner et al., 2000), but also in small companies and firms that
have low market notoriety; young firms and companies associated with new technology
(Carter and Lynch, 2001), and firms where boards are essentially composed of insiders
(Chance et al., 2000). The latter authors also find that some companies use stock option
repricing more than once, and when this happens, firms generally have lower
performance in the first year, but in the second year the stock option plans are mostly
inthemoney. If companies must change the exercise price more than once, the desired
incentive effect on the executive is reduced, and this exercise price will be close to the
present market stock price of the firm. This is the reason why the options are generally
inthemoney within two years. Pollock et al. (2002) also note that more visible CEOs,
and CEOs with high stock firm ownership, will have less power to negotiate small
spread between new exercise price and firm stock market value because the market will
think that they will be extracting personal benefits with these changes. Grossman and
Cannella Jr (2006) add the information that CEOs/Chairs that have a significant
ownership tend to maintain the reward policies based on fixed compensation across the
years.
There are some companies that do not contemplate the possibility of being
repriced in their stock option plans. What happens in these situations when the firm’s
3 Restricted stocks are stocks with restrictions on when they can be exercised. The restriction usually lifts in 3 to 5 years when stock vests.
14
stock market price falls to a level at which executives cannot exercise these options and
the desirable incentive effect disappears? Do companies lose their executives? In most
cases, when firms do not want to lose their executives, they give them a new stock
option plan with a more attractive exercise stock price (Chen, 2004).
Some authors defend that stock option repricing should not be used because
firms are paying bad executives, and shareholders are giving these executives a second
chance to get a remuneration that they do not deserve. This is the case of Garvey and
Milbourn (2006), who defend that executives are sometimes paid for “luck” and
sometimes for firm performance. In most cases when companies have bad
performances, shareholders make arrangements to pay these executives and for them to
stay in the job. In the case of emerging markets the situation is different. Executives
from firms with bad performances generally do not have a second chance and lose their
job (Gibson, 2003). In this way, with the exception of emerging markets, the number of
years that a CEO remains in his job (tenure) can be conditioned by firm performance,
and, according to GomezMejia et al., (2001), by the business risk and whether the
company is a family company.
Whether repricing is a good or bad methodology to motivate executives again is
not consensual in terms of finance research. We know that in some cases the reason
why companies present low performance is not associated with bad management but
due to external factors that the CEO cannot influence and this must be one of the
reasons why some authors defend that the repricing methodology must continue to be
considered in executive compensation contracts (Acharya et al., 2000).
One of the fields of research where the relationship between executive
compensation and performance was also analysed was in Hedge Funds, because the
compensation of these executives (also called fund advisors) is, in part, related to fund
performance. The fund advisors will receive an additional compensation when the fund
market price is higher than a certain level called the “highwater mark”. In this way,
executives will try to develop efforts to receive additional compensation and will
sometimes make speculative investments to guarantee that the fund price will rise to the
desirable values that will guarantee they receive this additional compensation. To
achieve these goals, executives sometimes develop speculative investments. The
existence of speculative investments was detected by Golec and Starks (2004) when
15
they investigated the effect of the introduction of regulation, by the American Congress,
to control fund advisors’ compensation and found that after the legislation, most of the
mutual funds changed their risk positions. Authors like Goetzmann et al. (2003) also
add that the existence of “highwater marks” in hedge fund compensation contracts can
be a limit of the performance fee because fund advisors will probably only develop
efforts to achieve this established limit.
Another interesting study associated with hedge funds was developed by
Khorana (2001), who analysed the effect of fund managers’ replacement on future
hedge fund performance. When managers that have implemented low quality decisions
that lead to low fund performance are replaced, hedge fund performance improves.
When managers who have implemented good quality decisions that lead to better fund
performance than average are withdrawn from the management of these funds, hedge
fund performance decreases.
Executive compensation has also been analysed, not only in relation to firms’
performances, but also to firms’ relative performances 4 . Traditional models of pay for
performance give additional remuneration to executives when companies have better
results than the previous year, and to evaluate firms’ performance a vast group of
variables is generally used (Hermalin and Wallace, 2001). In a pay for relative
performance model, the firm’s performance is compared not to last year’s firm results
but to the performance of the main rival firms. The pay for relative performance
methodology is a more demanding model of compensation than the simple traditional
method. In the first case, executives only receive additional compensation if the firm
where they work is one of the best in their area of the market. But the situation can be
difficult when the company has products with a high level of market competition.
Generally, these firms are forced by the market to have low performance and this can
negatively affect executive compensation (Aggarwal and Samwick, 1999).
Pay in terms of relative performance has been defended by authors like Garvey
and Milbourn (2004) as a mechanism for removing the influence of the marketwide
(não falta uma palavara aqui?) on executive compensation, and young and less wealthy
executives have been reported as delegating the stock options risk immunisation to their
4 Relative performances are the firm’s performances compared not with the performances of the
firm in the last few years but with the performance of the main rivals on the market.
16
company. But at present, Rajgopal et al. (2006) and the same authors Garvey and
Milbourn (2006) defend that this theory is not credible because executives can replicate
such indexation in their private portfolios and do not need the help of the companies to
fix the risk associated with receiving compensation based on stock options.
Also regarding executive compensation and firm’s relative performance,
Yermack (2006) documents that firms that give their CEOs the possibility of using firm
airplanes for their private benefit, present, on average, 4% lower performance than the
market average. Perry and Zenner (2001) also report that the introduction of legislation
in America in 1992 limiting the deductibility of executive cash compensation greater
than 1 million dollars (Internal Revenue Code 162 (m)), has made the affected
companies change to a compensation policy based on firm performance. If executives
can extract higher compensation from firms based on salaries without being indexed to
efforts to making firm stock price rise, this will generally be the best compensation
policy for executives. It does not matter what performances firms have because they
will get the same fixed amount in money. This new legislation introduces a fairer way
of paying top executives. They will receive higher fixed compensation in cash, up to
one million dollars, but the other part must be indexed to their personal effort level to
make the firm grow.
In companies where layoff policies are announced, the CEOs receive, on
average, 20% more compensation than companies that do not adopt these policies
(Brookman et al., 2007). Essentially, CEOs that have the hard job of reducing human
resources costs to improve firm performance will receive more than other CEOs.
Another interesting study was developed by Brickley et al. (1999), which documents
that a good way of inducing CEOs to increase firm performance is to give them the
possibility of being on the company board after their retirement. Bernartzi (2003)
analysed why firm employees place a significant part of their compensation in the
firm’s stock and found that, generally, they make these investments because they
believe that good past firm performance will continue in the future, but, according to the
author, past firm performance is not an index of good future firm performance.
Controlling for firm size, performance, and other factors, Balkin et al. (2000)
document that CEO shortterm compensation is related to the degree of innovation,
measured by the number of patents and R&D spending, and in the case of low
17
technology firms no relationship exists between innovation and short, or longterm CEO
compensation. Makri et al. (2006) also find that when a firm’s technological intensity
increases, the total CEO compensation incentive will be more closely aligned with the
impact of the firm’s invention on other related inventions and with the firm's
commitment to scientific research, and bonus compensation will be more related to
financial results.
Tosi, Werner et al. (2000) report that firm performance variance only explains less
than 5% of CEO total compensation and it is effectively the size of the firms that has the
biggest influence on executive wealth (40%). Authors like Beer et al. (2004) report that
some firms are abandoning payforperformance policy, because the costs of these
programs will be higher than the associated benefits, and are using other methodologies
like effective leadership, clear objectives, coaching, or training because, in their point of
view, these are better and more efficient investments.
To conclude on executive compensation and firm performance, it may seem that
pay in terms of performance is one of the best methodologies when shareholders want
to increase their wealth because this will reduce agency cost. But this is not true. Most
cases of fraudulent bankruptcy that have occurred in America since the NASDAQ crash
have been indirectly related to executive stock options. Some of these executives had
such a large part of their compensation indexed to firm stock market price that they
made fraudulent accounting transactions and lied to the market with the intention of
successively raising the firm’s stock price to the level where they could exercise their
stock options and receive the high amounts of compensation. The SarbanesOxley Act
has become an important instrument in correcting some of the problems involved.
Because of the associated problems, the use of stock options is decreasing and the use
of other long term compensation components is increasing, such as restricted stock.
Restricted stocks differ from stock options in the sense that they are stocks that are
granted to executives but which can only be sold in the long term. They are less
“dangerous” than stock options because executives will incorporate the new price of
firm stock daily and in the case of the stock options they can lose everything if they can
not raise the firm’s stock market price to the level where stock options can be exercised
and this is, in our point of view, the main problem behind those cases of executives
18
lying to the market and making fraudulent accounting transactions. Quite simply, if they
cannot exercise the options, they will get nothing.
4. Executive Compensation and Agency Theory
Regarding exchanges where principal delegates work for agents, the agency
theory precursors try to develop methods to solve contractual problems associated with
opportunistic agents.
In recent years, agency theory has been analysed in a vast number of situations
in the area of executive compensation following the analyses of Ross (1973) and Jensen
and Meckling (1976). Most of these studies investigate the executive opportunism
associated with the existence of asymmetric information between the executive and the
shareholders.
The asymmetric information problem can be categorised in the following ways:
hidden information, also called the Adverse Selection Problem, and hidden actions, also
called the Moral Hazard Problem. The Moral Hazard Problem may also appear as the
Double Moral Hazard Problem 5 .
Adverse Selection is associated with the fact that executives sometimes have
hidden information that can be omitted when the company makes the compensation
contract to get personal advantages in the future. This hidden information can generally
be associated with the following: executives have access to privileged information about
the firm’s environment; they are experts in the area and shareholders cannot evaluate
their personal knowledge, or it can be associated with the fact that the cost that the
shareholder must pay to get this information is higher than the associated benefits.
Hidden actions, also called moral hazard problems, are described by Katz and
Rose (1998) as: one party, the agent, performs actions that affect the other party
(principal); the principal cannot observe the agent’s actions and also the principal and
agent agree as to what action the executive must develop. The real difference between
the moral hazard and the adverse selection problem is based on the fact that in moral
hazard, executives will develop actions that the principal cannot observe or measure,
5 The two terms have been adopted from literature on insurance.
19
and this action makes the principal lose money, and in an adverse selection problem,
executives have more information than shareholders on some points, which they can use
for their own benefit during the contractual time.
The moral hazard problem can also assume the form of a Double Moral Hazard
problem. Gupta and Romano (1998) define this problem with the example of a
production process involving two parties, where it is difficult to know what the
contribution of each party is to the final product, and each of the parties, during the
production process, may develop a group of actions that cannot be observed by the other
party.
If executives have information that shareholders do not have, shareholders incur
the risk that executives opt not to inform, inform partially or falsify the information and
this will influence executive pay (Goldberg and Idson, 1995).
A significant number of mechanisms have been described to reduce agency
problems between executives and shareholders: the existence of institutional investors
in the company or a large number of blockholders; the use of outside directors on the
boards; the existence of debts in the firm; the managerial labour market and market for
corporate control (Agrawal and Knoeber, 1996). Authors like Burns and Kedia (2006)
defend that the use of stock options is also a good mechanism to align shareholders’ and
executives’ interests, reducing agency costs. Authors like Ang et al. (2000) have
described measures of absolute and relative agency costs to firms with different
ownership and management structures.
One of the abovementioned mechanisms for reducing the agency cost is the
existence of institutional investors, or blockholders, in the firm. They will defend their
interest more than individual shareholders because their power in terms of firm
ownership is greater. They will also have more control over executive acts and make an
effort not to increase the level of executive compensation (Hartzell and Starks, 2003).
The authors also describe that institutional investors are generally attracted to
companies where there is a strong relationship between pay and performance.
Another mechanism for reducing agency problems related to executive
compensation is the presence of outside directors on the company Board or
Compensation Committee. The theories that argue that executives that are on the Board
or Compensation Committee will try to extract personal benefits in terms of
20
compensation are not consensual. Authors like Hallock (1997) effectively found that
when executives are on two different boards at the same time (interlocked), they can
positively influence their personal compensation and gain more than noninterlocked
executives. On the other hand, Anderson and Bizjak (2003) document that executives
that are also on the Compensation Committee do not experience a decrease in their
personal compensation when they leave these Committees, meaning that they do not
extract opportunistic advantages in terms of compensation. If agency problems exist in
the firm, one of the solutions presented by Borokhovich et al. (1997) is to have boards
with more outside than inside directors. When CEOs are less related to the other board
elements, the probability of their influencing the other members of the board to increase
their personal compensation is less than when boards are composed essentially of inside
directors.
Bank loans have also been defended by authors like Almazan and Suarez (2003)
and Elston and Goldberg (2003) as another mechanism for monitoring executive
opportunism and reducing agency costs. According to the former authors, optimal
executive compensation contracts must be created based on three essential elements:
firm performance, incentives that can highly motivate the executives and bank loans. If
banks give loans to the company, they will frequently monitor the executives’ actions to
guarantee that that money will be paid in the future. They will also force executives not
to extract high compensation from companies to guarantee that firms will have enough
money to pay the loans.
Another study that relates executive compensation and bank loans was carried
out by Osano (2002), who documents that an interesting way to reduce the number of
bank loans that are vested but not paid, and increase the bank market value, is to give
stock options to the executives because they will make efforts for these to increase the
bank stock market price and exercise stock options.
The managerial labour market is another way of reducing agency costs related to
executive compensation. The value that executives receive as compensation must be
congruent with their ability and knowledge and the existence, or not, of workers with
the same knowledge on the market. If there are a significant number of experts in the
area, executives will probably act more in line with shareholders’ interests and will
accept less compensation, because they will know that a lot of other executives want
21
their job. When the number of workers on the labour market with the same knowledge
is small, executives will probably ask for better compensation because they know that
just a small number of people have the same knowledge. Murphy (2003) complements
this information by saying that if a large group of big companies competes on the
market for high quality executives, the compensation contracts that they give to their
executives force the other companies to use the same compensation structure.
The market for corporate control has also been defended as a mechanism to align
shareholders’ and executives’ interests in the sense that if executives are not supervised
by the market, they can probably extract higher compensation then when the market
monitors their actions (Agrawal and Knoeber, 1996). Rajan and Wulf (2006) also
complement this information by saying that if there is high external monitoring,
shareholders will not need to offer their executives perks because their interests will be
aligned with executives’ interests ( Rajan and Wulf, 2006).
The use of stock options has also been defended as a mechanism to reduce
agency problems because it forces executives not to misreport (Kedia, 2006). If
executives receive a significant part of their compensation in stock options, they will try
to develop the best practices to align shareholder and executive interests, and increase
the firm’s stock price to the level at which they will be able to receive this
compensation component in the future. But can we increasingly motivate the executive
to the level where agency problems disappear? If shareholders give their executives
more stock options than they want, in order to motivate them to increase firm
performance, they will diversify their personal portfolio by selling firm stock that they
already have (Ofek and Yermack, 2000).
The main problem is how executives can increase the firm stock market price to
the necessary level to exercise the options. Executives sometimes manipulate the firm’s
accounting, or the reported earnings, to influence the firm’s stock price in positive terms
and achieve their goals, and these situations are more frequent in companies where CEO
compensation is indexed to firm stock price (Bergstresser and Philippon, 2006). Povel
et al. (2007), Yermack (1997), Hu and Noe (2001) and Narayanan (1999) also
document that some executives choose the time to send positive information to the
market to get personal benefits, and that this occurs more frequently in good times. The
relationship between fraud and good times is stronger when investors’ monitoring is
22
low. In an IPO context, Lowry and Murphy (2007) show that executives can also
influence the IPO offer price and its timing.
The relationship between asymmetric information and executive compensation
has also been analysed in a hedge fund context. This is the case of Coles et al. (2000),
who investigated the closed endfunds advisors’ premiums and detected that some
premiums were higher when the executive, also called fund advisor, was related to fund
performance, when the fund assets managed by the fund advisor were highly
concentrated in this fund, and when the executive managed other funds with a weak
relationship between executive compensation and fund performance.
A small group of other studies associated agency problems with executive
compensation. This is the case of Bernardo, Cai and Luo (2001) and Bernardo (2004),
who describe that there are agency problems associated with division managers’
compensation. Division managers sometimes omit the quality of their projects to the
CEO to increase their reputation when this project exceeds expectations and ask for
better compensation in the future. Generally, it is the senior executives who receive
more compensation based on firm performance, who have greater ownership and are
involved in projects that can more greatly influence firm stock price (Barron and
Waddell, 2003). Another study was developed by Goldman (2004), who found that
when the CEO makes the financial budget distribution to the firm department, agency
problems can exist. The reason for this is associated with the fact that when CEOs
receive a significant part of their compensation in stock options, they will send the
highest part of the financial budget to departments that will guarantee that the firm’s
stock price will rise and they can exercise their stock options.
Finally, Aboody and Lev (2000) also describe that one of the biggest sources of
agency problems is the Research and Development department (R&D). Researchers
from these departments can have a significant amount of information about the products
or services that they are developing that nobody else knows, not even the firms’ CEOs.
If they believe that this product, or service, will be a success in the future, they will buy
a significant amount of company stock and will sell these shares at a good profit when
this information is sent to the market.
23
5. Executive Compensation and Mergers and Acquisitions
Interest in the way firms pay their executives and their relationship with firm
mergers and acquisitions has been increasing over the last few years. The existing
literature is essentially related to the use of some compensation components as a
mechanism to defend the company, or the executives, from takeover. This research has
made the following findings: it is not indifferent to give stock or stock options in terms
of executive effort to firm’s acquisitions; some of the executives can benefit in financial
terms from successful acquisitions, and the threat of takeover is a mechanism to reduce,
or control, the level of executive compensation. There is further research that points out
the firm’s benefits from the use of stock options to guarantee successful acquisitions
and the importance of the existence of outside directors on the board as an element to
control executive compensation and reduce the takeover hypothesis. Next, we will
describe each of these studies.
The use of stock options as a compensation mechanism for employees has been
reported has defend companies and top executives from takeovers has been analysed by
authors like Pagano and Volpin (2005). The reason why stock options are an efficient
way to protect companies from takeovers is because when employees have company
stock options, when the takeover appears they will defend the organisation and CEO
stability so as not to lose their future remunerations. Another benefit associated with the
use of stock options as a compensation mechanism for top executives is that when they
receive this kind of compensation they will be more careful and generally only buy
good companies. In other words, if shareholders give their CEO a significant part of
their compensation indexed to the firm’s stock market price, when CEOs decide to buy
a company, they generally only acquire firms with future significant growth
perspectives (Datta et al., 2001). Only these companies will guarantee that the firm’s
stock price will increase in the future and that the stock options the executive has can be
exercised. If CEOs buy bad companies, the firm’s stock price will probably decrease in
the future and CEOs will be not able to exercise these options. According to Deutsch et
al. (2007) the best way to improve shareholders' value is to give stock, or stock options,
not only to CEOs but also to outside directors of the boards. This will guarantee that not
only CEOs, but also other board members, will try to buy only good companies and
24
improve shareholders’ wealth. If they buy bad companies, their personal wealth will
also be penalised and not only that of the shareholders. Sanders (2001b) adds that giving
stock or stock options to executives has been described as having the same incentive
effect but, according to the author, the risk properties of these two compensation
components are asymmetrical and they have diametrically opposite effects on firms'
acquisition.
In most cases, when CEOs make successful mergers and acquisitions, they receive
an additional premium from the company that is generally based on cash, or in some
cases, on bonus (Hartzell et al. 2004, Bliss and Rosen, 2001 and Wright et al. 2002).
The size of this payment to CEOs has been described by authors like Grinstein and
Hribar (2004) as related to the power that they have inside the company. High powered
CEOs are able to extract significantly higher compensation from shareholders than
lower powered executives (Coombs and Skill, 2003) and, in this way, the premium that
they receive is related to their power and the intensity of monitoring activities (Wright
et al., 2002). In other words, CEOs can extract high premiums associated with
successful acquisitions if their power inside the company is higher and if the monitoring
level of their actions is small. Otherwise, they can get additional compensation but
perhaps not as much as they desired.
When a company buys another company, what happens to the executives of the
acquired company? Hartzell et al. (2004), describe that if executives from the acquired
company stay in the firm, they normally receive more compensation than before the
acquisition, but a significant number of these executives leave the company in the next
three years and the company gives them severance pay. Most of the other executives
that do not stay in the company retire, and only a few continue in other companies with
executive functions.
Aggrawal and Knoeber (1998) defend that the threat of takeover is also a
mechanism to control the increase in the level of managerial compensation, but this
threat can divert managerial effort from productive to defensive activities (Chakraborty,
and Arnott, 2001). Essentially, when top executives are afraid that another company
will try to buy their company and they will lose their job, they do not try to increase
their compensation so much. But if they are protected, the “whitecollar” workers, in
particular, will try to raise their compensations (Bertrand and Mullainathan, 2002).
25
The fear associated with losing their job makes executives resist tender offers. When
takeover occurs and generally when they decide to resist, they gain from this situation
(Cotter and Zenner, 1994). A methodology for managing executives’ fear of takeovers
is to adopt antitakeover mechanisms (Field and Karpoff, 2002 and Borokhovich et al.,
1997) as is the case of golden parachutes. This is a repellent methodology to reduce the
possibility of successful acquisitions in the sense that the acquirer firm must pay not
only the firm’s value, but also a significant amount of money to the CEO to leave the
company.
The use of golden parachutes has been defended as positively related to top
executive compensation (Field and Karpoff, 2002 and Borokhovich et al., 1997) in the
sense that executives that have this kind of financial protection receive higher
compensation than executives without golden parachutes. Davidson, Pilger and Szakmary (1998) also document that Compensation Committee composition determines
the way the market accepts the adoption of golden parachutes. If the Compensation Committee has more insiders or affiliated outsiders, the market reacts negatively,
reducing the firm’s stock price, but if the Compensation Committee has a higher
percentage of independent outsiders, the firm’s market stock price generally grows.
Boyle et al. (1998) also add the information that when executives have small firm
ownership, ownership is negatively related to the number of extraordinary antitakeover
mechanisms adopted by firms, but when their ownership is higher, the number of
extraordinary antitakeover provisions is positively related to executive ownership.
Essentially, when executives have significant firm ownership, they will try to protect
their wealth by introducing alternative choices to get money if takeover is a success.
The use of golden parachutes has also been described as a factor that negatively affects the firm’s stock price (Borokhovich et al., 1997). The adoption of the golden parachutes mechanism can be understood by the market as an act to protect low quality
executives. Evans et al. (1997) confirm this idea and document that banks that adopt golden parachutes effectively present, on average, lower performance than banks with the same average size that do not adopt this protection. In fact, executives with golden parachutes should not be worried either about takeovers or about poor performance.
They are protected, in financial terms, from both situations. If the performance is bad
and shareholders want to get rid of the executives, they are obliged to pay them the
26
golden parachute sum. The same amount will be received if another company buys
their company. In other words, with golden parachutes, executives will be financially
protected from a situation that can negatively affect their wealth.
The existence of a significant outside director on the board of the companies has
also been defended by authors like Harford (2003) as an antitakeover mechanism.
Companies with a significant number of outside directors have fewer probabilities of
being acquired because they will align the interests of executives more with
shareholders’ interests, reducing agency costs. If outside directors do not monitor
executive actions, takeovers will appear and they will probably lose their board seat in
the future. Conversely, Chatterjee et al. (2003) document that target firms managed by
independent directory boards are likely to ignore the takeover attempt and not refocus
their firms' strategy.
Kroll et al., (1997) further point out that in a managercontrolled firm, the
acquisition announcements of another firm result in negative excess returns to
shareholders, but in the case of firms controlled by the owner, the acquisition
announcements result in positive excess returns to shareholders. This means that when
executives control the firm, they can extract personal benefits from the acquisition, but
not in the other situations.
In a spinoff context, Sewar and Walsh (1996) document that the act of selecting
a new CEO to manage the new firm and the design of their compensation contract are
not strongly correlated with the positive firm stock price reaction to the spinoff
announcements. Certo et al. (2001) also report that founder management has been
described as having a positive impact on IPO underpricing, and executive stock options
and stock option ownership interacted to influence the premiums that investors applied
to IPO firms (Certo et al., 2003). Finally, Hambrick and Finkelstein (1995) document that in management
controlled firms, where a single major firm owner does not exist, the predominant
compensation policy is to maximise CEO compensation, but when the firms are
externallycontrolled or have a major nonmanager owner, the predominant
compensation policy is generally to minimise CEO pay, subject to the ability to
attract/retain a good or satisfactory CEO.
27
6. Executive Compensation and Dividend Policy
Until now there has been little research on the relation between executive
compensation and firm dividend policy. The existing bibliography analyses the
relationship between executive compensation, dividends and firm growth, the impact of
the use of the stock options in terms of dividend policies in the USA and Japan and the
impact of the 2003 tax cuts in the USA.
Authors like Smith and Watts (1992) document that those big companies
generally pay higher dividends and higher compensation to executives than small
companies.
When a firm announces that this year it will pay more dividends than last year,
the market normally reacts positively and the firm’s stock price increases. What Lippert
et al. (2000) found is that when executives are paid with a significant part of their
compensation based on stock options, the traditional increase in the firm’s stock market
price is less than in the situation where executives are not paid with stock options. The
authors present two explanations for this situation. One explanation is that pay for
performance and high dividends are both mechanisms that control executive
opportunism. The second reason is based on the behaviour finance theory which says
that when executives have a higher financial and psychological investment in a certain
project, they are more likely to believe that the project will be a success in the future
and will probably inform the market incorrectly about future firm performance. If the
market believes that there is incorrect executive signalling, it will discount this
information from the firm’s stock price and expect the firm’s stock price growth, with
the associated dividend increase, to be less.
The impact of the use of stock options in terms of dividend polices was also
analysed on the Japanese market by Kato, Lemmon et al. (2005). Contrary to the results
of Lippert et al., 2000, they did not find that the adoption of stock based compensation
changed the firm dividend policy.
When top executives have significant ownership, and the firms where they work
have a significant level of agency problems, this is an incentive to increase the firm’s
dividend payment (Fenn and Liang, 2001). If executives have a significant number of
stocks from the firms, and agency problems are high, they will be able to extract more
28
money from the company when they increase the dividends. Brown et al. (2007) also
found that executives with higher ownership were more likely to increase dividends
after the 2003 tax cut in the USA.
7. Executive Compensation and Capital Structure
The present research on the relation between a firm’s capital structure and
executive compensation focuses on the influence of executive ownership in a firm’s
debt structure, the level of executive compensation and the firm’s capital structure, and
on the effect of the market as a mechanism of executive control and the firm’s value
maximisation.
One of the interesting findings that relate executive compensation to a firm’s
capital structure is that executives with significant ownership generally choose short
term debts for their company and executives with small ownership choose longterm
debt (Datta et al., 2005). Essentially, when part of executives’ compensation is
dependent on the future evolution of the firm’s stock price, they will choose shortterm
debts because they are afraid that the market will penalise the firm’s stock price and
their wealth will be negatively affected. Executives that do not hold company stock do
not have these preoccupations and can have debts with higher maturity.
In Japan, firms that give stock options to their executives have, on average,
lower levels of debt than companies that do not adopt this kind of compensation
component (Kato et al., 2005). This suggests that firms with high levels of debt avoid
stock option compensation so as not to reduce agency costs associated with debt. The
authors found only slight evidence that these companies adopt stock option plans to
improve firm performance. Calcagno and Renneboog (2007) complement these ideas
and defend that when a company has risk debt, it is best to give executives incentives
based on the firm’s performance, because they will try to improve the firm’s
performance to the level where they can exercise stock options, rebalancing the capital
structure. Cadenillas et al. (2004) complement this information, defending that the best
way to make executives adopt policies to maximise the firm’s value is to grant stock
with high leverage to good managers and stock options with low leverage, or unlevered,
to not so good managers because the former will have the knowledge and will make
29
efforts to achieve the best methodology to increase firm stock value, while the latter will
not. Lewellen (2006) also describes that firm leverage increases stock volatility, and
higher stock option ownership tends to increase the volatility costs of debt.
Effectively, optimal capital structure has been defended in corporate finance
literature as a way of maximising stock market value, and when companies signal to the
market that they are changing the capital structure, this is generally understood as a
positive sign of the firm’s future performance. These results are congruent with Born
and McWilliams (1997)’s findings, which documented that firm performance decreases
over the years when the firm changes equity to debt. Authors like Berger et al., (1997)
also documented that when executives are not monitored by the market, they will not
choose the optimal capital structure that maximises firm value. Coles et al. (2006) add
the idea that if there is higher sensitivity of CEO wealth to stock volatility, they will
increase firm leverage to get personal advantages and Efendi et al. (2007) also affirm
that the likelihood of a misstated financial statement increases greatly when the CEO
raises new debt so as not to affect executive wealth.
Finally, Phillips (1995) defends that managers’ incentives to maximise
shareholders’ wealth increase following firm recapitalisation; Sundaram and Yermack
(2007) find that CEOs with high debt incentives manage their firms in more
conservative ways and Cumming, Fleming and Suchard (2005) also report that top
executives are better remunerated than venture capitalists in Australia, who raise capital
for the firm to develop their own businesses, meaning that they are extracting wealth
from how they invest capital in the company.
8. Risk and Executive Compensation
Authors like Ross (2004) describe the conditions under which incentive
schedules make agents more or less risk averse.
The most important research that relates executive compensation to risk is
associated with the granting of executive stock options or in some cases with restricted
stock options. The main question about this relationship is exactly how much
compensation indexed to firm stock price shareholders must give to their executives to
motivate them to increase the firm’s stock price. When shareholders give the executive
30
a small, or significant, part of their compensation based on stock options, they will not
guarantee that these executives will receive this amount and the executive cannot refuse
this risk. Only executives with the ability to make the firm’s stock market increase to
the level that will enable them to exercise their options, or make hedging strategies, will
transform risk compensation into fixed compensation.
Another component of executive compensation associated with risk is restricted
stock. Restricted stock is stock that is granted to executives but that cannot be sold for a
certain number of years, normally between 3 and 10 years. The most important
difference between stock options and restricted stock compensation is in terms of the
risk: stock options can only be exercised if the firm’s stock market value increases to a
certain level, otherwise executives will not receive this value, and in the second case, if
executives stay in the company until they are able to sell the stock, they will receive
some value. Whether the value that executives receive is high or low depends on their
ability to make the firm’s stock price grow. Essentially, stock options are a riskier
compensation component then restricted stock because if they do not increase the firm’s
stock price to exercise stock price, they will receive nothing and in the case of restricted
stock they can always receive something.
Authors like Tian (2004) defend that the use of stock options to motivate
executives only works up to a certain level and if this level is exceeded, the incentive
effect decreases. Chen, Steiner and Whyte (2006) effectively documented that since
America’s bank deregulation, banks have increased the use of employed stock option
inducing executive risktaking. Garvey and Milbourn (2004) add the information that
when executives are young, or have low firm stock ownership, they generally delegate
market risk immunisation of their personal compensations based on stock options to
their company. Jin (2002) also documents that CEO incentives decrease with the risk
associated with the firms but not with market risks.
There are several solutions proposed by authors to manage the relationship
between granting stock options to executives and the risk that they are able to take.
One solution proposed by Brisley (2006) to balance the risk assumed by executives is
to grant not traditional stock options but “progressive performance vesting” stock
options. Essentially, “progressive performance vesting” stock options allow fixed
numbers of options to vest periodically independent of stock price performance. In this
31
way, executives can exercise a certain amount of stock options during a fixed period
and not only at the end. Another methodology is described by Johnson and Tian (2000),
who develop a pricing model of stock options with a strike price indexed to a
benchmark and defend that this model filters out common risks beyond the executive's
control, thereby increasing the efficiency of incentive contracts. Another was defended
by Calvet and Rahman (2006), who argue that the best choice is to give CAPMbased
indexed stock options to executives, which are stock options indexed to the Capital
Asset Price Model, because with this methodology executives will not be involved in
investment projects with high idiosyncratic risk and the compensation model considers
wealth diversification and degree of risk aversion.
According to Prendergast (2002), the empirical investigations that analyse the
existence of negative relationship between risk and incentives are not greatly successful,
because some of these tests have a positive, and not negative, relationship between
uncertainty and incentives. The authors defend that the literature on this relationship
fails because it does not incorporate an important effect of uncertainty on incentives,
which is the employees’ responsibility. When companies work in a certain context,
firms will define what exactly employees have to do and then monitor their actions, but
when the context is uncertain, the responsibility is delegated to the workers and to
reduce their opportunistic actions they will index their compensation to final output.
Miller, Wiseman and GomezMejia (2002) analysed the effects of unsystematic
and systematic firm risk in terms of CEO compensation risk bearing and total
compensation and found that pay in terms of performances and the greater earnings
potential associated with this contingent form of executive pay are highest when
executives can control the performance outcomes. LarrazaKintana et al. (2007) also
add the information that employment risk and variability in compensation correspond to
greater risk taking, while downside risk and the intrinsic value of stock options
correspond to lower risk taking.
Finally, Tufano (1996), in the context of the gold mining industry, found that in
firms where executives have a significant number of stock options, executives do not
manage gold price risk so well, but in firms where executives have a significant amount
of firm stock, they manage gold price risk better, suggesting that executive risk aversion
affects the policies adopted by executives to manage corporate risk.
32
9. References
Aboody, D. and Lev, B. (2000). Information asymmetry, R&D, and insider gains. The Journal of Finance, Vol. 55 (6), 27472766.
Acharya, V., John, K. and Sundaram, R. (2000). On the optimality of resetting
executive stock options. Journal of Financial Economics, Vol. 57(1), 65101. Adams, R., Almeida, H. and Ferreira, D. (2005). Powerful CEOs and their impact on
corporate performance. The Review of Financial Studies, Vol. 18(4), 14031432. Aggarwal, R. and Samwick, A. (1999). Executive compensation, strategic competition,
and relative performance evaluation: theory and evidence. The Journal of Finance, Vol. 54(6), 19992043.
Aggarwal, R. and Samwick, A. (2003). Performance incentives within firms: the effects
of managerial responsibility. The Journal of Finance, Vol. 58(4), 16131649.
Agrawal, A. and Knoeber, C. (1996). Firm performance and mechanisms to control
agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, Vol. 31(3), 377397.
Agrawal, A. and Knoeber, C. (1998). Managerial compensation and the threat of
takeover. Journal of Financial Economics, Vol. 47(2), 219239. Almazan, A. and Suarez, J. (2003). Managerial compensation and the market reaction to
bank loans. Review of Financial Studies, Vol. 16(1), 237261. Anderson, R. and Bizjak, J. (2003). An empirical examination of the role of the CEO
and the Compensation Committee in structuring executive pay. Journal of
Banking & Finance, Vol. 27(7), 13231348. Ang, J., Cole, R. and Lin, J. (2000). Agency costs and ownership structure. The Journal
of Finance, Vol. 55 (1), 81106. Ang, J., Lauterbach, B. and Schreiber, B. (2002). Pay at the executive suite: how do US
banks compensate their top management teams? Journal of Banking & Finance, Vol. 26, 11431163.
Baliga, B., Moyer, R., and Rao, R. (1996). CEO duality and firm performance: What's
the fuss? Strategic Management Journal, Vol. 17 (1), 4153.
33
Balkin, D.,Markman, G. and GomezMejia, L. (2000). Is CEO pay in hightechnology
firms related to innovation? Academy of Management Journal, Vol. 43 (6), 1118 1129.
Barron, J. and Waddell, G. (2003). Executive rank, pay and project selection. Journal of Financial Economics, Vol. 67(2), 305349.
Bebchuk, L. and Fried, J. (2004). Pay without performances: the unfulfilled promise of
executive compensation. Harvard University Press. Beer, M., Cannon, M., Baron, J., Dailey, P., Gerhart, B., Heneman III, H., Kochan, T.,
Ledford, G. and Locke, E. (2004). Promise and peril in implementing payfor
performance. Human Resource Management, Vol. 43(1), 348. Benartzi, S. (2003). Excessive extrapolation and the allocation of 401(k) accounts to
company stock. The Journal of Finance, Vol. 56(5), 17471764. Berger, P., Ofek, E. and Yermack, D. (1997). Managerial entrenchment and capital
structure decisions. The Journal of Finance, Vol. 52(4), 14111438. Bergstresser, D. and Philippon, T. (2006). CEO incentives and earnings management.
Journal of Financial Economics, Vol. 80 (3), 511529. Bernardo, A., Cai, H. and Luo, J. (2001). Capital budgeting and compensation with
asymmetric information and moral hazard. Journal of Financial Economics, Vol. 61(3), 311344.
Bernardo, A. (2004). Capital budgeting in multidivision firms: information, agency, and
incentives. The Review of Financial Studies, Vol. 17(3), 739767. Bertrand, M. and Mullainathan, S.(2003). Enjoying the quiet life? Corporate governance
and managerial preferences. The Journal of Political Economy, Vol. 111(5), 1043 1075.
Bliss, R. and Rosen, R. (2001). CEO compensation and bank mergers. Journal of Financial Economics, Vol. 61(1), 107138.
Born, J. and McWilliams, V. (1997). Equityfordebt exchange offers: theory, practice,
and evidence. The Financial Review, Vol. 32(2), 273291. Borokhovich, K., Brunarski, K. and Parrino, R. (1997). CEO contracting and
antitakeover amendments. The Journal of Finance, Vol. 52(4), 14951517.
34
Boyle, G., Carter, R. and Stover, R. (1998). Extraordinary antitakeover provisions and
insider ownership structure: the case of converting savings and loans. The Journal of Financial and Quantitative Analysis, Vol. 33(2), 291304.
Boyd, B. and Salamin, A. (2001). Strategic reward systems: A contingency model of
pay system design. Strategic Management Journal, Vol. 22 (8), 777792.
Brenner, M., Sundaram. R. and Yermack, D. (2000). Altering the terms of executive
stock options. Journal of Financial Economics, Vol. 57(1), 103128. Brickley, J., Linck, J. and Coles, J. (1999). What happens to CEOs after they retire?
New evidence on career concerns, horizon problems, and CEO incentives. Journal
of Financial Economics, Vol. 52(3), 341377. Brisley, N. (2006). Executive stock options: early exercise provisions and risktaking
incentives. The Journal of Finance, Vol. 61 (5), 24872509. Brookman, J., Chang, S. and Rennie, C. (2007). CEO cash and stockbased
compensation changes, layoff decisions, and shareholder value. The Financial Review, Vol. 42 (1), 99 119.
Brown, J. Liang, N. and Weisbenner, S. (2007). Executive financial incentives and
payout policy: firm responses to the 2003 dividend tax cut. The Journal of
Finance, Vol. 62(4), 19351965. Burns, N. and Kedia, S. (2006). The impact of performancebased compensation on
misreporting. Journal of Financial Economics, Vol. 79(1), 3567. Cadenillas, A., Cvitanic, J. and Zapatero, F. (2004). Leverage decision and manager
compensation with choice of effort and volatility. Journal of Financial
Economics, Vol. 73(1), 7192. Calcagno, R. and Renneboog, L. (2007). The incentive to give incentives: on the
relative seniority of debt claims and managerial compensation. Journal of Banking & Finance, Vol. 31(6), 17951815.
Calvet, A. and Rahman, A. (2006). The subjective valuation of indexed stock options and their incentive effects. The Financial Review, Vol. 41(2), 205227.
Carpenter, M. (2000). The price of change: The role of CEO compensation in strategic
variation and deviation from industry strategy norms. Journal of Management,
Vol. 26, (6), 11791198.
35
Carpenter, M., Sanders, W. and Gregersen, H. (2000). International assignment
experience at the top can make a bottomline difference. Human Resource Management, Vol. 39 (2/3), 277285.
Carpenter, M., Sanders, Wm and Gregersen, H. (2001). Bundling human capital with organizational context: The impact of international assignment experience on
multinational firm performance and CEO pay. Academy of Management Journal, Vol. 44(3), 493511.
Carpenter, M. and Sanders, Wm. (2002). Top management team compensation: The
missing link between CEO pay and firm performance? Strategic Management
Journal, Vol. 23(4), 367375. Carpenter, M. and Sanders, Wm. (2004). The effects of top management team pay and
firm internationalization on MNC performance. Journal of Management, Vol. 30(4), 509528.
Carter, M. and Lynch, L. (2001). An examination of executive stock option repricing. Journal of Financial Economics, Vol. 61 (2), 207225.
Certo, S., Covin, J., Daily, C. and Dalton, D.(2001). Wealth and the effects of founder
management among IPOstage new ventures. Strategic Management Journal, Vol.
22(6/7), 641658. Certo, S., Daily, C., Cannella Jr, A. and Dalton, D.(2003). Giving money to get money:
How CEO stock options and CEO equity enhance IPO valuations. Academy of Management Journal, Vol. 46(5), 643653.
Chakraborty, A. and Arnott, R. (2001). Takeover defences and dilution: A welfare
analysis. The Journal of Financial and Quantitative Analysis, Vol. 36(3), 311344. Chance, D., Kumar, R. and Todd, R. (2000). The 'repricing' of executive stock options.
Journal of Financial Economics, Vol. 57(1), 129154. Chatterjee, S., Harrison, J. and Bergh, D. (2003). Failed takeover attempts, corporate
governance and refocusing. Strategic Management Journal, Vol. 24(1), 8796. Chen, M. (2004). Executive option repricing, incentives and retention. The Journal of
Finance, Vol. 53(3), 11671200. Chen, C., Steiner, T. and Whyte, A. (2006). Does stock optionbased executive
compensation induce risktaking? An analysis of the banking industry. Journal of Banking & Finance, Vol. 30(3), 915945.
36
Coles, J., Suay, J. and Woodbury, D. (2000). Fund advisor compensation in closedend
funds. The Journal of Finance, Vol. 55(3), 13851414. Coles, J., Naveen, D. and Naveen, L. (2006). Managerial incentives and risktaking.
Journal of Financial Economics, Vol. 79 (2), 431468. Conyon, M. Peck, S. and Sadler, G. (2001), Corporate tournaments and executive
compensation: Evidence from the UK. Strategic Management Journal, Vol. 22 (8), 805815.
Coombs, J. and Skill, M. (2003). Managerialist and human capital explanation for key
executive pay premiums: A contingency perspective. Academy of Management
Journal, Vol. 46(1), 6373. Coombs, J. and Gilley, K. (2005). Stakeholder management as a predictor of CEO
compensation: main effects and interactions with financial performance. Strategic Management Journal, Vol. 26(9), 827840.
Core, J., Holthausen, R. and Larker, D. (1999). Corporate governance, chief executive
officer compensation, and firm performance. Journal of Financial Economics, Vol. 51(3), 371406.
Core, J. and Guay, W. (2001). Stock option plans for nonexecutive employees. Journal
of Financial Economics, Vol. 61(2), 253287. Cotter, J. and Zenner, M. (1994). How managerial wealth affects the tender offer
process. Journal of Financial Economics, Vol. 35(1), 6397. Cumming, D., Fleming, G. and Suchard, J. (2005). Venture capitalist valueadded
activities, fundraising and drawdowns. Journal of Banking & Finance, Vol. 29 (2),
295331.
Datta, S. IskandarDatta, M. and Raman, K. (2001). Executive compensation and
corporate acquisition decisions. The Journal of Finance, Vol. 56(6), 22992336. Datta, S., IskandarDatta, M. and Raman, K. (2005). Managerial stock ownership and
the maturity structure of corporate debt. The Journal of Finance, Vol. 60 (5), 23332350.
Davidson, W., Pilger, T. and Szakmary, A. (1998). Golden parachutes, board and
committee composition, and shareholder wealth. The Financial Review, Vol.
33(4), 1732.
37
Deckop, J., Merriman, K. and Gupta, S. (2006). The effects of CEO pay structure on
corporate social performance. Journal of Management, Vol. 32 (3), 329342. Deutsch, Y., Keil, T. and Laamanen, T. (2007). Decision making in acquisitions: The
effect of outside directors' compensation on acquisition patterns. Journal of Management, Vol. 33 (1), 3056.
Devers, C., Cannella JR, A., Reilly, G. and Yoder, M. (2007). Executive compensation:
A multidisciplinary review of recent developments. Journal of Management, Vol. 33 (6), 10161072.
Dow, J. and Raposo, C. (2005). CEO compensation, change, and corporate strategy. The
Journal of Finance, Vol. 60(6), 2701272. Efendi, J., Srivastava, A. and Swanson, E. (2007). Why do corporate managers misstate
financial statements? The role of option compensation and other factor. Journal of Financial Economics, Vol. 85(3), 667708.
Elston, J. and Goldberg, L. (2003). Executive compensation and agency costs in
Germany. Journal of Banking & Finance, Vol. 27(7), 13911410. Evans, J., Noe, T. and Thornton, J. (1997). Regulatory distortion of management
compensation: The case of golden parachutes for bank managers. Journal of
Banking & Finance, Vol. 21(6), 825848. Fenn, G. and Liang, N. (2001). Corporate payout policy and managerial stock
incentives. Journal of Financial Economics, Vol. 60 (1), 4572. Field, L. and Karpoff, J. (2002). Takeover defences of IPO firms. The Journal of
Finance, Vol. 57 (5), 18571889.
Frydman, C. and Sacks, R. (2005). Historical trends in executive compensation 1936
2003. MIT Sloan School of Management Working paper. Garvey, G. and Milbourn, T. (2004). Incentive compensation when executives can
hedge the market: evidence of relative performance evaluation in the cross section.
The Journal of Finance, Vol. 58 (4), 15571581. Garvey, G. and Milbourn, T. (2006). Asymmetric benchmarking in compensation:
executives are rewarded for good luck but not penalized for bad. Journal of Financial Economics, Vol. 82(1), 197225.
Gibson, M. (2003). Is corporate governance ineffective in emerging markets? Journal of Financial and Quantitative Analysis, Vol. 38(1), 230250.
38
Goetzmann, W., Ingersoll, J. and Ross, S. (2003). Highwater marks and hedge funds
management contracts. The Journal of Finance, Vol. 58 (4), 16851718. Goldberg, L. and Idson, T. (1995). Executive compensation and agency effects. The
Financial Review, Vol. 30(2), 313335. Goldman, E. (2004). The impact of stock market information production on internal
resource allocation. Journal of Financial Economics, Vol. 71(1), 143167. Golec, J. and Starks, L. (2004). Performance fee contract change and mutual fund risk.
Journal of Financial Economics, Vol. 73(1), 93118. GomezMejia, L., NunezNickel, M. and Gutierrez, I. (2001). The role of family ties in
agency contracts. Academy of Management Journal, Vol. 44(1), 8595. GomezMejia, L., LarrazaKintana, M. and Makri, M. (2003).The determinants of
executive compensation in familycontrolled public corporations. Academy of Management Journal, Vol. 46 (2), 226237.
Grinstein, Y. and Hribar, P. (2004). CEO compensation and incentives: evidence from
M & A bonuses. Journal of Financial Economics, Vol. 73(1), 119–143. Grossman, W. and Cannella Jr, A. (2006). The impact of strategic persistence on
executive compensation. Journal of Management, Vol. 32 (2), 257278.
Gupta, S. and Romano, R. (1998). Monitoring the principal with multiple agents. RAND Journal of Economics, Vol. 29(2), 427442.
Hallock, K. (1997). Reciprocally interlocking boards of directors and executive
compensation. Journal of Financial and Quantitative Analysis, Vol. 32(3), 331 426.
Hambrick, D. and Finkelstein, S. (1995). The effects of ownership structure on
conditions at the top: The case of CEO pay raises. Strategic Management Journal, Vol. 16 (3), 175194.
Harford, J. (2003). Takeover bids and target directors’ incentives: the impact of a bid on
directors’ wealth and board seats. Journal of Financial Economics, Vol. 69(1), 51 83.
Hartzell, J. and Starks, L. (2003). Institutional Investors and executive compensation. The Journal of Finance, Vol. 58 (6), 23512374.
Hartzell, J., Ofek, E. and Yermack, D. (2004). What's in it for me? CEOs whose firms
are acquired. Review of Financial Studies, Vol. 17(1), 3761.
39
Hayward, M., Rindova, V. and Pollock, T. (2004), Believing one´s own press: the
causes and consequences of CEO celebrity. Strategic Management Journal, Vol. 25(7), 637653.
Henderson, A. and Fredrickson, J. (2001). Top management team coordination needs and the CEO pay gap: A competitive test of economic and behavioural views.
Academy of Management Journal, Vol. 44 (1), 96117. Hermalin, B. and Wallace, N. (2001). Firm performance and executive compensation in
the savings and loan industry. Journal of Financial Economics, Vol. 61(1), 139 170.
Hu, J. and Noe, T. (2001). Insider trading, and managerial incentives. Journal of Banking & Finance, Vol. 25(4), 681716.
Jensen, M. and Meckling, W. (1976). Theory of the firm: managerial behaviour, agency
costs and ownership structure. Journal of Financial Economics, Vol. 13(4), 305
360.
Jin, L. (2002). CEO compensation, diversification, and incentives. Journal of Financial Economics, Vol. 66(1), 2963.
Johnson, S. and Tian, Y. (2000). Indexed executive stock options. Journal of Financial
Economics, Vol. 57(1), 3564. Kato, H., Lemmon, M., Luo, M. and Schallheim, J. (2005). An empirical examination
of the costs and benefits of executive stock options: evidence from Japan. Journal of Financial Economics, Vol. 78(2), 435461.
Katz, M. and Rosen, H. (1998). Microeconomics. Irwin/McGrawHill, 3rd edition, 577.
Kedia, S. (2006). Estimating product market competition: methodology and application. Journal of Banking & Finance, Vol. 30, 875894.
Khorana, A. (2001). Performance changes following top management turnover:
evidence from openend mutual funds. Journal of Financial and Quantitative
Analysis, Vol. 36(3), 371393. Kole, S. (1997). The complexity of compensation contracts. Journal of Financial
Economics, Vol. 43 (1), 79104. Kroll, M., Wright, P., Toombs, L. and Leavell, H. (1997). Form of control: A critical
determinant of acquisition performance and CEO rewards. Strategic Management Journal, Vol. 18 (2), 8596.
40
LarrazaKintana, M.,Wiseman, R., GomezMejia, L. and Welbourne, T. (2007).
Disentangling compensation and employment risks using the behavioural agency
model. Strategic Management Journal, Vol. 28(10), 10011019. Lewellen, K. (2006). Financing decisions when managers are risk averse. MIT Sloan
School of Management Working Paper 443803
Lippert, R., Nixon, T. and Pilotte, E. (2000). Incentive compensation and the stock price
response to dividend increase announcements. The Financial Review, Vol. 35(4), 6994.
Lowe, K., Milliman, J., Cieri, H., Dowling, P. (2002). International compensation
practices: A tencountry comparative analysis. Human Resource Management, Vol. 41 (1), 4566.
Lowry, M. and Murphy, K. (2007). Executive stock options and IPO underpricing. Journal of Financial Economics, Vol. 85(1), 3965.
Magnan, M., StOnge, S. (1997). Bank performance and executive compensation: A
managerial discretion perspective. Strategic Management Journal, Vol. 18 (7), 573581.
Makri, M., Lane, P. and GomezMejia, L. (2006). CEO incentives, innovation, and
performance in technologyintensive firms: a reconciliation of outcome and
behaviorbased incentive schemes. Strategic Management Journal, Vol. 27(11), 10571080.
Mehran, H. (1995). Executive compensation structure, ownership, and firm
performance. Journal of Financial Economics, Vol. 38(2), 163184.
Milbourn, T. (2003). CEO reputation and stockbased compensation. Journal of Financial Economics, Vol. 68(2), 233262.
Miller, J.,Wiseman, R. and GomezMejia, L. (2002). The fit between CEO compensation design and firm risk. Academy of Management Journal, Vol. 45(4),
745756.
Morgan, A. and Poulsen, A. (2001). Linking pay to performance compensation
proposals in the S&P 500. Journal of Financial Economics, Vol. 62(3), 489523. Murphy, K. (1999). Executive compensation. Handbook of Labor Economics, in: O.
Ashenfelter & D. Card (ed.), Handbook of Labor Economics, edition 1, Vol.3, chapter 38, pages 24852563, Elsevier.
41
Murphy, K. (2003). Stockbased pay in new economy firms. Journal of Accounting and
Economics, Vol. 34(1), 129147. Narayanan, M. (1996). Form of compensation and managerial decision horizon.
Journal of Financial and Quantitative Analysis, Vol. 31(4), 467491. Narayanan, R. (1999). Information production, insider trading, and the role of
managerial compensation. The Financial Review, Vol. 34(4), 119144. Ofek, E. and Yermack, D. (2000). Taking stock: equitybased compensation and the
evolution of managerial ownership. The Journal of Finance, Vol. 55(3), 1367 1348.
Osano, H. (2002). Managerial compensation contract and bank bailout policy. Journal of Banking & Finance, Vol. 26(1), 2549.
Pagano, M. and Volpin, P. (2005). Managers, workers, and corporate control. The Journal of Finance, Vol. 60(2), 841868.
Perry, T. and Zenner, M. (2001). Pay for performance? Government regulation and the
structure of compensation contracts. Journal of Financial Economics, Vol. 62(3), 453488.
Phillips, G. (1995). Increased debt and industry product markets: an empirical analysis.
Journal of Financial Economics, Vol. 37(2), 189 238. Pollock, T., Fischer, H. and Wade, J. (2002). The role of power and politics in the
repricing of executive options. Academy of Management Journal, Vol. 45 (6), 11721182.
Povel, P., Singh, R. and Winton, A. (2007). Booms, busts, and fraud. The Review of
Financial Studies, Vol. 20(4), 12191254. Prendergast, C. (2002). The tenuous tradeoff between risk and incentives. The Journal
of Political Economy, Vol. 110 (5), 10711102. Rajagopalan, N. (1997). Strategic orientations, incentive plan adoptions, and firm
performance: Evidence from electric utility firms, Vol. 18 (10), 761785. Rajan, R. and Wulf, J. (2006). Are perks purely managerial excess? Journal of
Financial Economics, Vol. 79(1), 133. Rajgopal, S., Shevlin, T. and Zamora, V. (2006). CEOs' outside employment
opportunities and the lack of relative performance evaluation in compensation
contracts. The Journal of Finance, Vol. 61(4), 18131844.
42
Ross, S. (1973). The economic theory of agency: the principal’s problem. American
Economy Review, Vol. 63 (2), 134139. Ross, S. (2004). Compensation, incentives and duality of risk aversion and riskiness.
The Journal of Finance, Vol. 59 (1), 207225. Sanders, Wm (2001a). Incentive alignment, CEO pays level, and firm performance: A
case of "Heads I win, tails you lose? Human Resource Management, Vol. 40(2), 159170.
Sanders, Wm. (2001b). Behavioral responses of CEOs to stock ownership and stock
option pay. Academy of Management Journal,Vol. 44 (3), 477 492.
Sanders, Wm, and Boivie, S. (2004). Sorting things out: valuation of new firms in
uncertain markets. Strategic Management Journal, Vol. 25(2), 167186. Seward, J. and Walsh, J. (1996). The governance and control of voluntary corporate
spinoffs. Strategic Management Journal, Vol. 17(1), 2539.
Schnatterly, K. (2003). Increasing firm value through detection and prevention of white
collar crime. Strategic Management Journal, Vol. 24 (7), 587614. Smith, C. and Watts, R. (1992). The investment opportunity set and corporate financing,
dividend, and compensation policies. Journal of Financial Economics, Vol. 32 (3),
263292.
Sundaram, R. and Yermack., D. (2007). Pay me later: inside debt and its role in
managerial compensation. The Journal of Finance, Vol. 62(4), 15511588. Tian, Y. (2004). Too much of a good incentive? The case of executive stock options.
Journal of Banking & Finance, Vol. 28 (6), 12251245.
Tosi, H.,Werner, S., Katz, J. and GomezMejia, L.(2000). How much does performance matter? A metaanalysis of CEO pay studies. Journal of Management, Vol. 26 (2), 301339.
Tufano, P. (1996). Who manages risk? An empirical examination of risk management
practices in the gold mining industry. The Journal of Finance, Vol. 51(4), 1097 1137.
Tuschke, A. and Sanders, Wm. (2003). Antecedents and consequences of corporate
governance reform: the case of Germany. Strategic Management Journal, Vol. 24
(7), 631650.
43
Ueng, C., Wells, D. and Lilly, J. (2000). CEO influence and executive compensation:
large firms vs. small firms. Managerial Finance, Vol 26(8), 3 12. Wade, J., Porac, J., Pollock, T. and Graffin, S.(2006). The burden of celebrity: the
impact of CEO certification contests on CEO pay and performances. Academy of Management Journal, Vol. 49 (4), 643660.
Werner, S., Tosi, H. and GomezMejia, L.(2005). Organizational governance and
employee pay: how ownership structure affects the firm's compensation strategy. Strategic Management Journal, Vol. 26 (4), 377384.
Wright, P., Kroll, M., Lado, A. and Ness, B. (2002). The structure of ownership and
corporate acquisition strategies. Strategic Management Journal, Vol. 23 (1), 41 53.
Wright, P., Kroll, M. and Elenkov, D. (2002). Acquisition returns, increase in firm size,
and Chief Executive Officer compensation: The moderating role of monitoring.
Academy of Management Journal, Vol. 45(3), 599608. Yermack, D. (1997). Good timing: CEO stock options award and company news
announcements. The Journal of Finance, Vol. 52(2), 449476. Yermack, D. (2006). Flights of fancy: corporate jets, CEO perquisites, and inferior
shareholder returns. Journal of Financial Economics, Vol. 80(1), 211242.
44
CHAPTER 2
New vs. Old Economy: Trends and Determinants of Executive
Compensation 6
6 We are grateful to Prof. Amir Licht, from the Interdisciplinary Center Herzliya, paper discussant of this paper at the Financial Management Association/Asia Finance Association Annual Conference, in Hong Kong, in July 2007. We are also grateful to anonymous attendees of the conference for their helpful comments.
45
1. Introduction
In this new study, we examine the determinants and the forms of executive
compensation of new versus old economy US firms over time. We focus our research
on providing answers to the following questions: Are the determinants of executive
compensation for new versus old economy firms the same? Are there significant
differences between the compensation value and the forms of executive compensation
in new versus old economy firms? Does the form of executive compensation change
after the NASDAQ crash and the enactment of the SarbanesOxley act?
According to Anderson et al. (2000) and Murphy (2003), new economy firms are
defined as firms competing in the computer, software, Internet, telecommunications, or
networking fields and as per several researchers such as Ittner et al. (2003), Anderson et
al. (2000), Murphy (2003), Stathopoulos et al. (2004) and most recently Chen and Hung
(2006), new economy firms are fundamentally different in terms of many of the
characteristics they possess compared to the old economy firms.
The present studies on executive compensation have some limitations which
offer a motivation for the current inquiry. One of the limitations is that there is only one
study, by Stathopoulos et al. (2004), which analyses the executive compensation for
both new and old economy firms, but it is confined to the English market only. The
other limitation is that most studies focus on the preponderance of stock options as a
form of executive compensation, without analysing the remaining components of the
compensation. With the exception of the study developed by Murphy (2003), all other
studies on executive compensation analyse short periods of time; therefore, the
conclusions achieved by these studies must be validated for a longer period of time.
Given that there are inherent differences between new and old economy firms,
we believe that executive compensation between old and new firms may be influenced
by different factors, and the trends in terms of compensation may also be different over
time. Based on Narayanan and Seyhun (2005) and Murphy (2003)’s findings, we also
believe that the NASDAQ crash in 2000 and the SarbanesOxley act in 2002
significantly changed the form of compensation for executives in the new and old
economies because a significant number of restrictions in terms of corporate governance
were developed.
46
We also deal with the old problem in executive compensation literature as to
what is the best variable to measure the impact of the firm size: LN (assets), LN (market
value), LN (sales) or these variables without a natural logarithm. Effectively, firm size
is described as one of the most important variables in explaining executive
compensation; nevertheless, in the mind of the researchers, there is generally a doubt as
to whether using one of the above variables, at the exclusion of the others, will produce
inferior results. To solve this problem, we use the Principal Component method and
extract a factor that is the best combination of the three variables to measure the firm
size.
Our results reveal that the number of executives in new economy firms is
considerably smaller than the number of executives in old economy firms. Most of the
new economy executives are from firms associated with Prepackaged Software (26.02%), Semicondutor and Related Devices (17.29%), Computer Programming, Data
Processing (9.46%) and Telecommunications (7.50%).
Executives from new economy firms always receive on average more than those
from old economy firms, but the difference in total compensation has fallen in recent
years. Executive compensation in the new economy firms consists of more than 50%,
and in the case of old economy firms more than 30%, of stock options between 1999
and 2001. After that period, with the NASDAQ Crash and the introduction of the
SarbanesOxley Act in 2002, new and old economy firms instituted a change in the
structure of the components of the executive compensation, reducing the use of stock
options and increasing the use of bonus and restricted stocks.
We also find that the factors that explain executive compensation in new versus
old economy firms are generally different, and in the case of the variables that are the
same, our tests generally rejected the hypothesis that the coefficients related to these
common factors are equal. New economy total executive compensation is influenced by
firm size, the ratio of the number of stock options vested but not exercised and
executive stock ownership, whereas the old economy total executive compensation is
influenced by firm size, executive ownership, oneyear total return to shareholders and
the 5year annual growth rate of firm net income.
47
2. Literature Review and Research Questions
Anderson et al. (2000), Ittner et al. (2003), Murphy (2003) and most recently
Chen and Hung (2006) analyse the issue of executive compensation but only for new
economy firms. Stathopoulos et al. (2004) investigate the executive compensation both
for new and old economy firms in England, but only for a period of two years.
All the studies that analyse new economy executive compensation focus their
attention basically on one component of the executive compensation stock options at
the expense of other components of executive compensation. As discussed earlier, in
this inquiry we take a comprehensive and current view of issues related to executive
compensation.
Use of stock options has been a common and predominant form of executive
compensation; therefore, several studies have focused on this form of compensation.
There are multiple reasons for firms to award stock options to their executives.
Stathopoulos et al. (2004) believe that stock options are given to new economy
executives to align shareholders’ interests with executives’ goals, reduce agency costs,
achieve beneficial tax gains for the company and the employees, and attract and retain
executives with significant knowledge of new technologies. Ittner et al. (2003) are of
the view that new economy firms give stock options to executives because the firms
have difficulty generating enough cash flow to pay high salaries to executives.
Murphy (2003) thinks that a large group of big companies compete in the market
for high quality executives. The compensation contracts that they give to this kind of
executive force the other companies to use the same structure of compensation:
inclusion of stock options as a major component of compensation package. Another
reason invoked is related to what the author calls the "perceivedcostview"; in other
words, there is the wrong perception that executives compensated with stock options
constitute a cheap form of compensation
From prior studies, we know that until 2000, executives from new economy
firms were predominantly paid with stock options. However, the most important
fraudulent bankruptcy cases, such as Enron and WorldCom, are indirectly related to the
significant number of stock options granted to the executives. We believe that the
NASDAQ crash and the SarbanesOxley Act changed the components of compensation
48
for executives in both new and old economy firms due to the introduction of new rules
in the market place. The new rules contain provisions which have a significant impact
on the benefits and compensation of public company executives. Therefore, given the
above historical context we find reason to test empirically whether the NASDAQ crash
in 2000 and the SarbanesOxley Act in 2002 changed the structure of executive
compensation, and if they did, whether or not the impact of these major events was the
same for the executive compensation of new versus old economy firms.
According to Ittner et al. (2003), Murphy (2003) and Stathopoulos et al. (2004),
new economy companies differ from old economy companies because they present
higher growth rates of sales increase; they spend more money on research and
development; they present low ratios of booktomarket value; they offer lower
dividends per share and a high volatility of share returns. They still hold a smaller
number of employees, a reduced market value and smaller accounting returns than old
economy firms. In addition, they provide larger compensation relative to capital
ownership, have a higher percentage of stock options and a higher percentage of the
volume of stock options not exercised related to the total number of company shares.
Furthermore, Chen and Hung (2006) express the idea that companies listed on
NASDAQ have small boards, and the founder is usually the CEO. They have a large
shareholding of insiders, and the CEOs usually accept lower cash compensation and
higher option compensation in order to convince shareholders of managerial
commitment and future profitability.
Based on these differences, we also develop the idea that if new and old
economy firms have different characteristics, these characteristics can generally lead to
differences in terms of the factors that explain executive compensation. We also believe
that even if some of the factors are equal, their intensity can be different.
49
3. Data, Sample Selection and Statistics
3.1. Data and Sample Selection
Data is from the Standard and Poor´s ExecuComp database 7 , which collects
information about the five most well paid executives from firms listed on S&P Indexes.
We use Unbalanced Panel Data, and our final sample is composed of 67,437
observations of executive compensation for the 13year period from 1992 to 2004. We
retrieve compensation package details for up to the top five executives in each firm,
including salary, bonus, exante value of options, restricted stock award, Longterm
Incentive plan (LTIP), other annual compensation, all other compensation and several
variables associated with governance and finance.
To develop the sample used in this study, we apply a few restrictions. First, we
remove all the executives whose sum of salary and bonus was equal to zero, in other
words, those who received neither salary nor bonus during the year, receiving instead
some other remuneration types. We also exclude all those executives whose total
compensation is equal to zero. To achieve this final sample we exclude 272 items of
compensation related to executives that received compensation from more than one
company. We delete the compensation values from the company for which the
executives worked less time.
Using the Consumer Price Index (CPI) compiled by the Bureau of Labor
Statistics, with 1982 as the base year, we adjust the monetary variables for inflation.
In order to distinguish between executives from new and old economy firms, we
use the methodology of Murphy (2003), who considers firms from the new economy
with SIC codes 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961,
7370, 7371, 7372 and 7373 and firms from the old economy with SIC codes lower than
4000 unless categorised as new economy firms.
7 The ExecuComp version is from 062006.
50
3.2. Statistics
Table 1 presents the number of observations (compensation items) for each SIC
code of new and old economy firms. For example, there are 123 compensation items
from the Computer and Office Equipment industry which represent 0.77% of the total
compensation items in that industry (% of the group) and 0.18% of the total sample of
observations from old and new economy firms. Our sample has nearly 76%
observations from the old economy and 24% observations from the new economy.
Moreover, we can see from Table 1 that our sample observations are dominated by
executives associated with companies from PrePackaged Software, Semiconductor,
Related Devices and Telecommunications industries.
Table 1: Number of Items of Compensation by SIC Code
SIC Code SIC Code Description Number of Items of
Compensation % of the Group
% of Total (Old+ New) Economy)
PANEL A: New Economy 3570 Computer and Office Equipment 123 0.77% 0.18% 3571 Electronic Computers 590 3.68% 0.87% 3572 Computer Storage Devices 595 3.71% 0.88% 3576 Computer Communication Equipment 981 6.12% 1.45% 3577 Computer Peripheral Equipment 412 2.57% 0.61% 3661 Telephone & Telegraph Apparatus 972 6.07% 1.44% 3674 Semiconductor and Related Devices 2770 17.29% 4.11% 4812 Wireless Telecommunication 423 2.64% 0.63% 4813 Telecommunications 1201 7.50% 1.78% 5045 Computers and Software Wholesalers 288 1.80% 0.43% 5961 Electronic MailOrder Houses 562 3.51% 0.83%
7370 Computer Programming, Data Processing 1515 9.46% 2.25%
7371 Computer Programming Service 182 1.14% 0.27%
7372 Prepackaged Software 4168 26.02% 6.18%
7373 Computer Integrated Systems Design 1238 7.73% 1.84% Total New Economy 16020
PANEL B: Old Economy < 4000 and not new ecomomy
51417 76.24%
Total New+Old Economy
67437
Notes: To distinguish between executives from new and old economy firms, we used the methodology of Murphy (2003), which considers firms from the new economy those with SIC code 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961, 7370, 7371, 7372 and 7373 and firms from the old economy those with SIC codes less than 4000 and not yet categorised with the new economy.
51
Table 2 presents the average total compensation of executives and the ttest of
independence of means of executives from new versus old economy firms during the
period from 1992 to 2004.
Table 2 Mean Total Executive Compensation Levels for New and Old Economy Firms (19922004)
Our sample includes data from the five most well paid executives associated with the firms listed in the S&P500, S&PMidCap and S&PSmallCap during the period of 1992 to 2004. All the data is from the ExecuComp database. To distinguish between executives from new and old economy firms we used the methodology of Murphy (2003), which considers firms from the new economy those with SIC codes 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961, 7370, 7371, 7372 and 7373 and firms from the old economy those with SIC codes less than 4000 and not yet categorised with the new economy. Total compensation is the sum of salary, bonus, stock options, restricted stocks, longterm incentive plans (LTIP), other annual compensation and all other compensation listed under All other Compensation in the Summary Compensation Table. Values are in thousands of dollars. All the values are adjusted for inflation and are in 2004 dollars.
New Economy Old Economy T test of mean
Year N Mean N Mean Mean Difference t Sig.
1992 485 1477,41 2412 1260,39 217,02 2,134 0,033
1993 887 1399,42 3778 1183,84 215,58 3,177 0,002
1994 958 1668,20 4058 1302,55 365,65 4,640 0,000
1995 1026 828,22 4166 1306,67 521,56 4,529 0,000
1996 1226 2339,72 4319 1542,72 797,01 5,414 0,000
1997 1398 2721,97 4335 1815,28 906,67 5,558 0,000
1998 1495 3438,06 4419 2805,12 1641,51 2,664 0,008
1999 1577 4996,27 4301 2145,37 2850,90 7,368 0,000
2000 1507 6660,90 4091 2570,57 4090,85 9,109 0,000
2001 1417 5159,96 3880 2438,90 2721,07 6,032 0,000
2002 1366 3114,48 3942 2165,15 949,34 5,283 0,000
2003 1373 2346,03 3926 2033,53 312,50 2,863 0,000
2004 1305 2618,61 3790 2420,20 208,42 1,384 0,166
52
From Table 2 we see that, with the exception of the year 1993, the total average
compensation of executives from new and old economies increases from 1992 to 2000
(Nasdaq crash), decreases until 2003 and starts to increase again, but slowly, in 2004.
We can also verify that the mean total compensation difference between the two groups
of executives is very high in the years 1999 and 2000 but drops drastically after 2002,
and it is small in 2004.
Table 3 summarises the evolution of the various compensation components (as a
percentage of total compensation) of the new versus old economy firms during the
period from 1992 to 2004 8 . We use the IndependentSamples Ttest to compare the
means of executive compensation components and Levene's test for equality of
variances between the two subsamples of old versus new economy firms. We also
perform the same test to compare whether the difference between the values of each
component of compensation in the years 2001 to 2002 (NASDAQ crash effect) and
years 2003 to 2002 (SarbanesOxley Act effect) is statistically significant.
8 Last year of information available from Execucomp database when we started this study.
53
Table 3 Executive Components as a Percentage of Total Compensation in New and Old Economy
Firms (19922004)
The table displays the percentage of average value of the compensation components for the top five executives, CEOs and directors, in firms that belong to the S&P indexes. Salary is the executive salary for the year. Bonus is the dollar value of bonus (cash and noncash) earned by the executive officer during the fiscal years. Stock Options is the aggregate value of stock options granted to the executive during the fiscal year as valued using S&P`s BlackScholes methodology. Restricted Stocks is the value of restricted stock granted during the year (determined as of the date of the grant). LTIP is the amount paid out to the executive under the company's longterm incentive plan. These plans measure company performance over a period of more than one year (generally three years). Values are in thousands of dollars. All the figures are adjusted for inflation and are in 2004 dollars.
PANEL A: Top Five Executives
N Salary Bonus Stock Options Restricted Stocks LTIP Year
New Old New Old New Old New Old New Old New Old
1992 485 2412 45.50% 49.46% 17.92% 18.04% 28.71% 20.43%* 1.69% 3.88%* 2.64% 3.10%*
1993 887 3778 43.80% 48.59%* 18.28% 18.69% 30.52% 20.21%* 1.52% 3.72%* 1.69% 2.88%*
1994 958 4058 40.04% 45.65%* 18.44% 20.29% 34.09% 22.67%* 1.67% 3.27%* 1.22% 2.61%*
1995 1026 4166 38.94% 46.05%* 19.40% 20.18% 33.61% 21.08%* 1.69% 3.90%* 2.03% 3.13%*
1996 1226 4319 35.54% 42.34%* 16.00% 19.41% 40.36% 25.54%* 2.54% 3.89%* 1.55% 3.40%*
1997 1398 4335 35.29% 39.31%* 15.03% 19.94%* 42.92% 27.78%* 1.62% 3.84%* 1.15% 3.89%*
1998 1495 4419 36.37% 39.51%* 13.71% 17.42%* 43.74% 30.85%* 1.42% 3.87%* 0.78% 2.95%*
1999 1577 4301 31.51% 37.57%* 12.48% 18.45%* 50.17% 32.23%* 1.31% 3.72%* 0.83% 2.72%*
2000 1507 4091 27.91% 37.01%* 11.92% 18.21%* 55.02% 32.89%* 1.53% 3.90%* 0.54% 2.44%*
2001 1417 3880 29.72% 37.88%*/(**) 9.46%(*) 14.86%*(*) 54.16% 35.98%*/(*) 1.90% 3.88%* 0.45%/(*) 1.98%*/(*)
2002 1366 3942 35.17% 37.52%* 11.42% 17.39%* 47.29% 31.98%* 1.98% 5.37%* 0.38% 2.18%*
2003 1373 3926 32.93%(*) 38.08%* 15.84%(*) 18.03%*/(**) 40.56%(*) 27.92%*/(*) 3.75%(*) 7.06%*/(*) 0.40%(*) 3.08%*/(*)
2004 1305 3790 29.72% 33.14%* 15.47% 21.51%* 41.22% 26.75%* 6.60% 9.58%* 0.42% 3.41%*
54
Table 3 (cont.)
PANEL B: CEOs
N Salary Bonus Stock Options Restricted Stocks LTIP Year
New Old New Old New Old New Old New Old New Old
1992 27 170 37.51% 39.02% 25.00% 19.30%** 23.70% 26.28% 2.84% 4.30% 6.20% 6.02%
1993 118 537 35.50% 43.12%* 19.84% 19.80% 37.65% 22.60%* 1.10% 4.80%* 2.18% 3.47%
1994 161 701 32.55% 41.02%* 19.25% 20.91% 40.58% 26.22%* 1.80% 4.25%* 1.39% 2.77%**
1995 167 732 33.31% 40.92%* 18.87% 21.22% 39.46% 26.22%* 1.41% 4.25%* 2.48% 3.32%
1996 175 745 31.12% 37.32%* 17.01% 20.28%** 42.22% 24.65%* 3.81% 4.09% 1.52% 3.77%*
1997 201 751 30.19% 33.96%*** 16.30% 20.74%* 45.67% 29.03%* 1.93% 4.36%* 1.22% 4.46%*
1998 220 761 30.23% 34.52%** 13.72% 18.50%* 49.28% 31.07%* 1.32% 4.36%* 0.76% 3.23%*
1999 278 765 28.14% 31.81%*** 12.56% 18.96%* 52.87% 34.42%* 1.50% 3.77%* 0.94% 3.15%*
2000 270 757 28.09% 31.96%*** 13.09% 18.56%* 53.54% 36.84%* 1.28% 4.64%* 0.39% 2.77%*
2001 249 703 25.59% 32.76%* 8.69%(*) 14.51%*/(*) 57.80% 36.37%*/(*) 2.44%(*) 4.70%* 0.64% 2.07%*
2002 237 700 28.11% 30.99% 10.02% 17.78%* 54.95% 40.48%* 2.07% 6.06%* 0.42% 2.89%*
2003 239 701 30.14% 31.83% 15.70%(*) 18.44%** 44.98%(*) 35.57%*/(*) 5.21%(*) 7.85%**/(*) 0.29%(*) 3.98%*/(**)
2004 239 711 27.34% 28.03% 14.71% 22.73%** 46.83% 32.77%* 7.19% 11.07%* 0.43% 3.73%*
PANEL C: Director s
N Salary Bonus Stock Options Restricted Stocks LTIP Year
New Old New Old New Old New Old New Old New Old
1992 206 1121 42.29% 46.38%** 19.77% 18.49% 29.91% 22.28%* 1.90% 3.76%* 2.88% 3.65%
1993 325 1584 41.24% 45.71%* 18.96% 19.17% 33.49% 22.06%* 1.40% 3.84%* 1.72% 3.08%*
1994 331 1594 37.47% 42.73%* 19.40% 21.03%*** 35.56% 24.87%* 2.06% 3.46%* 1.55% 2.84%*
1995 349 1593 35.39% 42.86%* 19.45% 20.77% 36.32% 22.71%* 1.71% 3.95%* 2.51% 3.45%***
1996 418 1606 34.18% 39.12%* 15.90% 20.03%* 41.76% 27.81%* 2.77% 3.93%* 1.24% 3.59%*
1997 465 1618 32.58% 36.56%* 15.54% 20.57%* 43.88% 29.62%* 2.08% 3.97%** 1.20% 3.94%*
1998 478 1611 34.72% 36.54% 13.81% 18.08%* 44.17% 32.76%* 1.39% 4.03%* 0.69% 3.09%*
1999 483 1492 30.09% 34.04%* 13.10% 19.24%* 49.50% 34.36%* 1.82% 3.65%* 0.99% 2.85%*
2000 464 1362 27.97% 33.59%* 11.70% 18.83%* 53.81% 34.42%* 1.69% 4.64%* 0.43% 2.52%*
2001 416 1229 28.07% 34.75%* 8.52%(*) 15.39%*/(*) 56.32% 37.72%*/(*) 2.04%(*) 4.11%*/(*) 0.56%(*) 1.99%*/(*)
2002 373 1180 29.70% 32.85%*** 10.32% 18.32%* 52.67% 34.88%* 1.76% 5.60%* 0.58% 2.55%*
2003 354 1144 32.15% 32.93% 15.24%(*) 19.51%*/(**) 43.27%(*) 31.10%*(*) 4.31%(*) 7.15%*(*) 0.30%(*) 3.30%*/(*)
2004 343 1065 29.74% 28.92% 15.62% 22.86%* 43.75% 28.49%* 6.66% 10.31%* 0.71% 3.61% *
Note: Difference between the old and new economy is statistically significant at 1% level *, 5% level ** and 10% level ***. In rows for years 2001 and 2003 we also describe whether the differences between each component of compensation between years 2001 and 2000 (Nasdaq crash effect) and 2003 related to 2002 (SarbanesOxley Act effect) are statistically significant. Significance is presented as ( ).
55
If we analyse the compensation for the top five executives, we see that beginning
in 1992, the most important component of compensation was salary for both new and
old economy firms. After 1996, stock options were the most important component of
executive compensation in new economy firms, and in the case of old economy firms
the use of stock options also increased. After the NASDAQ crash in 2000 and the
introduction of the SarbanesOxley Act in 2002, the component weights of executive
compensation significantly changed in both new and old economy firms. The use of
stock options decreased, still significant in both subsamples, but the use of bonus and
restricted stocks 9 increased. Results for two sample ttests show that in most cases, the
use of options and restricted stocks significantly changed after the NASDAQ crash and
the SarbanesOxley Act.
In our view, the change from stock options to restricted stock may be due to the
fact that both are compensation components associated with performance of a firm,
whereas salary is not dependent upon performance. More precisely, when a firm grants
stock options to the executives, these options can be cashed in only after a significant
number of years (generally 3 to 10 years) and only if the market price is higher than the
exercise price. Thus, executives have an incentive to manipulate the firm accounting
data to influence the stock price and to refrain from sending less positive information to
the market about the firm’s future performances (Povel, Singh and Winton (2007),
Yermack (1997) and Hu and Noe (2001)).
The main goal of the introduction of the SarbanesOxley Act was essentially to
reduce manipulative acts and fraudulent cases. Restricted stock can be a safer
compensation component than stock options because executives effectively receive
stock and not the possibility of buying stock in the future. In this way, they assume the
daily loss or gain if the stock price decreases or increases. Like restricted stocks, bonus
is also a comparatively safe component of executive compensation though not totally
free from possible manipulation of data by executives. On the other hand, salary is not a
compensation component related to firm performances. In other words, if the firm pays
9 Restricted stocks are stock subject to restrictions on sale and risk of forfeiture until vested by continued employment. Restricted stock typically vests in increments over a period of several years. Dividends or dividend equivalent rights may be paid, and award holders may have voting rights, during the restricted period.
56
more salary, this does not imply that executives will increase their efforts to have better
performances.
Results for the CEO and Director subsamples show that the CEOs receive more
compensation based on stock options but less salary and bonus than directors. Although
the importance of stock options diminished after the NASDAQ crash and the Sarbanes
Oxley Act, it is still higher in new economy firms than in old economy firms. Long
term Incentive Plans continue as a residual component of executive compensation in
both new and old economy firms.
4. Research Design
We now test whether or not the executive compensation in new versus old
economy firms is influenced generally by the same (or different) factors. If some of the
factors influence new and old economy executive compensation at the same time, we
expect that coefficient values will be different and this difference can be statistically
significant.
We used Unbalanced Panel Data and Fixed Effect Regression Model, also called
within estimator or Least Square Dummy Variable model.
4.1. Dependent Variables
The dependent variables are LN (Total Compensation) and LN (Short Term
Compensation) and LN (Option Ratio). LN (Total Compensation) is the total
remuneration gained by the executives and is the sum of salary, bonus, stock options,
restricted stocks, LTIP 10 , other annual compensation and all other compensation. This
variable, without logarithm, was used by Aggarwal and Samwick (1999) to evaluate the
contracts offered to executives in a context of strategic competition between products
and evaluation of relative performance, and by Fields and Fraser (1999) to unmask the
10 A Long Term Incentive Plan (LTIP) is any plan that provides compensation that intends to serve as an incentive for performance and that occurs over a period longer than one year, but not including restricted stock, stock option or stock appreciation rights plans.
57
commercial banks when they attributed compensations to link executives to
performances, and also by Chen and Hung (2006).
LN (Short Term Compensation) is the LN (salary+bonus). Salary and bonus are
considered shortterm remunerations, and they are usually received in money. We used
this variable like Stathopoulos, Espenlaub and Walker (2004) and Chen and Hung
(2006).
Finally we also used LN (Option Ratio). We define option ratio as the value of
options received by the executive divided by the total compensation and this variable
was also used by Chen and Hung (2006).
Each one of these dependent variables will be confronted separately with a group
of independent financial and governance variables with the intention of finding, in a
more trustworthy way, possible differences in executive compensation between new and
old economy firms.
Essentially, we analyse:
New Economy
Executive
Compensation
LN(Cash Compensation)
LN(Option Ratio)
LN(Total Compensation)
Old Economy
LN(Cash Compensation)
LN(Option Ratio)
LN(Total Compensation)
0 1
2
3 4
5 6
7 8
( ) *Firm Size Component * ( ) * ( ) ( ) ( ) Re 1
5 * ( ) (1993...2004)
LN Compensation LN Not Exercised Ratio LN Number Mtgs LN Tenure
LN Ownership Sharehold t Y GrowthNI Y LN BsVolatility
YearsDummy f
β β β β β β β β β
ε
= + +
+ + + + + + + +
+ + +
+ + +
The dependent variable LN (Compensation) can assume the values of LN (Total
Compensation), LN (Option Ratio) and LN (Short Term Compensation) and f is the
fixed effect.
58
4.2 Independent Variables
We use two sets of independent variables, financial and governance, as described below.
4.2.1. Financial Variables
Generally, the firm size in executive compensation literature is used as one of
the following variables: LN (Mktval), which is the natural logarithm of the market value
of the company, defined as the closing price for the fiscal year multiplied by the
company's common shares outstanding (millions of dollars) and was used by authors
such as Datta et al. (2005); LN (Sales), which is the natural logarithm of net annual
sales as reported by the company, and was used by authors such as Elston and Goldberg
(2003) and Aggarwall and Samwick (2003) and the LN (Assets), which is the natural
logarithm of the total assets as reported by the company that was used by Anderson and
Bizjack (2003), Elston and Goldberg (2003), Rogers (2002), Fenn and Liang (2001),
Chen (2004) and Yermack (2004). Some authors like Hallock (1997), Sridharan (1996),
Grinstein and Hribar (2004) and Bliss and Rosen (2001), also use these variables
without the natural logarithm. One of the problems in all these studies is that the authors
use one of these variables at the expense of the other variables. Researchers expect to
receive better results by using one variable and not the others in a specific situation. But
there is no sound reason for ignoring one variable and selecting another variable.
Because these variables are highly correlated, and cannot be introduced at the
same time to explain executive compensation, we used Principal Component Analysis
to extract a factor that contains information from the three variables and solves the old
problem of using size variable in executive compensation literature.
Table 4 describes the statistics of Principal Component Analysis.
59
The Principal Component Analysis methodology can be described, in this case,
as:
1 11 12 13
2 21 22 23
1 2 3
( ) ( ) ( )
( ) ( ) ( ) .........
( ) ( ) ( ) p p p p
y a LN Sales a LN Assets a LN Mktval
y a LN Sales a LN Assets a LN Mktval
y a LN Sales a LN Assets a LN Mktval
= + +
= + +
= + +
From Table 4 we can see that variables LN (Sales), LN (Assets), LN (Mktval)
are highly correlated.
Table 4 Statistics from Principal Component Analysis
Panel A: Correlation Matrix (a)
LN(Assets) LN(Sales) LN(Mktval)
LN(Assets) 1 0,92 0,83
LN(Sales) 0,92 1 0,75
LN(Mktval) 0,83 0,75 1
Panel B: Total Variance Explained
Component Initial Eigenvalues Extraction Sums of Squared Loadings
Total % of
Variance Cumulative
% Total % of
Variance Cumulative
% 1 2.669 88.982 88.982 2.669 88.982 88.982 2 0.266 8.878 97.860 3 0.064 2.140 100.000
To apply the Principal Components Analysis it is necessary to have a high
correlation between the variables. We use the KaiserMeyerOlkin (KMO) test, which
compares the correlation between the variables. From Table 4 we can see that variables
are highly correlated. We only find one factor with Initial Total Eigenvalues superior to
1 that explains 88.98% of the total variance and the vector is:
60
1 11 12 13 ( ) ( ) ( ) y a LN Sales a LN Assets a LN Mktval = + +
or
0,975* ( ) 0,945* ( ) 0,909* ( )
Firm Size Component LN Assets LN Sales LN Mktval
= + + +
We will use the natural logarithm of Firm Size Component to test the impact of
firm size on total compensation, option ratio, and short term compensation of new and
old economy firms.
We also used the variable LN(Not Exercised Ratio), which is the natural
logarithm of the number of unexercised options that the executive held at year end that
were vested divided by the aggregate number of stock options/stock appreciation rights
granted to explain executive compensation. We expect that the number of options
vested but not exercised has a negative relationship with total compensation and options
ratio, meaning that if the executive has stock options that are not exercised, the firm will
probably give fewer stock options in the future. We expect that effect will be more
pronounced in new economy firms because many researchers such as Anderson, Banker
and Ravidran (2000), Ittner et al. (2003), Murphy (2003) and Stathopoulos et al. (2004)
show that new economy firms grant more stock options to the executives.
To analyse the relationship between the risk and executive compensation in new
and old economies, we use the variable LN(Bs Volatility), like Chen (2004) and Palia
(2001), (but without natural logarithm), which is the natural logarithm of the standard
deviation volatility calculated over 60 months with Black and Scholes´ methodology.
We expect a negative relationship between firm risk and cash compensation because if
the volatility is higher, the firm can reward the executives with stock options because
the value will increase with stock return volatility.
We also use the variable Share Ret 1Y, which is the oneyear total return to
shareholders, including the monthly reinvestment of dividends. We expect that the
return to shareholders will have a negative relationship with executive compensation
because if the executive has enough return on investments, it is not necessary to give
61
more compensation to executives to align shareholder interests with the interests of the
executive.
We also use the variable LN (Ownership), which is the natural logarithm of the
percentage of the company's shares owned by the named executive officer. This variable
was used, with or without natural logarithm, by authors like Core et al. (1999), Barron
and Waddell (2003) and Chen (2004). We expect, like most of the previous research,
that the percentage of the company shares owned by the executive will have a negative
relationship with executive compensation. According to Chen and Hung (2006), higher
ownership indicates that managers' interests are more aligned with shareholders. Higher
insider ownership is often considered to have a positive impact on corporate
governance, and the more the managerial ownership, the less incentive pay they usually
receive. This is also because if the executive has stock in the firm, he/she will already
be more involved and concerned about improving the firm’s stock price, and it is not
necessary to increase the incentives to reduce agency costs.
To measure the impact of firm growth, we use the variable Growth NI 5Y,
which is the 5year least square annual growth rate of Net Income. Because new
economy firms, according to the prior authors, have lower cash flows and give more
stock options to executives, we expect that the net income will not be an important
factor in explaining executive compensation in new economy firms. In the case of old
economy firms, we expect that this relationship will positively influence total
compensation.
To control for the effect of time, we use one dummy for each year between
1993 and 2004, like Barron and Waddel (2003) and Grinstein and Hribar (2004). We
expect that the dummy year will be significant in explaining executive compensation,
particularly in the bubble period of 1998 to 2000 and relative to the number of options
granted to new economy firm executives.
62
4.2.2. Governance Variables
LN (Tenure) is the natural logarithm of the number of years that the executive
has been doing the job as CEO. This variable was applied by a significant number of
authors to explain executive compensation such as Chindambaran and Prabhala (2003),
Ryan Jr and Wiggins III (2004), Murphy (1986), Barro and Barro (1990), Hallock
(1997) and Chen (2004) with or without the natural logarithm. We expect a positive
relationship between executive compensation and tenure because in the real world we
observe that more experienced executives command higher compensation.
The influence of the board and composition of the Compensation Committee on
executive compensation is one of the most recent fields of research in the area of
executive compensation. Ryan Jr and Wiggins III (2004) find that CEO compensation is
related to the power and the influence that the CEO has on the board. They find
evidence that firms with external directors on the board pay more compensation based
on stock options and restricted stocks. Anderson and Bizjak (2003) also analyse whether
board independence promotes the shareholders' interests and whether the presence of
the CEO on the Compensation Committee is related to opportunist behaviour. They did
not find evidence that when the executive leaves the compensation committee, the
remuneration decreases.
To analyse the relationship between board members and executive
compensation, we use the variable LN (Number Mtgs), similar to Davidson III et al.
(1998), which is the natural logarithm of the number of board meetings held during the
indicated fiscal year. The number of board meetings is related to the performance of the
firm, and the ability of the executive to make decisions is affected by the number of
meetings of the board during the year. According to Davidson et al. (1998), board
members are more aligned with shareholders´ interests when they have more meetings
during the year. Because of that, we expect a negative relationship between the number
of meetings and executive compensation.
63
Expected correlations
Independent Variables LN(Total Compensation) LN(Option Ratio) LN(Short Term
Compensation) Firm Size Component + + +
LN(Not Exercised Ratio) +
LN(Bs Volatility + +
Share Ret 1Y + +
LN(Ownership)
Growth NI 5Y + + +
Year Dummy + + +
In Table 12 we present the Pearson Correlation of independent variables. We
find no variables with high correlations in any of the cases.
64
5. Empirical Results
5.1. Summary Statistics Table 5
Summary Statistics
Firm Size Component is a vector that measures the firm size and is a combination of the LN(Assets), LN(Mktval) and LN(Sales) variables. Using these three variables and applying the Principal Component Analysis, we extract a vector that is the best combination of these three variables to analyse the influence of firm size. LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested divided by the aggregate number of stock options/stock appreciation rights granted. LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes’ method. LN (Number Mtgs) is the natural logarithm of the number of board meetings. LN(Dirmtgfee) is the natural logarithm of the value that the executive received for attending board meetings. LN (Tenure) is the natural logarithm of the number of years that the CEO has been doing the job. LN (Ownership) is the natural logarithm of the percentage of the company shares owned by the executive. Sharehold Ret 1Y is the oneyear return to shareholder. Sales is the value of the sales of the year. Sales3LS is the 3year least squares annual growth rate of Sales. Growth NI 1Y is the year firm net Income. Growth I 5Y is the 5year least squares annual growth rate of net income. EarnPerShare is the Earnings per Share (Primary) Excluding Extraordinary Items and Discontinued Operations. ROEAVG is the Net Income before Extraordinary Items and Discontinued Operations divided by the average of the most current year's Total Common Equity and the prior year's Total Common Equity. This quotient is then multiplied by 100. Divyeld is the Dividends per Share by ExDate divided by Close Price for the fiscal year. This quotient is then multiplied by 100. Emp. is the number of firm employees. Retyrs is the number of years of credited service the executive has under the company's pension plan and ROA is the Net Income before Extraordinary Items and Discontinued Operations divided by Total Assets. This quotient is then multiplied by 100. Values are in thousands of dollars.
NEW OLD
Dependent Variables N Min. Max. Mean Std. Dev. N Min. Max. Mean Std.
Dev. Firm Size Component 15638 4,96 33,16 19,03 4,61 50839 0,46 34,95 19,92 4,22
LN (Not Exercise Ratio) 10605 12,61 14,80 0,31 1,40 34406 10,63 8,63 0,67 1,16
LN (Bs Volatility) 14883 1,98 1,44 0,54 0,42 48561 2,16 1,42 0,98 0,43
LN (Number Mtgs) 14768 0,00 3,58 1,99 0,43 49493 0,00 3,66 1,87 0,37
LN (Dirmtgfee) 10320 3,18 1,95 0,45 0,43 40157 2,20 2,09 0,33 0,42
LN (Tenure) 4032 2,64 4,03 2,23 0,81 13177 5,90 4,09 2,24 0,85
LN (Ownership) 3053 6,21 5,63 0,32 1,71 8441 5,81 4,61 0,35 1,63
Sharehold Ret 1Y 15161 99,13 177042,86 146,32 4326,90 50359 97,08 531566,66 90,72 5813,84
Sales 15995 0,00 102635,40 3605,55 10589,55 51369 0,00 272941,48 4327,00 13952,08
Sales3LS 15794 87,18 19079,17 47,04 281,36 51154 86,57 5437,74 12,91 55,88
Growth NI 5Y 5937 48,13 1091,80 40,10 64,45 28337 67,99 1843,33 15,51 44,07
Growth NI 1Y 15995 38468,00 24728,00 131,30 1541,37 51369 56121,90 25330,00 179,51 1064,54
EarnPerShare 15964 91,05 1124,00 0,58 20,32 51257 69,83 80,83 1,19 2,93
ROEAVG 15541 16671,05 527,99 9,65 215,44 50016 8563,86 786,15 7,07 120,20
Divyield 15790 0,00 94,12 0,35 2,63 51154 0,00 2777,78 1,69 30,66
Empl 15722 0,01 329,00 11,21 31,90 51004 0,00 750,00 14,48 34,01
Executive Age 1786 31,00 92,00 52,97 8,43 7628 33,00 91,00 56,89 7,75
Retyrs 15164 0,00 46,80 1,52 6,14 49000 0,00 64,70 8,24 12,09
ROA 15995 587,97 1100,00 1,62 39,95 51364 1314,89 218,75 2,84 22,49
65
Ittner, Lambert and Larker (2003), Murphy (2003) and Stathopoulos, Espenlaub
and Walker (2004) are of the view that new economy companies differ from old
economy companies because they present higher growth taxes of sales increase; they
spend more money on research and development; they present low ratios of bookto
market value, reduced dividends per share and a high volatility of share returns. They
still hold a smaller number of employees, a more reduced market value and smaller
accounting returns than old economy firms. In addition, they provide a larger
compensation proportion based on capital ownership, have a higher percentage of stock
options and a higher percentage of the volume of stock options not exercised related to
the total number of company shares.
Our data also shows that new economy firms have smaller net incomes; the
earning per share and the return on equity are smaller; the new economy executives are
younger; the increase in net income and the yearly returns to shareholders are higher in
new economy firms, and new economy firms have more board meetings, and on
average, board members are paid more per board meeting.
5.2. Determinants of Executive Compensation in New and Old Economy Firms
We now examine whether determinants of executive compensation, in new and
old economy firms, are the same. We first test for correlations among independent
variables, as discussed above, and find the values are relatively low. Tables 6 and 7
present the results of LSDV (least square dummy variables) regressions for the CEOs
and directors of the new economy firms. Tables 8 and 9 present the LSDV regression
results for the CEOs and the directors of the old economy firms.
The regression uses three separate dependent variables: LN(total compensation),
LN(option ratio), and LN(short term compensation). Each of the dependent variables is
potentially explained by various independent variables as discussed earlier. We use
unbalanced panel data because some executives do not necessarily stay with the same
firm throughout our sample period. Standard errors are corrected using period
Seemingly Unrelated Regression (SUR) Panel Corrected Standard Errors (PCSE) which
corrects for both period heteroskedasticity and general correlation of observations
within a given cross section (Beck and Katz, 1995).
66
Table 6 Fixed Effect Regression: Least Square Dummy Variables CEOs
New economy Executives We used Unbalanced Panel Data Fixed Effect Regression Analysis. The dependent variables are LN (Total Compensation), LN(Short Term Compensation) and LN(Option Ratio). Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. LN (Total Compensation) is the natural logarithm of total executive compensation. LN(Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a vector that measures the firm size and is a combination of the LN (Assets), LN (Market Value) and LN(Sales) variables. Using these three variables and applying the Principal Component Analysis, we extract a vector that is the best combination of these three variables to analyse the influence of firm size; LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested divided by the aggregate number of stock options/stock appreciation rights granted; Growth NI 5Y is the 5year least squares annual growth rate of net income; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes’ method; LN (Number Mtgs) is the natural logarithm of the number of board meetings; LN (Tenure) is the natural logarithm of the number of years that the CEO has been doing the job; LN (Ownership) is the natural logarithm of the percentage of the company shares owned by the executive. We also used a dummy year variable from 1993 to 2004 to control for the year effect. Shareholder Ret 1Y is the one year return to shareholders. To distinguish between executives from new and old economy firms, we used the methodology of Murphy (2003), which considers firms from the new economy those with SIC codes 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961, 7370, 7371, 7372 and 7373 and firms from the old economy those with SIC codes less than 4000 and not yet categorised with the new economy. Independent Variables
LN(Total Comp.)
t Statistics
LN(Option Ratio)
t Statistics
LN(Short Term Comp.)
t Statistics
Constant 3,9468 1,352 1,9562 1,1438 0,9214 0,3206 Firm Size Component 0,3028* 6,2359 0,1286* 4,184 0,0908*** 1,7004
LN(Not Exercised Ratio) 0,3179* 8,6617 0,1818* 6,8661 0,0481 1,2339
LN(Number Mtgs) 0,0275 0,2729 0,0005 0,0064 0,0185 0,1592 LN(Tenure) 0,5680 0,5656 0,1886 0,3275 1,3216 1,3628 LN(Ownership) 0,1131** 2,1567 0,0100 0,2450 0,0519 0,7276 Share Ret 1Y 0,0006 1,4682 0,0008* 3,4916 0,0001 0,1211 Growth NI 5Y 0,0014 1,4230 0,0003 0,4749 0,0005 0,5452 LN(Bs Volatility) 0,0912 0,2688 0,1023 0,4549 0,7729*** 1,9145 Year 1993 0,0358 0,0978 0,1256 0,8451 0,0926 0,2508 Year 1994 0,2765 0,7954 0,1278 0,8449 0,0647 0,1823 Year 1995 0,0831 0,2269 0,3816** 2,2137 0,0149 0,0392 Year 1996 0,1009 0,2691 0,1279 0,6924 0,2067 0,5412 Year 1997 0,2297 0,6046 0,3068*** 1,7328 0,1039 0,266 Year 1998 0,0992 0,2625 0,3574*** 2,0384 0,0537 0,1337 Year 1999 0,2952 0,7644 0,3330*** 1,9847 0,1910 0,4538 Year 2000 0,4027 0,9869 0,4372*** 2,3732 0,0340 0,0781 Year 2001 0,4264 1,0235 0,3557*** 1,803 0,3018 0,587 Year 2002 0,4445 1,0685 0,3512*** 1,8624 0,1653 0,3581 Year 2003 0,3654 0,8643 0,3711*** 1,8912 0,2043 0,4406 Year 2004 0,3820 0,8871 0,4691*** 2,2527 0,2061 0,4280 Number Obs. 309 309 309 Adjusted RSquare 86.12% 68.19% 64.22% * Significant at 1% level, ** significant at 5% level *** significant at 10%
67
Table 7 Fixed Effect Regression: Least Square Dummy Variables Directors
New Economy Executives We used Unbalanced Panel Data Fixed Effect Regression Analysis. The dependent variables are LN (Total Compensation), LN(Short Term Compensation) and LN(Option Ratio). Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. LN (Total Compensation) is the natural logarithm of total executive compensation. LN(Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a vector that measures the firm size and is a combination of the LN (Assets), LN (Market value) and LN(Sales) variables. Using these three variables and applying the Principal Component Analysis, we extract a vector that is the best combination of these three variables to analyse the influence of firm size; LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; Growth NI 5Y is the 5year least squares annual growth rate of net income; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes’ method; LN (Number Mtgs) is the natural logarithm of the number of board meetings; LN (Tenure) is the natural logarithm of the number of years that the CEO has been doing the Job; LN (Ownership) is the natural logarithm of the percentage of the company shares owned by the executive. We also used a dummy year variable from 1993 to 2004 to control for the year effect. Shareholder Ret 1Y is the one year return to shareholders. To distinguish between executives from new and old economy firms, we used the methodology of Murphy (2003), which considers firms from the new economy those with SIC codes 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961, 7370, 7371, 7372 and 7373 and firms from the old economy those with SIC codes less than 4000 and not yet categorised with the new economy. Independent Variables
LN(Total Comp.)
t Statistics
LN(Option Ratio)
t Statistics
LN(Short Term Comp.)
t Statistics
Constant 4,7355 1,7959 2,2375 0,8747 0,2003 0,0795 Firm Size Component 0,2702* 6,7345 0,1238* 4,4973 0,1027** 2,4695
LN(Not Exercised Ratio) 0,2989* 10,0186 0,1775* 8,0077 0,0219 0,7446
LN(Number Mtgs) 0,0144 0,1524 0,0327 0,4932 0,0229 0,2291 LN(Tenure) 0,6834 0,732 0,1200 0,129 1,5270*** 1,7208 LN(Ownership) 0,1243** 2,4721 0,0026 0,0713 0,0465 0,7461 Share Ret 1Y 0,0005 1,3772 0,0010* 4,5157 0,0002 0,5579 Growth NI 5Y 0,0003 0,5012 0,0003 0,5971 0,0007 1,181 LN(Bs Volatility) 0,0125 0,0433 0,0382 0,205 0,5566*** 1,8738 Year 1993 0,0467 0,3124 0,0064 0,0594 0,1191 0,853 Year 1994 0,3595** 2,4498 0,0504 0,4878 0,0110 0,0809 Year 1995 0,2164 1,3484 0,1958*** 1,6966 0,0199 0,128 Year 1996 0,2414 1,4126 0,0541 0,4663 0,0624 0,3895 Year 1997 0,3609** 2,0062 0,1736 1,4341 0,0124 0,0722 Year 1998 0,2468 1,2597 0,2085 1,6096 0,0371 0,1915 Year 1999 0,3369 1,5179 0,1443 1,0652 0,2088 0,8975 Year 2000 0,4746*** 1,8966 0,2836*** 1,8632 0,0666 0,271 Year 2001 0,6066** 2,2624 0,1381 0,8059 0,1668 0,4926 Year 2002 0,5463** 2,0517 0,1724 1,0773 0,2111 0,7711 Year 2003 0,4771*** 1,7628 0,1864 1,112 0,2647 0,9578 Year 2004 0,4653*** 1,6794 0,3029*** 1,690 0,2772 0,9443 Number Obs. 393 393 393 Adjusted RSquare 83.78% 65.34% 66.42% * Significant at 1% level, ** significant at 5% level *** significant at 10%
68
Table 8 Fixed Effect Regression: Least Square Dummy Variables – CEOs
Old economy Executives We used Unbalanced Panel Data Fixed Effect Regression Analysis. The dependent variables are LN (Total Compensation), LN(Short Term Compensation) and LN(Option Ratio). Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. LN (Total Compensation) is the natural logarithm of total executive compensation. LN(Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a vector that measures the firm size and is a combination of the LN (Assets), LN (Market value) and LN(Sales) variables. Using these three variables and applying the Principal Component Analysis, we extract a vector that is the best combination of these three variables to analyse the influence of firm size; LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested divided by the aggregate number of stock options/stock appreciation rights granted; Growth NI 5Y is the 5year least squares annual growth rate of net income; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes’ method; LN (Number Mtgs) is the natural logarithm of the number of the board meetings; LN (Tenure) is the natural logarithm of the number of years that the CEO has been doing the job; LN (Ownership) is the natural logarithm of the percentage of the company shares owned by the executive. We also used a dummy year variable from 1993 to 2004 to control for the year effect. Shareholder Ret 1Y is the oneyear return to shareholders. To distinguish between executives from new and old economy firms, we used the methodology of Murphy (2003), which considers firms from the new economy those with SIC codes 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961, 7370, 7371, 7372 and 7373 and firms from the old economy those with SIC codes less than 4000 and not yet categorised with the new economy. Independent Variables
LN(Total Comp.)
t Statistics
LN(Option Ratio)
t Statistics
LN(Short Term Comp.)
t Statistics
Constant 5,0932* 6,5267 1,3329 0,9941 6,1012* 6,8681 Firm Size Component 0,2008* 11,0886 0,0929* 3,7742 0,1242* 7,6246
LN(Not Exercised Ratio) 0,1985 16,6525 0,3203* 19,4289 0,0022 0,2013
LN(Number Mtgs) 0,0042 0,085 0,1280** 1,9981 0,054 1,186 LN(Tenure) 0,5738 1,4006 0,2986 0,5463 0,7091*** 1,9516 LN(Ownership) 0,0487** 2,1481 0,0979 3,3076 0,0399*** 1,9564 Share Ret 1Y 0,0006* 2,7898 0,0015* 5,1449 0,0011* 4,8791 Growth NI 5Y 0,0015* 2,6522 0,0008 1,1308 0,0007 1,3028 LN(Bs Volatility) 0,0353 0,3513 0,4405* 3,3646 0,1995 2,1821 Year 1993 0,1884** 1,9655 0,0583 0,3795 0,0763 1,0477 Year 1994 0,3493* 3,5811 0,2739*** 1,7161 0,0969 1,2526 Year 1995 0,2558* 2,5925 0,1096 0,6842 0,0404 0,5021 Year 1996 0,4075* 4,025 0,2054 1,2783 0,0787 0,9663 Year 1997 0,5131* 4,9632 0,3097*** 1,9068 0,0914 1,1104 Year 1998 0,5163* 4,8442 0,3003*** 1,8338 0,0799 0,9421 Year 1999 0,5751* 5,4467 0,3127*** 1,8855 0,1401 1,5819 Year 2000 0,6167* 5,6181 0,2266 1,3306 0,1587*** 1,7721 Year 2001 0,6158* 5,4672 0,3798** 2,1975 0,0844 0,9022 Year 2002 0,6959* 6,1528 0,2490 1,4237 0,2428* 2,601 Year 2003 0,6182* 5,3628 0,3054*** 1,7425 0,1938** 2,0428 Year 2004 0,7819* 6,5267 0,2729 1,5073 0,3220* 3,2358 Number Obs. 1363 1363 1363 Adjusted RSquare 89.34% 76.52% 86.72% * Significant at 1% level, ** significant at 5% level *** significant at 10%
69
Table 9 Fixed Effect Regression: Least Square Dummy Variables Directors
Old economy Executives We used Unbalanced Panel Data Fixed Effect Regression Analysis. The dependent variables are LN (Total Compensation), LN(Short Term Compensation) and LN(Option Ratio). Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. LN (Total Compensation) is the natural logarithm of total executive compensation. LN(Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a vector that measures the firm size and is a combination of the LN (Assets), LN (Market value) and LN(Sales) variables. Using these three variables and applying the Principal Component Analysis, we extract a vector that is the best combination of these three variables to analyze the influence of firm size; LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested divided by the aggregate number of stock options/stock appreciation rights granted; Growth NI 5Y is the 5year least squares annual growth rate of net income; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes’ method; LN (Number Mtgs) is the natural logarithm of the number of the board meetings; LN (Tenure) is the natural logarithm of the number of years that CEO has been doing the job; LN (Ownership) is the natural logarithm of the percentage of the company shares owned by the executive. We also used a dummy year variable from 1993 to 2004 to control for the year effect. Shareholder Ret 1Y is the oneyear return to shareholders. To distinguish between executives from new and old economy firms, we used the methodology of Murphy (2003), which considers firms from the new economy those with SIC codes 3570, 3571, 3572, 3576, 3577, 3661, 3674, 4812, 4813, 5045, 5961, 7370, 7371, 7372 and 7373 and firms from the old economy those with SIC codes less than 4000 and not yet categorised with the new economy. Independent Variables
LN(Total Comp.)
t Statistics
LN(Option Ratio)
t Statistics
LN(Short Term Comp.)
t Statistics
Constant 5,1720* 4,3219 1,0845 0,7971 6,3836* 4,0755 Firm Size Component 0,2042* 12,5828 0,1007* 4,9939 0,1467* 7,0728
LN(Not Exercised Ratio) 0,2198 19,9261 0,3454* 25,0717 0,0318* 2,4885
LN(Number Mtgs) 0,0062 0,1344 0,1681* 2,9918 0,1077** 2,0766 LN(Tenure) 0,5212 1,0532 0,4784 0,8453 0,9224 1,4328 LN(Ownership) 0,0578** 2,5457 0,0988* 3,6193 0,0574** 2,2395 Share Ret 1Y 0,0005** 2,3707 0,0015* 6,2385 0,0009* 4,1489 Growth NI 5Y 0,0013* 2,6242 0,00004 0,0742 0,0005 0,9295 LN(Bs Volatility) 0,0854 0,9394 0,3290* 3,0302 0,2454** 2,2947 Year 1993 0,0543 0,8618 0,1461*** 1,7231 0,0765 0,8606 Year 1994 0,0703 1,0810 0,2922* 3,2781 0,0076 0,0917 Year 1995 0,0274 0,4200 0,1584 1,8167 0,0821 0,9385 Year 1996 0,1123** 1,6494 0,2498* 2,8461 0,0865 0,9097 Year 1997 0,1826** 2,4383 0,2995* 3,2076 0,0967 0,9593 Year 1998 0,1516** 1,9714 0,2959* 3,2041 0,1550 1,4683 Year 1999 0,2248* 2,9956 0,3124* 3,2663 0,0736 0,6915 Year 2000 0,2780* 3,4906 0,2285** 2,2541 0,0909 0,8108 Year 2001 0,2570* 3,0847 0,3965* 3,795 0,1705 1,4396 Year 2002 0,3684* 4,406 0,2748* 2,5956 0,0022 0,0184 Year 2003 0,3039 3,5053 0,3013* 2,8026 0,0670 0,5512 Year 2004 0,4535* 4,7947 0,2938* 2,5243 0,0106 0,0831 Number Obs. 1711 1711 1711 Adjusted RSquare 87.42% 75.86% 76.79% * Significant at 1% level, ** significant at 5% level *** significant at 10% Note: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
70
Our results reveal that there are significant differences in factors that explain
executive compensation in new versus old economy subsamples, and generally these
differences are statistically significant based on tests of equality of regression
coefficients (Table 10 and 11 appendix). We find that the factors that explain CEO
compensation in new economy firms also explain Director compensation in new
economy firms but with smaller intensity. We observe the same phenomena in the old
economy firms. We also find that the impact of the common factor (size) to explain
executive compensation is significantly different for old versus new economy firms.
As expected, the firm's size is one of the most important variables in explaining
executive compensation. In the case of the CEO subsample, the size variable has a
stronger impact on the executive compensation in new economy firms than in old
economy firms. The difference in impact of the size variable on executive compensation
is statistically significant for new versus old economy firms. The results suggest that as
the firm size increases, new economy executives receive more than executives from the
old economy.
The firm’s size also influences the number of options granted to the executives in
new and old economy firms, but the relationship is stronger in new economy firms. The
results are congruent with the findings of Ittner et al. (2003), Murphy (2003) and
Stathopoulos et al. (2004), Anderson, Banker, and Ravidran (2000) that new economy
firms grant more stock options to executives.
Our results also reveal that cash compensation is more sensitive to firm size in
old economy firms than new economy firms. That is, large companies from the old
economy pay more in cash than new economy firms.
In the case of the directors, the relationship between firm size and compensation
is similar to the relationship for the CEOs, but the intensity of the coefficients is
smaller, meaning that the impact of the size variable on executive compensation is
smaller than in the CEO subsample.
The number of vested stock options that the executives have, but are not
exercised, only affects, in negative terms, the total compensation of the CEOs in new
economy firms. This means that when CEOs can not exercise the options that are
vested, the firm does not increase their incentives. The results also reveal that for both
new and old economy firms when executives have stock options that are vested but not
71
exercised, then firms prefer not to give more stock options. This relationship is more
pronounced in the case of the old economy, meaning that new economy firms are more
averse to granting stock options to executives when there are already unexercised
options. In the case of the directors, the relationship is the same as the CEOs, with the
exception of directors from old economy firms, who are compensated with cash when
the options can not be exercised.
The number of board meetings is negative and statistically significant only in the
case of CEOs and Directors from the old economy. The results are congruent with the
findings of Ryan and Wiggins (2001), and Chen and Hung (2006), who say that more
monitoring power can reduce the need to provide CEOs with incentive compensation.
The total compensation decreases for old economy executives if the board members
attend all the meetings, meaning that, as indicated by Davidson, Pilger and Szakmary
(1998), board members are more aligned with shareholders´ interests when they have
more meetings during the year, and therefore the CEO compensation is more controlled.
The age of the executive (Tenure) is only positive and statistically significant
with total compensation in the case of CEOs from new economy firms and also in the
case of cash compensation from directors. As we expected, more experience is
associated with more remuneration in new economy firms. The results of old economy
firms are not statistically significant, meaning that executive experience does not
influence old economy executive compensation. We also found that CEO tenure is
negatively related to cash compensation, meaning that experienced executives probably
receive compensation in forms other than cash.
Contrary to our expectations, and the results of Chen and Hung (2006), we find
that, with the exception of new economy CEOs, executive ownership has a positive, and
not negative, influence on the total compensation of the CEOs and the Directors of new
as well as old economy firms. The relationship is stronger, however, in the case of
CEOs and Directors of new economy firms. There is a negative association between
executive ownership and the number of options that executives receive. When the
executives have more stocks options from their company, their interests are already
aligned with the shareholders, and it is therefore not necessary to give more stock
options to reduce the agency problem. The same relationship exists in the case of
directors.
72
The oneyear return to shareholders is negatively related to the options granted to
CEO in both new and old economy firms, and also in the case of new economy
directors. In the case of old economy directors, it is negatively related to total
compensation, option ratio and short term compensation, and the results are congruent
with the theory that if shareholders are already satisfied with investment returns, they do
not need to pay more to executives because their interests are aligned with executives’
interests.
The increase in the firm’s Net Income in the last five years does not affect CEO
executive compensation in new economy firms. In the case of CEOs and directors from
the old economy, the influence is positive on total compensation. The results are
congruent with Ang et al. (2002) and Aggarwal and Samwick (2003), who find that
CEOs are paid more in terms of performance than other executives and with Ittner et al.
(2003), who find that new economy firms give stock options to executives because the
firms have difficulty generating enough cash flow to pay high salaries to executives.
As we expected, stock return volatility has a strong negative influence on CEO
cash compensation in old economy firms and a less strong influence in old economy
firms. Also in the case of the directors, the stock return volatility is negatively related to
cash compensation in new and old economy firms, and the relationship is stronger in
new economy firms. The results mean that if the volatility increases, firms will reward
their executives with more stock options and less cash as incentives.
We also find that time has a positive effect on total compensation and options
compensation.
73
6. Conclusion
We examine various questions regarding executive compensation of old versus
new economy firms: Are the determinants (and their intensity) of executive
compensation the same in new versus old economy firms? Are the forms of
compensation received by executives in the new and old economies, in terms of salary,
bonus, stock options, restricted stocks and LongTerm Incentive Plans (LTIP) the same
and did the form of compensation change after the 2000 crash of Nasdaq and the 2002
SarbanesOxley Act?
Our results reveal that the number of executives of new economy firms is
considerably smaller than the number of executives of old economy firms, and that most
of the new economy executives are from firms associated with Prepackaged Software (26.02%) Semicondutor and Related Devices (17.29%), Computer Programming, Data
Processing (9.46%) and Telecommunications (7.50%).
Our results also reveal that new economy executives receive more, on average,
than executives from the old economy, but the difference decreases in the last sample
years. In the bubble period, new economy executive compensation is composed of more
than 50% of stock options and in the case of old economy firms with more than 30% of
stock options. After that period, with the NASDAQ Crash and the introduction of the
SarbanesOxley Act we observe a significant change in the structure of the components
of executive compensation reducing the use of stock options and increasing the use of
bonus and restricted stocks.
We also find that the factors that explain executive compensation in new and old
economy firms are generally different, and in the case of the variables that are the same,
our tests generally reject the hypothesis of the equality of regression coefficients related
to these common factors. New economy total executive compensation is influenced by
firm size, the ratio of the number of stock options vested but not exercised, and
executive stock ownership. Old economy total executive compensation is influenced by
firm size, executive ownership, oneyear total return to shareholders and 5year annual
growth rate of firm net income.
74
7. References
Anderson, M., Banker, R., and Ravidran, S. (2000). Executive compensation in the
information technology industry. Management Science, Vol. 46(4), 530547
Anderson, R. and Bizjak, J. (2003). An empirical examination of the role of the CEO
and the Compensation Committee in structuring executive pay. Journal of Banking & Finance, Vol. 27(7), 13231348.
Ang, J., Lauterbach, B. and Schreiber, Z. (2002). Pay at the executive suite: how do US
banks compensate their top management teams? Journal of Banking & Finance, Vol. 26(6), 11431163.
Aggarwal, R. and Samwick, A. (1999). Executive compensation, strategic competition,
and relative performance evaluation: theory and evidence. The Journal of
Finance, Vol. 54(6), 19992043. Aggarwal, R. and Samwick, A, (2003). Performance incentives within firms: the effects
of managerial responsibility. The Journal of Finance, Vol. 58(4), 16131649. Barro, J., and Barro, R. (1990). Pay, performance and turnover of bank CEOs. Journal
of Labor Economics, Vol. 8 (4), 448481. Barron, M. and Waddel, G. (2003). Executive rank, pay and project selection. Journal
of Financial Economics, Vol. 67(2), 305349. Beck, N. and Katz, J. (1995). What to do (and not to do) with timeseries crosssection
data. American Political Science Review, Vol. 89(3), 663647.
Bliss, R. and Rosen, R. (2001). CEO compensation and bank mergers. Journal of Financial Economics, Vol. 61(1), 107138.
Chen, M. (2004). Executive option repricing, incentives and retention. The Journal of Finance, Vol. 59(3), 11671199.
Chen, Y. and Hung, C. (2006). Corporate governance and insider compensation:
NASDAQ technology firms. Working Paper presented at Financial Management
Association Conference in Salt Lake City.
Chidambaran, N. and Prabhala, N. (2003). Executive stock option repricing, internal
governance mechanisms and management turnover. Journal of Financial Economics, Vol. 69(1), 153189.
75
Core, J., Holthausen, R. and Larker, D. (1999). Corporate governance, Chief Executive
Officer compensation, and firm performance. Journal of Financial Economics, Vol. 51(3), 371406.
Davidson III, W., Pilger, T. and Szakmary, A. (1998). Golden Parachutes, Board and
Committee composition, and shareholder wealth. Financial Review, Vol.
33(4),1732. Datta, S, IskandarDatta, M. and Raman, K. (2005). Managerial stock ownership and
the maturity structure of corporate debt. The Journal of Finance, Vol. 60(5), 23332350.
Elston, J. and Goldberg, L. (2003). Executive compensation and agency costs in
Germany. Journal of Banking & Finance, Vol. 27(7), 13911410. Fenn, G. and Liang, N. (2001). Corporate payout policy and managerial stock
incentives. Journal of Financial Economics, Vol. 60, 4572.
Fields, L. and Fraser, D. (1999). On the compensation implications of commercial bank
entry into investment banking. Journal of Banking & Finance, Vol. 23, 1261 1276.
Grinstein, Y. and Hribar, P. (2004). CEO compensation and incentives: evidence from
M & A Bonuses. Journal of Financial Economics, Vol. 73(1), 119143. Hallock, K. (1997). Reciprocally interlocking boards of directors and executive
compensation. Journal of Financial and Quantitative Analysis, Vol. 32(3), 331 426.
Hu, J. and Noe, T. (2001). Insider trading and managerial incentives. Journal of
Banking & Finance, Vol. 25(4), 681716. Ittner, C., Lambert, R. and Larker, D. (2003). The structure and performance
consequences of equity grants to employees of new economy firms. Journal of Accounting and Economics, Vol. 34(1), 89127.
Murphy, K. (1986). Incentives, learning and compensation: a theoretical and empirical
investigation of managerial labour contracts. Rand Journal of Economics, Vol. 17(1), 5976.
Murphy, K. (2003). Stockbased pay in new economy firms. Journal of Accounting and
Economics, Vol. 34(1), 129147.
76
Narayanan, M. and Seyhun, H. 2005, Effect of SarbanesOxley Act on the influencing
of executive compensation. SSRN working paper nº 852964 Palia, D. (2001). The endogeneity of managerial compensation in firm valuation: A
solution. The Review of Financial Studies, Vol. 13(3), 735764. Povel, P., Singh, R. and Winton, A. (2007). Booms, busts, and fraud. The Review of
Financial Studies, Vol. 20(4), 12191255. Rogers, D. (2002). Does executive portfolio structure affect risk management? CEO
risktaking incentives and corporate derivatives usage. Journal of Banking & Finance, Vol. 26(2), 271295.
Ryan, H. and Wiggins, R. (2001). The influence of firm and managerspecific
characteristics on the structure of executive compensation. Journal of Corporate Finance, Vol. 7(3), 101123.
Ryan, H. and Wiggins, R. (2004). Who is in whose pocket? Director compensation,
Board independence, and barriers to effective monitoring. Journal of Financial Economics, Vol. 73(3), 497524.
Sridharan, U. (1996). CEO influence and executive compensation. The Financial Review, Vol. 31(1), 5166.
Stathopoulos, K., Espenlaub, S. and Walker, M. (2004). U.K. executive compensation
practices: new economy versus old economy. Journal of Management Accounting Research, Vol. 16(1), 5792.
Yermack, D. (1997). Good timing: CEO stock options award and company news
announcements. The Journal of Finance, Vol. 52 (2), 449476.
Yermack, D. (2004). Remuneration, retention, and reputation incentives for outside
Directors. The Journal of Finance, Vol. 59(5), 22812308.
77
8. Appendix
Table 10: T Test of Equality of Fixed Effect Regressions Coefficients CEOs
Panel A: LN (Total Compensation) New Economy Old Economy
Independent Variables N Coef. Std. Error N Coef. Std. Error
Sig.
(Constant) 309 3,9468 2,9192 1363 5,0932 1,0035 No
Firm Size Component 309 0,3028 0,0486 1363 0,2008 0,0181 *
LN(Not Exercised Ratio) 309 0,3179 0,0367 1363 0,1985 0,0119 *
LN (Number Mtgs) 309 0,0275 0,1006 1363 0,0042 0,0494 *
LN (Tenure) 309 0,5680 1,0042 1363 0,5738 0,4097 *
LN (Ownership) 309 0,1131 0,0525 1363 0,0487 0,0227 *
Share Ret 1Y 309 0,0006 0,0004 1363 0,0006 0,0002 *
Growth NI 5Y 309 0,0014 0,0010 1363 0,0015 0,0005 *
LN (Bs Volatility) 309 0,0912 0,3394 1363 0,0353 0,1006 *
Year1993 309 0,0358 0,3664 1363 0,1884 0,0958 *
Year1994 309 0,2765 0,3476 1363 0,3493 0,0975 *
Year1995 309 0,0831 0,3661 1363 0,2558 0,0987 *
Year1996 309 0,1009 0,3748 1363 0,4075 0,1012 *
Year1997 309 0,2297 0,3798 1363 0,5131 0,1034 *
Year1998 309 0,0992 0,3779 1363 0,5163 0,1066 *
Year1999 309 0,2952 0,3862 1363 0,5751 0,1056 *
Year2000 309 0,4027 0,4081 1363 0,6167 0,1098 *
Year2001 309 0,4264 0,4166 1363 0,6158 0,1126 *
Year2002 309 0,4445 0,4160 1363 0,6959 0,1131 *
Year2003 309 0,3654 0,4228 1363 0,6182 0,1153 *
Year2004 309 0,3820 0,4306 1363 0,7819 0,1198 No
* Significant at level 1%; ** Significant at level 5%; *** Significant at level 10% Note: In this table we analyse whether the coefficient values of the regression are statistically different in new and old economy firms, based on the t test. In most of the regression variables we found that these coefficients are statistically different at 1%(*).
78
Table 10 (Cont.)
PANEL B: LN (Option Ratio) New Economy Old Economy
Independent Variables N Coef. Std.
Error N Coef. Std. Error
Sig.
(Constant) 309 1,9562 1,7107 1363 1,3329 1,3409 *
Firm Size Component 309 0,1286 0,0307 1363 0,0929 0,0246 *
LN(Not Exercised Ratio) 309 0,1818 0,0265 1363 0,3203 0,0165 *
LN (Number Mtgs) 309 0,0005 0,0756 1363 0,1280 0,0640 *
LN (Tenure) 309 0,1886 0,5758 1363 0,2986 0,5466 *
LN (Ownership) 309 0,0100 0,0408 1363 0,0979 0,0296 *
Share Ret 1Y 309 0,0008 0,0002 1363 0,0015 0,0003 *
Growth NI 5Y 309 0,0003 0,0006 1363 0,0008 0,0007 *
LN (BS Volatility) 309 0,1023 0,2249 1363 0,4405 0,1309 *
YEAR1993 309 0,1256 0,1487 1363 0,0583 0,1536 *
YEAR1994 309 0,1278 0,1513 1363 0,2739 0,1596 *
YEAR1995 309 0,3816 0,1724 1363 0,1096 0,1602 *
YEAR1996 309 0,1279 0,1847 1363 0,2054 0,1607 *
YEAR1997 309 0,3068 0,1771 1363 0,3097 0,1624 *
YEAR1998 309 0,3574 0,1753 1363 0,3003 0,1637 *
YEAR1999 309 0,3330 0,1678 1363 0,3127 0,1659 *
YEAR2000 309 0,4372 0,1842 1363 0,2266 0,1703 *
YEAR2001 309 0,3557 0,1973 1363 0,3798 0,1728 *
YEAR2002 309 0,3512 0,1886 1363 0,2490 0,1749 *
YEAR2003 309 0,3711 0,1962 1363 0,3054 0,1753 *
YEAR2004 309 0,4691 0,2082 1363 0,2729 0,1810 *
* Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
Note: In this table we analyse whether the coefficient values of the regression are statistically different in new and old economy firms, based on the t test. In most of the regression variables we found that these coefficients are statistically different at 1%(*).
79
Table 10 (Cont.)
Panel c: LN (Short Term Compensation) New Economy Old Economy
Independent Variables N Coef. Std.
Error N Coef. Std. Error
Sig.
(Constant) 309 0.9214 2,8743 1363 6,1012 0,8883 *
Firm Size Component 309 0,0908 0,0534 1363 0,1242 0,0163 *
LN(Not Exercised Ratio) 309 0,0481 0,0390 1363 0,0022 0,0108 *
LN (Number Mtgs) 309 0,0185 0,1161 1363 0,0540 0,0455 *
LN (Tenure) 309 1,3216 0,9691 1363 0,7091 0,3633 *
LN (Ownership) 309 0,0519 0,0713 1363 0,0399 0,0204 *
Share Ret 1Y 309 0,0001 0,0005 1363 0,0011 0,0002 *
Growth NI 5Y 309 0,0005 0,0009 1363 0,0007 0,0005 *
LN (BS Volatility) 309 0,7729 0,4037 1363 0,1995 0,0914 *
YEAR1993 309 0,0926 0,3692 1363 0,0763 0,0729 *
YEAR1994 309 0,0647 0,3551 1363 0,0969 0,0773 *
YEAR1995 309 0,0149 0,3805 1363 0,0404 0,0805 *
YEAR1996 309 0,2067 0,3819 1363 0,0787 0,0815 *
YEAR1997 309 0,1039 0,3906 1363 0,0914 0,0823 *
YEAR1998 309 0,0537 0,4012 1363 0,0799 0,0848 *
YEAR1999 309 0,1910 0,4210 1363 0,1401 0,0886 *
YEAR2000 309 0,0340 0,4356 1363 0,1587 0,0895 *
YEAR2001 309 0,3018 0,5142 1363 0,0844 0,0936 *
YEAR2002 309 0,1653 0,4615 1363 0,2428 0,0934 NO
YEAR2003 309 0,2043 0,4637 1363 0,1938 0,0949 *
YEAR2004 309 0,2061 0,4815 1363 0,3220 0,0995 * * Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
Note: In this table we analyse whether the coefficient values of the regression are statistically different in new and old economy firms, based on the t test. In most of the regression variables we found that these coefficients are statistically different at 1%(*).
80
Table 11: T Test of Equality of Fixed Effect Regressions Coefficients – Directors
Panel A: LN (Total Compensation) New Economy Old Economy
Independent Variables N Coef. Std.
Error N Coef. Std.
Error
Sig.
(Constant) 393 4,7355 2,6369 1711 5,1720 1,1967 No
Firm Size Component 393 0,2702 0,0401 1711 0,2042 0,0162 *
LN(Not Exercised Ratio) 393 0,2989 0,0293 1711 0,2198 0,0110 *
LN (Number Mtgs) 393 0,0144 0,0946 1711 0,0062 0,0465 *
LN (Tenure) 393 0,6834 0,9336 1711 0,5212 0,4949 *
LN (Ownership) 393 0,1243 0,0503 1711 0,0578 0,0227 *
Share Ret 1Y 393 0,0005 0,0004 1711 0,0005 0,0002 *
Growth NI 5Y 393 0,0003 0,0007 1711 0,0013 0,0005 *
LN (Bs Volatility) 393 0,0125 0,2879 1711 0,0854 0,0909 *
YEAR1993 393 0,0467 0,1495 1711 0,0543 0,0630 *
YEAR1994 393 0,3595 0,1468 1711 0,0703 0,0650 *
YEAR1995 393 0,2164 0,1605 1711 0,0274 0,0653 *
YEAR1996 393 0,2414 0,1709 1711 0,1123 0,0681 *
YEAR1997 393 0,3609 0,1799 1711 0,1826 0,0749 *
YEAR1998 393 0,2468 0,1960 1711 0,1516 0,0769 *
YEAR1999 393 0,3369 0,2219 1711 0,2248 0,0751 *
YEAR2000 393 0,4746 0,2502 1711 0,2780 0,0796 *
YEAR2001 393 0,6066 0,2681 1711 0,2570 0,0833 *
YEAR2002 393 0,5463 0,2663 1711 0,3684 0,0836 *
YEAR2003 393 0,4771 0,2707 1711 0,3039 0,0867 *
YEAR2004 393 0,4653 0,2770 1711 0,4535 0,0946 No
* Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
81
Table 11 (Cont.)
Panel B: LN (Option Ratio) New Economy Old Economy Independent
Variables N Coef. Std. Error N Coef. Std.
Error
Sig.
(Constant) 393 2,2375 2,5580 1711 1,0845 1,3606 *
Firm Size Component 393 0,1238 0,0275 1711 0,1007 0,0202 *
LN(Not Exercised Ratio) 393 0,1775 0,0222 1711 0,3454 0,0138 *
LN (Number Mtgs) 393 0,0327 0,0663 1711 0,1681 0,0562 *
LN (Tenure) 393 0,1200 0,9305 1711 0,4784 0,5660 *
LN (Ownership) 393 0,0026 0,0365 1711 0,0988 0,0273 *
Share Ret 1Y 393 0,0010 0,0002 1711 0,0015 0,0002 *
Growth NI 5Y 393 0,0003 0,0006 1711 0,00004 0,0006 *
LN (Bs Volatility) 393 0,0382 0,1865 1711 0,3290 0,1086 *
YEAR1993 393 0,0064 0,1079 1711 0,1461 0,0848 *
YEAR1994 393 0,0504 0,1032 1711 0,2922 0,0891 *
YEAR1995 393 0,1958 0,1154 1711 0,1584 0,0872 *
YEAR1996 393 0,0541 0,1160 1711 0,2498 0,0878 *
YEAR1997 393 0,1736 0,1211 1711 0,2995 0,0934 *
YEAR1998 393 0,2085 0,1295 1711 0,2959 0,0924 *
YEAR1999 393 0,1443 0,1354 1711 0,3124 0,0956 *
YEAR2000 393 0,2836 0,1522 1711 0,2285 0,1014 *
YEAR2001 393 0,1381 0,1714 1711 0,3965 0,1045 *
YEAR2002 393 0,1724 0,1600 1711 0,2748 0,1059 *
YEAR2003 393 0,1864 0,1676 1711 0,3013 0,1075 *
YEAR2004 393 0,3029 0,1792 1711 0,2938 0,1164 *
* Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
82
Table 11 (Cont.) Panel C: LN (Short Term Compensation)
New Economy Old Economy Independent Variables
N Coef. Std. Error N Coef. Std.
Error
Sig.
(Constant) 393 0.2003 2,5203 1711 6,3836 1,5663 *
Firm Size Component 393 0,1027 0,0416 1711 0,1467 0,0207 *
LN(Not Exercised Ratio) 393 0,0219 0,0294 1711 0,0318 0,0128 *
LN (Number Mtgs) 393 0,0229 0,1002 1711 0,1077 0,0519 *
LN (Tenure) 393 1,5270 0,8874 1711 0,9224 0,6438 *
LN (Ownership) 393 0,0465 0,0623 1711 0,0574 0,0256 *
Share Ret 1Y 393 0,0002 0,0004 1711 0,0009 0,0002 *
Growth NI 5Y 393 0,0007 0,0006 1711 0,0005 0,0006 *
LN (Bs Volatility) 393 0,5566 0,2970 1711 0,2454 0,1069 *
YEAR1993 393 0,1191 0,1396 1711 0,0765 0,0889 *
YEAR1994 393 0,0110 0,1358 1711 0,0076 0,0834 *
YEAR1995 393 0,0199 0,1557 1711 0,0821 0,0875 *
YEAR1996 393 0,0624 0,1603 1711 0,0865 0,0951 *
YEAR1997 393 0,0124 0,1712 1711 0,0967 0,1008 *
YEAR1998 393 0,0371 0,1935 1711 0,1550 0,1056 *
YEAR1999 393 0,2088 0,2327 1711 0,0736 0,1064 *
YEAR2000 393 0,0666 0,2457 1711 0,0909 0,1121 *
YEAR2001 393 0,1668 0,3387 1711 0,1705 0,1184 NO
YEAR2002 393 0,2111 0,2738 1711 0,0022 0,1190 *
YEAR2003 393 0,2647 0,2763 1711 0,0670 0,1215 *
YEAR2004 393 0,2772 0,2936 1711 0,0106 0,1281 *
* Significant at level 1%; ** Significant at level 5%; *** Significant at level 10% Note: In this table we analyse whether the coefficient values of the regression are statistically different in new and old economy firms, based on the t test. In most of the regression variables we found that these coefficients are statistically different at 1%(*).
83
Table 12: Pearson Correlation of Independent Variables
Panel A: New Economy 1 2 3 4 5 6 7 8
1Firm Size Component 1,000 0,189 0,121 0,065 0,012 0,069 0,101 0,100
2LN(Not Exercised Ratio) 0,189 1,000 0,081 0,041 0,182 0,086 0,160 0,035
3LN ( Number Mtgs) 0,121 0,081 1,000 0,030 0,151 0,095 0,057 0,097
4LN (Tenure) 0,065 0,041 0,030 1,000 0,347 0,076 0,044 0,144
5LN (Ownership) 0,012 0,182 0,151 0,347 1,000 0,103 0,014 0,130
6Share Ret 1Y 0,069 0,086 0,095 0,076 0,103 1,000 0,102 0,028
7Growth NI 5Y 0,101 0,160 0,057 0,044 0,014 0,102 1,000 0,071
8LN (Bs Volatility) 0,100 0,035 0,097 0,144 0,130 0,028 0,071 1,000
Panel B: Old Economy 1 2 3 4 5 6 7 8
1Firm Size Component 1,000 0,050 0,085 0,046 0,064 0,008 0,044 0,354
2LN(Not Exercised Ratio) 0,050 1,000 0,049 0,033 0,122 0,072 0,077 0,035
3LN ( Number Mtgs) 0,085 0,049 1,000 0,052 0,143 0,026 0,047 0,047
4LN (Tenure) 0,046 0,033 0,052 1,000 0,264 0,044 0,034 0,102
5LN (Ownership) 0,064 0,122 0,143 0,264 1,000 0,049 0,081 0,030
6Share Ret 1Y 0,008 0,072 0,026 0,044 0,049 1,000 0,116 0,120
7Growth NI 5Y 0,044 0,077 0,047 0,034 0,081 0,116 1,000 0,189
8LN (Bs Volatility) 0,354 0,035 0,047 0,102 0,030 0,120 0,189 1,000
84
CHAPTER 3
Executive Compensation: NYSE and NASDAQ Listed Firms 11
11 We are grateful to Prof. Sara Robicheaux, from BirminghamSouthern College, discussant of this paper at the Eastern Finance Association Annual Meeting in St Petersburg Florida – USA April 2008. We are also grateful to other anonymous professors present at this conference for their helpful comments.
85
1. Introduction
Since the late 80s and early 90s, the world has witnessed the demise of centrally
controlled socialist economies and the rise of the free market capitalist system.
Privatisation of state controlled enterprises, removal of barriers of international trade,
free flow of information, capital and labour, and advances of technology have indeed
created a small, integrated global village producing unparalleled global economic
growth. As a consequence of the growth of the global economy, we have also witnessed
a significant increase in the market value of companies all over the world. Global
growth, market integration, new opportunities, and increased corporate profitability
have intensified the search and competition for executive talent across the world. Firms
now compete for highly qualified executives globally hoping that their knowledge will
be instrumental in increasing the share value of the firms that they will manage.
In this changing global economic and corporate environment, firms, particularly
new economy firms, started paying their executives based on performance, essentially
with stock options. With stock options at their disposal, executives found an added
motivation to increase the value of the stock to raise their chances of exercising their
options later and hence increasing their wealth.
However, in the year 2000, the NASDAQ crash slowed this economic growth
and also some financial scandals came to the fore, along with the bankruptcy of
companies such as Enron and WorldCom, resulting from fraudulent accounting
practices and executive selfdealings. Management, in some widely publicised cases,
distorted the accounting data to manipulate the stock price in order to enhance personal
compensation by exercising options. In order to solve the problems associated with
these scandals, the SarbanesOxley Act of 2002 was created in the USA. It introduced
sweeping changes in governance, reporting, and disclosure requirements of public firms
with the intent of improving accuracy, reliability and timeliness of the information
provided to investors.
Interest in research in executive compensation is recent, but growing. In this
paper, we extend the executive compensation research to NYSE versus NASDAQ listed
firms. We focus our attention on the following questions: Are the total values and the
factors that explain executive compensation for NYSE versus NASDAQ listed firms the
86
same? Is the composition of compensation given to these executives different? What
happens to compensation composition and values after the NASDAQ crash in 2000 and
the SarbanesOxley Act in 2002? We analyse data from the years 1992 to 2004.
Our results reveal that NYSE executives receive, on average, more than the
other executives and the differences in total compensation values are statistically
significant. We also find that the forms of compensation for NYSE versus NASDAQ
listed firms are different. In other words, the percentage that each compensation
component represents in total compensation is different in these subsamples and this
structure changes in all the cases, essentially after the SarbanesOxley Act. Our results
also reveal that the factors that explain CEO and Director compensation on the NYSE
and NASDAQ are generally different, and when some factors are the same, the
coefficients are statistically different.
2. Literature Review and Research Questions
The literature review reveals there is only one study similar and related to our
topic. Firth, Lohen, Ropstad and Sjo (1996), focusing on Norwegian Stock Exchange
listed firms, explore the determinants of Chief Executive Officer (CEO) compensation.
They find a modest positive relationship between CEO compensation and the average
wage level of the company and a strong positive relationship between CEO
compensation and firm size.
Not directly related to the problem of executive compensation and listed firms,
but as a parallel investigation, there are a small number of recent studies associated with
firms listed on the NYSE, NASDAQ and AMEX that may bear important implications
for the findings and conclusions of our study. For example, Sapp and Yan (2000) find
that some of the small firms listed on NASDAQ are changing to AMEX because the
transaction costs are smaller at AMEX (NASDAQ as a competitive multidealer system
and AMEX, like NYSE, a monopoly specialist system) and when they change their
liquidity generally improves.
Also Chung, Ness and Ness (1999, 2001) and Bacidore and Lipson (2001), find
evidence that transaction costs on NASDAQ are higher than on the NYSE and due to
this, some NASDAQ firms are also changing to the NYSE to reduce costs. Bacidore
87
and Lipson (2001) investigate the effect of opening and closing procedures on the
NYSE and NASDAQ by examining firms that moved from the NASDAQ to the NYSE
and find evidence that opening trades on the NYSE are about 20 percent less costly than
on NASDAQ and these savings on the NYSE opening increase with the size of the
firms.
If NASDAQ and NYSE have different transaction cost structures and the firms
that are listed there have different characteristics, we expect that the factors that explain
executive compensation will also be different. NASDAQ listed firms are essentially
technological firms with low levels of cash flows (Murphy, 2003) and NASDAQ
transaction costs are also generally higher than those of the NYSE. These fundamental
differences can also affect what the company can pay its executives.
Another point that we analyse is associated with the impact of the NASDAQ
crash (2000) and the enactment of the SarbanesOxley Act (2002) on executive
compensation. So far, we know that the NASDAQ crash caused a reduction in
compensation values, but what happened to the components of compensation packages?
Do they still have the same proportion of salary, bonus, stock options, restricted stocks
or LTIP 12 in terms of total compensation as before the NASDAQ crash? And also was
there a significant change in forms and weights of compensation after the Sarbanes
Oxley Act in 2002?
Effectively, after the NASDAQ crash, there were a series of financial scandals
associated with the bankruptcy of some of the large American companies, based on
fraudulent accounting practices and executive selfdealing. The SarbanesOxley Act of
2002 was established precisely on July 30 th to solve this problem. It introduced
sweeping changes in governance, reporting, and disclosure requirements of public firms
with the intent of improving accuracy, reliability, and timeliness of the information
provided to investors. This Act contains provisions which have a significant impact on
the benefits and compensation of public company executives. The major changes in this
area include the following provisions: to prohibit publiclytraded companies from
making or arranging loans for their directors and executive officers; to expedite
12 Generally, executive compensation is composed of two more components: “other annual compensation” and “all other compensations”. The first case includes the types of compensation not included in salary and bonus and in the second case, all other forms of compensation not included in the other categories. We do not analyse these two forms of compensation because they are residual components and also because they include a large diversity of compensation products.
88
Securities and Exchange Commission (SEC) reporting to insider traders; to prohibit
corporate directors and executive officers from trading employers` securities during
planned blackout periods with respect to those securities and to require an employee
retirement Income Security Act to cover individual account plans to provide a 30day
notice of blackout periods.
After the SarbanesOxley Act, each of the major US stock markets, the NYSE,
the NASDAQ, and the AMEX adopted new listing standards in an effort to strengthen
the corporate governance practices of listed companies, associated with director
independence, audit committees, compensation committees, nominating committees,
stock option plans, certification, directors/officers and disclosure and foreign issuers.
If a group of new rules of corporate governance were adopted by the NYSE and
NASDAQ, we expect that the way in which companies pay their executives would
change after these important changes.
3. Data, Sample Selection and Statistics
3.1. Data and Sample Selection
Data is from the Execucomp database, which collects information on the five
highestpaid executives from 1500 firms listed on the S&P 500 Index, S&P Mid Cap
Index, and S&P Small Cap Index.
We use Unbalanced Panel Data Analysis. The sample consists of 73,683
observations of compensation, related to the 5 highestpaid executives from S&P 1500
firms between the years 1992 and 2004. This sample is built excluding all executives
whose sum of salary and bonus, and also total compensation, was equal to zero. We
include only longer period compensation (and delete the shorter period compensation)
of executives who receive more than one compensation in the same year. There are a
few instances where an executive switched jobs and received two different
compensations in the same year.
Using the Consumer Price Index (CPI), compiled by the Bureau of Labor
Statistics, with 1982 as a base of 100, we adjust the monetary variables to the price level
of the year 2004. To select between NYSE and NASDAQ firms we use the variable
89
EXCHANGE from Execucomp, which classifies NYS as NYSE firms, NAS as
NASDAQ listed firms. Based on Chen and Hung (2006) we make the differentiation
between CEO and Director compensation because, generally, the average compensation
value of CEOs is higher than that of Directors, and also the factors that explain their
compensation can also be different.
3.2. Statistics
In this section we examine the question of whether the executives from firms listed
on the NYSE and NASDAQ are paid differently in terms of total value and
compensation components and whether these items (total compensation value and
components) change after the NASDAQ crash in 2000 and after the introduction of the
SarbanesOxley Act (SO) in 2002. We present the basic statistics in two steps. In the
first step, we analyse the evolution of total compensation through all the observations of
executive compensation between 1992 and 2004 for all top five executives and then for
CEOs and Directors. In the second step, we analyse the percentage that each executive
compensation component represents, in terms of total compensation, year by year. In
this way, we can see the most important executive compensation components and
changes, if any, year by year. We also compare values and components of executive
compensation between 2000 and 2001 (before and after the NASDAQ crash) and 2002
and 2003 (before and after SO) to observe whether the NASDAQ crash and enactment
of the SO act had any impact on executive compensation.
Table 1 presents IndependentSamples Ttest to compare the means of executive
compensation components and Levene's test for equality of variances between the two
subsamples of NYSE and NASDAQ listed firms for the period of 1992 to 2004. The
null hypothesis is that population means are equal; the alternative hypothesis is that
means are different.
90
Table 1 Average Total Compensation between 1992 and 2004 Adjusted for Inflation
In this table we describe the total average compensation, between 1992 and 2004, first for all the Top Five Executives and then for CEOs and Directors. Data is from the ExecuComp database. Mean average and mean differences are in thousands of dollars.
Means tTest Year N Mean Average Mean Difference Significance
PANEL A: Top Five Executives
NYSE 54778 2661.39 1992 to 2004
NASDAQ 18668 2029.29 632.10 *
PANEL B: CEOs NYSE 9305 5781.90 1992 to 2004 NASDAQ 3085 4011.09
1770.82 *
PANEL C: Directors NYSE 19865 4446.18
1992 to 2004 NASDAQ 6524 3023.34
1422.74 *
(*) Mean difference is statistically significant at: (*) 1*% level, (**) 5% level, (***) 10% level.
From table 1, we can see that the average compensation of the five highest paid
NYSE executives is far higher than the average total compensation of the five highest
paid NASDAQ executives, and these differences are statistically significant. Both CEOs
and directors of the NYSE listed firms receive, on average, much more than the CEOs
and directors of NASDAQ listed firms. Mean differences in compensation values are
generally significant.
Table 2 presents the average total compensation of executives of NYSE, and
NASDAQ listed firms, and the IndependentSamples Ttest to compare the means of
executive compensation components and Levene's test for equality of variances between
the two subsamples of NYSE and NASDAQ listed firms, each year for the period of
1992 to 2004. Yearly analysis gives us a better comparison because single average
value based on thirteen yearly observations could be influenced by outlier years. The
null hypothesis will be that all population means are equal; the alternative hypothesis is
that at least one of the means is different.
91
Table 2
Year ly Inflation adjusted total compensation trends of NYSE and NASDAQ listed firms between 1992 and 2004
Our sample includes data from the five most well paid executives associated with firms listed on the S&P500, S&P Mid Cap and S&P Small Cap during the period from 1992 to 2004. All the data are from the ExecuComp database. Total compensation is the sum of salary, bonus, stock options, restricted stocks, longterm incentive plans (LTIP), other annual compensation and all other compensation. To differentiate between executives from NYSE and NASDAQ, we used the EXCHANGE variable from the ExecuComp database, which has the following codes: NYS for NYSE firms and NAS for NASDAQ firms. Mean average and mean differences are in thousands of dollars.
NYSE (1) NASDAQ (2) Mean Difference Year
N Mean N Mean (1) and (2)
1992 430 2.584,865 53 1.249,570 1.335,295 *
1993 2499 1.865,280 445 1.091,859 773,421 *
1994 3090 1.717,387 686 1.127,851 589,536 *
1995 3274 1.783,652 777 1.334,982 448,670 *
1996 3398 2.177,596 820 1.605,229 572,367 *
1997 3695 2.805,092 960 1.853,349 951,743 *
1998 3907 3.218,114 1195 2.174,102 1.044,012 **
1999 4211 3.317,494 1513 2.960,016 357,478
2000 4533 3.978,887 1715 3.352,551 626,336 **
2001 4526 3.535,659 1702 3.080,971 454,688 ***
2002 4631 3.079,657 1780 2.100,072 979,585 *
2003 4769 2.812,288 1949 1.764,695 1.047,593 *
2004 4909 3.088,7350 2062 2.001,634 1.087,101 *
(*) Mean difference is significant at: (*) 1% level, (**) 5% level and (***) 10% level.
From table 2 we can see that executives from the NYSE receive, on average, more
than executives from NASDAQ each year during the sample period.
In table 3 we describe the percentage that each executive compensation
component represents in terms of total compensation for NYSE and NASDAQ firms
year by year between 1992 and 2004. In the row for year 2001 and 2003 we describe
whether the differences of the values from year 2001 and 2000 (NASDAQ crash effect)
and 2003 and 2002 (SarbanesOxley Act Effect) are statistically significant.
92
Table 3
Year ly Percentages of Each Compensation Component of NYSE and NASDAQ Listed Firms (19922004)
This table presents the percentages that each compensation component represents in terms of total compensation by year for the Top Five executives, CEOs and Directors. Salary is the executive salary for the year. Bonus is the dollar value of bonus (cash and noncash) earned by the executive officer during the fiscal years. Stock options are the aggregate value of stock options granted to the executive during the fiscal year as valued using S & P BlackScholes methodology. Restricted stocks are the value of restricted stock granted during the year (determined as of the date of the grant). LTIP is the amount paid out to the executive under the company’s longterm incentive plan. We also report in year 2001 and 2003 rows whether the changes between year 2000 and 2001 (NASDAQ crash) and between 2002 and 2003 (Sarbanes Oxley Act) are statistically significant. Panel A: Top Five (% of Total Compensation)
Salary Bonus Stock Options Restricted Stocks LTIP Year
NYS NAS NYS NAS NYS NAS NYS NAS NYS NAS
1992 48.69 50.36 19.42 17.76 18.31 24.81 4.58 2.13 4.04 1.19
1993 46.51 48.81 21.41 20.25 17.78 24.64 4.61 1.28 3.92 1.10
1994 43.86 47.34 21.81 20.21 21.31 25.37 4.22 2.19 3.82 0.86
1995 42.95 47.75 22.70 19.46 19.54 25.56 4.71 1.72 4.44 0.09
1996 39.95 44.01 22.38 18.30 23.26 29.82 4.95 2.32 4.38 0.97
1997 36.83 42.47 21.90 17.78 25.83 32.98 5.54 2.07 4.60 0.70
1998 36.07 41.28 19.82 16.04 23.26 29.82 5.68 1.65 4.22 0.50
1999 34.10 37.81 20.02 15.87 32.31 40.67 4.86 1.78 3.59 0.30
2000 32.46 35.78 19.80 16.41 33.25 42.17 5.60 1.98 3.56 0.30
2001 33.63* 36.46* 17.20* 11.87* 35.47* 45.32* 6.03* 2.21* 2.52* 0.40*
2002 33.87 38.74 19.59 14.30 31.08 40.63 7.14 2.34 2.83 0.40
2003 34.65 39.05 20.92* 16.70* 25.32* 35.99* 9.74* 3.55* 3.47* 0.80*
2004 30.48 35.63 23.59 18.15 23.96 36.46 13.32 5.07 4.02 0.71
93
Table 3 (cont.)
Panel B: CEOs (% of Total Compensation)
Salary Bonus Stock Options Restricted Stocks LTIP Year
NYS NAS NYS NAS NYS NAS NYS NAS NYS NAS
1992 36.39 55.29 22.78 12.54 22.89 21.83 6.41 0.68 6.42 5.95 1993 37.64 49.22 23.31 19.42 23.07 17.80 4.62 4.06 5.01 4.45 1994 33.85 43.80 23.22 21.16 28.44 23.10 4.51 4.39 5.14 2.99 1995 31.82 42.46 24.21 23.06 26.76 21.42 5.75 4.99 5.91 3.67 1996 28.59 38.82 22.52 24.13 31.89 23.88 5.96 3.90 5.91 3.99 1997 24.16 36.09 21.93 22.33 36.40 26.05 6.70 5.16 6.04 4.30 1998 23.24 35.02 20.22 20.12 40.27 32.35 5.90 4.49 5.36 3.39 1999 20.33 31.81 19.35 20.57 46.09 35.90 4.94 3.76 4.31 3.10 2000 18.43 30.16 17.93 20.35 47.24 36.92 7.28 4.23 4.19 2.48 2001 18.63 32.12 14.86* 15.40* 51.80** 39.99 7.09 5.52* 3.38* 1.57* 2002 20.03 28.90 17.15 18.97 45.98 38.48 7.90 6.67 4.44 1.78 2003 18.83 30.18 21.68* 21.02** 36.93* 31.08* 11.88* 9.16* 5.56* 2.61* 2004 16.39 27.09 22.99 24.09 36.52 29.73 14.32 11.04 4.79 3.64 Panel C: Directors (% of Total Compensation)
Salary Bonus Stock Options Restricted Stocks LTIP Year
NYS NAS NYS NAS NYS NAS NYS NAS NY NAS
1992 38.98 52.87 21.41 16.58 22.86 16.58 5.35 3.72 5.12 2.52 1993 38.23 49.52 23.22 16.58 22.86 16.58 5.34 3.99 4.63 3.56 1994 35.62 45.44 23.27 21.24 25.98 21.24 5.15 4.34 5.15 2.56 1995 33.59 44.98 23.29 19.59 25.58 19.59 5.91 4.52 5.61 3.08 1996 30.53 40.94 22.41 21.01 30.47 21.01 5.94 4.15 5.47 2.93 1997 26.17 38.95 21.86 23.67 34.62 23.67 6.74 4.76 5.35 2.83 1998 25.40 37.89 20.41 30.41 37.29 30.41 5.82 4.36 5.47 2.59 1999 22.99 34.70 20.23 33.93 42.37 33.93 5.38 3.25 3.97 2.72 2000 20.75 33.15 18.63 34.23 44.06 34.23 6.90 4.31 4.35 1.72 2001 20.73 35.49 16.15* 35.95* 48.53* 35.95 6.47** 5.26* 3.09* 1.35* 2002 21.29 31.57 18.80 36.40 42.08 36.40 8.15 5.96 4.41 1.58 2003 21.10* 33.55* 21.80* 29.74* 35.05* 29.74* 11.56* 7.97* 4.64* 1.86* 2004 18.79 28.84 23.03 28.10 34.25 28.19 14.08 10.56 4.50 2.95 Significant at: (*) 1% level, (**) 5% level and (***) 10% level
94
From table 3 we see that salary, in all the cases, is the most important executive
compensation component of the top five executives until 1998. Between 1999 and 2001,
stock options become the most important component of executive compensation for
NYSE and NASDAQ listed firms.
Examining the impact of the NASDAQ crash on the compensation of the top
five executives, we find that NASDAQ listed firms exhibit a decrease in the value that
bonus represents in total compensation from 16.41% in 2000 to 11.87% in 2001, and
this decrease is compensated with a small increase in terms of stock options (42.17% in
2000 and 45.32% in 2001) and restricted stocks (1.98% in 2000 to 2.31% in 2001).
NYSE listed firms maintain practically the same composition of compensation for the
top five executive subsample.
To evaluate the impact of the SarbanesOxley Act in 2002 on executive
compensation, we compare the results for 2002 to 2003 and observe that NYSE and
NASDAQ executives are paid with fewer stock options but with more restricted stocks
and bonuses. We also see that, in the case of the NYSE and NASDAQ, the importance
of restricted stocks in total compensation increases in the later years.
Results for panel B show that for NYSE listed firms, salary is the most
important compensation component between 1992 and 1995 and after that, stock
options are the most important component of CEO compensation. Essentially in the
later years, bonus also becomes an important component of CEO compensation.
In the case of NASDAQ listed firms, the most important compensation
component is also salary, but after 1996 stock options have practically achieved a
similar level of importance.
If we analyse the impact of the NASDAQ crash on CEO compensation,
comparing the change in percentage that each compensation component represents in
terms of total compensation between the year 2000 and 2001, we see that NYSE
executives receive more stock options and less bonus, and NASDAQ executives receive
more stock options, more salary and less bonus.
In terms of the impact of the SarbanesOxley Act, comparing results for 2002
and 2003, we see that NYSE executives receive fewer stock options and more bonus
and restricted stocks. NASDAQ executives also receive fewer stock options and more
bonus, salary and essentially restricted stocks.
95
If we compare the compensation components between CEOs and Directors, we
see that directors receive more salary than CEOs. NASDAQ listed firms reveal that
bonus is a more important component of compensation for Directors than for CEOs but
in the case of the NYSE this difference is small. In all the cases, CEOs receive more
stock options than Directors. The use of restricted stock becomes an important
component after the SarbanesOxley Act in 2002 for NYSE and NASDAQ listed firms.
4. Research Design
We first examine the determinants of executive compensation. We believe that if
firms listed on the NYSE are significantly different from firms listed on NASDAQ, the
factors that explain the executive compensation in these two groups may also be
different. In this section we make this analysis.
We used Unbalanced Panel Data analysis and the Fixed Effect Regression
Model, also called within estimator or the Least of Square Dummy Variable (LSDV).
4.1. Dependent Variables
The dependent variables are LN (Total Compensation) and LN (Short Term
Compensation) and LN (Option Ratio).
The dependent variable LN (Total Compensation) is the natural logarithm of the
sum of salary, bonus, stock options, restricted stocks, LTIP, other annual compensation
and all other compensation, LN (Short term Compensation) is the natural logarithm of
the sum of salary and bonus. The other dependent variable is the option ratio LN (option
ratio), which is the natural logarithm of stock options granted to the executive divided
by the total compensation.
Based on Fama and French (1997) we control for industry effect inserting the 48
industry classification dummies. We also control for time effect inserting one dummy
for each year between 1993 and 2004. We expect that time will have a positive effect on
explaining executive compensation during the analysed period.
96
Based on Ang, Lauterbach and Scheiber (2002) 13 and Cheng and Hung (2006),
we also separate the analysis for CEOs and Directors because these two groups can
have different characteristics in terms of compensation.
The model that we use is:
0 1
2
3
4 5
6 7 8
9
( ) * F irm S iz e C om p o n e n t * ( ) * ( ) * ( ) * ( ) * 1 * 5 * * * (1 9 9 3 ..2 0 0 4 ) *
LN C om p en s a t io n LN N o t E xe r c is e d R a t io LN BS Vo la t i l i ty LN N umm tg s LN Ten u r e T r s t r Sa le s ls P d ir p e n s In te r lo c k Ye a r s D umm ys In d
β β β β β β β β β β β β
= + +
+ +
+ +
+ + +
+ + + +
+ + +
+ u s t r y D umm ys f ε + +
LN (Compensation) can assume the values of LN (total compensation), LN (Option ratio) or
LN (Cash compensation) and f is the fixed effect.
4.2. Independent Variables
To explain the factors that affect the executive compensation of the NYSE and the
NASDAQ listed firms we use a group of financial and governance variables.
4.2.1. Financial Variables
Firm size has been reported as one of the most important variables in explaining
executive compensation. To measure the impact of firm size on executive
compensation, researchers generally use the variables Assets, Market Value or firm
Sales with our without natural logarithm. But which of these variables is best for
measuring the impact of firm size on executive compensation? There is no empirical
answer to this question. Researchers use only one of these variables, at the expense of
other variables, to capture the size effect which they believe will produce the results
most consistent with their research design. Each of the size variables has an impact on
executive compensation but these variables are highly correlated, and cannot be
13 This analysis was made with 166 American Banks between 1993 and 1996.
97
introduced at the same time to explain executive compensation in the regression model.
We therefore introduce a new technique by using the Principal Component Analysis
to extract a factor that contains information from the three variables and resolve this old
problem in executive compensation research.
The factor that measures firm size will be composed of the following variables:
Y1 = a11LN (Sales) +a12 LN (Assets) +a13LN (Market Value)
Essentially, Principal Component Analysis solves the problem of a number of
variables that are highly correlated and cannot be introduced at the same time in a
model. In this way, from table 4 we can see that variables LN (Sales), LN (Assets), LN
(Market Value) are highly correlated and from KaiserMeyerOlkin and Bartlett`s test
that this correlation is higher and statistically significant.
Table 4 Statistics from Principal Component Analysis
Panel A: Corr elation Matr ix (a) LN (Market Value) LN (Assets) LN (Sales)
LN (Market Value) 1 0.820 0.796 LN (Assets) 0.820 1 0.845 LN (Sales) 0.796 0.845 1 Panel B: Total Variance Explained
Component Initial Eigenvalues Extraction Sums of Squared
Loadings
Total % of
Variance Cumulative
% Total % of
Variance Cumulative
% 1 2.640 88.010 88.010 2.640 88.010 88.010 2 0.208 6.936 94.946 3 0.152 5.054 100.000
Panel C: KMO and Bartlett' s Test KaiserMeyerOlkin Measure of Sampling Adequacy. 0,763* Significant at 1% level
From table 4 we can see that there is only one factor with Initial Total
Eigenvalues superior to 1 that explains 88.01% of the total variance and the vector is:
Firm Size Component = 0.929* LN (Market Value) +0.947* LN (Assets) + +0.938* LN (Sales)
98
We expect that firm size component will have a positive relationship with all
three executive compensation dependent variables. Authors like Noguera and Highfield
(2007) also report that because larger firms have more complex operations and will be
more difficult to monitor, the use of incentivebased compensation is therefore practised
more in large firms than in small firms. We thus expect a stronger relationship between
dependent variable and firm size in the case of NYSE listed firms, which are essentially
large firms.
We also use the variable LN (Not Exercised ratio), which is the natural logarithm
of the number of unexercised options that the executive holds at the end of the year that
were vested, divided by the aggregate number of stock options/stock appreciation rights
granted. We expect a negative relationship between this ratio and total executive
compensation and the volume of stock options granted, but a positive relationship with
short term cash compensation. If executives cannot exercise their options, the company
will probably have to give additional compensation, essentially in cash, to increase their
motivation. We also expect that this relationship will be higher in NASDAQ listed firms
because, as (Murphy (2003), Anderson, Banker and Ravidran (1998), and Stathopoulos,
Espenlaub and Walker (2004)) note, new economy executives are compensated more
with stock options. They will lose most of their compensation if the exercise price
remains below the market price. So to protect them against this risk the executives are
rewarded with more stock options.
To analyse the relationship between the firm risk and the executives’ total
compensation, option ratio and short term compensation, we also use the variable LN
(BS Volatility), which is the natural logarithm of the standard deviation volatility
calculated over 60 months with Black and Scholes methodology. We expect a positive
relationship between the two dependent variables (total compensation and option ratio)
with firm risk and a negative relationship with short term compensation. If the volatility
is high, the firm’s stock price will also be high and companies will probably prefer to
compensate their executives with more stock options. In this way, firms will probably
reduce compensation based on cash compensation (short term compensation) and
increase compensation based on stock options. We also expect that the relationship
between stock return volatility and option ratio will be higher for NASDAQ listed firms
99
because, as discussed above, executives from new technology firms are more
compensated with stock options.
Trs1yr, oneyear total return to shareholders including the monthly reinvestment
of dividends, is also used in our investigation to analyse the impact of shareholders’
return on executive compensation. If shareholders receive a high return on their
investments in the company, they do not need to give more stock options to executives
to align executives’ interests with the owners’ interest to reduce the agency cost. Based
on this, we expect a negative relationship between option ratio, total compensation and
the oneyear shareholders’ return and a small positive relationship with cash
compensation in the sense that companies will probably give some money to
compensate executives’ efforts, but they do not need to give more stock options to
motivate them.
We also test the effect of firm growth on executive compensation. The variable
that we use to test this effect is the 5year least squares annual growth rate of sales
(Sale5ls). We expect, like Ryan and Wiggins (2001), that executives will receive higher
incentive pay in firms with higher growth opportunities.
In this investigation, we also control, like Barron and Waddel (2003) and
Grinstein and Hribar (2004), for the time effect on executive compensation in the sense
that some compensation changes can be associated merely with time effect. To control
for time we create a dummy variable for each year between 1993 and 2004 assuming
the value of 1 if the compensation is from the year and 0 if not. We believe that time
will have a strong effect on explaining executive compensation in all the cases. In boom
years we expect higher compensation and in recessionary years we expect lower
compensation. Moreover, during technology boom years we expect an increase in
compensation.
To control for industry effect on executive compensation, and based on Fama
and French (1997) industry classifications we also create, for each industry, a dummy
that assumes the value of 1 when the executive is associated with a specific industry
sector and 0 when the executive is not associated with that specific sector. We also
believe that some industries pay their executives more than others therefore creating
industry specific effects.
100
4.2.2. Governance Variables
In the past few years we have witnessed a significant number of studies that
have analysed the relationship between board activity, board composition and executive
compensation. Authors like Ryan Jr and Wiggins III (2004) find that CEO
compensation is related to the power and influence that s/he has on the board, and firms
with external directors on the board pay more compensation based on stock options and
restricted stocks. Anderson and Bizjak (2003) also analyse whether board independence
promotes shareholders´ interests and whether the presence of the CEO on the
Compensation Committee is related to opportunistic behaviour.
To analyse this relationship between board activity and executive compensation
we use the variable LN (Number Mtgs), which is the natural logarithm of the number of
board meetings held during the indicated fiscal year, and the dummy variable Interlock,
which assumes the value one when it is “true” that the executive serves on another
board and zero if not.
We expect the number of board meetings to be negatively related to executive
compensation because more control reduces the ability to increase compensation and it
aligns the interests of shareholders and executives. Davidson III, Pilger and Szakmary
(1998) and Ryan and Wiggins (2001) also conclude that more monitoring power can
reduce the need to provide CEOs with incentive compensation.
Like Core, Holthausen and Larker (1999) and Hallock (1997), we expect that if
executives are interlocked, they can influence their personal compensation in positive
terms.
The number of years that an executive is CEO (LN (Tenure)) has also been
documented as an important variable in explaining executive compensation. Authors
like Chidambaram and Prabhala (2003), Ryan Jr and Wiggins III (2004), Murphy
(1986) and Barro and Barro (1990) use this variable. Like Ryan and Wiggins, (2001),
Conyon and He (2004) and Kang et al. (2004), we expect that CEO entrenchment due to
tenure will lead to higher cash compensation and lower incentive compensation (stock
options).
Finally we use the dummy variable Pdirpens, which assumes the value equal to 1
when it is "TRUE" that the company pays a pension/retirement plan to a director. We
101
expect that if the company has a director pension plan, it may have the ability to pay
less to its executives.
Expected correlations
Dependent Variables Independent Variables LN (Total
Compensation) LN(Option ratio) LN(Short term Compensation)
Firm Size Component + + + LN (Not Exercised Ratio) + LN(Bs Volatility) + + LN(Number Mtgs) LN (Tenure) + + Trs1yr + Sales5ls + + + Pdirpens Interlock + + + Year Dummy + + +
102
5. Empirical Results
5.1. Summary Statistics
In table 5 we describe the statistics of key financial and corporate governance
variables that can help us to understand the differences between companies listed on the
NYSE and NASDAQ.
Table 5 Statistics from Regression Variables
The table displays some statistics from firms that belong to NYSE and NASDAQ listed firms. Firm Size Component is a factor extracted from Principal Component Analysis, composed of information from variables LN (Assets), LN (Market Value) and LN (Sales), which are used to analyse firm size. LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN (BS Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings held during the indicated fiscal year. LN (Tenure) is the natural logarithm of the number of years as CEO. Trs1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends. Sale5ls is the 5year least squares annual growth rate of sales. Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not. Interlock is a dummy that assumes the value of 1 when the named officer is involved in a relationship requiring disclosure in the "Compensation Committee Interlocks and Insider Participation" section of the proxy and 0 when not. Values are in thousands of dollars.
NYSE NASDAQ Independent variables N Min. Max. Mea
n Std. Dev. N Min. Max. Mean Std.
Dev. Firm Size Component 9384 10.94 35.97 22.75 4.06 2815 10.24 31.86 18.96 3.70
LN (Not Exercised Ratio) 9384 7.13 14.80 0.82 1.20 2815 10.10 8.92 0.81 1.34
LN(Bs Volatility) 9384 2.17 0.89 1.14 0.38 2815 1.90 1.26 0.70 0.44
LN(Number Mtgs) 9384 0.00 3.61 1.93 0.36 2815 0 3.09 1.83 0.39
LN (Tenure) 9384 5.90 4.06 2.09 0.90 2815 2.05 3.78 2.18 0.91
Trs1tr 9384 95.32 890.39 18.74 44.25 2815 94.01 24828 41.56 475.82
Sales5ls 9384 37.67 251.13 11.94 16.56 2815 36.35 903.16 26.83 37.12
Pdirpens 9384 0 1 0.20 0.40 2815 0 1 0,03 0.17
Interlock 9384 0 1 0,05 0.22 2815 0 1 0,05 0.22
We can see that, NYSE firms are, on average, bigger than NASDAQ firms. The
average number of executive stock options vested (but not exercised) and also the
103
average numbers of executives interlocked are practically the same in both situations.
NASDAQ listed firms generate a higher return to shareholders and sale increases than
NYSE firms. NYSE firms pay more to executive pension plans than NASDAQ listed
firms.
5.2. Determinants of Executive Compensation for NYSE and NASDAQ Listed
Firms
In this section we test the hypotheses that there are differences in forms and
determinants of executive compensation for firms listed on the NYSE and NASDAQ.
As we see from the table above, these two groups of companies have different
characteristics; therefore, we believe that the factors that explain executive
compensation in these two groups can also be different.
To determine what factors influence executive total compensation, option ratio
and short term compensation (the three dependent variables), we use Unbalanced Panel
Data and Least Squares Dummy Variable Regression (LSDV). In all the regressions,
Standard errors are corrected using period Seemingly Unrelated Regression (SUR) –
Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity
and general correlation of observations within a given cross section (Beck and Katz,
1995).
We also check whether correlation between independent variables is significant
based on the Pearson Correlation test and find that the correlation between the variables
is small.
As a first step we investigate whether the listing place (exchange) and the job
title of the executive influences the compensation. If listing exchange (NASDAQ versus
NYSE firms) and job title (CEO versus Director) are significant variables then we will
have an indepth analysis of executive compensation for NYSE versus NASDAQ listed
firms and also for CEOs versus Directors. We run a fixed effect regression on each of
the three dependent variables using all the explanatory variables stated above, including
year and industry dummy variables, and on top of that, we add two dummy variables
“NASDAQ Dummy” (1 if the executive is from a NASDAQ listed firm, otherwise zero)
and “Pexecdirp” (1 if the executive is a Director, otherwise zero). Regression results
104
show that both variables are significant. Based on the above finding, we proceed with
our analysis and run separate fixed effect regressions on NYSE CEOs, NASDAQ
CEOs, NYSE Directors, and NASDAQ Directors in order to identify the determinants
of executive compensation. These regressions will show us whether the factors that
explain the compensation of CEOs and Directors are the same for NYSE versus
NASDAQ listed firms. Results are presented in tables 6 through 10. We further
extended our analysis in tables 11 and 12 (shown in appendix) and perform the tests of
equality of coefficients for regressions on NASDAQ CEOs versus NYSE CEOs and
NASDAQ Directors versus NYSE Directors. That is, we examine whether the factor
intensity on executive compensation is the same for NASDAQ versus NYSE sub
samples for CEOs and Directors.
105
Table 6 Fixed Effect Regression Analysis of Compensation Determinants
Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a factor extracted from Principal Component Analysis, composed of information from variables LN (Assets), LN (Market Value) and LN (Sales), which are used to analyse firm size. LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted. LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method. LN (Number Mtgs) is the natural logarithm of the number of board meetings held during the indicated fiscal year. LN (Tenure) is the natural logarithm of the number of years as CEO. Trs1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends. Sales5ls is the 5year least squares annual growth rate of sales. Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not. Interlock is a dummy that assumes the value of 1 when the executive is on two different boards at the same time and 0 when not. Nasdaq is a dummy that assumes the value of 1 when the executive is in a firm listed on NASDAQ and 0 when not. Pexecdir is a dummy that assumes the value of 1 when the executive is a director and 0 when not. We also control for time effect in terms of executive compensation using a dummy for each year between 1993 and 2004 and for industry effect using the Fama and French (1997) 48 industry classifications.
Ln(Total Compensation) T Statist ics LN(Option
Ratio) T
Statistics LN(Short Term Compensation)
T Statistics
Constant 1,429* 3,312 2,72* 6,263 2,890* 6,522 Firm Size Component 0,172* 30,587 0,032* 5,513 0,105* 19,971 LN (Not Exercised Ratio) 0,240* 51,573 0,302* 62,794 0,002 0,496 LN(Bs Volatility) 0,184* 5,510 0,333* 9,725 0,123* 3,933 LN(Number Mtgs) 0,008 0,446 0,029 1,578 0,105* 6,251 LN (Tenure) 1,039* 5,316 0,504** 2,557 0,689* 3,412 Trs1yr 0,000001 0,042 0,0001* 3,604 0,0001* 4,527 Sales5ls 0,002* 4,747 0,001* 2,579 0,0004 1,481 Pdirpens 0,164* 7,087 0,047** 1,965 0,067* 3,072 Interlock 0,098* 3,146 0,094* 2,935 0,038 1,326 Year1993 0,065* 2,635 0,031 1,027 0,043* 2,210 Year1994 0,276* 10,337 0,231* 7,783 0,140* 6,679 Year1995 0,289* 10,409 0,161* 5,266 0,129* 5,699 Year1996 0,428* 14,787 0,342* 11,064 0,167* 7,088 Year1997 0,532* 17,728 0,346* 10,806 0,208* 8,277 Year1998 0,586* 19,268 0,427* 13,378 0,265* 10,206 Year1999 0,728* 22,971 0,528* 16,135 0,308* 11,154 Year2000 0,808* 23,548 0,458* 12,816 0,368* 12,385 Year2001 0,876* 25,256 0,603* 16,753 0,281* 8,888 Year2002 0,949* 27,137 0,579* 15,826 0,436* 13,778 Year2003 0,921* 25,044 0,513* 13,385 0,479* 14,535 Year2004 1,105* 28,990 0,573* 14,398 0,586* 17,049
NASDAQ Dummy 0,435 1,599 0,372 1,328 0,775* 3,052 Pexecdir Dummy 0,410* 1,925 0,049* 2,244 0,370* 18,632 Apparel Dummy 0,562 1,311 0,335 0,761 0,870** 2,092 Business Dummy 0,988** 2,200 0,016 0,035 0,844*** 1,891 Candy Dummy 0,304 0,612 0,462 0,790 0,408 1,075 Computer Dummy 1,070** 2,031 2,237* 4,103 0,547 1,060 Construction Dummy 0,333 1,160 0,317 1,053 0,287 1,047 Consumer Dummy 1,163** 2,242 0,511 0,927 0,738*** 1,722 Medical Dummy 0,578 1,596 0,474 1,173 0,466 1,477 Trading Dummy 0,650 1,409 0,192 0,417 0,913*** 1,890 Wholesales Dummy 2,467* 4,920 1,579* 3,081 0,840 1,635
N 12225 12225 12225 Adjusted Rsquare 84.00% 68.05% 76.30% *Significant at 1% level, ** significant at 5% level, *** significant at 10%. Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995).
106
Table 7 Fixed Effect Regression Analysis of the Determinants of CEO Compensation for NYSE
Listed Firms
Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a factor extracted from Principal Component Analysis, composed of information from variables LN (Assets), LN (Market Value) and LN (Sales), which are used to analyse firm size; LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings held during the indicated fiscal year; LN (Tenure) is the natural logarithm of the number of years as CEO; Trs1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; Sales5ls is the 5year least squares annual growth rate of sales; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy that assumes the value of 1 when the executive is on two different boards at the same time and 0 when not; Nasdaq is a dummy that assumes the value of 1 when the executive is in a firm listed on NASDAQ and 0 when not; Pexecdir is a dummy that assumes the value of 1 when the executive is a director and 0 when not. We also control for time effect in terms of executive compensation using a dummy for each year between 1993 and 2004 and for industry effect using the Fama and French (1997) 48 industry classifications.
LN(Total Compensation)
t Statistics
LN(Option Ratio)
t Statistics
LN(Short Term Compensation)
t Statistics
Constant 0.346 0.524 2.099 2.897 4.578* 7.225 Firm Size Component 0.165* 19.770 0.010 1.087 0.124* 18.165 LN (Not Exercised Ratio) 0.220* 35.360 0.344* 47.105 0.002 0.325 LN(Bs Volatility) 0.190* 4.374 0.188* 3.733 0.123* 3.508 LN(Number Mtgs) 0.002 0.097 0.037 1.363 0.071* 3.461 LN (Tenure) 1.633* 6.058 0.379 1.289 0.147 0.559 Trs1yr 0.00004 0.293 0.002* 10.267 0.001* 13.136 Sales5ls 0.001*** 1.921 0.003* 3.897 0.0002 0.277 Pdirpens 0.103* 3.633 0.079** 2.455 0.042*** 1.881 Interlock 0.083** 2.254 0.142* 3.274 0.001 0.022 Year1993 0.117* 2.937 0.002 0.039 0.058** 2.045 Year1994 0.351* 8.539 0.202* 3.367 0.152* 5.137 Year1995 0.362* 8.664 0.148** 2.426 0.091* 2.936 Year1996 0.512* 11.766 0.293* 4.766 0.136* 4.215 Year1997 0.609* 13.430 0.316* 5.011 0.158* 4.738 Year1998 0.663* 14.548 0.411* 6.441 0.239* 6.963 Year1999 0.800* 17.210 0.543* 8.436 0.298* 8.361 Year2000 0.827* 16.864 0.496* 7.337 0.308* 8.151 Year2001 0.899* 17.939 0.628* 9.236 0.245* 6.068 Year2002 0.989* 19.466 0.611* 8.874 0.399* 9.899 Year2003 0.987* 18.803 0.600* 8.452 0.368* 8.988 Year2004 1.151* 21.427 0.611* 8.434 0.496* 11.773 Apparel Dummy 0,168 0,446 0,008
0,255 0,431 0.589*** 1.916
Business Dummy 1.547 3.443 0,255 0,413
0,515 0,186 0,471 Candy Dummy 0,570 1,420 0,413 0,615 0.541*** 1.912 Computer Dummy 2.719* 4.417 1.711** 2.492 1.206** 2.142 Construct Dummy 1.380* 3.889 0,541 1,332 0,310 1,002 Medical Dummy 0,061 0,170 0.270 0,601 0.479*** 1.684 N 6124 6124 6124 Adjusted RSquare 86.35% 68.36% 81.77% *Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
107
Table 8 Fixed Effect Regression Analysis of the Determinants of CEO Compensation for
NASDAQ listed firms Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a factor extracted from Principal Component Analysis, composed of information from variables LN (Assets), LN (Market Value) and LN (Sales), which are used to analyse firm size; LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings held during the indicated fiscal year; LN (Tenure) is the natural logarithm of the number of years as CEO; Trs1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; Sales5ls is the 5year least squares annual growth rate of sales; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy that assumes the value of 1 when the executive is on two different boards at the same time and 0 when not; Nasdaq is a dummy that assumes the value of 1 when the executive is in a firm listed on NASDAQ and 0 when not; Pexecdir is a dummy that assumes the value of 1 when the executive is a director and 0 when not. We also control for time effect in terms of executive compensation using a dummy for each year between 1993 and 2004 and for industry effect using the Fama and French (1997) 48 industry classifications.
LN(Total Compensation)
t Statistics
LN(Option Ratio) t Statistics LN(Short Term
Compensation) t
Statistics
Constant 1.786* 2.658 2.927* 4.624 4.481* 5.926
Firm Size Component 0.248* 19.816 0.035* 2.889 0.116* 9.394
LN (Not Exercised Ratio) 0.295* 23.322 0.281* 21.437 0.011 0.896
LN(Bs Volatility) 0.100 0.948 0.223** 2.092 0.424* 3.913
LN(Number Mtgs) 0.011 0.270 0.047 1.168 0.046 1.130
LN (Tenure) 0.609** 2.213 0.598** 2.323 0.102 0.325
Trs1yr 0.00003 1.604 0.00002 1.125 0.00003 1.584
Sales5ls 0.0002 0.397 0.00008 0.146 0.001 1.107
Pdirpens 0.072 0.428 0.127 0.696 0.028 0.170
Interlock 0.098 1.455 0.031 0.471 0.129* 1.971
Year1993 0.036 0.258 0.380** 2.527 0.047 0.510
Year1994 0.193 1.382 0.392* 2.654 0.051 0.553
Year1995 0.012 0.084 0.326** 2.192 0.015 0.143
Year1996 0.122 0.835 0.394** 2.451 0.077 0.763
Year1997 0.139 0.955 0.415* 2.755 0.027 0.272
Year1998 0.148 1.012 0.492* 3.257 0.020 0.195
Year1999 0.287** 1.914 0.576* 3.740 0.004 0.039
Year2000 0.358** 2.342 0.584* 3.737 0.091 0.811
Year2001 0.360** 2.333 0.661* 4.183 0.062 0.536
Year2002 0.334** 2.181 0.647* 4.077 0.025 0.220
Year2003 0.220 1.419 0.570* 3.563 0.048 0.408
Year2004 0.391** 2.495 0.640* 3.971 0.132 1.098
N 1877 1877 1877
Adjusted RSquare 85.43% 74.42% 67.96%
(*) Significant at 1% level, (**) significant at 5% level, (***) significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995).
108
Table 9 Fixed Effect Regression Analysis of Determinants of Director Compensation for
NYSE listed firms
Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a factor extracted from Principal Component Analysis, composed of information from variables LN (Assets), LN (Market Value) and LN (Sales), which are used to analyse firm size; LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings held during the indicated fiscal year; LN (Tenure) is the natural logarithm of the number of years as CEO;Trs1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; Sales5ls is the 5year least squares annual growth rate of sales; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy that assumes the value of 1 when the executive is on two different boards at the same time and 0 when not; Nasdaq is a dummy that assumes the value of 1 when the executive is in a firm listed on NASDAQ and 0 when not; Pexecdir is a dummy that assumes the value of 1 when the executive is a director and 0 when not. We also control for time effect in terms of executive compensation using a dummy for each year between 1993 and 2004 and for industry effect using the Fama and French (1997) 48 industry classifications.
LN(Total Compensation) t
Statistics LN(Option Ratio) t Statistics LN(Short Term
Compensation) t Statistics
Constant 1.253 0.827 0.483 0.307 0.437 0.265 Firm Size Component 0.144* 18.820 0.026* 3.117 0.099* 13.391 LN (Not Exercised Ratio) 0.228* 40.640 0.333* 54.353 0.007 1.285 LN(Bs Volatility) 0.168* 4.200 0.297* 6.952 0.138* 3.638 LN(Number Mtgs) 0.008 0.351 0.052** 2.192 0.115* 5.516 LN (Tenure) 1.475** 2.196 0.408 0.586 1.872** 2.553 Trs1yr 0.0003** 2.185 0.001* 11.749 0.002* 14.304 Sales5ls 0.003* 4.831 0.002** 3.178 0.002* 2.605 Pdirpens 0.111* 4.356 0.059 2.156 0.054** 2.195 Interlock 0.093* 2.631 0.129* 3.359 0.007 0.218 Year1993 0.062** 2.171 0.009 0.242 0.048** 2.105 Year1994 0.282* 9.111 0.209* 5.929 0.161* 6.629 Year1995 0.310* 9.775 0.175* 4.750 0.115* 4.408 Year1996 0.482* 14.252 0.352* 9.548 0.189* 6.755 Year1997 0.606* 17.099 0.396* 10.100 0.206* 6.873 Year1998 0.669* 18.720 0.438* 11.096 0.302* 9.477 Year1999 0.806* 21.711 0.560* 14.068 0.354* 10.750 Year2000 0.871* 21.844 0.497* 11.603 0.389 10.852 Year2001 0.973* 23.855 0.620* 14.092 0.346* 8.964 Year2002 1.077* 25.895 0.588* 13.042 0.539* 13.989 Year2003 1.081* 25.054 0.592* 12.542 0.530* 13.481 Year2004 1.268* 28.436 0.610* 12.467 0.671* 16.439 Apparel Dummy 0,352 0,067 0,379 0,649 0,077 0,131 Business Dummy 1.474*** 1.719 1.045 1,174 2.207** 2.402 Candy Dummy 0,035 0,067 0,049 0,073 0,047 0,114 Computer Dummy 2.033 1.567 0,513 0,379 2.796** 1.989 Construct Dummy 0,896 1.252 1,237 1,619 1.259*** 1.674 Medical Dummy 0,083 0.168 0,685 1,182 0,355 0,793 Consumer Dummy 1.330** 2.287 0,930 1,455 1.314** 2.524 Fabricat Dummy 2.444* 4.298 1.217** 2.065 1.268** 2.118 Trading Dummy 0,152 0.168 1,03 1,102 2.435** 2.695 N 8281 8281 8281 Adjusted RSquare square
84.16% 66.57% 74.67% *Significant at 1% level, ** significant at 5% level, *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995).
109
Table 10 Fixed Effect Regression Analysis of the Determinants of Director Compensation
for NASDAQ listed firms
Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and based on the 1982 base of 100, we adjust to inflation the monetary variables reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: Firm Size Component is a factor extracted from Principal Component Analysis, composed of information from variables LN (Assets), LN (Market Value) and LN (Sales), which are used to analyse firm size. LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings held during the indicated fiscal year; LN (Tenure) is the natural logarithm of the number of years as CEO;Trs1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; Sales5ls is the 5year least squares annual growth rate of sales; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy that assumes the value of 1 when the executive is on two different boards at the same time and 0 when not; Nasdaq is a dummy that assumes the value of 1 when the executive is in a firm listed on NASDAQ and 0 when not; Pexecdir is a dummy that assumes the value of 1 when the executive is a director and 0 when not. We also control for time effect in terms of executive compensation using a dummy for each year between 1993 and 2004 and for industry effect using the Fama and French (1997) 48 industry classifications.
LN(Total Compensation) t Statistics LN(Option Ratio) t Statistics LN(Short Term
Compensation) t Statistics
Constant 1.295** 2.054 2.760* 5.708 3.487* 4.952
Firm Size Component 0.246* 21.748 0.043* 4.162 0.124* 11.724
LN (Not Exercised Ratio) 0.274* 26.534 0.256* 29.042 0.028* 2.861
LN(Bs Volatility) 0.210** 2.426 0.319* 4.601 0.219* 2.644
LN(Number Mtgs) 0.020 0.528 0.018 0.540 0.077** 2.181
LN (Tenure) 0.818* 3.034 0.566* 2.732 0.254 0.840
Trs1yr 0.00001 0.572 0.00002 1.395 0.00005** 2.645 Sales5ls 0.0004 0.949 0.0004 1.053 0.0003 0.793
Pdirpens 0.015 0.112 0.001 0.013 0.103 0.843
Interlock 0.083 1.281 0.003 0.057 0.102*** 1.704
Year1993 0.062 0.801 0.143** 1.815 0.030 0.589
Year1994 0.260* 3.168 0.155** 2.030 0.150* 2.670
Year1995 0.147** 1.660 0.130** 1.687 0.127*** 1.946
Year1996 0.222* 2.527 0.240* 3.132 0.042 0.705
Year1997 0.264* 2.958 0.201* 2.612 0.079 1.216
Year1998 0.267* 2.924 0.272* 3.504 0.144** 2.297
Year1999 0.383* 4.039 0.341* 4.217 0.122*** 1.715
Year2000 0.440* 4.340 0.323* 3.837 0.188** 2.474
Year2001 0.441* 4.311 0.453* 5.262 0.042 0.539
Year2002 0.459* 4.500 0.423* 4.925 0.172** 2.180
Year2003 0.289* 2.747 0.350* 3.954 0.149*** 1.785
Year2004 0.484* 4.470 0.430* 4.732 0.248* 2.856
N 2543 2543 2543
Adjusted RSquare 82.68% 72.46% 66.17%
*Significant at 1% level, ** significant at 5% level ,*** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
110
5.3. Analysis of the Results
As we expected, the above results show that executive compensation for NYSE
versus NASDAQ listed firms is explained by different factors. We also analyse in tables
11 and 12 whether the factor intensity (coefficients of the regressions of the CEOs and
Directors of NYSE and NASDAQ listed firms) is the same. We find that in all the cases
the values of the coefficients are significantly different.
Prior research has shown that firm size is one of the most important variables in
explaining executive compensation. From tables 7 and 8, we can see that the size
variable is significant in explaining variation in executive total compensation; however,
it has a stronger impact in the case of NASDAQ listed firms than that of NYSE listed
firms. Firm size has a positive influence on option ratio for CEOs in the case of
NASDAQ listed firms, meaning that size influences the number of stock options
granted to NASDAQ executives. This relationship is not statistically significant for the
NYSE. The firm’s size also positively influences the CEO’s short term executive
compensation (salary and bonus).
In the case of Directors, all three executive compensation dependent variables
are positively related to the size of the firm but the directors’ compensation for
NASDAQ listed firms is more sensitive to firm size than for the NYSE listed firms.
Not exercised ratio, which represents the number of options not exercised but
vested, is negatively related to CEOs´ total compensation and option ratio both for
NYSE and NASDAQ listed firms and, as we expected, it is positively related to short
term CEO compensation in both cases, meaning that when executives have stock
options they cannot exercise, the firms give them additional compensation, in cash, to
increase executive motivation. In the case of Directors, we also find a negative
relationship between total compensation and option ratio and a positive relationship
between shortterm compensation and option ratio, but this relationship is only
significant, in the latter case, in NASDAQ listed firms, meaning that Directors from
these firms cannot exercise the stock options that they have and, therefore, firms give
more cash compensation to increase their motivation.
The stock return volatility influences total compensation positively with the
exception of CEOs from NASDAQ, where the relationship is not statistically
111
significant. The option ratio also has a positive and significant relationship with firm
stock return volatility. The results are congruent with Yermack (1995) and Bryan and
Hwang’s (2000) findings. We also find a negative relationship between cash
compensation and stock return volatility, as did Core et al. (1999). In the case of
Directors, we find a positive relationship between total compensation, option ratio and
firm stock return volatility and a negative relationship with shortterm compensation in
both cases, meaning that when volatility increases, firms prefer to give more stock
options to the executive and less cash compensation. These relationships are stronger in
the case of NASDAQ than for the NYSE.
Noguera and Highfield (2007) are of the view that the board of directors is the
central internal mechanism of corporate governance in place at any corporation. Some
authors include the size of the board, others the composition of the board or the
influence of the CEO on the board as an explanatory variable of executive
compensation. Davidson III et al. (1998) use the number of board meetings and find
that it is negatively related to executive compensation because more control reduces the
ability to increase compensation and thus align the interests of shareholders and
executives. Also, Ryan and Wiggins (2001) conclude that more monitoring power can
reduce the need to provide CEOs with more incentive compensation. Our results are
congruent with previous findings in the case of NYSE CEO short term compensation
and NYSE and NASDAQ Directors in the case of option ratio and short term
compensation.
We also find that tenure strongly affects, in positive terms, CEO compensation
for NYSE listed firms but only slightly affects NASDAQ firms. Only in the case of
NASDAQ listed firms do we find that the option ratio is positively and significantly
related to CEO tenure. Our results are not consistent with the results of Ryan and
Wiggins (2001) and Conyon and He (2004), who find that CEO entrenchment due to
tenure would lead to higher cash compensation and lower incentive compensation
(stock options). We only find consistent results in the case of Directors’ tenure in which
case there is a strong positive relationship between tenure and total and shortterm
compensation for NYSE listed firms. In the case of NASDAQ listed firms, option ratio
is positively related to tenure but the results are not congruent with the previous
112
findings that executives will prefer a more certain compensation (cash compensation)
over less certain compensation (stock options).
The oneyear return to shareholders negatively affects CEO option ratio for
NYSE listed firms, and positively affects the cash compensation that these CEOs
receive. In the case of NASDAQ listed firms, the relationships are not statistically
significant. As we expected, if shareholders are satisfied with the return of their
investments, they do not need to give their executives more incentives based on stock
options. We do not find a negative relationship between cash compensation and one
year shareholders’ return, meaning that NYSE companies compensate their CEOs with
more cash compensation when they receive higher returns of investments. In the case of
Directors, we find a positive relationship with total compensation and short term
compensation for NYSE listed firms and a negative relationship with option ratio. In the
case of NASDAQ listed firms, this relationship is only positive and statistically
significant with cash compensation.
The sales increase in the last five years has a positive and statistically significant
relationship but only with the CEO total compensation and option ratio for NYSE listed
firms. The results are consistent with the findings of Ryan and Wiggins (2001) and
Anderson et al., (2000), which show that executives receive higher incentive pay in
firms with higher growth opportunities. However, these results are inconsistent with the
findings of Ghosh and Sirmans (2005), who find a negative relationship between
executive total compensation and firm growth opportunities. In the case of NASDAQ
listed firms, the relationship is not statistically significant. In the case of NYSE
Directors, the sales increase in the last 5 years influences, in positive terms, all the
dependent variables. NASDAQ director compensation is not affected by the sales
increase in the last five years.
As we expected, the existence of firm pension plans influences, in negative
terms, total compensation, option ratio and short term compensation, but only in the
case of NYSE listed firms, meaning that if the firm has already put money into
executives’ pension plans, they are able to justify a reduction in executive compensation
during the year. In the case of Directors, only the total and shortterm compensation in
NYSE listed firms are affected by the existence of Directors’ pension plans in the
company.
113
As we also expected, interlocked executives will receive more total
compensation and stock options, but the results are only statistically significant for
NYSE listed firms. The results are congruent with Core, Holthausen and Larcker (1999)
and Hallock (1997), who find that interlocked executives can positively influence their
personal compensation. In the case of Directors, we find a positive relationship with
total compensation and option compensation for NYSE listed firms and a positive
relationship with short term compensation for NASDAQ listed firms.
We also achieve interesting results in terms of the effect of the years on CEO
and Director compensation. The year effect strongly explains the total executive
compensation in NYSE listed firms for CEOs and Directors but not for NASDAQ listed
firms. Only after 1999 do we find a positive and statistically significant relationship in
NASDAQ listed firms, but in smaller terms than in NYSE listed firms. When we
analyse the option ratio, we find that both groups of CEOs are always influenced by the
effect of time, but the relationship is stronger for NYSE listed firms. In terms of short
term compensation, this relationship is only positive and statistically significant for
NYSE listed firms. In the case of Directors, the time effect strongly influences total
compensation, option ratio and short term compensation for NYSE listed firms but the
time effect has less significance for NASDAQ listed firms.
Finally, we also analyse the industry effect, using the 48 Fama and French
(1997) industry classifications and find that some of these dummies can also explain
executive compensation. This is the case of Business, which has a negative relationship
with CEO total compensation, and computer and construction, which have a positive
relationship with total compensation. The computer industry also has a positive
relationship with the NYSE CEO option ratio. We also find a negative and statistically
significant relationship with shortterm compensation and the apparel, candy and
computer industries and a positive relationship in the medical industry.
In the case of Directors, we find a negative relationship with the business
industry and total compensation for NYSE listed firms and a positive relationship
between total compensation and the consumer and manufacturing industries, also for
NYSE listed firms. In the case of option ratio, we find a positive relationship with the
manufacturing industry also for NYSE and a negative relationship with the Business
industry and cash compensation for NYSE. We also find a positive relationship between
114
cash compensation and the computer, construction, consumer, manufacturing and
trading industries for NYSE listed firms.
6. Conclusion
This is the first paper that analyses whether executive compensation for NYSE
and NASDAQ 14 listed firms is explained by the same factors. Using a oneway fixed
effect regression in an unbalanced panel sample for the period of 1992 to 2004 we also
investigate the trends in terms of executive compensation in NYSE and NASDAQ listed
firms, the forms of the compensation and whether the forms and weights of
compensation changed after the NASDAQ crash in 2000 and the SarbanesOxley Act in
2002.
Our results reveal that executive compensation is influenced by different factors
for NYSE and NASDAQ listed firms, and when some of the factors are the same, the
intensity of the coefficients is different, and this difference is statistically significant.
We also verify that NYSE and NASDAQ CEO and Director Compensation are
composed of different components. The percentage that salary represents in terms of
total compensation in NYSE listed firms is higher for Directors than CEOs. Bonus is a
more important compensation component for Directors than CEOs in NASDAQ listed
firms, but in the case of NYSE firms, the difference is small. In all cases, CEOs receive
more stock options than Directors. The used of restricted stock increases essentially in
the last few years. We also find that after the NASDAQ Crash in 2000, and essentially
after the SarbanesOxley Act in 2002, the forms and weights of CEO and Director
compensation change for NYSE and NASDAQ listed firms.
14 We did not analyse the factors that explain CEO and Director compensation in AMEX listed firms because the number of items of compensation is very small, making it impossible for us to make the regression analysis.
115
7. References
Anderson, M., Banker R., and Ravindran S. (2000). Executive compensation in the
information technology industry. Management Science, Vol. 46(4), 530 547.
Anderson, R. and Bizjak, J. (2003). An examination of the role of the CEO and the
compensation committee in structuring executive pay. Journal of Banking &
Finance, Vol. 27(7), 13231348. Ang, J., Lauterbach, B. and Schreiber, B. (2002). Pay at the executive suite: how do US
banks compensate their top management teams? Journal of Banking & Finance, Vol. 26, 11431163.
Bacidore, J. and Lipson, M. (2001). The effects of opening and closing procedures on
the NYSE and NASDAQ. SSRN Working Paper No. 257049.
Barro, J., and Barro, R. (1990). Pay, performance and turnover of bank CEOs. Journal of Labor Economics, Vol. 8 (4), 448481.
Barron, M. and Waddel, G. (2003). Executive rank, pay and project selection. Journal of Financial Economics, Vol. 67(2), 305349.
Beck, N., Katz, J. (1995). What to do (and not to do) with timeseries crosssection data. American Political Science Review, Vol. 89(3), 663647.
Bryan, S. and Hwang, L. (2000). CEO stockbased compensation: an empirical analysis
of incentive Intensity, relative mix, and economic determinants. Journal of Business, Vol. 73(4), 661693.
Canyon, M. and He, L. (2004). Compensation Committee and CEO compensation
incentives in U.S. entrepreneurial firms, Working paper. Chen, Y. and Hung, C. (2006) Corporate governance and insider compensation: Nasdaq
technology firms. Working paper presented at FMA Conference in Salt Lake City. Chidambaran, N. and Prabhala, N. (2003). Executive stock option repricing, internal
governance mechanisms and management turnover. Journal of Financial
Economics, Vol. 69(1),153189. Chung K, Ness, B and Ness R (2001). Can the treatment of limit orders reconcile the
differences in trading costs between NYSE and Nasdaq Issues? Journal of Financial and Quantitative Analysis, Vol. 36 (2), 267286.
116
Chung, K., Ness, B. and Ness, R. (2001) Are NASDAQ stocks more costly to trade than
NYSE stocks? Evidence after decimalization. SSRN Working Paper No. 277817. Core, J., Holthausen, R. and Larcker, D. (1999). Corporate governance, Chief Executive
Officer compensation, and firm performance. Journal of Financial Economics, Vol. 51(3), 371406.
Davidson, W., Pilger, T. and Szakmary, A. (1998). Golden parachutes, Board and
Committee composition and shareholder wealth. The Financial Review, Vol. 33(4), 1732.
Fama, E. and French, K. (1997). Industry costs of equity. Journal of Financial
Economics, Vol. 43, 153193. Firth, M., Lohne, J., Ropstad, R. and Sjo, J. (1996). The remuneration of CEOs and
corporate financial performance in Norway. Managerial and Decisions Economics, Vol. 17(3), 291301.
Gosh, C. and Sirmans, C. ( 2005). On REIT CEO compensation: does board structure
Matter? Journal of Real State Finance & Economics, Vol. 30 (4), 397428. Grinstein, Y. and Hribar, P. (2004). CEO compensation and incentives: evidence from
M&A bonus. Journal of Financial Economics, Vol. 73(1),119143.
Hallock, K. (1997). Reciprocally interlocking Boards of Directors and executive
compensation. Journal of Financial and Quantitative Analysis, Vol. 32(3), 331 344.
Kang, S., Kumar, P. and Lee, H. (2006). Agency and corporate investment: the role of
executive compensation and corporate governance. Journal of Business, Vol.
79(3), 11271148. Murphy, K. (1986). Incentives, learning and compensation: a theoretical and empirical
investigation of managerial labour contracts. Rand Journal of Economics, Vol. 17(1), 5976.
Murphy, K. (2003). Stockbased pay in new economy firms. Journal of Accounting and Economics, Vol. 34(1), 129147.
Narayanan, M. and Seyhun, H. 2005, Effect of SarbanesOxley Act on the influencing
of executive compensation. SSRN working paper nº 852964
117
Noguera, M. and Highfield, M. (2007). CEO incentivebased compensation and REIT
performance. Financial Management Association Conference, Working paper,
Orlando, Florida (USA).
Ryan Jr, H. and Wiggins III, R. (2001). The influence of firm and manager specific
characteristics on the structure of executive compensation. Journal of Corporate Finance, Vol. 25(7), 101123.
Ryan Jr, H. and Wiggins R. (2004). Who is in whose pocket? Director compensation,
board independence, and barriers to effective monitoring. Journal of Financial Economics, Vol. 73(3), 497524.
Sapp, T. and Yan, X. (2000). Should small Nasdaq firms switch to AMEX? SSRN
working paper no. 238981.
Stathopoulos, K., Espenlaub, S. and Walker, M. (2004). U.K. executive compensation
practices: new economy versus old economy. Journal of Management Accounting Research, Vol. 16(1), 5792.
Yermack, D. (1995). Do corporations award CEO stock options effectively? Journal of Finance and Economics, Vol. 39(23), 237269.
118
8. Appendix
Table 11 T Test of Equality of Fixed Effect Regressions Coefficients – CEOs
(NYSE versus NASDAQ)
Panel A: LN (Total Compensation) NYSE NASDAQ t test
Dependent Variables N Coef. Std. Error N Coef. Std. Error Sig.
Constant 6124 0,346 0,661 1877 1,786 0,672 *
Firm Size Component 6124 0,165 0,008 1877 0,248 0,013 *
LN (Not Exercised Ratio) 6124 0,220 0,006 1877 0,295 0,013 *
LN(Bs Volatility) 6124 0,190 0,043 1877 0,100 0,105 *
LN (Number Mtgs) 6124 0,002 0,024 1877 0,011 0,040 *
LN (Tenure) 6124 1,633 0,270 1877 0,609 0,275 *
Trs1yr 6124 0,00004 0,000 1877 0,00003 0,000 *
Sales5ls 6124 0,001 0,001 1877 0,0002 0,001 *
Pdirpens 6124 0,103 0,028 1877 0,072 0,168 *
Interlock 6124 0,083 0,037 1877 0,098 0,067 *
Year1993 6124 0,117 0,040 1877 0,036 0,140 *
Year1994 6124 0,351 0,041 1877 0,193 0,139 *
Year1995 6124 0,362 0,042 1877 0,012 0,145 *
Year1996 6124 0,512 0,043 1877 0,122 0,146 *
Year1997 6124 0,609 0,045 1877 0,139 0,145 *
Year1998 6124 0,663 0,046 1877 0,148 0,147 *
Year1999 6124 0,800 0,046 1877 0,287 0,150 *
Year2000 6124 0,827 0,049 1877 0,358 0,153 *
Year2001 6124 0,899 0,050 1877 0,360 0,154 *
Year2002 6124 0,989 0,051 1877 0,334 0,153 *
Year2003 6124 0,987 0,052 1877 0,220 0,155 *
Year2004 6124 1,151 0,054 1877 0,391 0,157 *
Apparel Dummy 6124 0,168 0,376
Busines Dummy 6124 1,547 0,449
Candy Dummy 6124 0,570 0,406
Computer Dummy 6124 2,719 0,616
Construct Dummy 6124 1,380 0,355
Medical Dummy 6124 0,061 0,357 * Difference is statistically significant at 1% (*), 5% level (**) or 10% level(***)
119
Table 11 (Cont.)
PANEL B: Option Ratio NYSE NASDAQ T test
Dependent Variables N Coef. Std. Error N Coef. Std. Error Sig,
Constant 6124 2,099 0,724 1877 2,927 0,633 *
Firm Size Component 6124 0,010 0,010 1877 0,035 0,012 *
LN (Not Exercised Ratio) 6124 0,344 0,007 1877 0,281 0,013 *
LN(Bs Volatility) 6124 0,188 0,050 1877 0,223 0,107 *
LN(Number Mtgs) 6124 0,037 0,027 1877 0,047 0,040 *
LN (Tenure) 6124 0,379 0,294 1877 0,598 0,257 *
Trs1yr 6124 0,002 0,0001 1877 0,00002 0,00002 *
Sales5ls 6124 0,003 0,001 1877 0,00008 0,001 *
Pdirpens 6124 0,079 0,032 1877 0,127 0,183 *
Interlock 6124 0,142 0,043 1877 0,031 0,066 *
Year1993 6124 0,002 0,061 1877 0,380 0,150 *
Year1994 6124 0,202 0,060 1877 0,392 0,148 *
Year1995 6124 0,148 0,061 1877 0,326 0,149 *
Year1996 6124 0,293 0,061 1877 0,394 0,161 *
Year1997 6124 0,316 0,063 1877 0,415 0,151 *
Year1998 6124 0,411 0,064 1877 0,492 0,151 *
Year1999 6124 0,543 0,064 1877 0,576 0,154 *
Year2000 6124 0,496 0,068 1877 0,584 0,156 *
Year2001 6124 0,628 0,068 1877 0,661 0,158 *
Year2002 6124 0,611 0,069 1877 0,647 0,159 *
Year2003 6124 0,600 0,071 1877 0,570 0,160 *
Year2004 6124 0,611 0,072 1877 0,640 0,161 *
Apparel Dummy 6124 0,008 0,431
Business Dummy 6124 0,255 0,515
Candy Dummy 6124 0,413 0,615
Computer Dummy 6124 1,711 0,687
Construct Dummy 6124 0,541 0,406
Medical Dummy 6124 0,270 0,449
Difference is statistically significant at 1%(*), 5% level (**) or 10% level(***)
120
Table 11 (cont.)
PANEL C: LN(Short Term Compensation) NYSE NASDAQ t test
Dependent Variables N Coef. Std. Error N Coeff. Std. Error Sig.
Constant 6124 4.578 0,634 1877 4,481 0,756 *
Firm Size Component 6124 0,124 0,007 1877 0,116 0,012 *
LN (Not Exercised Ratio) 6124 0,002 0,005 1877 0,011 0,012 *
LN(Bs Volatility) 6124 0,123 0,035 1877 0,424 0,108 *
LN(Number Mtgs) 6124 0,071 0,021 1877 0,046 0,041 *
LN (Tenure) 6124 0,147 0,262 1877 0,102 0,314 *
Trs1yr 6124 0,001 0,0001 1877 0.00003 0,0001 *
Sales5ls 6124 0,0002 0,001 1877 0,001 0,001 *
Pdirpens 6124 0,042 0,022 1877 0,028 0,166 *
Interlock 6124 0,001 0,031 1877 0,129 0,065 *
Year1993 6124 0,058 0,029 1877 0,047 0,092 *
Year1994 6124 0,152 0,03 1877 0,051 0,093 *
Year1995 6124 0,091 0,031 1877 0,015 0,103 *
Year1996 6124 0,136 0,032 1877 0,077 0,100 *
Year1997 6124 0,158 0,033 1877 0,027 0,101 *
Year1998 6124 0,239 0,034 1877 0,020 0,102 *
Year1999 6124 0,298 0,036 1877 0,004 0,110 *
Year2000 6124 0,308 0,038 1877 0,091 0,112 *
Year2001 6124 0,245 0,04 1877 0,062 0,115 *
Year2002 6124 0,399 0,04 1877 0,025 0,114 *
Year2003 6124 0,368 0,041 1877 0,048 0,118 *
Year2004 6124 0,496 0,042 1877 0,132 0,120 *
Apparel Dummy 6124 0,588 0,307
Business Dummy 6124 0,186 0,394
Candy Dummy 6124 0,541 0,283
Computer Dummy 6124 1.206 0,563
Construct Dummy 6124 0,310 0,310
Medical Dummy 6124 0,479 0,284
121
Table 12 T Test of Equality of Fixed Effect Regressions Coefficients – Directors
Panel A: LN (Total Compensation) NYSE NASDAQ t test
Dependent Variables N Coef. Std. Error N Coef. Std. Error Sig.
Constant 8281 1,253 1,515 2543 1,295 0,630 *
Firm Size Component 8281 0,144 0,008 2543 0,246 0,011 *
LN (Not Exercised Ratio) 8281 0,228 0,006 2543 0,274 0,010 *
LN(Bs Volatility) 8281 0,168 0,040 2543 0,210 0,087 *
LN(Number Mtgs) 8281 0,008 0,022 2543 0,020 0,037 *
LN (Tenure) 8281 1,475 0,672 2543 0,818 0,270 *
Trs1yr 8281 0,0003 0,0001 2543 0,00001 0,00002 *
Sales5ls 8281 0,003 0,001 2543 0,0004 0,0004 *
Pdirpens 8281 0,111 0,026 2543 0,015 0,135 *
Interlock 8281 0,093 0,035 2543 0,083 0,065 *
Year1993 8281 0,062 0,029 2543 0,062 0,078 *
Year1994 8281 0,282 0,031 2543 0,260 0,082 *
Year1995 8281 0,310 0,032 2543 0,147 0,089 *
Year1996 8281 0,482 0,034 2543 0,222 0,088 *
Year1997 8281 0,606 0,035 2543 0,264 0,089 *
Year1998 8281 0,669 0,036 2543 0,267 0,091 *
Year1999 8281 0,806 0,037 2543 0,383 0,095 *
Year2000 8281 0,871 0,040 2543 0,440 0,101 *
Year2001 8281 0,973 0,041 2543 0,441 0,102 *
Year2002 8281 1,077 0,042 2543 0,459 0,102 *
Year2003 8281 1,081 0,043 2543 0,289 0,105 *
Year2004 8281 1,268 0,045 2543 0,484 0,108 *
Apparel Dummy 8281 0,352 0,568
Business Dummy 8281 1,474 0,858
Candy Dummy 8281 0,035 0,517
Computer Dummy 8281 2,033 1,298
Construct Dummy 8281 0,896 0,716
Medical Dummy 8281 0,083 0,496
Consumer Dummy 8281 1,330 0,581
Fabricant Dummy 8281 2,444 0,569
Trading Dummy 8281 0,152 0,904
* Difference is statistically significant at 1%(*), 5% level (**) or 10% level(***)
122
Table 12 (Cont.)
Panel B: LN (Option Ratio) NYSE NASDAQ t test
Dependent Variables N Coef. Std. Error N Coef. Std. Error Sig.
Constant 8281 0,483 1,571 2543 2,760 0,483 *
Firm Size Component 8281 0,026 0,008 2543 0,043 0,009 *
LN (Not Exercised Ratio) 8281 0,333 0,006 2543 0,256 0,009 *
LN(Bs Volatility) 8281 0,297 0,043 2543 0,319 0,069 *
LN(Number Mtgs) 8281 0,052 0,024 2543 0,018 0,033 *
LN (Tenure) 8281 0,408 0,697 2543 0,566 0,207 *
Trs1yr 8281 0,001 0,0001 2543 0,00002 0,00002 *
Sales5ls 8281 0,002 0,001 2543 0,0004 0,0003 *
Pdirpens 8281 0,059 0,027 2543 0,001 0,109 *
Interlock 8281 0,129 0,038 2543 0,003 0,054 *
Year1993 8281 0,009 0,036 2543 0,143 0,079 *
Year1994 8281 0,209 0,035 2543 0,155 0,076 *
Year1995 8281 0,175 0,037 2543 0,130 0,077 *
Year1996 8281 0,352 0,037 2543 0,240 0,077 *
Year1997 8281 0,396 0,039 2543 0,201 0,077 *
Year1998 8281 0,438 0,039 2543 0,272 0,078 *
Year1999 8281 0,560 0,040 2543 0,341 0,081 *
Year2000 8281 0,497 0,044 2543 0,323 0,084 *
Year2001 8281 0,620 0,044 2543 0,453 0,086 *
Year2002 8281 0,588 0,045 2543 0,423 0,086 *
Year2003 8281 0,592 0,047 2543 0,350 0,089 *
Year2004 8281 0,610 0,049 2543 0,430 0,091 *
Apparel Dummy 8281 0,379 0,585
Business Dummy 8281 1,045 0,890
Candy Dummy 8281 0,049 0,666
Computer Dummy 8281 0,513 1,356
Construct Dummy 8281 1,237 0,764
Medical Dummy 8281 0,685 0,580
Consumer Dummy 8281 0,930 0,639
Fabricant Dummy 8281 1,217 0,590
Trading Dummy 8281 1,030 0,935
* Difference is statistically significant at 1%(*), 5% level (**) or 10% level(***)
123
Table 12 (Cont.)
PANEL C: LN (Short Term Compensation) NYSE NASDAQ t test
Dependent Variables N Coef. Std. Error N Coef. Std. Error Sig.
Constant 8281 0,437 1,653 2543 3,487 0,704 * Firm Size Component 8281 0,099 0,007 2543 0,124 0,011 * LN (Not Exercised Ratio) 8281 0,007 0,005 2543 0,028 0,010 * LN(Bs Volatility) 8281 0,138 0,038 2543 0,219 0,083 * LN(Number Mtgs) 8281 0,115 0,021 2543 0,077 0,035 *
LN (Tenure) 8281 1,872 0,734 2543 0,254 0,302 *
Trs1yr 8281 0,002 0,000 2543 0,00005 0,00002 *
Sales5ls 8281 0,002 0,001 2543 0,0003 0,0003 *
Pdirpens 8281 0,054 0,024 2543 0,103 0,122 *
Interlock 8281 0,007 0,033 2543 0,102 0,060 *
Year1993 8281 0,048 0,023 2543 0,030 0,051 *
Year1994 8281 0,161 0,024 2543 0,150 0,056 *
Year1995 8281 0,115 0,026 2543 0,127 0,065 *
Year1996 8281 0,189 0,028 2543 0,042 0,059 *
Year1997 8281 0,206 0,030 2543 0,079 0,065 *
Year1998 8281 0,302 0,032 2543 0,144 0,063 *
Year1999 8281 0,354 0,033 2543 0,122 0,071 *
Year2000 8281 0,389 0,036 2543 0,188 0,076 *
Year2001 8281 0,346 0,039 2543 0,042 0,079 *
Year2002 8281 0,539 0,039 2543 0,172 0,079 *
Year2003 8281 0,530 0,039 2543 0,149 0,083 *
Year2004 8281 0,671 0,041 2543 0,248 0,087 *
Apparel Dummy 8281 0,077 0,592
Business Dummy 8281 2,207 0,919
Candy Dummy 8281 0,047 0,411
Computer Dummy 8281 2,796 1,406
Construct Dummy 8281 1,259 0,752
Medical Dummy 8281 0,355 0,448
Consumer Dummy 8281 1,314 0,521
Fabricant Dummy 8281 1,268 0,599
Trading Dummy 8281 2,435 0,986
* Difference is statistically significant at 1%(*), 5% level (**) or 10% level(***)
124
Table 13: Pearson Correlation of the Independent Variables
Panel A: NYSE 1 2 3 4 5 6 7 8 9
1 Firm Size Component 1,000 0,023 0,308 0,268 0,042 0,030 0,043 0,152 0,058
2 LN(Not Exercised Ratio) 0,023 1,000 0,001 0,033 0,125 0,036 0,024 0,013 0,042
3 LN (Bs Volatility) 0,308 0,001 1,000 0,057 0,132 0,057 0,188 0,271 0,036
4 LN ( Number Mtgs) 0,268 0,033 0,057 1,000 0,030 0,043 0,021 0,151 0,060
5 LN (Tenure) 0,042 0,125 0,132 0,030 1,000 0,016 0,030 0,121 0,134
6 Trs1yr 0,030 0,036 0,057 0,043 0,016 1,000 0,042 0,025 0,002
7 Sales5ls 0,043 0,024 0,188 0,021 0,030 0,042 1,000 0,139 0,095
8 Pdirpens 0,152 0,013 0,271 0,151 0,121 0,025 0,139 1,000 0,002
9 Interlock 0,058 0,042 0,036 0,060 0,134 0,002 0,095 0,002 1,000
Panel B: NASDAQ 1 2 3 4 5 6 7 8 9
1 Firm Size Component 1,000 0,127 0,305 0,054 0,025 0,001 0,079 0,115 0,059
2 LN(Not Exercised Ratio) 0,127 1,000 0,024 0,001 0,187 0,027 0,068 0,034 0,005
3 LN (Bs Volatility) 0,305 0,024 1,000 0,168 0,101 0,098 0,198 0,188 0,035
4 LN (Number Mtgs) 0,054 0,001 0,168 1,000 0,080 0,008 0,039 0,075 0,043
5 LN (Tenure) 0,025 0,187 0,101 0,080 1,000 0,003 0,079 0,039 0,096
6 Trs1yr 0,001 0,027 0,098 0,008 0,003 1,000 0,000 0,006 0,004
7 Sales5ls 0,079 0,068 0,198 0,039 0,079 0,000 1,000 0,067 0,003
8 Pdirpens 0,115 0,034 0,188 0,075 0,039 0,006 0,067 1,000 0,019
9 Interlock 0,059 0,005 0,035 0,043 0,096 0,004 0,003 0,019 1,000
125
CHAPTER 4
Executive Compensation: An Examination of S&P Listed Firms 15
15 We are grateful to Professor Sheng Huang, from Washington University, discussant of this paper at the Midwest Finance Association in Texas (USA) in February of 2008 and to other anonymous professors present at the conference for their helpful comments. We are also grateful to Professor Mário Augusto, from the University of Coimbra, discussant of the paper at the 18 th Luso Spanish Conference on Management in February in Porto.
126
1. Introduction
In this age and era of technological advances, regulatory changes, and
increased crossborder flow of information, capital and labour, firms search for
executive talent globally to find the best fit for jobs. The emergence of the global
village, removal of barriers to free trade, and the rise of Internet technology have indeed
created an increased demand for corporate executives. Perhaps, this is the reason that
there is a new and growing interest among academics as well as practitioners to explore
the numerous dimensions of executive compensation. Researchers are examining
questions related to the determinants and forms of executive compensation and potential
differences in compensation in numerous situations including new versus old economy
firms and men versus women executives. Firm size has been defended by a significant
number of authors as being one of the most important factors that influence executive
compensation. Therefore, research inquiries related to executive compensation can
produce misleading results if the size effect is not accounted for. For example,
comparison of compensation between men versus women executives can be more
meaningful if size effect is accounted for.
The motivation to develop this investigation is based on the findings of Lambert
et al. (1990), who analyse the relationship between the percentage that executive
compensation changes and the percentage that firm size changes and found that this
relationship is positive and significant but is smaller than when absolute values and not
percentages are used to analyse the relationship. According to the authors, the result
means that the changes in executive compensation are not primarily influenced by firm
size. To analyse the impact of firm size in terms of executive compensation, most
authors adjusted for size effect in multiple regression setting using variables like Sales,
Market Value, Assets or the natural logarithm of these values, but we go a step further
in our study. Instead of inserting one additional size variable in multiple regressions, we
focus on three different size groups of large firms (S&P 500), medium size firms (S&P
Mid Cap), and small firms (S&P Small Cap), individually.
Based on the findings of Lambert et al. (1990) that changes in executive
compensation are not primarily influenced by changes in terms of firm size, we control
for firm size and industry effect and analyse whether the factors that influence executive
127
compensation (and the forms of compensation) are the same for the S&P500, S&P Mid
Cap and S&P Small Cap index listed firms between the period of 1992 and 2004 and
also if this changed after the NASDAQ crash. We also examine whether the forms of
executive compensation changed after the NASDAQ crash in 2000 and the Sarbanes
Oxley Act in 2002. Based on Narayanan and Seyhun (2005) and Murphy (2003)’s
findings, we also believe that the NASDAQ crash in 2000 and the SarbanesOxley Act
in 2002 significantly changed the form of compensation because a significant number of
restrictions in terms of corporate governance were developed.
We believe that, controlling for firm size and industry effect, the factors that
explain executive compensation may be different because each S&P index represents
different groups of firms with different characteristics such as net income, sales and
dividends. We want to focus on each of the three sets of S & P firms – small, medium,
and large firms and compare and contrast the differences in executive compensation, if
any, among these different size firms. We also separate the analysis between CEOs and
Directors based on the investigation of Cheng and Hung (2006), who found significant
differences between these two groups.
Our motivation to develop this investigation is also due to the fact that we
believe that the changes introduced by the SarbanesOxley Act in 2002 and also the
NASDAQ crash in 2000 changed the structure of executive compensation not only in
terms of total values, but also in terms of the percentage or fraction that each
compensation component represents in the total compensation.
Our results reveal that the NASDAQ crash in 2000 and, especially, the
SarbanesOxley Act in 2002, did indeed change the forms of the components of
executive compensation. We also find that the factors that explain executive
compensation in S&P500, S&P Mid Cap and S&P Small Cap listed firms are generally
different: if some variables are equal, their intensity is different and this difference is
generally statistically significant. We also document that the NASDAQ crash changes
the influence of the factors that explain executive compensation.
128
2. Literature Review and Research Questions
Firm size has been used by a vast number of authors to analyse diverse situations
in the executive compensation area. Datta, IskandarDatta and Raman (2001), use firm
size to explain how executive compensation determines corporate acquisitions.
Bushman, Indjejikian and Smith (1996) analyse individual performance and the CEOs’
annual incentive plans and find that firm size is positively related to salary and bonuses
and negatively related to longterm compensation. Bertrand and Hallock (1999) find
that the gender pay gap among highestpaid executives is positively related to firm size.
Some authors also analyse the impact of firm size on the use of resetting, also
called repricing, of stock options plans. Bens et al. (2003) find a negative relationship
between stock option repurchase and firm size; Brenner et al. (2000), Carter and Lynch
(2001,2003) and Chance et al. (2000), find stock option resetting is more common in
smaller firms. The same results are discovered by Chidambaran and Prabhala (2003).
Also Chance et al. (2000) say that firms that reprice stock options plans are generally
younger, associated with high technology and have outofthemoney stock options
plans.
Ueng et al. (2000) describe that CEO pay in large firms is a function of the
influence of this CEO on the board, the firm size and firm’s performances but do not
find that CEO influence on the board in small companies affects CEO pay. Kostiuk
(1990), analysing the impact of firm size on executive compensation, describes that
there is significant variability in the level of compensation between firms of the same
size, which may indicate that CEOs and other executives from the firms listed on
S&P500, MidCap and Small Cap have different forms and values of compensation.
Hermalin and Wallace (2001), like Aggarwall and Samwick (1999), use the
variable size to determine the relationship between compensation and performance and
find a positive relationship between these two variables and stock options granted, and a
negative relationship with the use of golden parachutes, restricted stocks and with
supplemental pension plans. Ryan and Wiggins (2004) also document a positive
relationship between total compensation and equity based compensation. Kato et al.
(2005) find the same results on the Japanese market. Morgan and Poulsen (2001) find a
129
positive relation between firm size and the vote returns for managementsponsored
compensation proposals.
As we can see, firm size can explain a lot of situations in terms of executive
compensation, but what happens if we fix this effect in terms of executive
compensation, making separate analyses for firms listed on S&P500, S&P Mid Cap and
S&P Small Cap?
In this paper, we analyse the following two major sets of questions: Is the value
of total compensation packages across S&P firms the same? Are the forms of total
compensation packages across S&P firms the same? Has there been a change in the
forms of compensation among S&P firms since the NASDAQ crash (2000) and the
enactment of the SarbanesOxley Act (2002)? If there has been a change, is it the same
across the S&P firms?
If firms listed on S&P500, S&P Mid Cap and S&P Small Cap indexes are
different, we expect that the factors that explain executive compensation in these firms
can also be different, and if some of the variables are the same, the intensity of these
factors can be different and statistically significant.
The second set of questions explores the following questions: Are the
determinants of executive compensation the same across S&P firms? Is the intensity
(impact) of the factors explaining executive compensation across S & P firms the same?
Have the factors that explain executive compensation changed since the NASDAQ
Crash in 2000?
The NASDAQ crash is also effectively associated with a group of financial
scandals and the bankruptcy of some large American companies, based on fraudulent
accounting practices and executive selfdealing. To solve some of these corporate
governance problems, the SarbanesOxley Act was created in the USA in 2002,
introducing changes in governance, reporting, and disclosure requirements of public
firms with the intent of improving accuracy, reliability, and timeliness of the
information provided to investors.
With these significant changes in terms of corporate governance rules, we
believe that the percentage that each compensation component (salary, bonus, stock
options, restricted stocks, longterm incentive plans) represents in total compensation
has changed since the NASDAQ crash in 2000 and essentially since the SarbanesOxley
130
Act in 2002, in these three groups. We also expect that these changes, in terms of the
structure of executive compensation, will be different in S&P500, S&P Mid Cap and
S&P Small Cap listed firms.
3. Data, Sample Selection and Statistics
3.1. Data and Sample Selection
We use data from the Standard ExecuComp database, which contains
information about executive compensation for public U.S. companies from 1992 to
2004. The dataset includes all the S&P index listed firms. Together, these firms
represent over 80% of the total market capitalisation of U.S. public firms 16 . This
database contains information about the five most well paid executives from each of the
firms in the database. In terms of compensation, the database categorises the various
components into salary, bonus, exante value of options, restricted stocks award, long
term incentive plan (LTIP), other annual compensation and all other compensation.
We use unbalanced panel data because not all executives stay with the same firm
during the sample period. The sample constitutes 79,650 observations of compensation
related to the 5 most highly paid executives from these 1500 firms 17 from 1992 to 2004.
This sample is built excluding the entire executive observations whose sum of salary
and bonuses, and also total compensation, was equal to zero. We also exclude
observation from executives who have remunerations more than once in the same year,
deleting the observations with lower remunerations than the executive receives.
Using the Consumer Price Index (CPI), compiled by the Bureau of Labor
Statistics, and using 1982 as base year, we adjust the monetary variables to the price
level of the year 2004.
16 The data is from 1992 to 2004 because 1992 is the first year of information in the ExecuComp database and 2004 was the last complete year of information when we started this work. In this database, there are executives with data for all of the years and others with only a small number of years of information. 17 The three groups of executives that we are studying (S&P500, S&PMidCap and S&PSmallCap listed firms), were achieved with the variable SPCODE index from ExecuComp database, which classifies the firms listed in the S&P500 with the code “SP”, the firms listed in the S&PMidCap 400 Index with the code “MD” and the firms listed in the S&PSmallCap Index with the code “SM”.
131
3.2. Statistics
Table 1 presents the statistics of average total compensation of executives for
firms listed in the S&P500, S&P Mid Cap and S&P Small Cap index. We use OneWay
Analysis of Variance to test the null hypothesis of the equality of three means against
the alternative hypothesis that at least one mean is different. 18
18 We use the Bonferroni correction, which is a multiplecomparison correction used when several dependent or independent statistical tests are being performed simultaneously.
132
Table 1
Descriptive Statistics
Mean Total Compensation Levels Adjusted for Inflation by Year (19922004) Top Five
This table presents the average total compensation of the five most well paid executives associated with firms listed in the S&P500, S&P Mid Cap and S&P Small Cap index during the period from 1992 to 2004 and the results for OneWayAnova to test whether the mean differences are statistically different. Data are from ExecuComp database. Monetary variables are adjusted to inflation and are stated in 2004 dollars. Total compensation is the sum of salary, bonus, stock options, restricted stocks, other annual compensation and all other compensation. Salary is the executive salary for the year. Bonus is the dollar value of bonus (cash and non cash) earned by the executive officer during the fiscal years. Stock option is the aggregate value of stock options granted to the executive during the fiscal year as valued using the Black Scholes methodology. Restricted stocks are the value of restricted stock granted during the year (determined as of the date of the grant). LTIP is the amount paid out to the executive under the company’s longterm incentive plan. Other annual compensation is the dollar value of other annual compensation not properly categorised as salary or bonus. All other compensation is the amount listed under “All other Compensation” in the Summary Compensation Table. Mean average and mean difference are in thousands of dollars.
S&P500 (1) S&PMidCap(2) S&PSmallCap(3) Mean Difference Mean Test (ANOVA) (Bonferroni Test)
Year
N Mean N Mean N Mean (1) and (2)
(1) and (3) (2) and (3) (1) and (2)
(1) and (3)
(2) and (3)
1992 1416 1963.48 678 875.84 539 801.26 1087.64 1314.67 227,.02 * * ***
1993 2007 1951.39 1044 908.05 916 960.44 1043.34 1267.07 223.73 * * **
1994 2106 2179.17 1132 1019.63 1061 731.25 1159.54 1447.93 288.39 * * *
1995 2167 2380.44 1231 1084.13 1220 712.24 1296.30 1668.19 371.89 * * *
1996 2250 2975.18 1370 1254.44 1386 814.20 1720.75 2160.98 440.23 * * *
1997 2354 3937.39 1524 1511.26 1674 900.35 2426.12 3037.04 610.91 * * *
1998 2448 4807.43 1637 161.16 1964 958.87 3192.26 3848.56 656.30 * *
1999 2515 5557.27 1703 1872.61 2131 1000.30 3684.65 4556.97 872.32 * * *
2000 2583 6697.17 1820 2152.70 2223 1347.02 4544.46 5350.15 805.67 * * **
2001 2595 6039.35* 1845 1962.25* 2337 1157.00* 4077.10 4882.35 805.25 * * *
2002 2636 4663.58 1934 2062.78 2539 1038.81 2600.81 3624.77 1023.97 * * *
2003 2690 4332.43* 2012 1744.33* 2696 1002.22 ***
2588.11 3330.21 742.10 * * *
2004 2672 4767.34 1957 1975.63 2671 1181.24 2791.71 3586.10 794.39 * * *
Note: Difference is statistically significant at (*) 1% level, (**) 5% level and (***) 10% level.
From table 1, we can see that, on average, the top five executives from the
S&P500 firms consistently receive higher compensation during the period of 1992 to
2004. The means differences of total compensation between S&P500, S&P Mid Cap,
133
and S&P Mid Cap are significant each year during the sample period except for S&P
Mid Cap and S&P Small Cap listed firms in 1998.
Table 2, summarises the time series percentage changes that each compensation
component represents in terms of total compensation for all S & P listed firms during
the period from 1992 to 2004 19 . We also perform the OneWay ANOVA analysis of
variance to test the hypothesis that the means of each component of executive
compensation for S&P500, S&P Mid Cap and S&P Small Cap index listed firms are
equal. In most cases, we reject the null hypotheses that the means are equal 20 .
We also test in Table 2 whether the differences in executive compensation
between year 2001 in relation to year 2000 (NASDAQ crash effect) and the year 2003
in relation to 2002 (SarbanesOxley Act effect) are statistically significant. Based on a t
test we find that in most cases, these differences are statistically significant.
19 Last year of information available from Execucomp database when we started this investigation. 20 We also use the Levene's test for homogeneity of variance and Bonferroni correction, which is a multiplecomparison correction used when several dependent or independent statistical tests are being performed simultaneously.
134
Table 2
Executive Compensation Components as a Percentage of Total Compensation for S&P 500, S&P Mid Cap and S&P Small Cap Listed Firms (19922004)
This table presents the percentages that each compensation component represents in terms of total compensation by year. Our sample includes data from the five most well paid executives associated with firms listed in the S&P500, S&P Mid Cap and S&P Small Cap during the period from 1992 to 2004. Salary is the executive salary for the year. Bonus is the dollar value of bonus (cash and noncash) earned by the executive officer during the fiscal years. Stock Option is the aggregate value of stock options granted to the executive during the fiscal year as valued by using S & P’s Black Scholes methodology. Restricted stocks are the value of restricted stock granted during the year (determined as of the date of the grant). LTIP is the amount paid out to the executive under the company’s longterm incentive plan. Other annual compensation is the dollar value of other annual compensation not properly categorised as salary or bonus. All other compensation is the amount listed under “All other Compensation” in the Summary Compensation Table. We describe S&P500 listed firms as SP, S&P Mid Cap listed firms as MD and S&P Small Cap firms as SM.
Panel A: Top Five (% of Total Compensation)
Salary Bonus Stock Options Restricted Stocks LTIP Year
SP MD SM SP MD SM SP MD SM SP MD SM SP MD SM
1992 41.67 55.90 59.54 20.38 17.25 18.40 22.63 16.46 13.95 5.25 3.33 2.45 4.78 2.76 1.54
1993 40.66 51.02 56.26 22.45 20.28 19.51 21.90 16.30 15.68 4.88 3.55 2.56 4.33 3.47 1.30
1994 37.90 48.30 53.57 22.35 20.79 20.59 25.39 20.06 17.68 4.47 3.79 2.63 4.69 2.78 0.94
1995 36.02 47.15 54.78 23.21 21.86 20.24 24.50 18.75 15.96 5.34 3.78 2.23 5.39 2.96 1.53
1996 33.02 43.28 51.32 21.81 23.31 19.12 29.74 20.71 20.31 5.57 4.17 2.59 5.07 3.20 1.71
1997 29.54 40.63 48.21 20.92 21.18 20.65 33.45 24.17 22.22 6.23 4.76 2.46 5.30 3.25 1.82
1998 27.84 40.26 47.00 19.84 18.39 17.87 35.72 29.43 26.00 6.36 4.13 2.87 5.08 2.52 1.74
1999 25.10 36.51 45.83 19.16 19.27 18.21 41.93 33.00 27.18 5.12 3.85 2.83 4.12 2.43 1.28
2000 23.05 35.04 44.19 18.31 20.30 18.37 42.53 33.63 29.29 7.58 3.86 2.45 4.42 2.10 1.05
2001 23.71 (*)
36.3 44.84 15.95 (*)
16.44 (*)
14.80 (*)
46.09 (*)
35.92 (*)
31.06 (*)
6.97 (*)
4.82 (*)
3.36 (*)
3.10 (*)
1.46 (*)
1.06
2002 25.77 34.95 45.52 18.18 18.95 17.23 39.32 34.20 27.74 6.46 5.65 3.89 3.91 1.42 0.91
2003 25.07 (*)
36.09 (*)
46.91 21.18 (*)
19.99 (**)
17.93 (**)
32.49 (*)
28.91 (*)
24.05 (*)
10.94 (*)
7.71 (*)
4.96 (*)
4.77 (*)
2.02 (*)
1.02 (*)
2004 21.80 32.46 42.09 23.08 22.98 20.22 31.98 26.45 23.92 13.28 10.06 7.12 4.79 3.13 1.29
135
Panel B: CEOs (% of Total Compensation)
Salary Bonus Stock Options Restricted Stocks LTIP Year
SP MD SM SP MD SM SP MD SM SP MD SM SP MD SM
1992 36.39 55.29 63.11 22.78 12.54 19.32 22.89 21.83 12.16 6.41 0.68 0.00 6.42 5.95 1.11
1993 37.64 49.22 52.17 23.31 19.42 21.22 23.07 17.80 16.71 4.62 4.06 4.05 5.01 4.45 0.96
1994 33.85 43.80 50.59 23.22 21.16 21.06 28.44 23.10 19.31 4.51 4.39 3.61 5.14 2.99 1.10
1995 31.82 42.46 51.36 24.21 23.06 21.53 26.76 21.42 18.21 5.75 4.99 2.32 5.91 3.67 1.88
1996 28.50 38.82 47.18 22.52 24.13 20.67 31.98 23.88 22.04 5.96 3.90 2.50 5.91 3.99 2.56
1997 24.16 36.09 43.77 21.93 22.33 23.10 36.40 26.05 23.53 6.70 5.16 2.96 6.04 4.30 2.35
1998 23.34 35.02 40.81 20.22 20.12 19.15 40.27 32.35 29.01 5.90 4.49 4.03 5.36 3.39 2.31
1999 20.33 31.81 40.49 19.35 20.57 19.15 46.09 35.90 30.01 4.94 3.76 3.78 4.31 3.10 1.48
2000 18.43 30.16 38.65 17.93 20.35 19.94 47.24 36.92 31.54 7.28 4.23 3.63 4.19 2.48 1.36
2001 18.63 32.12 39.44 14.86
(*) 15.40 (*)
14.99 (*)
51.80 (**)
39.99 35.01 (**)
7.09 5.52 3.92 3.38 (*)
1.57 (*)
1.16 (*)
2002 20.03 28.90 39.63 15.92 18.97 17.88 45.98 38.48 31.65 7.90 6.67 4.41 4.44 1.78 1.28
2003 18.83 30.18 40.25 17.35
(*) 21.02 (**)
18.91 36.93 (*)
31.08 (*)
29.04 (**)
11.88 (*)
9.16 (*)
5.99 (*)
5.56 (*)
2.61 (*)
1.29
2004 16.39 27.09 35.95 16.30 24.09 21.28 36.52 29.73 27.50 14.32 11.04 8.36 4.79 3.64 1.60
Panel C: Directors (% of Total Compensation)
Salary Bonus Stock Options Restricted Stocks LTIP Year
SP MD SM SP MD SM SP MD SM SP MD SM SP MD SM
1992 38.98 52.87 56.55 21.41 17.89 20.33 23.29 18.90 14.50 5.35 3.72 2.64 5.12 2.52 1.17
1993 38.23 49.52 54.31 23.22 21.12 21.05 22.86 16.58 16.10 5.34 3.99 2.97 4.63 3.56 1.08
1994 35.62 45.44 50.59 23.27 22.35 21.22 25.98 21.24 19.84 5.15 4.34 3.11 5.15 2.56 1.11
1995 33.59 44.98 51.29 23.29 22.80 21.00 25.58 29.59 17.87 5.91 4.52 2.23 5.61 3.08 1.58
1996 30.53 40.94 46.49 22.41 25.64 19.91 30.47 21.01 23.74 5.94 4.15 3.43 5.47 2.93 1.53
1997 26.17 38.95 44.59 21.86 23.66 22.56 34.62 23.67 23.32 6.74 4.76 2.87 5.35 2.83 2.60
1998 25.40 37.89 43.13 20.41 19.61 18.69 37.29 30.41 28.07 5.82 4.36 3.33 5.47 2.59 1.40
1999 22.99 34.70 42.30 20.23 20.47 18.93 42.37 33.93 29.02 5.38 3.25 3.15 3.97 2.72 1.24
2000 20.75 33.15 41.86 18.63 20.47 19.71 44.06 34.23 29.29 6.90 4.31 2.93 4.35 1.72 1.17
2001 20.73 (*)
35.49 (**)
41.11 (*)
16.15 (*)
16.58 (*)
15.92 (*)
48.53 (*)
35.95 32.68 (*)
6.47 (***)
5.26 (*)
3.84 (*)
3.09 (*)
1.35 (*)
0.98 (**)
2002 21.29 31.57 41.11 18.80 19.06 18.89 42.08 36.40 29.72 8.15 5.96 4.05 4.41 1.58 0.96
2003 21.10 33.55 41.24 21.81
(*) 21.12 (*)
19.90 35.05 (*)
29.74 (*)
26.74 (*)
11.56 (*)
7.97 (*)
5.73 (*)
4.64 1.86 (*)
1.13 (*)
2004 18.79 28.84 37.10 23.02 24.01 21.78 34.25 28.19 26.26 14.08 10.56 7.47 4.50 2.95 1.51
Note: Differences between years 2001 and 2000 (presented in year 2001 line) and between 2003 and 2002 (presented in 2003 line) are statistically significant at (*) 1% level, (**) 5% level and (***)10% level.
136
If we analyse the top five executives, we conclude that the most important
component of executive compensation in S&P Small Cap listed firms is salary. In the
case of S&P500 listed firms, until 1997, the most important component of executive
compensation is also salary, but after 1997 stock options become the most important
component, representing over 46% of the total compensation in year 2001. If we
analyse S&P Mid Cap listed firms, we see a decrease in salary (as a fraction of the total
compensation) from 55.90% in 1992 to 32.46% in 2004 and an increase of stock options
from 16.46% in 1992 to 33.63% in 2000 and 26.45% in 2004.
Essentially after year 2002 (SarbanesOxley Act), firms give fewer stock options
and more restricted stock and bonuses and generally these differences are statistically
significant (see the significance levels in table 2 comparing the values of 2003 with
those of 2002). Salary continues to decrease in all the cases but it is still the most
important component of executive compensation in S&P Small Cap listed firms
(42.09% of total compensation in 2004).
If we analyse CEOs and directors separately, we conclude that CEOs receive
more stock options than Directors and this compensation component represents over
50% of total executive compensation in S&P500 listed firms and over 30% in the case
of S&P Mid Cap and S&P Small Cap listed firms. The weight of stock options
decreases from 1992 to 2004 but still represents over 30% in S&P500 and S&P Mid
Cap companies and over 20% in S&P Small Cap companies in 2004.
Directors now receive more salary than CEOs and this is more significant in
small companies. They also receive more bonuses than CEOs, but the difference is not
so pronounced as is in the case of salary. In the case of stock options, CEOs receive, in
the later years, a little more than Directors and this is also the case for restricted stocks.
From these results we can conclude that essentially after the introduction of
SarbanesOxley in 2002, the structure of executive compensation changed, reducing the
use of stock options and increasing the use of bonus and restricted stocks.
137
4. Research Design
We control for firm size making an individual analysis for each of the three S &
P listed index firms S&P 500, S&P Mid Cap, and S&P Small Cap index. We attempt
to identify the determinants of executive compensation in each of S & P subgroups. We
also perform separate analysis for CEOs and Directors, as did Cheng and Hung (2006),
because both groups of executives have different characteristics.
We use unbalanced panel data with Fixed Effect Regression Model, also called
within estimator or Least Squares Dummy Variable (LSDV).
4.1. Dependent Variables
We use three different classifications for executive compensation, the dependent
variables, which are LN (Total Compensation), LN (Short Term Compensation) and LN
(Option Ratio).
LN (Total Compensation) is the total of the remunerations received by the
executives and is the sum of salary, bonus, stocks options, restricted stocks, LTIP, other
annual compensation and all other compensation. We used this variable like Chen and
Hung (2006). This variable, without logarithms, was also used by Aggarwal and
Smawick (1999) to evaluate the contracts offered to executives in a context of strategic
competition between products and evaluation of relative performance, and by Fields and
Fraser (1999) to unmask the commercial banks when they attributed compensations to
link executives to the performances. LN (Short Term Compensation) is the LN
(salary+bonus). Salary and bonus are considered shortterm remunerations, and they are
usually received in money. We used this variable in a similar way to Aggarwal and
Samwick (1999), Stathopoulos et al. (2004) and Chen and Hung (2006). Finally, we
also used LN (Option Ratio). We define option ratio as the value of options received by
the executive divided by the total compensation and this variable was also used by Chen
and Hung (2006).
Each one of the above dependent variables will be confronted separately with a
group of independent financial and governance variables with the intention of finding,
138
in a more reliable way, possible differences of executive compensation between
S&P500, S&P Mid Cap and S&P Small Cap listed firms.
Essentially, we analyse:
S&P500
LN(Cash Compensation)
LN(Option Ratio)
LN(Total Compensation)
S&PMidCap
LN(Cash Compensation)
LN(Option Ratio)
LN(Total Compensation
Executive
Compensation
S&PSmallCap LN(Cash Compensation)
LN(Option Ratio)
LN(Total Compensation)
0 1
2 3
4 5
6 7 8 9
( ) * ( ) * ( ) * 1 * ( ) * ( ) * ( ) * * * * * (1993...2004)
LN Compensation LN Not Exercised Ratio LN Ajex Tra yr LN Bs Volatility LN Number Mtgs LN Tenure ROA Pdirpens Interlock Industry Dummys Years Dummy
f
β β β β β β β β β β β β
= + + + + +
+ + + + + + +
+ + + + + ε
The dependent variable LN (Compensation) can assume the values of LN
(Total Compensation), LN (Option Ratio) and LN (Short Term Compensation) and f is the fixed effect.
4.2. Independent Variables
In order to explore the determinants of executive compensation for firms from
the S&P 500, S&P Mid Cap and S&P Small Cap listed indexes, we use the following
independent variables.
139
4.2.1. Financial Variables
To analyse the relationship between firm performance and executive
compensation, we use the variables ROA, which is the net income before extraordinary
items and discontinued operations divided by Total Assets. This variable was employed
by Zenner (2001) as one of the most commonly used variables to analyse the
relationship between executive compensation and performance. We expect a positive
relationship between executive compensation and firm performance, because better
returns can lead to higher compensation.
We also use the variable LN(Not Exercised Ratio), which is the natural
logarithm of the number of unexercised options that the executive held at year end that
were vested, divided by the aggregate number of stock options/stock appreciation rights
granted. We expect this variable to have a negative relationship with total compensation
and options ratio, meaning that if the executive has stock options that are not exercised,
the firm will probably give fewer stock options in the future.
We also analyse the relationship between the volatility of firm stock return and
executive compensation. To do this we use the variable LN (BS Volatility), which is the
natural logarithm of the standard deviation volatility calculated over 60 months with the
methodology of Black and Scholes. This variable was used by authors like Chen (2004)
without natural logarithm. If the volatility of firms’ stock returns increases, the value of
stock options will also increase and eventually shareholders will prefer to provide
incentives for their executives with more stock options and with less salary and bonus
(shortterm compensation). In this way, we expect a positive relationship between risk
and option ratio, an inherent positive relationship with total compensation, and a
negative relationship with shortterm compensation.
We also analyse the relationship between the return to shareholders and
executive compensation. We use the variable (Trs1yr), which is the oneyear total return
to shareholders, including the monthly reinvestment of dividends. We expect that, if
shareholders receive a high return from their investments in the company, they do not
need to give more stock options to executives to align their interests with executives’
interests. Based on this, we expect a negative relationship between option ratio, total
compensation and the oneyear shareholders’ return and a small positive relationship
140
with cash compensation in the sense that companies will probably give some money to
compensate executives’ efforts, but they do not need to give more stock options to
motivate them.
LN (Ajex) is the natural logarithm of the ratio used to adjust pershare data for
all stock splits that have occurred subsequent to the end of the company’s fiscal year.
We expect that executives might influence the determination of this ratio and they will
probably do so in the sense of having personal benefits. Based on this, we expect a
positive relationship between LN (Ajex) and the dependent variables.
We also use a dummy variable for each year between 1993 and 2004, as did
Barron and Waddel (2003) and Grinstein and Hribar (2004), to control for the effect of
the time in terms of executive compensation because we expect that time will be one of
the important factors in explaining the evolution of executive compensation during the
analysed period. Based on Fama and French (1997) industry classification we also used
dummy variables to fix the industry effect.
4.2.2. Governance Variables
We also tested the influence of a number of variables related to corporate
governance in executive compensation.
One of the most important topics in terms of corporate governance and executive
compensation is the influence of the Board and Compensation Committee’s
composition on executive compensation. This relationship was analysed by Ryan Jr and
Wiggins III (2004), who found that the determination of CEO compensation was related
to the power and influence that s/he has on the Board. They found evidence that firms
with external directors on the Board pay more compensation based on stock options and
restricted stocks. Anderson and Bizjak (2003) also analysed whether board
independence promotes shareholders’ interests and whether the presence of the CEO on
the Compensation Committee is related to opportunistic behaviour. They did not find
evidence that when executives leave the Compensation Committee, their remuneration
decreases.
To analyse this relationship we used two variables: LN (Number Mtgs) and
Interlock. LN (Number Mtgs) is the natural logarithm of the number of board meetings
141
held during the indicated fiscal year. The ability of the executive to take decisions is
affected by the number of board meetings during the year. According to Davidson,
Pilger and Szakmary (1998), board members align more with shareholders` interests
when they have more meetings during the year.
We also used the dummy variable Interlock, which assumes the value one if the
named executive is on two different Boards or Compensation Committees, and zero if
not. Based on the findings of Hallock (2007), we expect that the executives who are on
two boards (interlocked) will receive more compensation than the executives who are
not.
We also analyse the influence of the number of years of the tenure of a CEO
(LN (Tenure)) on executive compensation. This variable was applied by a significant
number of authors to explain executive compensation including Chindambaran and
Prabhala (2003), Ryan and Wiggins (2004), Murphy (1986) and Barro and Barro
(1990). We expect a positive relationship between executive compensation from
S&P500, S&P Mid Cap and S&P Small Cap and the number of years as CEO.
Finally, we test the influence in terms of compensation if the company has a
director pension/retirement plan (Pdirpens). This is a dummy variable that assumes the
value equal to 1 when it is “true” and 0 when it is not. We expect a negative relationship
between Pdirpens and executive compensation because, if the firm already pays a
pension plan for the executive, this will give the firm reason not to increase the present
compensation.
142
Expected correlations
Dependent Variables Independent Variables LN(Total
Compensation) LN(Option ratio) LN(Short term Compensation)
LN (Not Exercised Ratio) + LN(Ajex) + + + Trs1yr + LN(Bs Volatility) + + LN(Number Mtgs)
LN (Tenure) + +
ROA + + + Pdirpens Interlock + + + Year Dummy + + +
Table 14 describes Pearson correlations between independent variables and we
do not find high correlations between these variables.
143
5. Empirical Results
5.1. Summary Statistics
Table 3 presents some statistics to better understand the differences between the
firms listed on S&P500, S&P Mid Cap and S&P Small Cap indexes.
Table 3 Statistics from S&P500, S&P Mid Cap and S&P Small Cap Listed Firms
The table displays descriptive statistics from firms that belong to the S&P 500, MidCap 400 and SmallCap 600 indexes. LN (BS Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method. LN (Number Mtgs) is the natural logarithm of the number of the board meetings. LN(Shrownpc) is the natural logarithm of executive ownership. LN (Tenure) is the natural logarithm of the number of years as CEO. LN (Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted. Trs1tyr is the oneyear total return to shareholders, including the monthly reinvestment of dividends. ROA is the net income before extraordinary items and discontinued operations divided by total assets. Interlock is a dummy variable that assumes the value 1 when true, indicating that the executive is involved in a relationship requiring disclosure in the Compensation Committee Interlocks and Insider participation and 0 when not. Empl is the number of firm employees. Divyield is the dividends per share by exdate divided by close price for the fiscal year. Ni5ls is the 5year least squares annual growth rate of net income. Page2 is the executive age. Pdirpens is a dummy that assumes the value equal to 1 when it is true that the company pays into a directors’ pension plan and 0 when not.
S&P500 S&P MidCap S&P SmallCap
LN (BSs Volatility) 1.148 1.040 0.881 LN(Number Mtgs) 1.968 1.870 1.829 LN(Shrownpc) 0.437 0.419 0.342 LN(Tenure) 2.112 2.161 2.119 LN (Not Exercised Ratio) 0.736 0.690 0.677 Trs1yr 26.838 28.166 25.903 ROA 5.357 5.036 4.605 Interlock 0.020 0.030 0.0314 Empl 40.758 10.610 4.939 Divyield 1.668 1.501 1.271 Ni5ls 18.613 19.333 18.664 Page 2 56.431 55.910 55.300 Pdirpens 0.183 0.0970 0.0671
From table 3, we can conclude that S&P500 listed firms have a higher stock
return volatility and the boards have more meetings during the year. Executives from
large companies also have more stock firm ownership than executives from S&P Mid
144
Cap and S&P Small Cap listed firms. The number of years of the executives’ tenure
(Tenure) and also the executive average age is similar in all the three groups.
Executives from large companies have more stock options vested but not exercised than
executives from other subgroups. In terms of the oneyear return to shareholders the
medium size companies produce higher returns to shareholders.
The ROA is higher in S&P 500 firms than S&P Mid Cap or S&P Small Cap
listed firms and the executives from Small Cap listed firms are more interlocked.
Large companies give more dividends than small and medium size companies,
however; the medium size companies exhibit higher increase in net income.
Finally, firms from S&P500 give better pension plans to executives than other
companies.
5.2. Determinants of Executive Compensation in S&P500, S&P Mid Cap and S&P
Small Cap Listed Firms
Based on the summary statistics described in table 3, we develop the idea that if
S&P500 listed firms are so different in terms of growth of net incomes, dividends, etc,
from S&P Mid Cap and Small Cap companies, the factors that explain executive
compensation in these three groups will probably be different.
We use the Hausman Test to detect whether it is better to use a fixed or random
effect regression model. The results are favourable for using fixed effect regression
analysis, also called within method or Least of Square Dummy Variable Regression
(LSDV). In all the regressions, Standard errors are corrected using period Seemingly
Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for
both period heteroskedasticity and general correlation of observations within a given
cross section (Beck and Katz, 1995). In this section, the results of the multivariate
analyses are described in tables 4 to 11, and in tables 12 and 13 we describe whether the
difference between the coefficients values of these Fixed Effect Regressions are
statistically significant.
145
Table 4 Fixed Effect Regression Analysis of Determinants of CEO Compensation for
S&P500 Listed Firms Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and 1982 as base year, we adjust the monetary variables for inflation reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN(Ajex) is the natural logarithm of the ratio used to adjust pershare data for all stock splits that have occurred subsequent to the end of the company’s fiscal year; Tts1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings. LN (Tenure) is the natural logarithm of the number of years as CEO. ROA is the net income before extraordinary items and discontinued operations divided by total assets. Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not. Interlock is a dummy variable that assumes the value equal to 1 when the executive is on two different boards at the same time and 0 when not. We control for time effect inserting one dummy for each year between 1993 and 2004 and for industry effect using one dummy for each industry per Fama and French (1997) industry classification. To distinguish between executives from S&P500, S&PMidcap and S&PSmallCap, we used the SPCODE variable from the ExecuComp database that considers firms from S&P500 those with code SP, S&PMidCap with code MD and S&PsmallCap with code SM.
LN (Total Compensation) t Statistics LN (Option
Ratio) t
Statistics LN (Short Term Compensation)
t Statistics
Constant 8.840* 22,735 3,256 1.171 8,560* 4,391 LN (Not Exercised Ratio) 0.245* 26.930 0.315* 35.246 0.007 0.920 LN(Ajex) 0.227* 8.313 0.035 1.354 0.028 1.203 Trs1yr 0.001* 3.749 0.001* 8.031 0.001* 10.004 LN(Bs Volatility) 0.028 0.430 0.192* 3.100 0.317* 5.600 LN(Number Mtgs) 0.033 0.972 0.046 1.438 0.074* 2.373 LN(Tenure) 0.265 0.276 1,027 0,897 0,666 0.830 ROA 0.001 1.330 0.0006 0.630 0.006* 6.952 Pdirpens 0.043 1.136 0.101* 2.788 0.027 0.868 Interlock 0.136** 2.574 0.108** 2.124 0.015 0.326 Year1993 0.108** 2.524 0.002 0.025 0.043 1.431 Year1994 0.314* 6.793 0.226* 3.763 0.139* 4.143 Year1995 0.368* 7.715 0.162* 2.589 0.113* 3.096 Year1996 0.574* 11.504 0.292* 4.429 0.182* 4.758 Year1997 0.739* 13.508 0.334* 5.049 0.276* 6.874 Year1998 0.846* 15.478 0.437* 6.614 0.402* 9.556 Year1999 1.072* 18.803 0.613* 9.102 0.478* 10.070 Year2000 1.163* 18.899 0.556* 7.801 0.589* 11.830 Year2001 1.296* 20.491 0.669* 9.276 0.552* 9.912 Year2002 1.301* 20.282 0.620* 8.416 0.673* 12.170 Year2003 1.317* 19.764 0.568* 7.478 0.707* 12.825 Year2004 1.510* 22.735 0.622* 8.160 0.856* 15.541 Candy Dummy 0,500 1,148 0,396 0,644 0,512*** 1,694 Computer Dummy 0,242 0,133 3,019 1,402 1,312 0,866 Electronic Dummy 0,758 1,112 0,398 0,599 1,480** 2,358
N 3841 3841 3841 Adjusted R square 80.30% 71.07% 74.49%
*Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
146
Table 5 Fixed Effect Regression Analysis of Director Compensation Determinants in
S&P500 Listed Firms Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and 1982 as base year, we adjust the monetary variables for inflation reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN(Ajex) is the natural logarithm of the ratio used to adjust pershare data for all stock splits that have occurred subsequent to the end of the company’s fiscal year; Tts1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings; LN (Tenure) is the natural logarithm of the number of years as CEO. ROA is the net income before extraordinary items and discontinued operations divided by total assets; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy variable that assumes the value equal to 1 when the executive is on two different boards at the same time and 0 when not. We control for time effect inserting one dummy for each year between 1993 and 2004 and for industry effect using one dummy for each industry per Fama and French (1997) industry classification. To distinguish between executives from S&P500, S&PMidcap and S&PSmallCap, we used the SPCODE variable from the ExecuComp database that considers firms from S&P500 those with code SP, S&PMidCap with code MD and S&PSmallCap with code SM.
LN(Total Compensation)
t Statistics LN(Option Ratio t
Statistics LN(Shor t Term Compensation)
t Statistics
Constant 8.035* 2,930 3,093 1,083 7,982* 3,490 LN (Not Exercised ratio) 0.246* 30.827 0.299* 41.115 0,004 0,552 LN(Ajex) 0.195* 7.679 0.064* 2.901 0.019 0,757 Trs1yr 0.001* 4.607 0.001* 8.084 0.001* 9.880 LN(Bs Volatility) 0.117*** 1.896 0.244* 4.606 0,367* 6.196 LN(Number Mtgs) 0.003 0.098 0.044 1.544 0.100* 3.277 LN(Tenure) 0.063 0.052 1,026 0,822 0.051 0.521 ROA 0.004* 4.799 0.001 0.954 0.006* 6.341 Pdirpens 0.057*** 1.695 0.057*** 1.901 0.020 0.611 Interlock 0.103** 2.085 0.093** 2.068 0,026 0,563 Year1993 0.091* 2.684 0.020 0.475 0.063** 2.282 Year1994 0.298* 7.867 0.214* 5.465 0.181* 6.151 Year1995 0.361* 9.029 0.179* 4.141 0.167* 5.066 Year1996 0.543* 12.605 0.334* 7.405 0.245* 6.886 Year1997 0.749* 16.131 0.392* 8.518 0.331* 8.599 Year1998 0.884* 18.931 0.468* 10.230 0.468* 11.070 Year1999 1.111* 22.397 0.628* 13.501 0.553* 12.203 Year2000 1.263* 23.493 0.565* 11.171 0.678* 13.612 Year2001 1.403* 25.109 0.680* 13.181 0.656* 12.024 Year2002 1.426* 25.156 0.623* 11.797 0.820* 15.083 Year2003 1.427* 24.303 0.585* 10.648 0.869* 16.015 Year2004 1.621* 27.645 0.632* 11.367 1.015* 18.474 Candy Dummy 0,024 0,045 0,050 0,074 0,064 0,150 Computer Dummy 0,212 0,094 3,398 1,452 0,904 0,478 Consumer Dummy 0,829 1,013 0,940 1,201 1,426*** 1,861 Eelectronic Dummy 0,841 1,092 0,350 0,486 1,515** 2,087
N 5207 5207 5207
Adjusted RSquare 78.75% 70.40% 67.06% *Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
147
Table 6 Fixed Effect Regression Analysis of Determinants CEO Compensation for S&P
Mid Cap Listed Firms Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and 1982 as base year, we adjust the monetary variables for inflation reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN(Ajex) is the natural logarithm of the ratio used to adjust pershare data for all stock splits that have occurred subsequent to the end of the company’s fiscal year; Tts1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings; LN (Tenure) is the natural logarithm of the number of years as CEO. ROA is the net income before extraordinary items and discontinued operations divided by total assets; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy variable that assumes the value equal to 1 when the executive is on two different boards at the same time and 0 when not. We control for time effect inserting one dummy for each year between 1993 and 2004 and for industry effect using one dummy for each industry per Fama and French (1997) industry classification. To distinguish between executives from S&P500, S&PMidcap and S&PSmallCap, we used the SPCODE variable from the ExecuComp database that considers firms from S&P500 those with code SP, S&PMidCap with code MD and S&PSmallCap with code SM.
LN (Total Compensation)
t Statistic
s
LN (Option Ratio
t Statistics
LN (Shor t Term Compensation)
t Statistics
Constant 3.753* 5.994 2.934* 4.959 5.482* 12.203
LN (Not Exercised Ratio) 0.225* 19.624 0.353* 26.917 0.010 1.195
LN(Ajex) 0.186* 4.341 0.157* 3.241 0.058*** 1.756 Trs1yr 0.00002 1.096 0.00001 0.702 0.0003*** 1.722 LN(Bs Volatility) 0.050 0.574 0.229** 2.358 0.114*** 1.674 LN(Number Mtgs) 0.038 0.904 0.036 0.770 0.028 0.855 LN(Tenure) 1.435* 5.630 0.914* 3.788 0.357*** 1.898 ROA 0.009* 6.288 0.0002 0.146 0.012* 10.873 Pdirpens
0.040 0.604 0.056 0.757 0.065 1.319 Interlock 0.052 0.831 0.076 1.069 0.059 1.250 Year1993 0.121 0.694 0.045 0.285 0.138** 2.088 Year1994 0.336*** 1.954 0.101 0.631 0.203* 3.178 Year1995 0.338*** 1.944 0.009 0.056 0.230* 3.518 Year1996 0.507* 2.907 0.160 1.000 0.300* 4.579 Year1997 0.633* 3.625 0.139 0.856 0.384* 5.715 Year1998 0.747* 4.280 0.318*** 1.961 0.466* 6.955 Year1999 0.918* 5.228 0.412** 2.546 0.538* 7.902 Year2000 1.090* 6.116 0.321** 1.911 0.658* 9.251 Year2001 1.149* 6.432 0.472* 2.841 0.610* 8.357 Year2002 1.236* 6.893 0.428** 2.566 0.743* 10.072 Year2003 1.222* 6.804 0.394** 2.334 0.749* 9.919 Year2004 1.403* 7.784 0.430** 2.547 0.898* 11.583
N 1997 1997 1997
Adjusted RSquare 72.64% 67.70% 70.19%
*Significant at 1% level, ** significant at 5% level, *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
148
Table 7 Fixed Effect Regression Analysis of Determinants of Director Compensation for
S&P Mid Cap Listed Firms Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and 1982 as base year, we adjust the monetary variables for inflation reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN(Ajex) is the natural logarithm of the ratio used to adjust pershare data for all stock splits that have occurred subsequent to the end of the company’s fiscal year; Tts1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings. LN (Tenure) is the natural logarithm of the number of years as CEO. ROA is the net income before extraordinary items and discontinued operations divided by total assets. Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not. Interlock is a dummy variable that assumes the value equal to 1 when the executive is on two different boards at the same time and 0 when not. We control for time effect inserting one dummy for each year between 1993 and 2004 and for industry effect using one dummy for each industry per Fama and French (1997) industry classification. To distinguish between executives from S&P500, S&PMidcap and S&PSmallCap, we used the SPCODE variable from the ExecuComp database that considers firms from S&P500 those with code SP, S&PMidCap with code MD and S&PsmallCap with code SM.
LN (Total Compensation) t Statistics LN (Option
Ratio t
Statistics LN (Short Term Compensation)
t Statistics
Constant 3.421* 5.983 2.437* 4.211 4.739* 9.878 LN (Not Exercised Ratio) 0.245* 23.625 0.333* 30.245 0.008 0.956 LN(Ajex) 0.193* 5.021 0.217* 5.374 0.035 1.105 Trs1yr 0.0000003 0.015 0.00001 0.730 0.000** 2.245 LN(Bs Volatility) 0.036 0.509 0.364* 4.898 0.153** 2.571 LN(Number Mtgs) 0.048 1.242 0.004 0.104 0.066** 2.051 LN(Tenure) 1.659* 6.627 0.820* 3.234 0.700* 3.332 ROA 0.010* 8.550 0.001 0.784 0.009* 10.037 Pdirpens 0.086 1.445 0.010 0.153 0.065 1.300 Interlock 0.023 0.380 0.100 1.524 0.045 0.874 Year1993 0.064 1.003 0.065*** 0.931 0.052 1.160 Year1994 0.300* 4.673 0.139 1.954 0.146* 3.205 Year1995 0.321* 4.696 0.058* 0.859 0.179* 3.631 Year1996 0.518* 7.550 0.225* 3.278 0.277* 5.522 Year1997 0.675* 9.510 0.214* 2.901 0.372* 7.415 Year1998 0.776* 10.749 0.311* 4.351 0.438* 8.411 Year1999 0.909* 12.622 0.413* 5.841 0.498* 0.935 Year2000 1.064* 13.522 0.294* 3.613 0.643* 11.119 Year2001 1.156* 14.567 0.446* 5.713 0.604* 10.116 Year2002 1.258* 15.689 0.386* 4.852 0.757* 12.512 Year2003 1.224* 15.221 0.381* 4.694 0.760* 12.346 Year2004 1.435* 17.390 0.429* 5.230 0.889* 13.803
N 2669 2669 2669 Adjusted Rsquare 70.49% 64.98% 64.63%
*Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995).
149
Table 8 Fixed Effect Regression Analysis of Determinants of CEO Compensation for S&P
Small Cap Listed Firms
Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and 1982 as base year, we adjust the monetary variables for inflation reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN(Ajex) is the natural logarithm of the ratio used to adjust pershare data for all stock splits that have occurred subsequent to the end of the company’s fiscal year; Tts1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings; LN (Tenure) is the natural logarithm of the number of years as CEO. ROA is the net income before extraordinary items and discontinued operations divided by total assets; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy variable that assumes the value equal to 1 when the executive is on two different boards at the same time and 0 when not. We control for time effect inserting one dummy for each year between 1993 and 2004 and for industry effect using one dummy for each industry per Fama and French (1997) industry classification. To distinguish between executives from S&P500, S&PMidcap and S&PSmallCap, we used the SPCODE variable from the ExecuComp database that considers firms from S&P500 those with code SP, S&PMidCap with code MD and S&PSmallCap with code SM.
LN (Total Compensation)
t Statistics
LN (Option Ratio t Statistics LN (Shor t Term
Compensation) t
Statistics Constant 6.492* 5.054 4.527* 3,233 2,158** 2,238 LN (Not Exercised Ratio) 0.243* 22.273 0.332* 25.975 0.008 0.826 LN(Ajex) 0.309* 6.236 0.310* 5.380 0.046 1.110 Trs1yr 0.001* 3.773 0.001* 5.404 0.001* 7,095 LN(Bs Volatility) 0.161** 2.194 0.020 0.237 0.119*** 1,957 LN(Number Mtgs) 0.070*** 1.811 0.064 1.144 0.017 0.515 LN(Tenure) 0.375 0,699 2.190* 3.791 3.443* 8.541 ROA 0.003* 3.187 0.002*** 1.914 0.004* 5.605 Pdirpens 0.092 1.096 0.067 0.702 0.119*** 1.697 Interlock 0.098 1.458 0.214* 2.839 0.060 1.021 Year1993 0.183 1.191 0.005 0.018 0.146 1.431 Year1994 0.039 0.254 0.015 0.053 0.289* 2.832 Year1995 0.012 0.078 0.056 0.201 0.276* 2.603 Year1996 0.092 0.586 0.014 0.050 0.313* 2.983 Year1997 0.155 0,988 0.015 0.053 0.361* 3.434 Year1998 0.214 1.357 0.005 0.019 0.450* 4.260 Year1999 0.328** 2.084 0.060 0.213 0.502* 4.715 Year2000 0.410* 2.579 0.038 0.137 0.565* 5.244 Year2001 0.496* 3.123 0.135 0.482 0.533* 4.931 Year2002 0.508* 3.191 0.164 0.585 0.629* 5.779 Year2003 0.533* 3.350 0.141 0.506 0.665* 6.104 Year2004 0.770* 4.803 0.119 0.424 0.807* 7.369 Apparel ( Dummy) 0,619 1,266 1,330** 2,514 1,582* 4,082
N 2209 2209 2209
Adjusted RSquare 75.49% 71.91% 70.60%
*Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
150
Table 9 Fixed Effect Regression Analysis of Determinants of Director Compensation for
S&P Small Cap Listed Firms
Data is from the ExecuComp database from 1992 to 2004. We used Unbalanced Panel Data Fixed Effect Regression Analysis. Using the Consumer Price Index (CPI), compiled by the Bureau of Labor Statistics, and 1982 as base year, we adjust the monetary variables for inflation reporting the values to the year 2004. Dependent variables are LN (Total Compensation), LN (Short Term Compensation) and LN (Option Ratio). LN (Total Compensation) is the natural logarithm of total executive compensation. LN (Short Term Compensation) is the natural logarithm of Salary and Bonus. LN (Option Ratio) is the natural logarithm of the value of options granted to the executive divided by total compensation. The independent variables are: LN(Not Exercised Ratio) is the natural logarithm of the number of unexercised options that the executive held at year end that were vested, divided by the aggregate number of stock options/stock appreciation rights granted; LN(Ajex) is the natural logarithm of the ratio used to adjust pershare data for all stock splits that have occurred subsequent to the end of the company’s fiscal year; Tts1tr is the oneyear total return to shareholders, including the monthly reinvestment of dividends; LN (Bs Volatility) is the natural logarithm of standard deviation volatility calculated over 60 days with the Black Scholes method; LN (Number Mtgs) is the natural logarithm of the number of board meetings; LN (Tenure) is the natural logarithm of the number of years as CEO. ROA is the net income before extraordinary items and discontinued operations divided by total assets; Pdirpens is a dummy that assumes the value equal to 1 when it is true and zero when not; Interlock is a dummy variable that assumes the value equal to 1 when the executive is on two different boards at the same time and 0 when not. We control for time effect inserting one dummy for each year between 1993 and 2004 and for industry effect using one dummy for each industry per Fama and French (1997) industry classification. To distinguish between executives from S&P500, S&PMidcap and S&PSmallCap, we used the SPCODE variable from the ExecuComp database that considers firms from S&P500 those with code SP, S&PMidCap with code MD and S&PsmallCap with code SM.
LN (Total Compensation)
t Statistics
LN (Option Ratio)
t Statistics
LN (Shor t Term Compensation)
t Statistics
Constant 6.091* 4.318 3,305** 2,202 1.785 1.511
LN (Not Exercised Ratio) 0.235* 25.758 0.318* 32.193 0.009 1.141 LN(Ajex) 0.281* 6.772 0.331* 7.313 0.050 1.400 Trs1yr 0.0005* 4.011 0.0008* 5.955 0.001* 8.423 LN(Bs Volatility) 0.178* 2,825 0.158** 2.242 0.055 1.015 LN(Number Mtgs) 0.017 0.501 0.025 0.662 0.078* 2.543 LN(Tenure) 0,467 0.749 1.818* 2.747 3.524* 6.769 ROA 0.004* 4.698 0.0002 0.256 0.005* 7.583 Pdirpens 0.079 1.030 0.085 1.038 0.041 0.616 Interlock 0.084 1.346 0.139** 2.097 0.077 1.398 Year1993 0.024 0.326 0.066 0.824 0.035 0.685 Year1994 0.216* 2,891 0.159*** 1.868 0,176 1,398 Year1995 0.184* 2.406 0.098 1.161 0.166* 2.904 Year1996 0.324* 4.243 0.246* 3.053 0.210* 3,680 Year1997 0.415* 5.448 0.191** 2.314 0.257* 4,212 Year1998 0.426* 5.501 0.183** 2.211 0.359* 6.196 Year1999 0.571* 7.307 0.236* 2.815 0.397* 6.546 Year2000 0.654* 8.013 0.201** 2.370 0.448* 7.201 Year2001 0.766* 9.470 0.308* 3.572 0.440* 7.028 Year2002 0.818* 10.053 0.333* 3.802 0.574* 8.958 Year2003 0.830* 10.088 0.305* 3.480 0.582* 8.964 Year2004 1.098* 12.982 0.317* 3.519 0.760* 11.497 Apparel Dummy 0,695* 1,297 0,972*** 1,670 0,955 2,173
N 3003 3003 3003
Adjusted RSquare 72.29% 69.64% 66.96% *Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995)
151
In tables 12 and 13, we analyse whether the regression coefficients, which explain
the variation in executive compensation, are the same for each of three S&P index
firms. That is, we are interested in determining whether explanatory factors have a
similar impact on response variables across all S&P firms. In most cases, the
coefficients are different and this difference is statistically significant at 1% level.
5.3. Impact of NASDAQ Crash on the Determinants of Executive Compensation
In tables 10 and 11 we analyse whether the NASDAQ crash changed the
determinants and the associated impact on total executive compensation.
152
Table 10
Fixed Effect Regression Analysis of CEO Total Compensation Determinants Before and After NASDAQ Crash in 2000 for S&P Listed Firms
S&P500 S&P MidCap S&PSMallCAP
1992 to 2000 2001 to 2004 1992 to 2000 2001 to 2004 1992 to 2000 2001 to 2004 Independent
Variables Ln(Total Comp.)
(t statistics)
Ln(Total Comp.) (t statistic)
Ln(Total Comp.)
(t statistic)
Ln(Total Comp.)
(t statistic)
Ln(Total Comp.)
(t statistic)
Ln(Total Comp.)
(t statistic)
Constant 9,434* (2,970) 14,339*
(4,955) 7,193* (29,674)
8,734* (39,759)
7,863* (3,358)
7,584* (40,093)
LN (Not Exercised Ratio)
0,246* (22,937)
0,300* (13,835)
0,201* (12,825)
0,260* (12,383)
0,236* (15,017)
0,284* (15,752)
LN(Ajex) 0,283* (8,791)
0,269* (2,771)
0,236* (3,864)
0,144 (1,444)
0,271* (3,989)
0,337* (2,863)
Trs1yr 0,001* (4,496)
0,0004 (1,144)
0,00007 (0,317)
0,00001 (1,056)
0.0005** (2,431)
0,00005 (0,189)
LN(Bs Volatility) 0,177** (2,087)
0,045 (0,234)
0,070 (0,610)
0,345 (1,619)
0.345* (3,378)
0,052 (0,320)
LN(Number Mtgs) 0,065 (1,311)
0,099*** (1,844)
0,066 (1,122)
0,067 (0,994)
0.010 (0,159)
0,113** (2,012)
LN(Tenure) 0,442 (0,373)
2,385 (1,644)
0.072 (0,079)
ROA 0,011* (5,971)
0,004* (3,265)
0,011* (4,637)
0,002 (0,876)
0.003* (2,585)
0,006* (3,345)
Pdirpens 0,002 (0,041)
0,141 (1,281)
0,053 (0,701)
0,112 (0,567)
0.029 (0,255)
0,010 (0,031)
Interlock 0,080 (1,310)
0,015 (0,129)
0,063 (0,804)
0,008 (0,0559)
0.169 (1,291)
0,018 (0,189)
Year1993 0,114* (2,660)
0,144 (1,988)
0.150 (1,031)
Year1994 0,323* (7,038)
0,329** (1,988)
0.066 (0,456)
Year1995 0,387* (8,076)
0,330** (1,965)
0.021 (0,143)
Year1996 0,610* (11,62)
0,478* (2,828)
0.144 (0,966)
Year1997 0,790* (14,209)
0,608* (3,571)
0.238 (1,600)
Year1998 0,871* (15,711)
0,727* (4,280)
0.267*** (1,764)
Year1999 1,078* (18,279)
0,892* (5,207)
0.389** (2,576)
Year2000 1,147* (17,538)
1,086* (6,201)
0.456* (2,959)
Year2001
Year2002 0,003 (0,085)
0,097** (2,364)
0,015 (0,431)
Year2003 0,087** (2,079)
0,149* (3,337)
0,064*** (1,689)
Year2004 0,286* (5,646)
0,397* (7,012)
0,269* (5,936)
Candy Dummy 0,475 (1,098)
Computer Dummy 0,149 (0,066)
N 2391 1450 1187 951 1089 1173
Adjusted R Square
83.13% 81.31% 76.04% 77.32% 78.47% 81.22%
*Significant at 1% level, ** significant at 5% level *** significant at 10% Note 1: Standard errors are corrected using period Seemingly Unrelated Regression (SUR) – Panel Corrected Standard Errors (PCSE): correction for both period heteroskedasticity and general correlation of observations within a given cross section (Beck and Katz, 1995).
153
Table 11 Fixed Effect Regression Analysis of Determinants of Directors´ Total Compensation Before and After NASDAQ Crash for S&P Listed Firms
S&P500 S&P MidCap S&PSmallCap
1992 to 2000 2001 to 2004 1992 to 2000 2001 to 2004 1992 to 2000 2001 to 2004 Independent Variables Ln(Total
Comp.) (t statistics)
Ln(Total Comp.)
(t statistic)
Ln(Total Comp.) (t statistic)
Ln(Total Comp.)
(t statistic)
Ln(Total Comp.)
(t statistic)
Ln(Total Comp.) (t statistic)
Constant 8,522** (2,441)
13,848* (4,889)
7,205* (57,585)
3,191* (3,396)
6,569** (2,509)
7,348* (56,320)
LN (Not Exercised Ratio)
0,251* (27,623)
0,294* (19,719)
0,226* (21,824)
0,291* (15,055)
0,264** (20,984)
0,271* (18,994)
LN(Ajex) 0,244* (8,511)
0,297 (3,945)
0,247 (6,402)
0,127 (1,554)
0,234* (4,326)
0,225* (2,590)
Trs1yr 0,001* (4,929)
0,000** (0,394)
0,0001 (0,788)
0,00002 (0,885)
0,0008* (4,371)
0,0002 (1,002)
LN(Bs Volatility) 0,083 (1,089)
0,290 (2,040)
0,067 (0,908)
0,021 (0,101)
0,372* (4,226)
0,113 (0,836)
LN(Number Mtgs) 0,125* (2,916)
0,060 (1,367)
0,020 (0,499)
0,020 (0,325)
0,012* (0,243)
0,071 (1,538)
LN(Tenure) 0,271 (0,195)
2,492** (1,695) 0,336*
(0,316)
ROA 0,014* (7,914)
0,0002 (0,242)
0,012* (8,150)
0,005* (3,725)
0,003 (2,906)
0,008* (5,449)
Pdirpens 0,011 (0,293)
0,099 (1,140)
0,057 (0,964)
0,293*** (1,802)
0,196 81,852)
0,179 (0,653)
Interlock 0,010 (0,180)
0,087 (0,923)
0,039 (0.639)
0,059 (0,478)
0,107 (1,209)
0,010 (0,120)
Year1993 0,103* (3,103)
0,086*** (1.948)
0,009 (0,131)
Year1994 0,312* (8,499)
0,315* (6.147)
0,252 (3,380)
Year1995 0,387* (9,872)
0,308* (6,147)
0,223 (2,918)
Year1996 0,587* (13,613)
0,478* (9,400)
0,391 (5,088)
Year1997 0,802* (17,212)
0,624* (11,553
0,505 (6,509)
Year1998 0,914* (19,510)
0734* (13,736)
0,511 (6,404)
Year1999 1,139* (22,668)
0,834* (14,596)
0,669 (8,254)
Year2000 1,257* (22,473)
1.027* (16.44388)
0,760 (8,932)
Year2001
Year2002 0,075* (2,638)
0,149* (3,838)
0,040 (1,525)
Year2003 0,108* (3,118)
0,153* (3,446)
0,109* (3,664)
Year2004 0,293* (7,375)
0,367* (6,528)
0,295* (7,651)
Apparel Dummy 0,747 (0,927)
Candy Dummy 0,007 (0,013)
Computer Dummy 0,280 (0,107)
Consumer Dummy 0,244 (0,323)
N 2241 2547 1343 1655 1909
Adjusted R square 80.46% 79.52% 70.00% 75.64% 83.52%
154
5.4. Analysis of the Results
In tables 4 through 11 we compare compensation for the CEOs and Directors of
S&P 500, S&P Mid Cap and S&P Small Cap listed firms. We find that the factors that
explain their compensation are not all the same and, in the case of the factors that are
the same, the intensity of the coefficients is different and this difference is statistically
significant (tables 12 and 13).
The number of stock options vested has a negative influence on executive
compensation in S&P 500, S&P Mid Cap and S&P Small Cap listed firms, and also on
the number of stock options that are granted to the executive. Essentially, when
executives have stock options that go unexercised due to market price staying below the
exercise price, companies reduce compensation based on stock options and increase
cash compensation with the purpose of not losing the executives and giving them
incentive again.
The ratio used to adjust pershare data for all stock splits has a negative
influence on total compensation for all the executives of S&P listed firms; however, it
has higher impact on small size companies. In the case of S&P Mid Cap and S&P Small
Cap firms, this ratio also negatively affects the number of stock options and the cash
compensation that executives receive.
The oneyear return to shareholders positively affects, on a small scale, both the
total and cash compensation of CEOs and directors of S&P 500 listed firms. But it
impacts only cash compensation for all S&P firms. The oneyear return to shareholder
is negatively related to the option ratio for S&P 500 and S&P Small Cap listed firms
meaning that when shareholders are satisfied with the returns generated by the firm on
their investment, they do not feel the need to give more incentive to the executives with
stock options to align executive interests with their own.
The volatility of stock returns has a positive relationship with the number of
stock options and a negative relationship with cash compensation for the executives of
S&P 500 and S&P Mid Cap listed companies, meaning that when volatility is high,
companies pay more with stock options and less with cash compensation. However,
only large companies have a positive and statistically significant relationship. We find
a positive relationship between stock return volatility and total compensation in the case
155
of S&P Small Cap CEOs and a negative relationship with S&P500 Directors. Option
ratio is positively influenced by stock return volatility in all the cases except CEOs and
Directors from S&P SmallCap listed firms. Stock return volatility is negatively related
to cash compensation in all the situations excluding S&P Small Cap Directors. We can
conclude that when volatility is higher, shareholders prefer to grant stock options to
executives and reduce the use of cash compensation.
The number of board meetings is generally negatively related to cash
compensation of CEOs and Directors for S&P500 firms, but for Directors only for S&P
Mid Cap and S&P Small Cap firms. Our findings are congruent with those of Ryan and
Wiggins (2001), and Chen and Hung (2006), in the sense that monitoring power can
reduce the necessity to motivate executives with more compensation to reduce agency
problems. According to Davidson, Pilger and Szakmary (1998), board members are
more aligned with shareholders´ interests when they have more meetings during the
year, and therefore CEO compensation is more controlled. We only find a positive
relationship between total compensation and the number of meetings in the case of
CEOs from S&P Small Cap listed firms, meaning that when the number of board
meetings is higher, CEOs from these firms will receive more.
We also analyse the impact of the number of years as CEO (Tenure) on total
compensation, option ratio and short term compensation. The number of years that
executives are leading the company (Tenure) has a strong and positive influence on
CEO and Directors’ total compensation in the case of medium size companies. The
number of stock options that they receive is also positively influenced by CEO tenure.
In the case of large size companies, the relationship is negative, meaning that when
executives have more years of experience as CEO, compensation decreases, as does the
number of stock options that they receive. The number of stock options granted is
positively related to CEO and Directors’ tenure in S&P Mid Cap firms. We also find a
positive relationship between cash compensation of CEO and Directors and tenure for
S&P Mid Cap and S&P Small Cap listed firms. We can conclude, like Chen and Hung
(2006) and Chung and Pruitt (1996), that tenure influences executive compensation.
ROA has a positive influence on total compensation in the case of all CEOs and
Directors, with the exception of CEOs from S&P500, but has a negative influence on
the number of stock options granted to executives for small size companies.
156
As we expected, the effect of the existence of a firm’s pension plan on total
compensation is negative, but only for CEOs and Directors from S&P500 and S&P
Small Cap listed firms. In the case of S&P Mid Cap listed firms, the coefficients are not
statistically significant. The results imply that when firms already contribute to a
pension plan for the executives, they do not increase the compensation. In the case of
CEOs and Directors of S&P 500 firms, the number of stock options granted to the
executives is also negatively influenced by the existence of a pension plan.
We find, like Hallock (1997), that CEO interlocking, being a member of two
different boards at the same time, positively influences the total compensation of CEOs
of S&P 500 firms. The number of options that executives receive is also positively
related to interlocking for CEOs and Directors of S&P 500 and only CEOs for S&P
Small Cap Listed firms. We can conclude that CEOs and Directors from big companies
can essentially extract benefits from interlock relationships.
Time has a positive effect on total compensation, the number of options and
shortterm compensation. Based on Fama and French (1997) industry classification, we
also analysed the effect of the industry in terms of pay to executives.
In tables 10 and 11 we find interesting results about the impact of the NASDAQ
crash on the variables that explain executive compensation. In the case of CEOs of S&P
500 firms, after the NASDAQ crash, the variables oneyear return to shareholders and
volatility are not significant, and the variable number of board meetings changes its sign
from positive to negative while explaining total compensation. This relationship is
statistically significant. In the case of S&P listed firms the most important changes are:
LN (Ajex) is not statistically significant after the NASDAQ crash but ROA becomes
statistically significant. In the case of S&P Small Cap firms, after the NASDAQ crash,
volatility is not statistically significant but the number of meetings becomes significant.
With regard to the subsample of S&P 500 Directors, LN (Ajex), the number of
board meetings and ROA are not statistically significant after the NASDAQ crash. In
the case of S&P Mid Cap firms, the number of board meetings and executive
interlocking are not statistically significant after the NASDAQ crash. The dummy
variable if firms have pension plans for executives (Pdirpens) changes from positive
influence before the NASDAQ crash to negative influence after the crash. Finally, in the
case of S&P Small Cap firms, the oneyear return to shareholders and the number of
157
board meetings are not statistically significant after the NASDAQ crash, but ROA is
now positively related to total compensation and this relationship is statistically
significant.
6. Conclusion
In this paper we analyse whether the total value and the components of executive
compensation across S&P500, S&P Mid Cap and S&P Small Cap listed firms are
statistically the same or different during the period from 1992 to 2004. We also examine
whether the total compensation and forms of compensation change after the NASDAQ
crash and the SarbanesOxley Act in 2002. Furthermore, we control for firm size and
industry effect based on the Fama and French (1997) industry classifications and
analyse whether the determinants that explain executive compensation are the same or
different across S&P large, medium, and small firms.
Our results reveal that the mean executive compensation and the component
weights (forms of compensation) are significantly different for firms across S&P500,
S&P Mid Cap and S&P Small Cap indexes. Total compensation and forms of
compensation change after the NASDAQ crash and enactment of the SO Act. In
general, salary and stock options decrease and the use of restricted stocks and bonuses
essentially increase after the SarbanesOxley Act, among S&P500 and also S&P Small
Cap firms, implying the effectiveness of the Act. Corporate salaries and stock option
awards were subject to a lot of public debate before the introduction of the SO Act and
it appears that firms made significant changes in their compensation packages in the
spirit of the Act.
Our results also reveal that, controlling for firm size and industry effect, the
factors that explain CEO and Director compensation in S&P500, S&P Mid Cap and
S&P Small Cap listed firms are, in general, not all the same and if some factors are the
same, the intensity of the coefficients is significantly different. We also find that the
NASDAQ crash changed the influence of some of the factors that explain executive
compensation.
158
7. References
Aggarwall, R. and Samwick. A.(1999). Executive compensation, strategic competition
and relative evolution: theory and evidence. The Journal of Finance, Vol. 54(6), 19992043.
Aggarwall, R. and Samwick, A. (2003). Performance Incentives within the firms: The
effects of managerial responsibility. The Journal of Finance, Vol. 58(4), 1613 1649.
Anderson, R. and Bizjak, J. (2003). An examination of the role of the CEO and the
compensation committee in structuring executive pay. Journal of Banking &
Finance, Vol. 27(7), 13231348. Barro, J., and Barro, R. (1990). Pay, performance and turnover of bank CEOs. Journal
of Labor Economics, Vol. 8 (4), 448481. Barron, M. and Waddel, G. (2003). Executive rank, pay and project selection. Journal
of Financial Economics, Vol. 67(2), 305349. Beck, N. and Katz, J.(1995). What to do (and Not to Do) With Timeseries Cross
Section Data. American Political Science Review. Vol. 89(3), 663647. Bens, D. , Nagar, V., Skinner, D. and Wong, M. (2003). Employee stock options, EPS
dilution and stock repurchases. Journal of Accounting and Economics, Vol. 36(1 3), 5190.
Bertrand, M. and Hallock, K. (2001). The gender gap in top corporate jobs. Industrial & Labour Relations Review, Vol. 55 (1), 321.
Brenner, M., Sundaram, R. and Yermack, D. (2000). Altering the terms of executive
stock options. Journal of Financial Economics, Vol. 57(1), 103128. Bushman, R., Indjejikian, R. and Smith, A. (1996). CEO Compensation: The role of
individual performance evaluation. Journal of Accounting and Economics, Vol. 21(1), 161193.
Carter, M. and Lynch, L. (2001). An examination of executive stock option repricing. Journal of Financial Economics, Vol. 61(2), 207225.
Carter, M. and Lynch, L. (2003). The effect of stock option repricing on employee
turnover. Journal of Accounting and Economics, Vol. 37(1), 91112.
159
Chance, D., Kumar, R. and Todd, R. (2000). The “repricing” of executive stock options.
Journal of Financial Economics, Vol. 57(1), 129154. Chen, M. (2004). Executive option repricing, incentives and retention. The Journal of
Finance, Vol. 59(3), Vol. 11671199. Cheng, Y. and Hung, C. (2006). Corporate Governance and Insider Compensation:
NASDAQ Technology firms. Financial Management Conference Working Paper, Salt Lake City (UTAHUSA).
Chidambaran, N. and Prabhala, N. (2003). Executive stock option repricing, internal
governance mechanisms and management turnover. Journal of Financial
Economics, Vol. 69(1), 153189. Datta, S., IskandarDatta, M. and Kaman, K.(2001). Executive Compensation and
Corporate Acquisitions Decisions. The Journal of Finance, Vol. 56 (6), 2299 2326.
Davidson, W., Pilger, T. and Szakmary, A. (1998). Golden parachutes, board and
committee composition and shareholder wealth. The Financial Review, Vol. 33(4), 1732.
Fama, E. and French, K. (1997). Industry costs of equity. Journal of Financial
Economics, Vol. 43(2),153193. Fields, L. and Fraser, D. (1999). On the compensation implications of commercial bank
entry into investment banking. Journal of Banking & Finance, Vol. 23, 1261 1276.
Grinstein, Y. and Hribar, P. (2004). CEO compensation and incentives: evidence from
M&A bonus. Journal of Financial Economics, Vol. 73(1),119143. Hallock, K. (1997). Reciprocally interlocking boards of directors and executive
compensation. Journal of Financial and Quantitative Analysis, Vol. 32(3), 331 426.
Hermalin, B. and Wallace, N. ( 2001). Firm performance and executive compensation in
the savings and loan industry. Journal of Financial Economics, Vol. 61(1), 139 170.
Kato, H., Lemmon, M., Luo, M. and Schallheim, J. (2005). An empirical examination
of the costs and benefits of executive stock options: Evidence from Japan. Journal of Financial Economics, Vol. 78(2), 435461.
160
Kostiuk, P. (1990). Firm size and executive compensation. The Journal of Human
Resources, Vol. 25 (1),90105. Lambert, R., Larcker, D. and Weigelt, K. (1991). How sensitive is executive
compensation to organizational size? Strategic Management Journal, Vol. 12, 395402.
Morgan, A. and Poulsen, A. (2001). Linking pay to performance. Journal of Financial Economics, 62(3), Vol. 489523.
Murphy, K. (1986). Incentives, learning and compensation: a theoretical and empirical
investigation of managerial labour contracts. Rand Journal of Economics, Vol.
17(1), 5976. Murphy, K. (2003). Stockbased pay in new economy firms. Journal of Accounting and
Economics, Vol. 34(1), 129147. Narayanan, M. and Seyhun, H. 2005, Effect of the SarbanesOxley Act on the
influencing of executive compensation. SSRM working paper no. 852964
Ryan, H. and Wiggins, R. (2001). The influence of firm and managerspecific
characteristics on the structure of executive compensation. Journal of Corporate Finance, Vol. 7(3), 101123.
Ryan Jr, H. and Wiggins, R.(2004). Who is in whose pocket? Director compensation,
board independence, and barriers to effective monitoring. Journal of Financial Economics, Vol. 73(3), 497524.
Stathopoulos, K., Espenlaub, S. and Walker, M. (2004). U.K. executive compensation
practices: new economy versus old economy. Journal of Management Accounting
Research, Vol. 16(1), 5792. Ueng, C., Wells, D. and Lilly, J. (2000). CEO influence and executive compensation:
large firms vs. small firms. Managerial Finance, Vol. 26(8), 3 12.
161
8. Appendix
Table 12 T Test of Equality of Fixed Effect Regressions Coefficients – CEOs
In these tables we compare, with the t test, whether the values of CEO regression coefficients are equal or not, and whether the difference is statistically significant. (*) means that the difference is statistically significant at 1%; ** at 5% level and (***) that it is statistically significant at 10%. Panel A: LN (Total Compensation)
CEOS&P500 (1) CEOS&PMidCap (2) CEOS&PSmallCap (3) T test f
equality of Coef. Independent
Variables N Coef. Std.
Error N Coef. Std. Error N Coef. Std.
Error
(1) vs. (2)
(1) vs. (3)
(2) vs. (3)
Constant 3841 8.840 2,334 1997 3.753 0,626 2209 6.492 1,285 * *
LN (Not Exercised Ratio) 3841 0.245 0,009 1997 0.225 0.011 2209 0.243 0,011 * * *
LN(Ajex) 3841 0.227 0,027 1997 0.186 0.043 2209 0.309 0,0001 * * *
Trs1yr 3841 0.001 0,0002 1997 0.00002 0,00002 2209 0.001 0,074 * * *
LN(Bs Volatility) 3841 0.028 0.066 1997 0.050 0,088 2209 0.161 0.074 * * *
LN(Number Mtgs) 3841 0.033 0.034 1997 0.038 0.088 2209 0.070 0.038 * * *
LN(Tenure) 3841 0.265 0.961 1997 1.435 0.042 2209 0.375 0.537 * * *
ROA 3841 0.001 0.001 1997 0.009 0.255 2209 0.003 0.0008 * * *
Pdirpens 3841 0.043 0.038 1997 0.040 0.001 2209 0.092 0.083 * * *
Interlock 3841 0.136 0.053 1997 0.052 0.066 2209 0.098 0.067 * * *
Year1993 3841 0.108 0.043 1997 0.121 0.062 2209 0.183 0.153 * * *
Year1994 3841 0.314 0.046 1997 0.336 0.173 2209 0.039 0.155 * * *
Year1995 3841 0.368 0.048 1997 0.338 0.172 2209 0.012 0.156 * * *
Year1996 3841 0.574 0.051 1997 0.507 0.174 2209 0.092 0.158 * * *
Year1997 3841 0.739 0.055 1997 0.633 0.174 2209 0.155 0.157 * * *
Year1998 3841 0.846 0.055 1997 0.747 0.175 2209 0.214 0.158 * * *
Year1999 3841 1.072 0.057 1997 0.918 0.174 2209 0.328 0.157 * * *
Year2000 3841 1.163 0.062 1997 1.090 0.176 2209 0.410 0.159 * * *
Year2001 3841 1.296 0.063 1997 1.149 0.178 2209 0.496 0.159 * * *
Year2002 3841 1.301 0.064 1997 1.236 0.179 2209 0.508 0.159 * * *
Year2003 3841 1.317 0.067 1997 1.222 0.179 2209 0.533 0.159 * * *
Year2004 3841 1.510 0.066 1997 1.403 0.180 2209 0.770 0.160 * * *
Candy Dummy 3841 0,500 0,436 1997
Computer Dummy 3841 0,242 1,823 1997
Electronic Dummy 3841 0,758 0,681 1997
APPAREL Dummy 3841 2209 0,619 0,488
Note: * Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
162
Table 12 (Cont.) PANEL B: LN (Total Option Ratio)
CEOS&P500 (1) CEOS&PMidCap (2) CEOS&PSmallCap (3) Equality of Coef. Independent Variables
N Coeff. Std. Error N Coef. Std.
Error N Coef. Std. Error
(1) vs. (2)
(1) vs. (3)
(2) vs. (3)
Constant 3841 3,256 2,780 1997 2,934 0,592 2209 4,527 1,400 * * *
LN (Not Exercised Ratio) 3841 0,315 0,009 1997 0,353 0,013 2209 0,332 0,013 * * *
LN(Ajex) 3841 0,035 0,026 1997 0,157 0,048 2209 0,310 0,058 * * *
Trs1yr 3841 0,001 0,000 1997 0,00001 0,000 2209 0,001 0,000 *
LN(Bs Volatility) 3841 0,192 0,062 1997 0,229 0,097 2209 0,020 0,084 * * *
LN(Number Mtgs) 3841 0,046 0,032 1997 0,036 0,046 2209 0,064 0,044 * * *
LN(Tenure) 3841 1,027 1,145 1997 0,914 0,241 2209 2,191 0,578 * * *
ROA 3841 0,0006 0,001 1997 0,0002 0,001 2209 0,002 0,001
Pdirpens 3841 0,101 0,036 1997 0,056 0,074 2209 0,067 0,095
Interlock 3841 0,108 0,051 1997 0,076 0,071 2209 0,214 0,075 *
Year1993 3841 0,002 0,063 1997 0,045 0,159 2209 0,005 0,277 * * *
Year1994 3841 0,226 0,060 1997 0,101 0,160 2209 0,015 0,280 * * *
Year1995 3841 0,162 0,063 1997 0,009 0,160 2209 0,056 0,279 * * *
Year1996 3841 0,292 0,066 1997 0,160 0,160 2209 0,014 0,277 * * *
Year1997 3841 0,334 0,066 1997 0,139 0,162 2209 0,015 0,278 * * *
Year1998 3841 0,437 0,066 1997 0,318 0,162 2209 0,005 0,279 * * *
Year1999 3841 0,613 0,067 1997 0,412 0,162 2209 0,060 0,279 * * *
Year2000 3841 0,556 0,071 1997 0,321 0,168 2209 0,038 0,279 * * *
Year2001 3841 0,669 0,072 1997 0,472 0,166 2209 0,135 0,280 * * *
Year2002 3841 0,620 0,074 1997 0,428 0,167 2209 0,164 0,280 * * NO
Year2003 3841 0,568 0,076 1997 0,394 0,169 2209 0,141 0,280 * * *
Year2004 3841 0,622 0,076 1997 0,430 0,169 0,119 0,280 * * *
Candy Dummy 3841 0,396 0.615
ComputerDummy 3841 3,019 2,153
ElectronicDummy 3841 0,398 0,665
ApparelDummy 2209 1,330 0,529
Note: * Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
163
Table 12 ( Cont.) PANEL C: LN (Short Term Compensation)
CEO S&P500 (1)
CEO S&PMidCap (2)
CEO S&PSmallCap (3)
Equality of Coef. Independent
Variables N Coef. Std.
Error N Coef. Std. Error N Coef. Std. Error
(1) vs. (2)
(1) vs. (3)
(2) vs. (3)
CONSTANT 3841 8,560 1,950 1997 5.482 0,449 2209 2,158 0,964 * * *
LN (Not Exercised Rat) 3841 0,007 0,008 1997 0.010 0,009 2209 0.008 0,009 * NO *
LN (Ajex) 3841 0,028 0,0233 1997 0.058 0,0329 2209 0.046 0,041 * * *
Trs1yr 3841 0,001 0,0001 1997 0.0003 0,000015 2209 0.001 0,0001 * * *
LN (Bs Volatility) 3841 0,317 0,0566 1997 0.114 0,068 2209 0.119 0,061 * * *
LN (Number Mtgs) 3841 0,074 0,0311 1997 0.028 0,033 2209 0.017 0,033 * * *
LN (Tenure) 3841 0,666 0,8022 1997 0.357 0,188 2209 3.443 0,403 * * *
ROA 3841 0,006 0,0009 1997 0.012 0,001 2209 0.004 0,0007 * * *
Pdirpens 3841 0,027 0,0306 1997 0.065 0,05 2209 0.119 0,07 * * *
Interlock 3841 0,015 0,0461 1997 0.059 0,048 2209 0.060 0,058 * * *
Year1993 3841 0,044 0,0307 1997 0.138 0,066 2209 0.146 0,102 * * *
Year1994 3841 0,139 0,034 1997 0.203 0,064 2209 0.289 0,102 * * *
Year1995 3841 0,113 0,036 1997 0.230 0,065 2209 0.276 0,106 * * *
Year1996 3841 0,182 0,038 1997 0.300 0,065 2209 0.313 0,104 * * *
Year1997 3841 0,276 0,04 1997 0.384 0,067 2209 0.361 0,105 * * *
Year1998 3841 0,402 0,042 1997 0.466 0,067 2209 0.450 0,105 * * *
Year1999 3841 0,478 0,047 1997 0.538 0,068 2209 0.502 0,106 * * *
Year2000 3841 0,589 0,049 1997 0.658 0,071 2209 0.565 0,108 * * *
Year2001 3841 0,552 0,0557 1997 0.610 0,073 2209 0.533 0,108 * * *
Year2002 3841 0,673 0,0553 1997 0.743 0,074 2209 0.629 0,109 * * *
Year2003 3841 0,707 0,055 1997 0.749 0,075 2209 0.665 0,109 * * *
Year2004 3841 0,856 0,055 1997 0.898 0,078 2209 0.807 0,387 * * *
Candy Dummy 3841 0,512 0,302
Computer Dummy 3841 1,315 1,519
Electronic Dummy 3841 1,480 0,627
Note: * Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
164
Table 13 T Test of Equality of Fixed Effect Regressions Coefficients – Directors
In these tables we compare, with the t test, whether the Director regression coefficients are different, and whether the difference is statistically significant. (*) means that the difference is statistically significant at 1%; ** at 5% level and (***) that is statistically significant at 10%. Panel A: LN (Total Compensation)
S&P500 (1) S&PMidCap(2) S&PSmallCap (3) Equality Coefficients Independent
Variables N Coef. Std.
Error N Cof. Std. Error N Coef. Std.
Error
(1) vs. (2)
(1) vs. (3)
(2) vs. (3)
CONSTANT 5207 8,035 2,742 2669 3,421 0,572 3003 6,091 0,536 *
LN (Not Exercised Rat) 5207 0,246 0,008 2669 0,245 0,010 3003 0,235 0,009 * * *
LN (Ajex) 5207 0,195 0,026 2669 0,193 0,038 3003 0,281 0,041 * * *
TRS1YR 5207 0,001 0,000 2669 0,0000003 0,000 3003 0,0005 0,0001 * * *
LN (Bs Volatility) 5207 0,117 0,062 2669 0,036 0,072 3003 0,178 0,063 * * *
LN (Number Mtgs) 5207 0,003 0,031 2669 0,048 0,039 3003 0,017 0,034 * * *
LN (Tenure) 5207 0,063 1,200 2669 1,659 0,250 3003 0,467 0,622 * * *
ROA 5207 0,004 0,001 2669 0,010 0,001 3003 0,004 0,001 * * *
Pdirpens 5207 0,057 0,034 2669 0,086 0,059 3003 0,079 0,077 * * *
Interlock 5207 0,103 0,050 2669 0,023 0,061 3003 0,084 0,062 * * *
Year1993 5207 0,091 0,034 2669 0,064 0,064 3003 0,024 0,072 * * *
Year1994 5207 0,298 0,038 2669 0,300 0,064 3003 0,216 0,075 * * *
Year1995 5207 0,361 0,040 2669 0,321 0,068 3003 0,184 0,077 * * *
Year1996 5207 0,543 0,043 2669 0,518 0,069 3003 0,324 0,076 * * *
Year1997 5207 0,749 0,046 2669 0,675 0,071 3003 0,415 0,076 * * *
Year1998 5207 0,884 0,047 2669 0,776 0,072 3003 0,426 0,077 * * *
Year1999 5207 1,111 0,050 2669 0,909 0,072 3003 0,571 0,078 * * *
Year2000 5207 1,263 0,054 2669 1,064 0,079 3003 0,654 0,082 * * *
Year2001 5207 1,403 0,056 2669 1,157 0,079 3003 0,766 0,081 * * *
Year2002 5207 1,426 0,057 2669 1,258 0,080 3003 0,818 0,081 * * *
Year2003 5207 1,427 0,059 2669 1,224 0,080 3003 0,830 0,082 * * *
Year2004 5207 1,621 0,059 2669 1,435 0,083 3003 1,098 0,085 * * *
Candy Dummy 5207 0,024 0,529
Computer Dummy 5207 0,212 2,254
Consumer Dummy 5207 0,829 0,818
Electronic Dummy 5207 0,841 0,771
Apparel Dummy 3003 0,695 0,536
Note: * Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
165
Table 13 (Cont.) PANEL B: LN (Option Ratio)
S&P500 (1) S&PMidCap(2) S&PSmallCap (3) Equality Coefficients
Independent Variables N Coef. Std.
Error N Coef. Std. Error N Coef. Std.
Error
(1) vs. (2)
(1)vs. (3)
(2) vs. (3)
CONSTANT 5207 3,093 2,855 2669 2,437 0,579 3003 3,305 1,501 * * *
LN (Not Exercised Rat.) 5207 0,299 0,007 2669 0,333 0,011 3003 0,318 0,01 * * *
LN (Ajex) 5207 0,064 0,022 2669 0,217 0,04 3003 0,331 0,045 * * *
Trs1yr 5207 0,001 0,0007 2669 0,00001 0,00002 3003 0,0008 0,0001 * * *
LN (Bs Volatility) 5207 0,244 0,053 2669 0,364 0,074 3003 0,158 0,068 * * *
LN (Number Mtgs) 5207 0,044 0,028 2669 0,004 0,04 3003 0,025 0,037 * * *
LN (Tenure) 5207 1,026 1,249 2669 0,820 0,254 3003 1,818 0,662 * * *
ROA 5207 0,001 0,001 2669 0,001 0,001 3003 0,0002 0,001 * * *
Pdirpens 5207 0,057 0,03 2669 0,010 0,064 3003 0,085 0,082 * * *
Interlock 5207 0,093 0,045 2669 0,100 0,066 3003 0,139 0,066 * * *
Year1993 5207 0,020 0,041 2669 0,065 0,070 3003 0,066 0,08 * * **
Year1994 5207 0,214 0,039 2669 0,139 0,071 3003 0,159 0,085 * * *
Year1995 5207 0,179 0,043 2669 0,058 0,067 3003 0,098 0,085 * * *
Year1996 5207 0,334 0,045 2669 0,225 0,069 3003 0,246 0,081 * * *
Year1997 5207 0,392 0,046 2669 0,214 0,074 3003 0,191 0,083 * * *
Year1998 5207 0,468 0,046 2669 0,311 0,071 3003 0,183 0,083 * * *
Year1999 5207 0,628 0,047 2669 0,413 0,071 3003 0,236 0,084 * * *
Year2000 5207 0,565 0,051 2669 0,294 0,081 3003 0,201 0,085 * * *
Year2001 5207 0,680 0,052 2669 0,446 0,078 3003 0,308 0,086 * * *
Year2002 5207 0,623 0,053 2669 0,386 0,080 3003 0,333 0,088 * * *
Year2003 5207 0,585 0,055 2669 0,381 0,081 3003 0,306 0,088 * * *
Year2004 5207 0,632 0,056 2669 0,429 0,082 3003 0,317 0,09 * * *
Candy Dummy 5207 0,05 0,669
Computer Dummy 5207 3,398 2,341
Electronic Dummy 5207 0,940 0,782
Apparel Dummy 5207 0,350 0,72 3003 0,972 0,582
* Significant at level 1%; ** Significant at level 5%; *** Significant at level 10%
166
Table 13 ( Cont.) PANEL C: LN (Short Term Compensation)
S&P500 (1) S&P MidCap(2)
S&P SmallCap (3) Equality Coefficients
Independent Variables
N Coef. Std. Error N Coef. Std. Error N Coef. Std.
Error (1)
vs. (2)
(1) vs.(3)
(2) vs. (3)
CONSTANT 5207 7,982 2,287 2669 4,739 0,48 3003 1,785 1,181 * * *
LN (Not Exercised Rat) 5207 0,004 0,008 2669 0,008 0,009 3003 0,009 0,008 * * No
LN (Ajex) 5207 0,019 0,025 2669 0,035 0,032 3003 0,050 0,035 * * *
Trs1yr 5207 0,001 0,0001 2669 0,0004 0,0002 3003 0,001 0,0001 * * *
LN (Bs Volatility) 5207 0,367 0,059 2669 0,153 0,059 3003 0,055 0,054 * * *
LN( Number Mtgs) 5207 0,100 0,031 2669 0,066 0,032 3003 0,078 0,03 * * *
LN (Tenure) 5207 0,521 1,000 2669 0,700 0,210 3003 3,524 0,521 * * *
ROA 5207 0,006 0,001 2669 0,009 0,001 3003 0,005 0,001 * * *
Pdirpens 5207 0,020 0,033 2669 0,065 0,05 3003 0,041 0,066 * * *
Interlock 5207 0,026 0,047 2669 0,045 0,051 3003 0,077 0,055 ** * *
Year1993 5207 0,063 0,028 2669 0,052 0,044 3003 0,035 0,052 * * *
Year1994 5207 0,181 0,029 2669 0,146 0,045 3003 0,176 0,056 * * *
Year1995 5207 0,167 0,033 2669 0,179 0,049 3003 0,166 0,057 * * *
Year1996 5207 0,245 0,036 2669 0,277 0,05 3003 0,210 0,057 * * *
Year1997 5207 0,331 0,038 2669 0,372 0,05 3003 0,257 0,061 * * *
Year1998 5207 0,468 0,042 2669 0,438 0,052 3003 0,359 0,058 * * *
Year1999 5207 0,553 0,045 2669 0,498 0,053 3003 0,397 0,061 * * *
Year2000 5207 0,678 0,050 2669 0,643 0,058 3003 0,448 0,062 * * *
Year2001 5207 0,656 0,055 2669 0,604 0,06 3003 0,440 0,063 * * *
Year2002 5207 0,820 0,054 2669 0,757 0,061 3003 0,574 0,064 * * *
Year2003 5207 0,869 0,054 2669 0,760 0,062 3003 0,582 0,065 * * *
Year2004 5207 1,015 0,055 2669 0,889 0,064 3003 0,760 0,066 * * *
Candy Dummy 5207 0,064 0,425
Computer Dummy 5207 0,904 1,891
Electronic Dummy 5207 1,426 0,766
Apparel Dummy 5207 1,515 0,726 3003 0,955 0,44
167
Table 14: Pearson Correlation of the Independent Variables Panel A: S&P500 1 2 3 4 5 6 7 8 9
1 LN( Not Exercised Ratio) 1,000 0,016 0,069 0,023 0,015 0,142 0,011 0,004 0,015
2LN(Ajex) 0,016 1,000 0,203 0,086 0,075 0,186 0,126 0,157 0,025
3Trs1yr 0,069 0,203 1,000 0,093 0,021 0,032 0,125 0,035 0,001
4LN(Bs Volatility) 0,023 0,086 0,093 1,000 0,028 0,123 0,165 0,306 0,049
5LN(Number Mtgs) 0,015 0,075 0,021 0,028 1,000 0,006 0,121 0,165 0,021
6B89LN(Tenure) 0,142 0,186 0,032 0,123 0,006 1,000 0,061 0,132 0,110
7ROA 0,011 0,126 0,125 0,165 0,121 0,061 1,000 0,042 0,004
8Pdirpens 0,004 0,157 0,035 0,306 0,165 0,132 0,042 1,000 0,014
9Interlock 0,015 0,025 0,001 0,049 0,021 0,110 0,004 0,014 1,000
Panel B: S&PMidCap 1 2 3 4 5 6 7 8 9
1 LN( Not Exercised Ratio) 1,000 0,115 0,019 0,014 0,007 0,124 0,015 0,047 0,009
2LN(Ajex) 0,115 1,000 0,042 0,057 0,119 0,161 0,145 0,013 0,104
3Trs1yr 0,019 0,042 1,000 0,105 0,017 0,002 0,008 0,013 0,001
4LN(Bs Volatility) 0,014 0,057 0,105 1,000 0,015 0,108 0,173 0,271 0,016
5LN(Number Mtgs) 0,007 0,119 0,017 0,015 1,000 0,051 0,145 0,113 0,022
6LN(Tenure) 0,124 0,161 0,002 0,108 0,051 1,000 0,038 0,073 0,151
7ROA 0,015 0,145 0,008 0,173 0,145 0,038 1,000 0,005 0,033
8Pdirpens 0,047 0,013 0,013 0,271 0,113 0,073 0,005 1,000 0,014
9Interlock 0,009 0,104 0,001 0,016 0,022 0,151 0,033 0,014 1,000
PANEL C: S&PSmallCAP 1 2 3 4 5 6 7 8 9
1 LN(Not Exercised Ratio) 1,000 0,017 0,031 0,038 0,057 0,152 0,023 0,023 0,084
2LN (Ajex) 0,017 1,000 0,153 0,094 0,070 0,076 0,166 0,077 0,073
3Trs1yr 0,031 0,153 1,000 0,110 0,034 0,002 0,134 0,039 0,023
4LN (Bs Volatility) 0,038 0,094 0,110 1,000 0,077 0,041 0,178 0,154 0,100
5LN (Number Mtgs) 0,057 0,070 0,034 0,077 1,000 0,114 0,113 0,039 0,109
6LN (Tenure) 0,152 0,076 0,002 0,041 0,114 1,000 0,021 0,027 0,139
7ROA 0,023 0,166 0,134 0,178 0,113 0,021 1,000 0,011 0,045
8Pdirpens 0,023 0,077 0,039 0,154 0,039 0,027 0,011 1,000 0,050
9Interlock 0,084 0,073 0,023 0,100 0,109 0,139 0,045 0,050 1,000
168
Conclusions and Possible Extensions
In this dissertation we organise the literature on executive compensation from
the last 13 years based on the best publications in the area. In the other parts, we present
three essays associated with executive compensation. These three essays analyse
whether the factors that explain executive compensation are different or not in firms
from the new and old economy, NYSE and NASDAQ and also in S&P500,
S&PMidCap and S&PSmallCap, respectively. We also analyse for these three groups
whether total executive compensation and the form of this compensation are the same or
not for CEOs and Directors and whether they changed after the NASDAQ crash in 2000
and SarbanesOxley Act in 2002.
Our results reveal that in the case of new and old economy firms, new economy
executives receive, on average, much more than executives from the old economy,
primarily due to stock options, but in the last few years the difference in compensation
between the executives of both groups has been decreasing. The NASDAQ crash and
the SarbanesOxley Act were significant in terms of the way in which executives are
paid in both new and old economy firms. Firms in both groups have reduced the use of
stock options and have instead increased the use of bonuses and restricted stocks. We
also find that the factors that explain executive compensation in new and old economy
firms are generally different, and in the case of the variables that are the same, such as
firm size component, the intensity of the factors is different. Our results also reveal that
new economy executives receive more, on average, than executives from the old
economy, but the difference decreases in the last sample years. In the bubble period,
new economy executive compensation is composed of more than 50% of stock options
and in the case of old economy firms of more than 30% of stock options. After that
period, with the NASDAQ Crash and the introduction of the SarbanesOxley Act, we
observe a significant change in the structure of the components of executive
compensation – a reduction in the use of stock options and an increase in the use of
bonus and restricted stocks.
In the case of S&P500, S&P Mid Cap and S&P Small Cap executives our results
reveal that the mean executive compensation and the component weights (forms of
compensation) are significantly different. Total compensation and forms of
169
compensation change after the NASDAQ crash and enactment of the SO Act. In
general, salary and stock options decrease and the use of restricted stocks and bonuses
essentially increases after the SarbanesOxley Act, among S&P500 and also S&P Small
Cap firms, implying the effectiveness of the Act. Corporate salaries and stock option
awards were subject to a lot of public debate before the introduction of the SO Act and
it appears that firms made significant changes in their compensation packages in the
spirit of the Act.
We also find that that in S&P500, S&P Mid Cap and S&P Small Cap firms the
factors that explain executive compensation are generally not the same. As in the new
and old economy situation, if some of the factors are simultaneously significant for both
cases, the intensity of these factors is generally different and this difference is
statistically significant. In this case, we also find that the NASDAQ crash changed the
influence of some of the factors that explain executive compensation, and also the way
in which executives are paid.
In the case of NYSE and NASDAQ listed firms, executive compensation is
generally driven by different factors and the way in which executives are paid is also
different. The percentage that salary represents in terms of total compensation in NYSE
listed firms is higher for Directors than for CEOs. Bonus is a more important
compensation component for Directors than for CEOs in NASDAQ listed firms, but in
the case of NYSE firms, the difference is small. In all cases, CEOs receive more stock
options than Directors. The used of restricted stock increases essentially in the last few
years. We also find that after the NASDAQ Crash in 2000, and essentially after the
SarbanesOxley Act in 2002, the forms and weights of CEO and Director compensation
change for NYSE and NASDAQ listed firms.
The limitations of this investigation arise from the fact that we do not know
whether the achieved results will be the same when we divide executives by gender.
Another limitation of this study is that it does not analyse the three most important stock
exchanges in America (NYSE, NASDAQ and AMEX), but only the first two. The
reason for this is based on the fact that, in the latter, the number of executive
compensation observations is too small and makes it impossible to run regression
analyses with a significant number of variables.
170
This investigation can be extended to analyse whether men’s and women’s
compensation are explained by the same factors or not. Another interesting point to
investigate is to analyse the relationship between gender, executive compensation and
firms’ performances, and also gender, executive compensation and dividend policy. In
other words, can firms managed by women CEOs be more profitable than firms
managed by men CEOs? Are dividend policy, executive remuneration and gender
related?