"aw.“
1.5% .mu
. 5m”?

Jana.
I... .2,
s.

ol!

.. hr.“
n 1.. I
231...“.

..

ply 2.9.3

: f A .11..)
: J. :n.

: .1;

L

11....3,

 

 

l'; ('57.? T‘

A. 00:

543557¢ﬁ

This is to certify that the
dissertation entitled

DECONSTRUCTING COMPENSATION:
THE DIMENSIONS OF EXECUTIVE PAY

presented by

Cynthia Elizabeth Devers

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree In Business Administration

 

 

ﬂew/74M W;

Major Professor’ s Signature

April 18, 2003

 

Date

MSU is an Afﬁnnative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRCJDateDuepss—sz

DECONSTRUCTING COMPENSATION: THE DIMENSIONS OF EXECUTIVE PAY
By

Cynthia Elizabeth Devers

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Management

2003

ABSTRACT
DECONSTRUCTING COMPENSATION: THE DIMENSIONS OF EXECUTIVE PAY

By

Cynthia Elizabeth Devers

Past research has essentially resulted in equivocal ﬁndings regarding the
relationships among compensation schemes and executive behavior or ﬁrm performance.
In response to this problem, I propose that the current operationalization of compensation
measures (i.e. pay mix or pay risk) in extant research may be responsible for the inability
to ﬁnd meaningful relationships, as the manner in which these measures are constructed
largely ignores their effects on executive’s perceptions of wealth. Consequently, I
develop a model that distinguishes among the objective dimensions embedded in the
compensation contract (implicit risk, horizon length, horizon ﬂexibility, and performance
standards) and the perceptual constructs (projected value of compensation, endowed
wealth, problem ﬂaming, perceptions of control, and wealth orientation) that emerge in
response to these dimensions, in order to disentangle the potential incentive properties of
various forms of compensation.

I then derive and test hypotheses ﬁom this model regarding the manner in which
executives actually value, and hence, perceive (frame) future incentive pay and how these
perceptions affect executives’ wealth orientation. Findings demonstrate general support
for the model and suggest that executives’ perceptions of and responses to their
contingent pay can be dynamic, and that although incentive alignment may initially be
achieved via the inclusion of contingent pay in the compensation package, this alignment

may not hold constant throughout the compensation horizon. Results also reveal that the

sense of control executives’ perceive they have over a form of incentive compensation
moderates the inﬂuence of framing on their wealth orientation. These ﬁndings and

implications for future research are discussed.

I dedicate this dissertation to my family. I thank my parents for their past and
present guidance and for supporting me in any endeavor I pursue. I also thank Lisa,
Kylie, and Kenna for their encouragement and patience throughout my time in the
doctoral program, during the completion of this dissertation, and always. Without such

support, this would have never been possible.

iv

ACKNOWLEDGEMENTS

I express my gratitude to my dissertation chair, Robert M. Wiseman, and
committee members Gerry McNamara, Donald E. Conlon, and John A. Wagner III for
their assistance and invaluable support throughout this process. I also extend my sincere
thanks to John R. Hollenbeck for his advice on the collection of data and methodology

used in my analyses.

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... ix
LIST OF FIGURES ........................................................................................................ x
CHAPTER 1
INTRODUCTION .......................................................................................................... 1
CHAPTER 2
COMPENSATION RESEARCH ................................................................................... 3
CHAPTER 3
COMPENSATION DESIGN ......................................................................................... 8
Wealth Orientation — How Well Does Executive Compensation Work? ................. 9
Why Wealth Orientation? ......................................................................................... 11
How Do Executives Value Compensation? .............................................................. 15
What are the Effects of Executive Compensation? .................................................. 24
Perceptions of Control Over Pay .............................................................................. 29
How Well Does Executive Compensation Work? — Inﬂuences
on Wealth Orientation ....................................................................................... 32
CHAPTER 4
METHODOLOGY ......................................................................................................... 34
Pilot Study 1 ............................................................................................................. 34
Variables ............................................................................................................. 35
Pilot Study 2 ............................................................................................................. 40
Measures ............................................................................................................. 41
Study 1 ...................................................................................................................... 42
Methods .............................................................................................................. 42
Measures ............................................................................................................. 43
Results ................................................................................................................ 48
Study 2 ...................................................................................................................... 56
Methods .............................................................................................................. 56
Measures ............................................................................................................. 57
Results ................................................................................................................ 59
CHAPTER 5
DISCUSSION ................................................................................................................. 63

vi

Implications and Future Research ............................................................................ 63

Conclusion ................................................................................................................ 67
REFERENCES ............................................................................................................... 69
APPENDICES

Appendix A ......................................................................................................... 75
Appendix B ......................................................................................................... 77
Appendix C ......................................................................................................... 79
Appendix D ......................................................................................................... 87
Appendix E ......................................................................................................... 89
Appendix F ......................................................................................................... 97
Appendix G ......................................................................................................... 99
Appendix H ......................................................................................................... 107
Appendix I .......................................................................................................... 109

vii

LIST OF TABLES

Table 1: Pilot Study 1 - t-test Results ............................................................................ 37
Table 2: Pilot Study 2 - t-test Results ............................................................................ 42
Table 3: Implicit Risk .................................................................................................... 44
Table 4: Likelihood ....................................................................................................... 45
Table 5: Descriptive Statistics and Correlations ........................................................... 49
Table 6: Descriptive Statistics and Correlations ........................................................... 49
Table 7: Projected Value of Compensation .................................................................. 50
Table 8: Endowed Wealth ............................................................................................. 51
Table 9: Mean PVC & Endowed Wealth ...................................................................... 52
Table 10: Problem Framing ........................................................................................... 53
Table 11: Study 1 - Wealth Orientation ......................................................................... 56
Table 12: Study 2 — Descriptive Statistics and Correlations .......................................... 59
Table 13: Study 2 - t-test Results .................................................................................... 60
Table 14: Study 2 — Wealth Orientation ......................................................................... 61

viii

LIST OF FIGURES

Figure 1: The Full Model ................................................................................................ 14
Figure 2: Decision Points ................................................................................................ 35
Figure 3: Actual Spread .................................................................................................. 40
Figure 4: Problem Framing and Wealth Orientation ...................................................... 40
Figure 5: Mean PVC & Endowed Wealth ...................................................................... 52
Figure 6: Problem Framing and Wealth Orientation Means .......................................... 54
Figure 7: Spread, PVC, Endowed Wealth, & Aspiration ............................................... 54
Figure 8: Interaction Effect ............................................................................................. 62

ix

CHAPTER 1
INTRODUCTION

If the system worked perfectly, executive pay would rise when the boss

delivered the goods for shareholders. And it would fall when corporate

performance declined But it doesn't always happen that way

(Business Week Online, 2000).

The quote above exempliﬁes the manner in which executive compensation is
currently portrayed in the popular press. In fact, rarely a day goes by without hearing
condemnation of the excessive compensation awarded to high proﬁle leaders of poor
performers such as Kmart and AT&T. These stories along with the WorldCom, Tyco,
and Enron scandals have spawned vociferous calls for compensation reform in Corporate
America. More speciﬁcally, much of this rhetoric contains the notion that the design of
executive compensation packages played a primary role in propagating such scandals.
For example, Brandon Rees, a research analyst with the AF L-CIO's ofﬁce of investment
states, “it seems like every company that had accounting problems also had troublesome
compensation practices" (BusinessWeek Online, 2002) In addition, Stone (2002)
charges that compensation practices offer executives incentives to manipulate numbers to
offset downturns in business. All this has some scholars questioning whether incentive
compensation motivates executives to do more harm than good (Abowd & Kaplan,

1 999).

Although many believe executives are unduly overpaid, there are others who

claim executive compensation works as intended (Loewenstein, 2000; Perry & Zenner,

2000). These advocates argue that a fairly strong link exists between executive pay and

ﬁrm performance (Murphy, 1999). In fact, Hall and Liebman (1998) ﬁnd empirical

evidence that “CEO compensation is highly responsive to ﬁrm performance” (pg. 685).
Nonetheless, the effects of executive compensation remain a hotly contested issue in the
popular press, with shareholders crying foul while compensation committees scramble to
defend the pay packages awarded to executives.

In response to this debate Abowd and Kaplan (1999) raised several questions that
that they claim remain unanswered regarding executive compensation:

1) How much does executive compensation cost the ﬁrm?

2) How much is executive compensation worth to the recipient?
3) How well does executive compensation work?

4) What are the effects of executive compensation?

5) How much executive compensation is enough?

6) Could executive compensation be improved?

Taking the stance that the purpose behind the study of executive compensation is
to ultimately design efﬁcient and effective incentive systems, it would ﬁrst seem
important to examine how executives actually perceive and respond to their
compensation packages. However, it is surprising how little we know about the
motivating properties of executive compensation. Consequently, in an attempt to
determine how executives do perceive and respond to compensation, this study is
designed to answer three of the questions raised in Abowd and Kaplan (1999): how well

does compensation work?, how do executives value compensation? and what are the

effects of executive compensation?

CHAPTER 2

COMPENSATION RESEARCH

One of the factors complicating the study of compensation is that executive pay
packages can contain several different types of pay (e.g., base salary, cash bonus awards,
stock options, restricted stock grants, long-terrn incentive pay, deferred compensation,
etc), which are generally combined into unique pay packages. Further, each type of pay
can take on several different designs. Stock options, for instance, have many variations
including ﬁxed-price, perforrnance-based, indexed, ﬁxed value, ﬁxed grant, and mega-
grant, to name just a few (c.f., Hall, 2000). Consequently, stock options can vary in how
and when they are awarded, how and when they become vested, and how the exercise
price is determined. Though stock options may have the most ﬂexibility in their designs,
all types of pay can be designed in different ways. Even base salary can be paid in the
year it was earned or deferred to a future year. Thus, each type of pay can exhibit a
variety of designs, and when combined with other types of pay in unique ways, can create
very unique solutions to the problem of risk sharing between executives and principals.
As such each executive’s particular pay scheme is one intended to serve the needs of the
executive, the shareholders, the particular circumstances of the executive’s employment
situation, and the restrictions imposed on the compensation agreement by exogenous
factors. Consequently, considerable differences exist in compensation contracts among
executives (Murphy, 1999) making compensation research more difﬁcult.

We currently have evidence to show that executives respond to stock options
differently than they respond to equity ownership (Sanders, 2001 ), yet much research still

fails to recognize variation within pay type. When the differences in risk properties

among different forms of pay are acknowledged, one of two measures of compensation
risk are typically used: pay mix - the proportion of (contingent) variable pay to total
compensation, often referred to as “the mix of ﬁxed and incentive components” (Gray &
Carmella 1997, pg. 518), or the proportion of long-term to short-term or cash
compensation (F inkelstein & Boyd, 1998). Consequently, many studies have empirically
examined the inﬂuence of one or the other of these variables on various executive
decisions, including but not limited to capital investment (Chakraborty, Kazarosian, &
Trahan, 1999), acquisitions and divestitures (Sanders, 2001), or ﬁrm performance
(Lambert, Larcker, & Baker, 1987).

These two variables are attractive to researchers, as they are easily constructed
using widely available secondary data sources. However; these data are generally
produced from proxy statement data that classiﬁes pay too broadly, thus allowing
examination of only a narrow set of compensation design conﬁgurations (c.f., Wiseman,
Gomez-Mejia, & F ugate, 2000). Consequently, discrete forms of compensation are
routinely aggregated into coarsely constructed categories that largely obscure the effects
of pay risk on executive decisions.

To illustrate, using pay mix as a variable requires the researcher to split total
compensation into two general categories: contingent (variable) and non-contingent pay
(ﬁxed). Consequently, many scholars, investigating the incentive propensity of
compensation routinely group restricted stock, stock options, long-term performance
plans, stock ownership, and long-term bonuses into a single measure of contingent pay or
long-term pay since these forms of pay are considered to have substitutable incentive

effects (Agrawal & Mandelker, 1987; Jensen & Murphy, 1990, Mehran, 1995) and

subsequently examine the effects of these ratios on executive decisions and/or ﬁrm
performance.

Several challenges to the use of coarse-grained measures of compensation have
arisen. In general these arguments suggest that coarse-grained measures ignore
differences in the incentive properties of the individual forms of pay, and thus obscure
dissimilar or even contradictory effects of different forms of pay on executives’
perceptions and ultimately, executive behavior (of. Wiseman, Gomez-Mejia & Fugate,
2000). For example, Bryan, Hwang, and Lilien suggest stock options and restricted stock
each have differential payoﬁ' functions and that in high growth ﬁrms restricted stock
awards “can exacerbate executives’ aversion to risk taking” and that stock option awards
will act as a more efﬁcient incentive than restricted stock (2000: 663). Their argument
notes that these two stock-based forms of pay have differential payoff functions and
which may result in opposing inﬂuences on agents’ decisions. Similarly, Sanders (2001)
found stock option value (measured at award) and stock ownership to have differential
effects on executives’ acquisition and divestiture decisions. This empirical evidence
suggests that executives may attach separate values to each form of pay contained in their
compensation packages, perhaps viewing the package not in aggregate, but as a portfolio
of distinct, potential pay opportunities, in which each have varying value and risk
potential. Sanders’ evidence further questions the wisdom of combining superﬁcially
similar forms of pay into one measure.

In sum, using coarse-grained categories of pay fail to account for the manner in
which executives perceive each particular type of pay, especially when the forms of pay

being combined have competing risk characteristics. Speciﬁcally, broad measures, such

as pay mix, combine multiple forms of pay into a single measure and hence exaggerate
potential noise in the associations between these measures and other variables of interest.
In addition, these measures confound two critical dimensions of risk: the value at risk and
the probability of receiving this value. This is problematic as it can undermine empirical
tests of theory, adversely affect the predictability of models of compensation, and, as Hall
and Murphy (2002) ﬁnd, lead researchers to draw misleading conclusions. In other
words, traditional measures of executive compensation may hinder our ability to identify
robust inﬂuences on behavior and performance, and thus prevent researchers from
observing important differences that, if recognized explicitly, might provide stronger
predictions of how executive pay inﬂuences behavior.

In this study I suggest that measures of compensation should reﬂect executives’
projected value of ﬁrture pay for each form of compensation in the pay package, rather
than ex ante estimates of shareholder cost. For example, executives’ projected value of
stock options and option-like pay forms can vary depending upon the point in the horizon
and/or type of measurement. To illustrate, at award, a set of options given “at the
money” will be assigned some positive recorded value based typically on the Black-
Scholes (1973) formula. The Black-Scholes valuation is based on the assumption that
options will held until expiration - generally 10 years; however, in practice, the majority
of executives exercise their options well before expiration, many immediately upon
vesting (Huddart & Lang, 1996). This causes some to argue that option grants will
encourage risk-seeking behavior (Sanders, 2001) as they offer upside potential without
downside risk. However, consistent with the concept of instant endowment (Thaler,

1980; Thaler & Johnson, 1990), others have argued that once the market price of options

exceeds the strike price, and agents endow this value, the options take on perceived (real)
value (Wiseman & Gomez-Mejia, 1998). In fact, Hall and Murphy suggest that the
proclivity of executives to exercise options early (three to fours years in) results ﬁom risk
aversion on the part of executives who are attempting to “lock in” gains following
increases in market price (2000: 28). Therefore, when options achieve a positive spread,
and that potential value is acknowledged executives’ will sense downside risk to that
value, which most likely intensify, as the vesting date grows closer. Consequently, I
suggest that options can have differential risk characteristics, depending and when they
are measured. However, there is no standard measurement period employed by
compensation researchers investigating the inﬂuence of stock options on executive
behavior. Empirically this is problematic as the perceived value of options, hence the
effects of options on executives’ perceptions can change, depending upon the time period
of the horizon researchers are using to construct their measures. Consequently, if we
wish to truly test how executives respond to various compensation schemes, we must use
executives’ measures of their projected value of compensation rather than ex ante cost to

shareholder ﬁgures in our models.

CHAPTER 3

COMPENSATION DESIGN

Compensation design involves the construction of a contract between the agent
and the principal. As such, compensation designs contain several features that are
typically required of any contractual relation (MacNeil, 1978). Speciﬁcally, this includes
the amounts of compensation that may be awarded, the performance standards (internal
or external) used in awarding different forms of compensation, the implicit risk
associated with a particular form of compensation, and the length and the ﬂexibility of
the time horizon, including when performance will be assessed and compensation
awarded. Using these as a guide, I argue that the various forms of executive
compensation can be modeled along these elements and that these elements give rise to
ﬁve perceptual dimensions (projected value of compensation, endowed wealth, problem
ﬂaming, perceptions of control, and wealth orientation) that affect executives’
perceptions of compensation.

It is widely accepted that compensation schemes specify the degree of risk sharing
between agents and principals (Coffee, 1988). That is, compensation contracts specify
the amount of risk that each party to the contract will bear as a result of the compensation
agreement between the agent and the principal (Holrnstrom, 1979; Shavell, 1979). When
an agent's compensation is contractually guaranteed over a speciﬁed time regardless of
his/her performance (e.g., base salary), the agent perceives little risk, while the principal
bears all the risk associated with the cost of that contract. That is, the principal is
obligated to pay regardless of agent efforts on behalf of the principal. Conversely, if

agents are paid solely on a contingent basis such that the agent's compensation is

dependent upon achieving certain performance objectives important to the principal, then
the agent bears all of the risk in the contract. Generally, the majority of compensation
contracts involve risk sharing by both parties. This risk sharing property is a key element
in debates over the efﬁcacy of executive compensation design (c.f, Beatty & Zaj ac, 1994;
Gomez-Mejia & Wiseman, 1997). The concept of risk within a principal-agent relation
generally recognizes two dimensions. First, that something of value to the agent is "at
risk" of being lost, and second, the likelihood that this value could be lost (see Wiseman,
et al., 2000, for a review of compensation risk). Whether the executive expects to receive
this value would therefore appear to be a critical factor in examining how a given
compensation design is perceived by the agent and how this well this design works
(Wiseman & Gomez-Mejia, 1998).

mm Orien_t2_1tion - How Well Does Execgtive Compengtion Work_?

Agency theory proposes that principals and agents have competing attitudes
toward risk. Principals can diversify their stake across ﬁrms, thus are considered risk
neutral (Eisenhardt, 1989). On the other hand, agents are overinvested in the ﬁrm (they
cannot diversify away employment and income concerns) and are considered
opportunistic, thus, as a means of protecting personal wealth (deﬁned as current income,
current assets, and fully anticipated future income, Sheﬂ'in & Thaler, 1988), will act in a
risk averse manner (Donaldson, 1961; Eisenhardt, 1989), “preferring lower risk options at
the expense of returns” (W iseman and Gomez-Mejia, 1998: 136) when making decisions
that affect the future performance of the ﬁrm. Consequently, in order to re-channel
opportunism and inﬂuence agents’ actions, agency theory advocates the use of

performance-contingent pay components in compensation contracts that align the

orientation of the agent with that of the principle (Tosi & Gomez-Mejia, 1989). Or in
other words, in order for compensation to work properly incentives are necessary to force
a shift in agents’ orientation ﬂom wealth maintenance (risk aversion) to one of increasing
wealth (focusing on the upside potential).

On the other hand, behavioral decision literature proposes that decision-makers
are loss averse; hence, their orientations will shift from wealth maintenance to wealth
enhancement throughout the compensation contract, in accordance to where performance
lies relative to a salient reference point. To illustrate, prospect theory (Kahneman &
Tversky, 1979) suggests decision makers ﬂame decisions around a wealth reference point
of current wealth (the status quo) and that prospects for positive changes to the status quo
create positive (gain) ﬂames, triggering risk aversion, while prospects for losses of
current wealth create a negative (loss) ﬂame, triggering risk-seeking behavior
(Kahneman & Tversky, 1979; Hodgkinson, et a1, 1999; Sitkin & Pablo, 1992). Hence,
performance-based incentives may work differently on executives’ wealth orientations
(i.e. whether they are concerned with increasing personal wealth or maintaining the
current level) throughout the compensation contract.

Both agency theory (incentive alignment hypotheses) and behavioral theory
(effects of problem ﬂaming) suggest that at any given time decision makers’ wealth
orientation (focus) runs along a single continuum (the intention to protect vs. the
intention to increase), which is consistent with cognitive theories that propose decision
makers have limited cognitive capacity to process information and thus favor the use of
simple heuristics and search “for rapid adequate solutions” (F iske & Taylor, 1991: 13).

However, they both differ in their proposed antecedents of executives’ behavior (risk

10

aversion vs. loss aversion). In response, Wiseman and Gomez-Mejia (1998) developed
the behavioral agency model (BAM) of executive risk-taking in an attempt to integrate
both views by proposing compensation affects executive decisions through the risk
executives bear (perceive) ﬂ'om various forms of pay. In this paper we take the BAM
back a step to focus on executives’ wealth orientation rather than risk bearing as it moves
us closer to understanding whether incentive alignment actually does occur (i.e. are
executives focusing on maintaining or increasing wealth).

Whv Wealth Orientation?

A number of empirical tests have supported prospect theory predictions
(Kahnemen & Tversky, 1974; Tversky and Kahneman, 1986, 1992; Thaler & Johnson,
1990; March & Shapira, 1992; Sullivan & Kida, 1995; Loehman, 1998; Kt’ihberger,
Schulte-Mecklenbeck, & Pemer, 1999). Although prospect theory has been a dominant
theory in behavioral risk-taking literature, some empirical evidence ﬁnds behavior that
directly contradicts its predictions (i.e. the house money effect, Thaler & Johnson, 1990;
threat-ridgity, Staw, et al, 1983).

To deal with these inconsistencies, I suggest that it is critical that we uncover
executives’ wealth orientation, an antecedent of choice behavior, rather than simply
measuring executives’ perceptions of risk of the behavior or the behavior directly. To
illustrate, empirical tests of prospect theory have primarily used risk-taking behaviors
(i.e. gambling choices, Kahneman & Tversky, 1979; Thaler & Johnson, 1990, investment
decisions, Sullivan and Kida, 1995, or acquisitions and divestitures, Sanders, 2001) or
perceptions of risk of a particular choice (Weber & Milliman, 1997; Sitkin & Weingart,

1995) as a dependent variable. I suggest that this is problematic, as we cannot

ll

automatically assume that behavior labeled as risky by the researcher is perceived as
equally risky by the subjects under study (McNamara & Bromiley, 1997, 1999). For
example, we truly do not know how risky executives consider acquisitions or divestitures
to be (Sanders 2001), or if they perceive them as risky at all (see Hayward’s and
Harnbrick’ 1997 discussion of managerial hubris). This is in line with March and Shapira
who state that “[f]rom a behavioral point of view, this contextual variation in risk-taking
seems to stem less ﬂom the revision of a coherent preference for risk (March, 1988) then
ﬂom a change in focus among a set of inconsistent and ambiguous preferences” (1987:
88). In fact, empirical evidence has shown that individuals vary in the manner in which
they perceive risks presented to them (Weber & Milliman, 1997; Cooper, Woo, and
Dunkelberg, 1988). Consequently, a behavior labeled as risky by one may actually be
perceived very differently by another, perhaps leading to the conﬂicting results we have
seen in empirical tests of prospect theory. Hence, we cannot be assured that researchers’
interpretations accurately match executives’ perceptions of their own behavior (i.e. is it
risk seeking or risk averse?) Nor can we extrapolate the wealth orientation that emerges
from a particular incentive or situation regardless of the choices before executives by
examining choice behavior or perceptions of risk of a particular choice.

Although referred to by various labels, the wealth orientation of decision makers
contemplating risky decisions across contexts has been a topic of discussion for several
years (see March & Shapira, 1987, 1992; Lopes, 1987; Sitkin & Pablo, 1992; Sitkin &
Weingart, 1995). Predictions here are generally patterned after prospect theory in that
they center on the idea that when performance is ahead of a salient reference point (a

positively ﬂamed or gain situation), the desire to maintain those gains dominate decision

12

makers’ concerns, thus, decision makers fear the danger of performance falling below
that reference point. Conversely, when performance is behind the salient reference point
(a negatively ﬂamed or loss situation), decision makers’ concerns shift toward locating
opportunities that will increase performance to meet aspirations (March & Shapira, 1987;
Lopes, 1987; Sitkin & Weingart, 1995). Although these predictions mirror prospect
theory, as stated earlier, empirical tests of prospect theory have resulted in contradictory
results. As such, I suggest that to more accurately assess how well various dynamic
compensation contexts work we must directly examine executives’ wealth orientation to
determine if and when changes occur and under which situations.

I suggest that wealth orientation inﬂuences risk behavior as previously predicted.
I also suggest that wealth orientation inﬂuences how executives may perceive each of the
options available to them under various situations, such that one will focus on either the
dangers or the opportunities inherent in each choice, depending on where performance
lies relative to target (March & Shapira, 1987). In addition, understanding changes in the
wealth orientation of executives may allow us to better predict the level or type of search
behavior executives may engage in response to various incentives and situations or if
they will search at all (i.e., an orientation toward increase would be likely to result in
greater search than an orientation toward maintenance). Finally, I propose that
understanding wealth orientation will allow us to move closer to integrating agency and
behavioral theory and help us better understand how executives perceive and are
considering responding to various compensation contexts and to better gauge the

effectiveness of compensation forms and the compensation contract.

13

Since the concern here is with the design of compensation as opposed to the
context surrounding a compensation agreement (e.g., Bloom and Milkovich, 1998), I
focus on the elements of the compensation agreement that can affect the agent's
perception of both the value of the compensation at risk, and the agent’s perceived
control over receiving this value. Although, I believe that context plays an important role
in inﬂuencing the design of compensation, the purpose here is to understand how
manipulating the elements of compensation design may inﬂuence agents’ reactions to

compensation.

 

Figure 1
Bormdary of the contract The Full Model

 

Carma _
Pater-mo: ‘ “PM Framing Wealth
Value of ’ Eﬂ’ect Orientation

Endowed Perceptions
lens!!!
Pafomnce
standard

 

 

 

 

 

14

To this end, I argue that within the compensation contract, wealth orientation is a
result of two distinct inﬂuences. The ﬁrst is the comparison of the projected value of
future compensation at risk to a reference point of either endowed wealth or an aspiration
for wealth (both will be explained in a subsequent section). The second is the perceived
control over receiving that compensation (see Figure l for the ﬂrll model).

The next section will address the manner in which agents calculate the projected
value of various forms of compensation and examine the effects of this value on two
perceptual constructs that affect wealth orientation through a reference point comparison
process.

How Do Executives Value Compenm

Typically, models of compensation segregate variable forms of pay, such as cash
bonuses, ﬂ'om ﬁxed forms of pay, such as base pay, under the assumption that variable
forms are more at risk than ﬁxed pay. On the surface this split seems reasonable since by
deﬁnition, the amount of variable forms of pay awarded should vary from period to
period, as it is generally made contingent upon the agent satisfying certain performance
requirements speciﬁed in the contract. In other words, unlike base salary, the precise
amount of money agents may ultimately realize ﬂ'om variable forms of pay is not
assuredl. Using this view, in our analyses, we must treat incentive pay such as stock
options as just that, a carrot, or an incentive, that may be received contingent upon
performance, as opposed to a sure reward for past performance. However, it is common
that researchers operationalize the proportion of pay considered "at risk" within the
contract into a single variable by combining the amount of ex post awards (e.g, prior cash

bonus awards) with ex ante estimates of long term awards (e.g., the Black-Scholes (1973)

15

model is often used to estimate the future contribution to wealth of stock options at the
time of award, which occurs several years prior to vesting). I argue that this practice is
problematic for several reasons. The most obvious is that combining ex ante pay
estimates and ex post amounts into a one variable confounds its meaning, causing one to
question whether it is an estimate of potential pay available to the agent, or a measure of
actual pay realized. Further, it is not clear that measures of actual pay awarded ever
provide reasonable proxies for the amount of potential pay promised in the contract. For
example, using cash bonus pay awarded in this manner assumes that this ex post amount
provides an accurate proxy for the potential anticipated amount of the bonus contained in
the contract at the beginning of the contract period. However, one cannot assume with
conﬁdence that all agents regularly reap the maximum award possible, as some will earn
the maximum allowable within their contracts, while others will not. Consequently,
measures that use cash bonus pay awarded likely underestimates the potential bonus pay
available to the agent at the start of the compensation contract.

Conversely, the value of stock options is generally operationalized using
estimates of the future value of these awards to agents taken at the time of the award. In
this sense, these measures are approximations of the value agents may later realize, given
a set of assumptions about the future (i.e. continued employment and stock price
appreciation, etc.), not all of which reﬂect the nature of stock optionsz. As a result, I
suggest that measures of variable pay that combine ex post awards with ex ante estimates

of value result in combining pay types that might actually have conﬂicting risk properties

 

‘ Note that employment risk could threaten even ﬁxed forms of pay.

2 For example, stock options are not exercisable until some futm'e date, and are never tradable in the same
sense that ﬁnancial options may be traded. The lack of a secondary market for stock options calls into
question the use of valuation models such as the Black-Scholes (c.f., Hall and Muphy, 2000).

16

for executives (i.e. there would be risk attached to ex ante estimates of contingent pay;
however there would be no risk associated with ex post measure of compensation, as it is
already paid). This holds serious implications for the study of the incentive properties of
compensation schemes, leading one to question previous empirical ﬁndings involving
variables constructed in this manner.

Furthermore, I also suggest that traditional measures of compensation value
generally do not distinguish between the projected value that an agent perceives s/he will
gain ﬂom different pay forms and the actual amount of pay that agent may receive or is
promised. By consciously lumping all forms of variable pay together, researchers assume
that these types of pay are fungible such that agents do not distinguish between different
forms of pay. That is, agents perceive and treat these forms of pay the same. However,
since, individuals segregate personal wealth into various mental accounts, including
current income, current assets, and fully anticipated future income (Sheﬂin & Thaler,
1988; Thaler, 1980) it is likely that agents make distinctions among various pay forms as
well. In fact, research ﬁndings are beginning to support this assumption. For example,
Lanaza-Kintana, Wiseman, Gomez-Mejia, & Welboume (1999) found that executives of
IPO ﬁrms primarily use base salary for month to month expenditures, while pay ﬂom
stock options is primarily put toward savings. This suggests all pay forms may not be
ﬂmgible; hence, executives may value distinct forms of pay differently depending upon
the amount contractually promised and the uncertainty attached to that pay form.

As such, I suggest that the projected value of contingent compensation includes
that pay which is contractually obligated (as performance targets are achieved or

expected to be achieved), but not yet paid to the agent. This includes cash bonuses and

17

long-term incentive pay for which the targets that trigger bonus awards, have been
achieved (or are expected to be achieved) - but must be maintained, the value (actual
spread) of unexercised, not-yet-vested stock options, the current market value of unvested
restricted stock awards, and future base pay. Although this projected value is expected, it
is not certain to be dispersed as it could eventually be reduced or lost due to termination
or declines in performance that jeopardize the award3.

In sum, I argue that agents calculate a projected value for each form of
compensation, which does not necessarily reﬂect the amount estimated by measures such
as the Black-Scholes, past awards, or the amount eventually awarded. Instead, this
projected value is driven in large part by characteristics of the pay itself as contained
within the compensation contract. That is, the projected value of compensation is the
value agents project they may eventually receive ﬂom a given form of pay at the
conclusion of the compensation contract. I argue that this projected value is a function of
four inﬂuences: the amount of pay promised in the compensation contract, the current
performance of the particular form of pay, the implicit risk attached to the particular form
of pay, and the length of the time horizon to ﬁnal award of that form of pay.

For example, the contracted value of contingent compensation reﬂects the amount
that is speciﬁed in the contract. In the case of a bonus (long- or short-term), this value
will typically fall between a speciﬁed ﬂoor (the minimum bonus value or threshold) and a
speciﬁed ceiling (the maximum value or cap) attainable by achieving performance
targets. Generally, in the case of options (or similar forms of compensation), this value

would have a ﬂoor (zero for options “underwater”) yet have an unlimited ceiling, unless

 

3 For reasons of simplicity we set aside issues of golden parachutes and other elements of a compensation
contract which would protect the agent from losses in the event of termination.

18

otherwise speciﬁed by the contract. Lastly, in the case of restricted stock, this value
would again have a ﬂoor (zero, prior to vesting as there generally is no strike price or
purchase required; however, this value could fall below zero if retained and the market
price declines below the taxes paid on the award); however, the ceiling is likely unlimited
(as in the case of options).

Current performance inﬂuences how executives value each individual form of
future compensation in their pay package. For example, in the case of a bonus, if
performance targets are met (or appear that they will be met) it is likely that the agent
will assign a positive ﬁgure to the projected value of the bonus, based upon the amount
speciﬁed in the contract (see above). In the case of options and restricted stock, it is
likely that agents compare the market performance of the stock to the strike price
speciﬁed in the contract and use this comparison to formulate the projected value of
compensation.

In addition, I suggest that executives then adjust their projected value of
compensation to accommodate two characteristics of the particular form of pay: its time
horizon (time before actual payment) and the inherent variability of the particular form of
pay over time (referred to as implicit risk). Gray and Carmella deﬁne compensation time
horizon “as the extent to which the total compensation package provides ﬁnancial
rewards to the executive based on long-term performance outcomes” and suggest that
incentives tied to executive compensation vary with the length of the compensation time
horizon. (1997:523). I agree that the time horizon of compensation is an important
dimension of compensation design; however, I challenge the view that pay tied to

multiyear performance targets will unequivocally encourage executives to act in a

19

manner that will increase the long-term performance of the ﬁrm. In my view, this
argument is void of any consideration of the risk properties of the time horizon.

To illustrate, base salary is typically paid beginning with the ﬁrst month of
employment and is paid regularly throughout the time ﬂame of the contract. On the other
hand, cash bonuses are generally awarded annually and thus not granted until the end of
year. Finally, agents may not beneﬁt from certain long-term forms of compensation for
several years. For example, gains from stock options cannot be realized until after those
options are vested, often three to four years from award. From a simple recognition of
the time value of money compensation received in the present has greater value than an
equal amount of compensation to be paid in the ﬂrture (i.e. agents assign greater value to
pay awarded in the current year (e.g., end-of-year bonus) than to pay promised in future
years (e.g., multi-year incentive pay).

Additionally, the ﬁrrther into the future a compensation award is promised, the
greater the chance that exogenous factors may intervene and prevent the award. For
example, if a cash bonus is contingent upon maintaining a speciﬁc stock price, the further
away the award is made, the greater the possibility that economic conditions could
deteriorate lowering stock price and eliminating the promised bonus.

As Shelly and Omer state, “[m]anagers are not indifferent to the timing of
compensation. In general, they are impatient (i.e. their discount rates are positive), which
makes current receipts more valuable than future receipts of the same size” (1996, pp.
45). Thus, at a minimum, one would expect agents to discount the projected value of

future compensation relative to pay awarded in the present.

20

In addition to the discounting for time, there is empirical evidence that agents
recognize and factor the implicit risk (the inherent variability of the outcome), into their
projected value of a form of future compensation (Stevenson, 1986; Shelley & Omer,
1996). That is, if variable pay is truly variable (whether due to endogenous or exogenous
factors), we expect agents to discount future pay over and above the discount for time,
based on their subjective assessment of pay variability and uncertainty of award, or the
implicit risk to the compensation that creates a sense of loss for the agent (Shelley, 1993).
This risk-adjustment essentially discounts the value of anticipated future pay by its
inherent volatility or uncertainty. At one extreme, future base pay exhibits little or no
volatility over the length of the contract (holding raises and cost of living adjustments
constant) and thus exhibits low implicit risk. At the other extreme, the spread in stock
options, being tied directly to stock price, may seem to have the most volatility relative to
other forms of pay and thus contain high implicit risk, such as stock options in the
dotcom industry craze in the late 1990’s. Thus, I suggest that executives ﬂtrther adjust
the projected value of compensation by its implicit risk. Forms of pay that exhibit high
implicit risk would be discounted more than forms of pay exhibiting less implicit risk.
For example, if the value of stock options (i.e., the spread) is rising steadily over time
during a period of economic muniﬁcence, we would expect agents to assign a much
lower discount to their projected value (and may even appreciate its value as agents often
did in the dot com industry during the late 1990's) than agents of ﬁrms with dismal
prospects, a declining stock price, and a hostile economic climate.

In support of this argument, Shelley & Omer (1996) used a laboratory study to

manipulate implicit risk and time separately, and found evidence that agents discount

21

future pay for both time and implicit risk independently. This ﬁnding implies that agents
are likely to discount highly variable forms of pay (such as pay linked to stock price, i.e.,
stock options) more heavily than pay that is relatively static (more certain) over time
(e.g., base salary or constant annual bonuses). While this discounting may not reﬂect
large differences under conditions of low inﬂation and short time horizons, it can
represent substantial discounts of long-term incentive pay under inﬂationary conditions.
The practical point of this argument is that agents may assign a very low value to
currently unexercisable stock options’, even those with a positive spread, if they are not
exercisable for several years, ﬁrm performance is volatile, and the economic environment
is unstable. This challenges current models of compensation, which assign a positive
value to stock options (even under-water options) ﬂom the time of award based on simple
assumptions about stock price appreciation (e.g. the Black-Scholes assumes 5% growth
rate).

In sum, the use of actual (or even estimated) amounts of pay requires researchers
to assume that agents assign a projected value to future pay that is commensurate with the
cost of that pay to shareholders. Challenging this assumption is evidence that agents
under-value some forms of pay (e. g., stock options) relative to the cost of that pay to
shareholders (Hall & Murphy, 2000). This differential discounting may explain why
different forms of pay are used for different purposes. The implication of this argument
is that agents may assign values to some forms of highly variable compensation

differently than less variable forms.

 

‘ It is likely too difﬁcult computationally for agents to disentangle each stock option award; hence it is
likely that they simply categorize options into those that are exercisable and those that are not yet vested.

22

This argument echoes those by Hall and Murphy (2000, 2002) that agents under-
value stock options relative to the value assigned them by models such as Black-Scholes.
Consistent with their view, agents appear to discount new stock option awards because
they generally cannot be exercised for several years, the stock price (the basis for
determining spread and ultimately the realized value) is volatile, and agents generally
don't hold options for the entire period assumed in models of valuation (i.e. agents
generally cash out options before they reach their theoretical maximum value). In this
model, as vesting of an option approaches, the effect of the time horizon diminishes, thus
the agent's discounting of that option is reduced. However, I suggest that the discounting
of the perceived value of options held never reaches zero, since exercisable options
continue to exhibit a positive level of implicit risk. As a result, the relation of perceived
value to the actual spread of an option is concave in relation to time. Once stock options
are vested, however, agents continue to discount the spread to accommodate for inherent
volatility of the stock (implicit risk) similar to what equity owners do in assessing the
value of their equity ownership. In sum, I argue that the generally accepted measures of
stock-based compensation (i.e. Black-Scholes and other similar measures) do not
accurately reﬂect executives’ valuations of pay. Consequently:

Hypothesis 1: Implicit risk is negatively associated with the projected
value of a form of compensation.

Hypothesis 2: The length of the compensation time horizon is negatively
associated with the projected value of a form of compensation.

The valuation process I describe has important implications for executive
compensation research as it suggests that incentive properties (the effects of

compensation) can be quite different across different forms of contingent pay, raising

23

questions about the effectiveness of various types of pay as motivational tools for agents.
More interestingly, it suggests that these properties can vary within forms of pay at
various points during their compensation horizon, particularly in the case of options and
restricted stock. This will be addressed below in the discussion of how the projected
value of compensation inﬂuences the perceptual constructs of endowed wealth and
problem ﬂaming.

What are the Effects of Executive Compensatigﬂ

Problem ﬂaming. As discussed above, research in behavioral decision theory
ﬁnds that across contexts agents’ risk preferences and risk taking behaviors are not
isomorphic (Shapira, 1995; Wiseman & Gomez-Mejia, 1998). In particular, advocates of
prospect theory suggest that decision makers ﬂame decisions around a reference point of
current personal wealth in hand (Kahneman & Tversky, 1979) and that prospects for
positive changes to this current wealth create a gain context, while prospects for losses to
this wealth create a loss context (Kahneman & Tversky, 1979; Hodgkinson, et a1, 1999;
Sitkin & Pablo, 1992).

In their discussion of problem ﬂaming, Wiseman and Gomez-Mejia, (1998),
expanded the concept of current personal wealth to include endowed wealth (fully
anticipated money, not just money in hand). Building on this deﬁnition, I suggest that in
relation to contingent compensation, endowed wealth is that portion of the projected
value of future compensation contained within the compensation agreement the executive
has already factored in to calculations of personal wealth.

This notion of endowed wealth derives ﬂom the previously established concepts

of mental accounting (Thaler, 1980, 1985; Kahneman & Tversky, 1984) and endowment

24

(Thaler & Johnson, 1980; Kahneman, Knetsch, & Thaler, 1990). As discussed, empirical
research on mental accounting has found that individuals do not view personal wealth as
fungible (Sheﬂin & Thaler, 1988; Thaler, 1980). Combining the mental accounting
process with notions of endowment effects (Thaler & Johnson, 1980; Kahneman,
Knetsch, & Thaler, 1990), I suggest that executives may endow any or all of their
projected value of a particular form of contingent compensation into any one of these
three accounts, subsequently raising their current perceived level of personal wealth
(W iseman & Gomez-Mejia, 1998). Although the endowment of future contingent
compensation has not been documented empirically, anecdotally, we witnessed many
executives and employees endow portions (or all) of the projected value of stock options
into their fully anticipated current or future income accounts (hence, personal wealth),
during the recent high tech market bubble of the late 1990’s. This is evidenced by the
fact that in that robust market, many willingly chose to trade salary for stock options.
However, when share prices severely declined these individuals began to feel burned,
with many demanding (and getting) higher salaries or raises to compensate for their
perceived losses of personal wealth.

Although this anecdotal evidence exists, other than Kahneman, Knetsch, and
Thaler (1990), there has been little discussion about how the endowment process
operates. I suggest that in the case of contingent compensation, endowed wealth is a
function of factors both endogenous and exogenous to the contract. For example, I argue
that wealth is endowed when executives endow (mentally shift) some or all of the
projected value of future compensation into any one of their personal wealth accounts. I

also suggest that endowment rises with positive changes in the current projected value of

25

compensation, although this adjustment process is likely not a one-to—one matching.
Following the logic of Kahneman and Tversky (1979), it is doubtﬁrl that executives
immediately adapt to increases in their projected value of compensation. Consequently,
it is likely that this new value is not immediately endowed into personal wealth accounts.
Rather, increases in endowed wealth lag increases in projected compensation by some
adjustment period. I also suggest that shifts in endowed wealth are asymmetric with
changes in the current projected value of compensation. That is, following notions of the
life-cycle hypothesis (Sheﬂin & Thaler, 1988) executives are likely to adapt easily to
higher levels of wealth, but will resist negative changes to wealth. As such, executives
are unlikely to immediately lower their endowed wealth reference points following
decreases in the projected value of compensation (Wiseman & Gomez-Mejia, 1998).
Therefore, it is likely that over time, the endowed wealth reference point will increase
incrementally with (although lag) increases in the current projected value of
compensation; however, it will fail to decrease in the same manner with declines in this
value (Kameda and Davis, 1990; Gooding, Goel & Wiseman, 1996)

I also argue that endowed wealth is driven by factors exogenous to the contract.
For example, Kahneman, Knetsch, and Thaler (1990) found that the endowment effect
seems to largely be driven by individual differences or preferences. As such, this
evidence suggests that if executives do endow any portion of the projected value of future
compensation, this process will occur at different rates across individuals. For example,
it is likely that this endowment is driven by the past performance of executives’ own
compensation (March & Shapira, 1992) or executives’ conﬁdence or hubris, stemming

ﬂom recent successes, media attention, or self-importance (Hayward & Hambrick, 1997).

26

However, since these individual level factors are derived external to the compensation
contract, a more complete analysis is beyond the scope of this paper. Therefore, I suggest
that when wealth is endowed ﬂorn a particular form of compensation a reference point of
endowed wealth, inﬂuenced by both changes in the current projected value of that
compensation form (W iseman & Gomez-Mejia, 1998) and individual differences
emerges. Hence:

Hypothesis 3a: The current projected value of compensation is positively
associated with the endowed wealth reference point.

Hypothesis 3b: When the current performance of a form of compensation

declines the endowed wealth reference point will decrease more slowly

than the projected value of compensation.

Furthermore, I argue that agents ﬂame decisions made on behalf of principals by
the expected effect of those decisions on personal wealth, such that executives compare
their current projected value of a form compensation to this reference point of endowed
wealth from that same pay form. Furthermore, when the projected value of compensation
exceeds the reference point of endowed wealth a gain context is created; conversely,
when the projected value of compensation falls below the reference point of endowed
wealth a loss context emerges.

Attainment discrepancy. Though occasionally used interchangeably with problem
ﬂaming as speciﬁed in prospect theory (Kahneman and Tversky, 1979), I suggest that
attainment discrepancy represents a speciﬁc type of problem ﬂaming. Attainment
discrepancy revolves around an aspiration for performance, and as such, reﬂects the
attainment of a wealth goal that is ﬁxed throughout the compensation contract, rather

than an appraisal of prospects for changes to perceptions of contributions to existing

personal wealth. The aspiration level reﬂects the amount of wealth the agent seeks ﬂom

27

a particular form of compensation and is essentially a predetermined goal for wealth. In
contrast, the endowed wealth reference point, as discussed previously, rests on counted
(endowed) wealth, which I argue includes a portion of the projected value of
compensation ﬂ'om current contingent compensation.

As it pertains to the compensation contract, the endowed wealth reference point,
does not emerge until this perceived contribution assumes a positive value and can
continuously adjust upward with this value throughout the compensation time horizon.
However, the formation of the aspiration level is largely driven by factors external to the
compensation contract. For example, I suggest that executives use social comparison
(O’Reilly, Main, & Crystal, 1988), and peer contingent compensation ﬁgures (Gomez-
Mejia & Wiseman, 1997) as a basis for determining compensation aspiration levels. To
illustrate, equity theory (Adams, 1963, 1965) would suggest that executives regularly
compare their compensation to that of a reference group of executives. Consequently, as
the compensation ﬁgures of referents rise, and this information is made available by
compensation consultants and the media, executives’ compensation aspirations will also
rise to match that ﬁgure (Wiseman & Gomez-Mejia, 1998). Furthermore, Lant (1992)
ﬁnds that there is typically an optimistic bias in the formulation of aspiration levels. This
indicates that agents may establish a baseline of expected wealth and then adjust this
ﬁgure upward to arrive at the ﬁnal aspiration level (Lant, 1992).

I argue that the aspiration level is vital in determining performance feedback or
attainment discrepancy (March & Shapira, 1992) in the case of contingent compensation
such that when the current projected value of compensation exceeds the aspiration level a

positive discrepancy exists and when the current projected value of compensation is

28

below the aspiration level a negative discrepancy exists (Lant, 1992). I suggest that
discrepancies have important implications for compensation design, as March & Shapira
(1992) ﬁnd that positive attainment discrepancies result in risk avoidance and negative
attainment discrepancies result in risk seeking.

However, I suggest that it would be cognitively difficult for executives to
explicitly separate and independently evaluate the effects of both problem ﬂaming and
attainment discrepancy simultaneously. Consequently, I suggest that executives’
problem ﬂaming (the perception of being either ahead or behind), depends upon where
current performance lies relative to the reference point that is salient at that particular
time (see March & Shapira, 1987, 1992 for a discussion of the shifting focus of attention
between multiple reference points). In other words, executives focus on one reference
point (endowed wealth or the aspiration level) at a time and this salient reference point
drives problem ﬂaming; however, the point that is salient may change over time. Hence:

Hypothesis 4a: Controlling for the endowed wealth reference point and

aspiration level, the projected value of compensation will have a positive

inﬂuence on ﬂaming.

Hypothesis 4b: Controlling for the projected value of compensation and

aspiration level, the endowed wealth reference point will have a negative

inﬂuence on ﬂaming.

Hypothesis 4c: Controlling for the projected value of compensation and

the endowed wealth reference point the aspiration level will have a

negative inﬂuence on ﬂaming.

Hypothesis 5: A maximum of one reference point will drive problem
ﬂaming.

Perceptions of Control Over ng
As noted above, executive’s wealth orientation is a ﬁmction of two primary

factors, the projected value of compensation that is at risk and the executive's perception

29

of control over receiving that value. Having described how executives may value future
compensation, I now turn to how they may assess risk to this value. In particular, I look
at how dimensions of the compensation agreement inﬂuences perceptions of control over
receiving anticipated pay. I argue that two dimensions of compensation design inﬂuence
agent perceptions of control: the performance measures used in gauging agent effort and
the ﬂexibility of the time horizon of the particular form of contingent compensation. As I
will argue, these two dimensions of the design of a compensation scheme inﬂuence agent
perceptions of control over the outcomes used by principals to determine whether to
award pay or not.

Performance standards. Contingent pay is frequently tied to either internal
standards weighting performance relative to some target endogenous to the ﬁrm or
external standards, weighting performance relative to a target exogenous to the ﬁrm.
Internal standards are often accounting based, while external standards are generally
market based. I argue here that both can have differential effects on executives’
perception of control such that internal standards will create greater perceptions of
control than external standards. To illustrate, internal standards (i.e. accounting
standards) can be highly inﬂuenced by executives’ actions, while external standards (i.e.
market- based standards) contain more noise and are more difﬁcult for executives to
directly inﬂuence (Lambert, 1993; Wiseman & Gomez-Mejia, 1998; Murphy, 1999),
particularly in the recent business environment. In addition, Murphy (2000) recently
found that ﬁrms adopting internal determined standards are more likely to show evidence
of income smoothing than ﬁrms with externally determined standards, supporting the

view that executives can more easily inﬂuence internal standards than external standards.

30

Finally, internal standards based on accounting measures are more easily manipulated by
executives through changes in debt structure, inventory management procedures,
accounting procedures and so forth. Consequently, I argue that the use of internally
focused measures of performance creates a greater perception of control for the agent
than external standards.

Hypothesis 6: Contingent compensation forms tied to internal standards

will create greater perceptions of control than contingent compensation

forms tied to external standards.

Commnsation time horizon ﬂexibilig. Earlier I discussed how the length of the
compensation time horizon (the length of the compensation contract for each particular
form of pay) inﬂuences the manner in which agents derive the projected value of each
particular compensation form. Speciﬁcally, longer time horizons are likely to reduce the
value of projected compensation (c.f., Shelley & Omer, 1996). In addition I suggest that
the ﬂexibility of the compensation time horizon has a separate effect that can inﬂuence
agents’ perceived sense of control over the compensation form. For example, although
common long-term performance incentives such as stock options vest on a speciﬁc future
date, once vesting is achieved, their exercisability horizon is ﬂexible as agents can opt to
hold options until they expire — usually 10 years (Murphy, 1999). Thus, during this
ﬂexible period, agents can exercise at their own discretion, essentially allowing them to
“time” the exercise of stock option to when it is most advantageous to do so.
Consequently in the event that an agent’s action has taken a longer/shorter period of time

to achieve success or be rewarded by the market, a ﬂexible horizon allows the executive

to strike at the most opportune time, affording him/her an increased perception of control.

31

On the other hand, different long-term incentives such as long-term performance
plans can consist of a ﬁxed horizon, in which the evaluation of performance-to-target
occurs on a speciﬁc date that is ﬁxed at the start of the compensation contract.
Consequently, under this arrangement, the agent has no discretion or control over when
the performance evaluation occurs. Hence, the timing of "cashing out" cannot be
manipulated as in the case of option-like incentives, thus I expect perceptions of control
to be diminished.

Hypothesis 7: Contingent compensation forms that incorporate a ﬂexible

horizon will create greater perceptions of control than contingent

compensation forms that incorporate an inﬂexible horizon.

How Well Does Executive Compens_ation Work? - Inﬂuences on Wealth Orientation

Many compensation scholars, following Shapira's (1995) work on risk taking,
have suggested that executives give primary weight to protecting or maintaining gains
and therefore, assert that agents are principally concerned with loss aversion and are
indifferent to uncertainty (Wiseman & Gomez-Mejia, 1998; Sanders, 2001).
Consequently, since agents are loss averse, their wealth orientation will focus on
maintaining real or perceived gains at the expense of further increasing (enhancing)
wealth (Wiseman and Gomez-Mejia, 1998). As such, I extend this argument to recognize
the inﬂuence of problem ﬂaming on wealth orientation.

For example, in the case of positively ﬂamed situations executives perceive they
are in a gain situation and the desire to maintain those gains dominate their concerns.
Conversely, when in the case of a negatively ﬂamed problem, executives’ concerns shift
toward increasing performance (March & Shapira, 1987; Lopes, 1987; Sitkin &

Weingart, 1995). Hence to test the loss aversion perspective:

32

Hypothesis 8: Problem ﬂaming is positively associated with wealth

orientation such that positive ﬂaming is positively associated with a

wealth maintenance orientation and negative ﬂaming is negatively

associated with a wealth maintenance orientation.

Moderating effects of pgrceptions of control. Furthermore I suggest that it is only
when agents lack the perception of control that wealth maintenance emerges as their
primary concern (Shapira, 1995). To illustrate, examining the Shapira study more
closely reveals that the subjects in this study viewed risk (the potential for loss)
differently ﬂorn gambling (probability or uncertainty) because they perceived as
managers that they had some sense of control over their situation. Implicit in this ﬁnding
is that the perception of control differentiates potential loss and uncertainty. Extending
this, as executives’ perception of control increases, the inﬂuence of a potential loss on
wealth orientation is decreased. In turn, as executives’ perception of control decreases,
the inﬂuence of a potential loss on wealth orientation is heightened. Therefore:

Hypothesis 9: The relationship between problem ﬂaming and wealth

orientation is moderated by perceptions of control such that high

perceptions of control will decrease the inﬂuence of problem ﬂaming on

wealth orientation and low perceptions of control will increase the
inﬂuence of problem ﬂaming on wealth orientation.

33

CHAPTER 4

METHODOLOGY

As stated earlier, it is difﬁcult to expect that results ﬂom previous empirical
studies that measure the effects of perceptions by using simple expected value gambling
contexts ﬁrlly generalize to executive decision making contexts, as they over inﬂate the
inﬂuence of downside risk on risk-taking behavior thus, they do not reﬂect typical agent
compensation situations. In addition, the majority of risky decision studies have been
static in nature involving choices between outcomes with clearly stated probabilities for
success (Hollenbeck, Ilgen, Phillips, & Hedlund, 1994). Consequently, I tested the model
using two dynamic scenario-based studies (see appendices A — 1) consisting of multiple
(seven) stages mirroring typical executive compensation contexts as suggested in
Hollenbeck et al, 1994. Before testing the complete model I tested the measures in each
of the scenarios in two pilot studies. These two pilot studies are detailed below.

Pilot Studv 1

Subjects. The subjects in this experiment were 30 senior undergraduate business
majors in a capstone management course at Michigan State University.

Desigg and procedures. Subjects voluntarily participated in the study during a
single class session. Subjects were given a packet containing a scenario-based
questionnaire and scenario background information. The scenario-based questionnaire
was a 2 (High Implicit Risk vs. Low Implicit Risk) x 2 (Long Horizon - 3 year vs. Short
Horizon - 18 months) x 7 (seven decision periods) factorial design stock option scenario.
Subjects were randomly assigned to conditions, completed background information, and

were instructed to read the scenario and complete the questionnaire by answering

34

questions at seven different time periods that varied the current performance of the stock
throughout the compensation time horizon (see below).
Variables

Cometed vglpe of stock options. For this stock option test, contracted value is
the strike price of the options awarded to the subject ($29.00).

Current ﬂormance. Current performance is the spread between the market
price and the strike price at each of the seven decision points. Current performance by
decision:

decision 1 = $4.00 per share; decisions 2 — 4 = $13.00 per share; decision

5 = $7.00 per share; decision 6 = $4.00 per share; decision 7 = $0.00 per
share (underwater).

 

Figure 2
Decision Points

50.0
47.5
45.0
42.5
40.0
37.5
35.0
32.5
30.0
27.5
25.0
22.5

otititiEtitiiiittitriitiitittiiEtttil

 

 

 

 

 

These decision points plotted on the three year horizon are illustrated in Figure 2 (the 18
month horizon is identical other than the differing dates).
Implicit risk. Following Shelly and Omer (1996), implicit risk was manipulated

in the scenario via the macroeconomic certainty/uncertainty that is attached to the

35

particular form of compensation (in this case stock options) was chosen to induce
implicit risk. Subjects in the high implicit risk condition read the following:

Your ﬁrm manufactures nonessential components for your customers. As

a result, economic conditions directly affect your ﬁrrn’s stock price.

Current economic conditions are quite unstable. In addition, your stock

price and others in your industry have been quite volatile (rising and

falling dramatically) over the last decade. This pattern is projected to

continue over the next few years. As such, industry analysts consider your

stock to be a high risk investment.

Subjects in the no implicit risk condition read:

Your ﬁrm manufactures critical components for your customers. As a

result, your ﬁnn’s stock price is sheltered ﬂom any negative effects of the

economy. Current economic conditions are quite stable. In addition, your

stock price and others in your industry have generally continued to rise

over the last decade. This pattern is projected to continue over the next

few years. As such, industry analysts consider your stock to be a veg safe

investment.

The phrases high risk and very safe were chosen as they reﬂect the manner in
which analysts rate stocks and bonds. Safe was used rather than no risk as it better
reﬂects the manner in which no risk is referred to in the investment community and
popular business press. For example: thestreet.com, a popular investment website, uses
the terms safety vs. risk when referring to how to balance your stock portfolio (Seymour,
2002); Standard and Poor’s fund ratings assess the safety of invested principal.
Morningstar.com discusses risky investments and states, “A bond fund with a risk score
of 1.2 may be riskier than the typical bond offering, but it is probably less risky than a
safe stock fund which has a risk score of 0.8. In addition, bonds rated A and above are
considered quite safe, while those just above default (CC) are referred to as highly risky.
In the Ask the Expert segment of CNN money.com discusses bonds in their safe category.

The implicit risk manipulation was checked via the following measure:

36

On the scale below please indicate your perception of the likelihood of
making your proﬁt goal...

Subjects indicated their response on a seven point scale with the following anchors: 1 =
low likelihood and 7 = high likelihood. The t-test results (see Table 1) between
conditions were signiﬁcant (t = 3.873, p<.01) and in the predicted direction (no implicit
risk, x = 6.1429, s.d. = .6901; high implicit risk, M = 4.00, s.d. = 1.29). Consequently,

the implicit risk manipulation performed as expected.

 

 

TABLE 1
Pilot Study 1 - t-test Results
IMPLICIT RISK
No Implicit High Implicit
Risk/l8mos Risk/18mos
Variable Mean s.d. Mean s.d. T
Likelihood 6.14 .69 4.00 1.29 3.87, p<.01

 

Horizon lenm. Two horizon length conditions (three years and 18 months) are
presented. A recent survey by management consultant ﬁrm Hewitt Associates reports
stock option vesting generally varies ﬂom 1-5 years, with 45% of companies using a
three-year vesting schedule; only 26% use a four year vesting schedule (Zall, 2000). The
three-year horizon was used, as it is the most common vesting period. In order to use a
shorter time horizon, yet match the proportion through the time horizon and still remain
in the typical vesting period of 1 — 5 years, an 18-month horizon was used.

Subjects in the long horizon condition read:

In three years (on December 31, 2004) you may purchase any or all of the
10,000 shares for $29.00 per share.

Subjects in the short horizon condition read:

37

In 18 months (on June 30, 2003) you may purchase any or all of the
10,000 shares for $29.00 per share.

In addition each decision provides the subjects with the current date and time until
exercise.

Projected vglue of CODTDCILSLIOD (PVC). Following Shelly and Omer (1996),
subjects’ expectations measured the projected (expected future, but at risk) outcome (in
this case future compensation). During each decision period the projected value of
compensation was measured using the following question:

At this point in time, what is the proﬁt per share you expect to earn at

exercise (when you are able to purchase and resell these shares) in XX
years (months)?: $ per share.

 

Endowed weglg During each decision period subjects’ level of
endowed wealth was measured using the following question:

At this point in time, what is the proﬁt per share you are conﬁdent enough

to use to secure ﬁnancing on a purchase you really want to make?:

$ per share.

To my knowledge this is the ﬁrst explicit empirical test of the difference between
projected (expected) and endowed wealth. Consequently, the term secure ﬁnancing was
used to illustrate subjects’ conﬁdence in the imminent addition to personal wealth and
the ﬁnality of the decision.

Aspiration level. The same aspiration level was assigned to subjects in all
conditions. Subjects in all conditions read:

YOUR GOAL

Your goal, set by the board of directors, is to increase your ﬁrrn’s stock

price to $35.00 a share (by December 31, 2004/June 30, 2003).

The following question, asked prior to beginning the scenario, serves a manipulation

check:

38

What is your proﬁt per share goal at exercise (when you are able to
purchase and resell these shares?: $ per share.

Problem ﬂaming (BF). As discussed above, it is cognitively difﬁcult for agents to
explicitly separate and independently evaluate what actually drives their feelings
completely (Fiske & Taylor, 1991). As such, a general perception of subjects’ feelings
(perceptions) of being ahead or behind (a ﬂaming effect) was measured with a single
scale, via the following measure:

At this point in time I feel as though the shares are performing. .

Subjects indicated their response on a seven point scale with the following

anchors: 1 = behind where I’d like them to be and 7 = ahead of where I’d

like them to be.

\_Ve_alth orientation. The following scale was constructed to assess subjects’
wealth orientation speciﬁc to this stock option scenario. In addition, the anchors of
maintaining the current stock price and increasing the current stock price speciﬁcally
addressed and tested the wealth maintenance (loss aversion) assumption of behavioral
decision theory. As such, wealth orientation was measured using the following item:

At this point in time I am more concerned about. ..

Subjects indicated their response on a seven point scale with the following
anchors: l= increasing price and 7 = maintaining price (wealth orientation was reverse
coded).

The PF measure and the wealth orientation measures are correlated. To test for
multicollinearity, a multiple regression equation that regressed wealth orientation on the

PF variable and a lagged wealth orientation (ﬂom the prior decision) was signiﬁcant

(both IVs were also signiﬁcant). The variance inﬂation factor for this equation was 1.78,

39

which is well below the recommended cutoff of 10.0. The actual spread and means for

problem ﬂaming and wealth orientation are shown in Figures 3 & 4.

 

 

 

 

Figure 3
Actual Spread 1

 

 

 

1”.
N
P
o
/"

 

 

 

 

 

 

 

 

i 1 2 3 4 5 6 7
l Decision period

 

 

 

 

 

Figure 4
l Problem Framing and Wealth Orientation

 

 

L
T

 

+Problemﬂamingmeun
\ +Wealth orierlationmean

—d

 

 

 

 

 

 

 

 

CANOJhUIODN
q

 

 

 

 

 

 

 

 

Pilot Studv 2
This study was designed to test the effects of performance standards and horizon

ﬂexibility on subjects’ perception of control.

40

Subjects. The subjects in this experiment were 33 senior undergraduate business

majors in a capstone management course at a large Michigan State University.

Measures

Desigp and procedures. Subjects voluntarily participated in the study during a
single class session. Subjects were given a packet containing a scenario-based
questionnaire and scenario background information. The scenario-based questionnaire
was a 2 (Internal vs. External Performance Standard) x 2 (Flexible vs. Non-Flexible
horizon) x 7 (seven decision periods) factorial design bonus scenario (see appendix 2).
Subjects were randomly assigned to conditions, completed background information, and
were instructed to read the scenario and complete the questionnaire across seven different
decision periods that varied the current performance of the bonus compensation
throughout the compensation time horizon (see below).

Pmptions of control. Subjects’ perceptions of control were measured via the
following measure:

On the scale below please indicate the level of control you feel you have

over the structure of this bonus (how much control do you feel you have

over your ﬁrm exceeding last period’s (the industry average) sales in the

allotted time ﬂame?)
Subjects indicated their response on a seven point scale with the following anchors: 1 =
low control and 7 = high control. T-test results between internal and external standard
conditions (see Table 2) were signiﬁcant (t = 4.51, p<.001) and in the predicted direction
(internal standard, M = 6.33, s.d. = .7071; external standard, M = 4.25, s.d. = 1.17). T-

test results between the ﬂexible and non-ﬂexible time horizon conditions (see Table 2)

41

were signiﬁcant (t = 2.92, p<.05) and in the predicted direction (ﬂexible horizon M =

6.33, s.d. = .71; non-ﬂexible standard, M = 5.13, s.d. = .35).

TABLE 2
Pilot Study 2 - Meet Results

 

INTERNAL VS EXTERNAL STANDARD

Internal/flexible Exgrnal/ﬂexible

Variable Mean s.d. Mean s.d. t

Perception of control 6.33 .71 4.25 1.17 4.51, p<.001

FLEXIBLE VS NON-FLEXIBLE HORIZON

Flexi Ie/ln rnal N n-Flexi le/Internal

Mean s.d. Mean s.d. t

Perception of control 6.33 .71 5.13 .35 2.92, p<.05

Consequently, the internal/extemal measure and horizon ﬂexibility manipulations
performed as expected. All other variables used were constructed identically to pilot
study 1.

The measures tested in pilot study one and pilot study two were used in two

studies that tested the complete model. These studies are described below.

Study 1
Methods

Subjects. Subjects in this experiment were 137 executive and weekend master of

business administration students in a management class at Michigan State University.

42

Over 95% of these students had received some form of contingent pay, with 55% having
' direct experience with stock options. Eighty-two percent of the subjects were men.
Desigp and procedures. Subjects voluntarily participated in the study during a
single class session. Subjects were given a packet containing a scenario-based
questionnaire and scenario background information. The scenario-based questionnaire
was a 2 (High Implicit Risk vs. No Implicit Risk) x 2 (Long Horizon - 3 years vs. Short
Horizon - 18 months) x 7 (seven decision periods) factorial design stock option scenario
(see appendix 1). Subjects were randomly assigned to conditions, completed
background information, and were instructed to read the scenario and complete the
questionnaire by answering questions at seven different time periods that varied the

current performance of the stock throughout the compensation time horizon (see below).

Measures

Contracted value of stock options. For the stock option scenario, contracted value
is the strike price of the options awarded to subjects ($29.00).

Cunerrpperfonnang Current performance is measured as the spread between the
market price and the strike price at each of the seven decision points. Current
performance by decision:

decision 1 = $4.00 per share; decisions 2 through 4 = $13.00 per share;

decision 5 = $7.00 per share; decision 6 = $4.00 per share; decision 7 =

$0.00 per share (these numbers are identical to those used in pilot study 1).

Implicit risk. Following Shelly and Omer (1996), implicit risk was manipulated

as a between groups factor identically to pilot study 1.

Subjects in the high implicit risk condition read the following:

43

Your ﬁrm manufactures nonessential components for your customers. As
a result, economic conditions directly affect your ﬁrrn’s stock price.
Current economic conditions are quite unstable. In addition, your stock
price and others in your industry have been quite volatile (rising and
falling dramatically) over the last decade. This pattern is projected to
continue over the next few years. As such, industry analysts consider your
stock to be a high risk investment.

Subjects in the no implicit risk condition read:
Your ﬁrm manufactures critical components for your customers. As a
result, your frnn’s stock price is sheltered ﬂom any negative effects of the
economy. Current economic conditions are quite stable. In addition, your
stock price and others in your industry have generally continued to rise
over the last decade. This pattern is projected to continue over the next
few years. As such, industry analysts consider your stock to be a very
safe investment.
Subjects indicated their response on a seven point scale with the following
anchors: 1 = risky form of compensation and 7 = safe form of compensation.

The implicit risk manipulation was checked via the following measure:

On the scale below please indicate your perception of the likelihood of
making your proﬁt goal. . .

Subjects indicated their response on a seven point scale with the following anchors: 1 =
low likelihood and 7 = high likelihood. The t-test results between conditions (see Table
3) were signiﬁcant (t = 5.23, p<.001) and in the predicted direction ( no implicit risk, M
= 5.67, s.d. = .98; high implicit risk, M = 4.66, s.d. = 1.26).

TABLE 3
Implicit Risk '

 

No Implicit Risk ﬂgg' b Implicig Risk

 

 

Variable Mean s.d. Mean s.d. T
Likelihood 5.67 .98 4.66 1.26 5.23, p<.001
1= low & 7 = high

44

In order to ensure that implicit risk manipulation was driving subjects’ responses
above it was necessary to ensure that subjects’ perceptions of implicit risk (their general
beliefs about the risk of stock options as a form of compensation) were replaced with the
level of the implicit risk inherent in each condition. As such, prior to the manipulation,
subjects were asked to provide their general assessment of the risk of stock options as a
form of compensation with the following item:

Please rate your feelings about the risk of stock options as a form of
compensation

Subjects’ indicated their response on a seven point scale with the following anchors: 1 =
risky form of compensation and 7 = safe form of compensation. This assessment of the
risk of stock options and the dichotomous variable implicit risk were entered into a
regression equation. The implicit risk manipulation demonstrated a signiﬁcant and
negative inﬂuence on likelihood responses (p<.001); however, subjects prior beliefs had
no signiﬁcant influence indicating that subject’s prior beliefs had no signiﬁcant inﬂuence

on the results of this study (see Table 4).

 

 

TABLE 4
Likelihood'
Model 1 Model 2

Riskiness of options (prior belief) .02 (.07) .02 (.06)
Implicit risk -1 .00‘" (.19)
Adjusted R2 -.or .16‘"
Increase in adjusted R2 .16“”

Df 135 134

 

‘ N= 137. mp<.001
Horizon length. Two horizon length conditions (three years and 18 months) were

presented.

45

Subjects in the long horizon condition read:

In three years (on December 31, 2004) you may purchase any or all of the
10,000 shares for $29.00 per share.

Subjects in the short horizon condition read:

In 18 months (on June 30, 2003) you may purchase any or all of the
10,000 shares for $29.00 per share.

In addition each decision provides the subjects with the current date and time until
exercise.

Promoted value of compeniation (PVC). Following Shelly and Omer (1996),
subjects’ expectations measured the projected (expected future, but at risk) outcome (in
this case future compensation). During each decision period subjects’ projected value of
compensation was collected using the following question:

At this point in time, what is the proﬁt per share you expect to earn at

exercise (when you are able to purchase and resell these shares) in XX
years (months)?: $ per share.

 

Endowed wealth. During each decision period subjects’ level of endowed wealth
was measured using the following questions.
At this point in time, what is the proﬁt per share you are conﬁdent enough

to use to secure ﬁnancing on a purchase you really want to make?:
$ per share.

 

Aspiration level. Subjects’ aspiration level was manipulated in the scenario.

Subjects in all conditions will read:

 

5 The following question was also used with the EMBA sample to measure the amount subjects had
endowed into their perceptions of personal wealth: Deﬁning personal wealth to include both current
income and ﬁdly anticipated ﬁrture income, at this point in time, what profit per share would you include in
your calculations of personal wealth? The two questions were signiﬁcantly and highly correlated (above
.84, p<.01, for each decision) and the average measure intraclass correlations were all above .90). One
question was desired to reduce subject fatigue; as such, the endowed wealth question above was chosen.

46

YOUR GOAL.

Your goal, set by the board of directors, is to increase your ﬁrm’s stock
price to $35.00 a share (by December 31, 2004/June 30, 2003).

Subjects were asked to complete the following item prior to beginning the

scenario as a manipulation check:

What is your proﬁt per share goal at exercise (when you are able to
purchase and resell these shares?: $ per share.

 

Interestingly less than half (48.6%) of subjects set a proﬁt per share goal of $6.00
per share at exercise; while 51.4% varied ﬂom the intended manipulation. As such,
subjects’ reported goals are used in the analysis as the aspiration level. More speciﬁcally,
consistent with Lant, 1992, 40.9% indicated that their goal exceeded $6.00 per share
(range $7.00 to $35.00), while 10.2% set a goal lower than $6.00 (range $0.00 to $5.90)
reﬂecting a positive bias in the setting of aspiration levels. This ﬁnding will be elaborated
on in the discussion section.

Problem ﬂaming (PF ). As discussed above, it is cognitively difﬁcult for agents to
explicitly separate and independently evaluate what actually drives their feelings
completely (F iske & Taylor, 1991). As such, problem ﬂaming was measured with a
single scale, via the following measure:

At this point in time I feel as though the shares are performing. . .

Subjects indicated their response on a seven point scale with the following
anchors: l = behind where I’d like them to be and 7 = ahead of where I’d like them to be.

mm orientation. The following scale was constructed to assess subjects’
wealth orientation speciﬁc to this stock option scenario:

At this point in time I am more concerned about. . ..

47

Subjects indicated their response on a seven point scale with the following anchors: l=
increasing price and 7 = maintaining price (wealth orientation was reverse coded)
Controls. I controlled for subjects’ gender and group membership (EMBA vs.
weekend MBA group). Group membership demonstrated a signiﬁcant positive inﬂuence
on endowed wealth and a signiﬁcant negative inﬂuence on wealth orientation. Group
membership was not signiﬁcant in any of the other the models and its removal did not
affect the results. Consequently, in the interest of parsimony, the results I report include
group membership only in models in which it was a signiﬁcant inﬂuence. Further,
gender did not demonstrate a signiﬁcant inﬂuence in any of the analyses and its removal
did not affect the results; hence, for it was dropped. In my initial estimations I also
controlled for subjects’ self-rating of risk (their propensity to take risk) using the item:

Please rate your willingness to undertake risky business propositions as
compared to others at or near your level in your ﬁrm.

Subjects indicated their responses on a seven-point scale with the anchors 1= much less
willing to accept risk and 7 = much more willing to accept risk (MacCrimmmon and
Wehrung, 1986). This variable was not signiﬁcant in any of the models; further, its
removal did not affect the results. Again, for the sake of parsimony, the results I report
do not include subjects’ self-rating of risk.

Finally, I calculated variance inﬂation factors to test for the possibility of

multicollinearity across the independent variables and found no violations in the analyses.

Results
Hypothesis 1 was tested using multiple regression analysis using data ﬂorn

individual decision periods and Hypothesis 3b, was tested with multiple regression across

48

decisions 1 — 4 (see explanations below). Repeated measures regression was used to test

all other hypotheses (see Cohen & Cohen, 1983, Chapter 11 and Hollenbeck, Ilgen, &

Sego, 1994 for thorough discussions of repeated measures regression). Tables 5 & 6

include the descriptive statistics and correlation matrix for variables included in the

 

 

 

 

 

study.
TABLE 5
Descriptive Statistics and Correlations‘
hdean ssh 1 2
I lrnplicit Risk .50 .50
2 Time Horizon Length .50 .50 -.02
3 Projected Value of Conﬁnsation D01 8.27 4.71 -.28" .058
' N=137. n p<.01 (2-tailed).
TABLE 6
Descriptive Statistics and Correlations‘
Mean s.d. 1 2 3 4 5 6 7
1 Group .35 .48
2 Projected Value 8.57 6.51 .00
3 Endowed wealth(lagged) 5.19 4.69 .10" .47”
4 Goal 7.97 4.34 -.04 .31“ .29"
5 Problem framing 4.03 2.19 .02 .54" .07" -. 9"
6 Wealthorientation 3.39 2.14 -.14 .23” .03 -.23” .53”
7 GroupxPVC 0.00 3.23 .30" -.03 .06 -.09 .03 .02

8 Group x problem ﬂaming 1.43

‘N=959. * p<.05;"p<.01(2-tailed).

2.33

.84” .19“I

.10”

208* .33” .02 .50”

Implicit risk and time horizon. Consistent with Hypothesis 1, implicit risk

exhibits a negative association with the PVC (p<.001). However; contrary to Hypothesis

2, the length of the compensation time horizon did not signiﬁcantly inﬂuence PVC. In

49

addition, although Hypotheses l and 2 were tested using data ﬂom decision 1. They
were also tested using data ﬂom each of the other six decision periods individually, via
multiple regression analysis. Results were similar across decision periods. Time horizon
length exhibited no signiﬁcant inﬂuence on PVC in any period (this was a test of the
difference between lengths of two time periods rather than a test of the time left to
exercise). However, implicit risk demonstrated a signiﬁcant and negative inﬂuence on
PVC in the ﬁrst four periods. In periods 5 -— 7 the inﬂuence of implicit risk on PVC was

not signiﬁcant. Results ﬂom decision 1 and decision 5 are shown in Table 7.

 

 

TABLE 7
Projected Value of Compensation ‘
Model 1 Model 2

Independent Variable (decision #1) (decision #5)
Implicit risk -2.65” (.77) -1.1 1 (.62)
Time horizon .51 (.77) .61 (.62)
Adjusted R2 .07“ .02

Qf 135 135

 

'L Unstandardized coefﬁcients are shown with standard errors in
parentheses. "p<.01; ‘p<.05

Endowed weaﬂ The relationship between PVC and endowed wealth was tested
using repeated measures regression based upon 959 observations. Variance was
partitioned into two orthogonal sources: between subjects and within subjects variance.
As shown in Table 8, between subjects variance accounts for 39.39% of the total
variance, while 60.61% of the total variance is attributable to within subject’s variance.
Consistent with Hypothesis 3a, the reference point of endowed wealth is strongly and
positively associated with the current PVC. The PVC was measured during each of the

seven decision periods; hence it is a within subjects factor and accounts for 56% of the

50

total variance and 92% of the within subjects variance (p<.01). In addition, although the
main effect for group (EMBA) was non-signiﬁcant, the interaction term of group and

projected value was signiﬁcant (p<.01).

 

 

TABLE 8
Endowed wealth '
Hierarchical Independent Unstandardized Total Incremental Incremental
Step Variable Coefﬁcients R2 R2 Variance Within Incremental
Subjects Variance Between
(60.61%)' Subjects (39.3%)"
Group
1 (EMBA) .67“ .01 .01 .03’
2 Projected value .54" .57” .56" .92"
Group x
3 Projected value .16" .58" .01 .02"

 

" N =959 (7 observations per I37 subjects) (df =959-l 3 7-k- I ). b N =1 3 7 ( I observation per I37 subjects)
(df=l37-k-l), " p<.05;"”"p<.01

Hypothesis 3b predicts that when the current performance of a form of
compensation declines the endowed wealth reference point will decrease more slowly
than the projected value of compensation.

Current value increases ﬂom decision one to decision two, but remains steady
over decisions two — four. It begins to decline in decision ﬁve. As such, decisions ﬁve —
seven were used to test hypothesis 3b. To test the hypothesis, three variables were
created to reﬂect the percentage change in endowed wealth ﬂom decisions four to ﬁve,
ﬁve to six and six to seven. Three variables were also created to reﬂect the percentage
change of PVC score in the same manner. The percentage change scores between PVC
and endowed wealth ﬂom decision four to decision ﬁve, and decision six to seven are
non-signiﬁcant. Interestingly however, ﬂom decision ﬁve to decision six endowed
wealth decreases at a rate of 40%, while PVC decreases at a rate of only 30%. This
difference is in the opposite direction and signiﬁcant (t = -3.40, p<.001). The means for

PVC and endowed wealth are shown in Table 9 and displayed in graphic form in Figure

51

5. As such, it appears that when the current performance of a form of compensation

declines the endowed wealth reference point does not decrease more slowly than the

projected value of compensation. Thus, Hypothesis 3b is not supported.

 

 

TABLE 9
Mean PVC & Endowed Wealth'
Mean Mean
% change % change

Decisions PVC EW t value
Dec. 1 to 2 78% 75% -.74
Dec.2to3 -1% 3% 195"
Dec. 3 to 4 -l% 3% 1.29
Dec.4t05 -31% -34% -l.14
Dec. 5 to 6 -30% -40% 340"”
Dec. 6 to 7 -87% -82% .70

 

'N= 121 (Dec.1t02,2to3), 123 (Dec.3to4), 117,(Dec.4t05),

95 (Dec. 5 to 6) m p<.001; ”p<.01; *p<.05

Figure 5
Mean PVC & Endowed Wealth

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Problem M' g and attainment discrewcy. Hypothesis 4a, 4b, 4c, and 5 were

tested using repeated measures regression based upon 959 observations. Again, variance

was partitioned into two orthogonal sources: between subjects and within subjects

52

variance. Subjects’ aspiration level was measured prior to decision 1 and remains
constant throughout the analysis, hence, it is a between subjects effect. Endowed wealth,
PVC, and problem framing were measured during each decision period; thus, they
account for within subjects variance. As shown in Table 10, between subjects variance
accounts for 11.58% of the total variance, while 88.42% of the total variance is
attributable to within subjects variance.

Hypothesis 4a predicts the projected value of compensation will have a positive
inﬂuence on ﬂaming. Conversely, Hypotheses 4b and 4c predict that the endowed
wealth reference point and the aspiration level will have a negative inﬂuences on
ﬂaming.

In order to formally test Hypotheses 4a, 4b, 4c, the problem ﬂaming variable was
regressed on subjects’ PVC, reference point of endowed wealth (lagged one period), and
initial aspiration level in a repeated measures regression equation. To support
Hypotheses 4a and 4b, and 4c the PVC should be positively and signiﬁcantly related to

problem ﬂaming while the endowed wealth reference point and the aspiration should be

 

 

TABLE 10
Problem Framing'
Hierarchical Independent Unstandardized Total Incremental Incremental Incremental
Step Variable Coefﬁcients R2 R2 Variance Within Variance Between
Subjects (88.42%)' Subjects 01.58%)"

Projected
1 value .24” .29" .29 .33"

Endowed
2 wealth lagged -.09“ .33” .04 .05"
3 Aspiration -.13" .39” .06 .51"

 

" N =959 (7 observations per 1 3 7 subjects) (df=959-I37-lt- 1). TN =1 3 7 ( l observation per 1 3 7 subjects)
(df=l37-k-l), ”p<.01;

53

signiﬁcantly and negatively related to problem ﬂaming. In support of Hypotheses 4a, 4b,

and 4c, problem ﬂaming is signiﬁcantly and positively inﬂuenced by the PVC (p<.01)

 

W6
ProblemFramiugundWeulthOrientatiouMeuns

 

 

 

 

 

N b gamma, j

mwhmcaxr

 

 

 

 

 

 

 

Figure 7

Spread,l’VC, WWe-ImaAspu-uion |

$14.25
$13.50
$12.75
$12.11)
$1 1.25
$10.50
$9.75
$9.00
$8.25
$7.50
$6.75
$6.00
$5.25
54.50
E $3.75
; $3.00
$2.25

 

+Spreud
+PVC

[ +A‘pﬂﬂlm _ “J:

 

 

 

 

$1.50
$0.75
$0.00

 

 

54

and signiﬁcantly and negatively inﬂuenced by the reference point of endowed wealth
(p<.01) and aspiration level (p<.01). Consequently, Hypotheses 4a and 4b, and 4c are
supported. Results of Hypotheses 4a & b and 4c presented in Table 10.

Subjects’ problem ﬂaming and wealth orientation scores by decision are graphed
in Figure 6. The current performance (actual spread), subjects’ means for PVC and
endowed wealth are graphed in Figure 7.

Regressing problem ﬂaming on subjects’ PVC, endowed wealth, and initial
aspiration level also serves as a test Hypothesis 5, which states a maximum of one
reference point drives subjects’ problem ﬂaming during the seven decision periods.
Since both reference points demonstrate signiﬁcant negative inﬂuences on problem
ﬂaming, more than one reference point is salient during each period. Consequently,
Hypothesis 5 is not supported. However, it is interesting to note that it appears that the
aspiration level has a much stronger effect on problem ﬂaming than the reference point of
endowed wealth as it explained over 51% of the between subjects variance (see Table
10).

Wealth orientation. The relationship between problem ﬂaming and wealth
orientation was tested using repeated measures regression based upon 959 observations.
As such, variance was partitioned into two orthogonal sources: between subjects and
within subjects variance. As shown in Table 11, between subjects variance accounts for
30.52% of the total variance, while 69.48% of the total variance is attributable to within
subjects variance. Wealth orientation and problem ﬂaming were measured during each
decision period, hence they are within subjects effects. Hypothesis 8 predicts that

problem ﬂaming demonstrates a positive inﬂuence on wealth orientation, such that

55

positive ﬂaming is positively associated with a wealth maintenance orientation and
negative ﬂaming is negatively associated with a wealth maintenance orientation. A test of
this hypothesis reveals that subjects’ problem ﬂaming is positively associated with

wealth orientation, explaining 41% of the within subjects variance (p<.01); thus
conﬁrming Hypothesis 8 (see Table 11).

TABLE 11
Study 1 -Wealth Orientation ‘

 

Hierarchical Independent Unstandardized Total Incremental Incremental Incremental

 

Step Variable Coeffeicents R2 R2 Variance Within Variance Between
Subjects (69.48)‘ Subjects (30.52) "
1 Group (EMBA) -.65" .02 .02 .06"
2 Problem Framing .52" .31" .29" .41 ‘”"
Group x problem
3 ﬂaminL -.18 .31" .00 .oo

 

° N =959 (7 observations per 13 7 subjects) (df=959-I3 7-k- l). b N =1 3 7 (I observation per 13 7 subjects)
(df=l37-k-1), "p<.01;

Study 2

Mam

Subjects. 102 masters of business administration students in a management class
participated in the study. Subjects’ mean age was 28 years and the mean for work
experience was 4.75 years. Over 83% of the subjects had experience with incentive-
based pay forms with 75% having direct experience with annual bonuses.

Desigh and Procedures. Subjects voluntarily participated in the study during a
single class session. As in pilot study 2, subjects were given a packet containing a
scenario-based questionnaire and scenario background information. The scenario-based
questionnaire was a 2 (Internal vs. External Performance Standard) x 2 (Flexible vs. Non-

Flexible horizon) x 7 (seven decision periods) factorial design bonus scenario. Subjects

56

were randomly assigned to conditions, completed background information, and were
instructed to read the scenario and complete the questionnaire across seven different
decision periods that varied the current performance of the bonus compensation
throughout the compensation time horizon (see below).
Measures

Problem ﬂaming. As in the pilot studies and study 1, problem ﬂaming was
measured via the following measure:

At this point in time I feel as though the bonus is. . .
Subjects’ indicated their response on a seven point scale with the following anchors: 1 =
behind where I’d like it to be and 7 = ahead of where I’d like it to be.

Wealth orientation. As in the pilot studies and study 1, wealth orientation was
measured via the following measure:

At this point in time I am more concerned about. . .
Subjects indicated their response on a seven point scale with the following anchors: 1=
increasing the bonus and 7 = maintaining the bonus (wealth orientation was reverse
coded).

Horizon ﬂexibilig. Horizon ﬂexibility was manipulated. Subjects in the ﬂexible
horizon read:

In December 2001 a long-term bonus plan was added to your

compensation package. The minimum length of the plan is three years,

but you have the option to extend it to four years. The plan works as

follows:

On December 31, 2004 your ﬁrm’s three-year unit sales ﬁgure will be

compared to the unit sales ﬁgure of the last three-year period. The plan

rewards you for achieving percentage increases in the three-year unit sales
according to the following schedule:

57

OPTION

If you wish, you can extend your plan for up to one additional year, in the

event sales are not performing to plan on December 31, 2004. This allows
you to the opportunity to cash out your bonus anytime between December
31, 2004 and December 31 , 2005 — at the time most advantageous to you.

Subjects in the non-ﬂexible horizon read:

In December 2001 a long-term bonus plan was added to your
compensation package. The length of the plan is three years. The plan
works as follows:

On December 31, 2004 your ﬁrrn’s three-year unit sales ﬁgure will be
compared to the unit sales ﬁgure of the last three-year period. The plan
rewards you for achieving percentage increases in the three-year unit sales
according to the following schedule:

Performance standards. Performance standards were manipulated. Subjects in

 

the internal standard condition read:

On December 31, 2004 your ﬁrm’s three-year unit sales ﬁgure will be
compared to the unit sales ﬁgure of the last three-year period. The plan
rewards you for achieving percentage increases in three-year unit sales
according to the following schedule:

Subjects in the external standard condition read:

On December 31, 2004 your ﬁrm’s three-year unit sales will be compared
with the average industry unit sales for the same period. Most of your
competitors also have bonus plans that are based on exceeding the
industry average unit sales. Consequently, you are unsure what the period
industry average unit sales ﬁgure will be on December 31, 2004. The plan
rewards you for achieving percentage increases over the three-year
industry average unit sales according to the following schedule:

Perceptions of control. Subjects’ perceptions of control were measured via the
following measure:

On the scale below please indicate the level of control you feel you have

over the structure of this bonus (how much control do you feel you have

over your ﬁrm exceeding last period’s (the industry average) sales in the

allotted time ﬂame?) Subjects’ indicated their response on a seven point
scale with the following anchors: l = low control and 7 = high control.

58

Controls. In my initial estimations I controlled for subjects’ gender and subjects’
self-rating of risk (their propensity to take risk) using the item:

Please rate your willingness to undertake risky business propositions as
compared to others at or near your level in your ﬁrm.

Subjects indicated their responses on a seven-point scale with the anchors 1= much less
willing to accept risk and 7 = much more willing to accept risk (MacCrimmmon and
Wehrung, 1986). Neither the gender or subjects’ self-rating of risk were signiﬁcant in
any of the models; further, their removal did not affect the results. Thus, for the sake of
parsimony, the results I report do not include subjects’ self-rating of risk.

Finally, I calculated variance inﬂation factors to test for the possibility of
multicollinearity across the independent variables and found no violations in the analyses.
_R_esth§

Table 12 includes the descriptive statistics and correlation matrix for variables

 

 

included in the study.
TABLE 12
Study 2 -Descriptive Statistics and Correlations '
Mean s.d. 1 2
1 Problem Framing 3.78 1.83
2 Perceptions of Control (POC) 4.71 1.46 .03
3 Problem ﬂamigx POC .07 2.64 .00 -.01

 

" Né7r4. * p<.05; u p<.01 (2-tailed).

Hypotheses 6 & 7 were tested via t-tests. These tests are described below.
Performance standards. Hypothesis 6 predicts that contingent compensation
forms tied to internal standards will create greater perceptions of control than contingent

compensation forms tied to external standards. T-test results between internal and

59

external standard conditions (see Table 13) were signiﬁcant (t = 3.57, p<.001) and in the
predicted direction (internal standard, M = 5.20, s.d. = 1.37, external standard, M = 4.22,
s.d. = 1.41). Consequently, Hypothesis 6 received strong support.

Horizon ﬂexibility. Similarly, Hypotheses 7 states that contingent compensation
forms that incorporate a ﬂexible horizon will create greater perceptions of control than
contingent compensation forms that incorporate an inﬂexible horizon. Tetest results
between the ﬂexible and non-ﬂexible time horizon conditions (see Table 13) were
marginally signiﬁcant (t = 1.86, p<.10) and in the predicted direction (ﬂexible horizon M
= 4.96, s.d. = 1.49; non-ﬂexible standard, M = 4.43, s.d. = 1.40). Consequently,
Hypothesis 7 is also supported. Results of Hypotheses 6 & 7 are presented in Table 13.

TABLE 13

Study 2 - t-test Results
INTERNAL VS EXTERNAL STANDARD

 

Internal External
Variable Mean s.d. Mean s.d. T
Perception of control 5.20 1.37 4.22 1.41 3.57 p<.001

 

FLEXIBLE VS NON-FLEXIBLE HORIZON

 

Flexi Ie were
Mean s.d. Mean s.d. t
Perception of control 4.96 1.49 4.43 1.40 1.86 p<. 10

 

Moderated repeated measures regression was used to test Hypothesis 9.
Moderation. The moderation effect was tested using repeated measures
regression based upon 714 observations. Variance was partitioned into two orthogonal

sources: between subjects and within subjects. As shown in Table 13, between subjects

60

variance accounts for 44.51% of the total variance, while 55.49% of the total variance is
attributable to within subjects variance. Hypothesis 9 predicts that relationship between
problem ﬂaming and wealth orientation is moderated by perceptions of control. Problem
ﬂaming and wealth orientation are within subject factors, while perceptions of control
(measured once at the start of the exercise) is a between subjects effect. Hypothesis 9
was tested by regressing subjects’ wealth orientation score on their problem ﬂaming
score, perceptions of control score, and the product term of both using hierarchical
repeated measures regression. This test reveals that the main effect of problem ﬂaming
on wealth orientation is positive and Signiﬁcant (p<.01). It also shows that the product
term of problem ﬂaming and perceptions of control has a negative and signiﬁcant

inﬂuence on wealth orientation explaining 2% of the within subjects variance (p<.01).

 

 

TABLE 14
Study 2 -Wealth Orientation '
Hierarchical Unstandardized Independent Total Incremental Incremental Incremental
Step Coeffeicients Variable R2 R2 Variance Within Variance Between
Subjects (55.49)‘ Subjects (44.512
Problem
1 .27" Framing .06“ .06" .12"
Perceptions of
2 -.09 Control .06 .00 .00
Problem
Framing x
Perceptions of
3 -.07""I Control .08” .01" .02"

 

" N= 714 (7 observations per 102 subjects) (df=959-102-k- I ). b N =1 02 (I observation per 102 subjects)
(df=102-k-l), ”p<.01;

This indicates that perceptions of control moderates the inﬂuence of problem ﬂaming on
wealth orientation such that high perceptions of control decrease the inﬂuence of problem
ﬂaming on wealth orientation and low perceptions of control increase the inﬂuence of

problem ﬂaming on wealth orientation. Table 14 displays the results of the moderated

61

relationships and the effects of the control variable. The interaction effect is illustrated in

 

 

 

 

 

 

 

Figure 8.
Figures
Interaction Effect

I If g— 4 — f ﬂ — .2.7

ﬂ _ _ __
I 8 #1312?

E
. 5
l E
l ’
1

 

 

 

; -1 so +1 so

 

62

 

CHAPTERS

DISCUSSION

I have developed and tested a model that disentangles the incentive properties of
various forms of compensation in an attempt to better understand how well compensation
works, how executives value compensation, and the effects of executive compensation on
executives’ wealth orientation. Drawing on behavioral decision theory and research on
corporate governance and compensation I distinguish between elements that are
embedded in the compensation contract and perceptual dimensions that emerge in
response to these elements and test how these dimensions can differentially effect
executives’ perceptions (c.f. Wiseman et al, 2000). The results presented above conﬁrm
broad support of the model and provide further empirical evidence that, as Wiseman and
Gomez-Mejia (1998) suggest, integrating behavioral decision theory and agency theory
can help us better understand executives’ perceptions of and responses to various

compensation schemes.

Implications and Future Research

The ﬁndings here suggest that executives calculate a value for each type of
compensation that is inﬂuenced by the implicit risk of that form of pay. This process
differs ﬂom valuation methods such as the Black-Scholes, which provide us with ex ante
estimates of shareholder cost. This ﬁnding has important implications for executive
compensation research as it suggests that the incentive properties can be quite different
across different forms of contingent pay and different risk conditions, which raises

questions about the effectiveness of our current methods of compensation valuation.

63

Endowed wealth. In line with the BAM (W iseman and Gomez-Mejia, 1998) I
also ﬁnd that executives actually do endow a portion of their future compensation (stock
options and bonuses) into calculations of personal wealth. This is an important ﬁnding as
it demonstrates that contingent pay forms, including stock-options, actually can assume
downside risk characteristics. This ﬁnding directly contradicts the agency theory
assumption that stock-options contain no downside risk and call for revising the way we
view the risk properties of contingent pay in executive compensation research.

In addition, I ﬁnd that factors deﬁned as lying outside the compensation
agreement can also play a signiﬁcant role in how executives endow wealth. In particular,
I ﬁnd that over 56% of the total variance in endowed wealth is between subjects variance,
conﬁrming the Kahneman, Knetsch, and Thaler (1990) ﬁnding that the endowment effect
seems, in large part, to be driven by individual differences or preferences. Consequently,
a more speciﬁc and thorough examination of how compensation is perceived by
executives should incorporate individual-level factors.

thimistic aspirations. Although the results of this study demonstrate that
executives do endow a portion of the value of future compensation, as Lant, (1992) ﬁnds
they also indicate that executives are quite optimistic in the manner in which they set
future aspirations for wealth accumulation. In addition, in both studies the aspiration level
seems to have acted as a ceiling for the amount of wealth an executive endowed (see
Figure 6 for an illustration of this relationship ﬂom study 1). As such, in this study it
appears that the aspiration level may have inﬂuenced the amount of wealth executives

endowed. Perhaps future research can better isolate the effects of the aspiration level on

64

endowment in order to understand what actually drives this process in dynamic
compensation situations.

Multiple reference mints. I also suggested that it is particularly important to
distinguish between the reference points executives use to ﬂame decisions. The ﬁndings
here reveal that managers use both their reference point of endowed and an aspiration
level to ﬂame decisions. Recognizing these reference points separately is critical as it is
possible for the current projected value of compensation to exceed the aspiration level
(initial wealth goal ﬂom a form of compensation) but to have fallen below the endowed
wealth reference point (e.g. option value had risen in excess of the aspiration level
substantially and this value is endowed, the spread subsequently falls below that
reference point, yet it remains above the aspiration level).

This situation would create competing inﬂuences, as the former would result in a
positive ﬂame, creating concern for wealth maintenance; however, the later would create
a loss ﬂame, invoking concern for increasing wealth. This is important as Sullivan and
Kida (1995) ﬁnd that typical reference dependent research ﬁndings do not hold under
circumstances in which choice alternatives fall between multiple points of reference (i.e.
managers exhibit both risk taking and risk avoidance behavior in these conditions). This
has major implications for the study of how compensation forms inﬂuence executive
choice behavior, and calls for new research that examines the effects of multiple points of
reference and their combined inﬂuences on executives’ wealth orientation and risk taking
behaviors.

Wealth orientation. The ﬁndings of this study also show that problem ﬂaming

signiﬁcantly inﬂuences executives’ wealth orientation. This is another important ﬁnding

65

I inf-Ii"

as understanding the changes in the wealth orientation of executives should allow us to
better predict the type of options executives are likely to search for in response to various
incentives and situations (i.e. an orientation toward increasing wealth would likely result
in a bias toward searching for more risk-laden choices than an orientation toward wealth
maintenance). In addition, it should allow us to move closer understanding how
executives perceive and are considering responding to various compensation contexts and
to better gauge the effectiveness of compensation forms and the compensation contract.
Perceptions of control. This study also reveals that executives’ perceptions of
control are an important moderating inﬂuence on the effect of problem ﬂaming on wealth
orientation. This is important, as the majority of recent studies examining decision
making have been conducted using models of probability in simple gambling situations
(chance events). However, I suggest that simple gambling contexts are quite different
ﬂom executives’ decision contexts, as in gambling, perceptions of control are largely
discounted (Shapira, 1995). Although some studies have shown that individuals believe
that they can exert some degree of control in chance events, research indicates that these
perceptions of control are due to the introduction of “manipulations that are suggestive of
skill, such as competition, choice, familiarity, and involvement” (Taylor & Brown, 1988).
Hence, I suggest that results ﬂom simple expected value gambling contexts that
offer no skill-like manipulation, such as those that involve subjects choosing between a
sure win or some less than sure “chance” they will win some amount much larger than
the sure win, or betting on the results of a coin ﬂipped by someone else, do not fully
generalize to executive decision making contexts, as they over inﬂate the inﬂuence of

downside risk on risk-taking behavior and are not reﬂective of how executives view their

66

situations (Shapira, 1995). As such, it is likely that the results of this research may not
generalize to executive compensation and decision making situations. Consequently, in
order to account for executives’ perceptions of control, I argue that we must examine the
effects of various types of compensation forms in dynamic situations that more closely

mirror executive compensation contexts.

Conclusion

Finally, I suggest that wealth orientation inﬂuences how executives may perceive
each of the options available to them, such that they may focus on either the dangers or
the opportunities inherent in each choice, depending on their wealth orientation at the
time of consideration. Hence, I suggest that wealth orientation inﬂuences executive risk
preferences in making decisions on behalf of principals, when those decisions hold
implications for the executive’s personal wealth. Speciﬁcally, as executives’ wealth
orientation shifts toward maintenance their risk preferences should become more
conservative, as they seek to protect personal wealth (real or projected). Consequently, I
suggest that future research should speciﬁcally examine the relationship between wealth
orientation and risk taking under various compensation contexts.

The results of these studies raise important questions about the practice of mixing
different forms of pay with varying wealth prospects into a single measure, as this is
likely to create variables that contain contrasting inﬂuences on executives. This is
problematic for the interpretation of research as it presents measurement problems that
hinder the capability to understand these inﬂuences on executives’ perceptions, behavior,
and/or ﬁrm performance. I hope that the ﬁndings of these studies will prompt

compensation researchers to revise the manner in which we create our measures, which

67

should allow us to better understand the incentive properties of pay forms in various
contexts and help us to more thoroughly investigate the relationship between

compensation and executive behavior.

68

REFERENCES

Abowd, J. M. and Kaplan, BS (1999). Executive compensation: six questions that need
answering. Journal of Economic Perspectives. 13(4): 145-168.

Adams, J. S. (1963). Toward and understanding of inequity. Journal of Abnormal and
Social Psychology, 67 422-436.

Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (ed.). Advances in
experimental social;sychology. New York: Academic Press.

Agrawal, A. and Mandelker, G.N. (1987). Managerial incentives and corporate
investment and ﬁnancing decisions. The Journalof Finance, 42: 823-837.

Beatty, RP. and Zajac, EJ. (1994). Managerial incentives, monitoring, and risk bearing:
A study of executive compensation, ownership, and board structure in initial public
offerings. Administrative Science Quarterly. 39: 313-355.

Black F. and Scholes M. (1973). The pricing of options and corporate liabilities. Journal
of Political Econory. 81: 637-659.

Bloom, M. and Milkovich, G.T. (1998). Relationships among risk, incentive pay, and
organizational performance. Academy of Management Journal, 41: 283-297.

Bryan, S., Hwang, L., and Lilen, S. (2000). CEO stock-based compensation: An
empirical analysis of incentive-intensity, relative mix, and economic determinants.
Journal of Business, 73: 661-693.

Coffee, J. (1988). Shareholders versus managers: The strain in the corporate web. In J .C,
Coffee, L. Lowenstein, and S. Rose_Acherman (Eds), Knights, raiders, targets: 40-
62. New York: Oxford Press.

Cooper, A. C., Woo, C. Y., and Dunkelburg, W. C. (1988). Entrepreneurs' perceived
chances for success. Journal of Business Venturing. 3(2) 97-108.

Chakraborty, A., Kazarosian, M., and Trahn, EA. (1999). Uncertainty in executive
compensation and capital investment: A panel study. Financial Management, 28:
126-139.

Donaldson, G. (1961). Corporate Debt Capacity: A Studv of Corporate Debt Policy arg
the Determipation of Corporate Debt Capacity. Boston: Division of Research,
Graduate School of Business Administration, Harvard University.

Eisenhardt, K. (1989). Agency theory: An assessment and review. Academy of
Management Review, 14(1): 57-74.

69

Finkelstein S. and Boyd, B.K. (1998). How much does the CEO matter? The role of
managerial discretion in the setting of CEO compensation. Academy of
Management Journal, 41: 179-199.

Fiske, S. and Taylor, S. (1991). Social Cognition. New York: Random House.

Gomez-Mejia, L. R. and Wiseman, RM. (1997). Reﬂaming Executive Compensation:
An assessment and outlook. Journal of Management, 23: 291-375.

Gooding, RZ, Goel S. and Wiseman RM. (1996). “Fixed versus variable points in the
risk-return relationship.” Journal of Economic Behavior and Organization, 29: 331-
350.

Gray. SR. and Cannella, A.A. (1997). The role of risk in executive compensation.
Journal of Ma_nagement, 23: 517-540.

Hall, B. J. (2000). What you need to know about stock options. Harvard Business
Review March-April: 121-129

 

Hall, B. J. and Murphy, K. J. (2002). Stock options for undiversiﬁed executives. Journal
of Accounting and Economics, Vol 33(2): 3-42.

Hall, B. J. and Murphy, K. J. (2000). Optimal Exercise Prices for Executive Stock
Options, American Economic Review. 90: 209-214.

Hall, Brian and Liebman, Jeffrey. (1998). Are CEOs Really Paid Like Bureaucrats?.
Quarterly Journal of Economics, 113, 3, 653-691 .

Hayward, M.L.A. and Hambrick, DC. (1997). Explaining the premium paid for large
acquisitions: Evidence of CEO hubris. Administrative Science Oaarterlv. 42: 103-
127.

Hodgkinson, G. P., Brown, N.J., Maule, A.J., Glaister, K.W. and Pearman, AD. (1999).
Breaking the frame: An analysis of strategic cognition and decision making under
uncertainty Strategic Management Journal, 20: 977-985.

Hollenbeck, J.R., Ilgen, DR, and Sego, DJ. (1994). Repeated measures regression and
mediational tests: Enhancing the power of leadership research. Leadership

M111, 5: 3-23.

Hollenbeck, J .R., Ilgen, D.R., Phillips, J .M., and Hedlund, J. (1994). Decision risk in
dynamic two-stage contexts: Beyond the status quo. Journal of Applied
Psychology, 79: 592-598.

Holrnstrom, B. (1979). Moral hazard and observability. Bell Journal of Economics. 10:
74-91 .

7O

Huddart, S. and Lang, M. (1996). Employee stock option exercises: An empirical
analysis. Journal of Accountingand Economics, 21: 5-43.

Jensen, M. C. and Murphy, K. J. (1990). Performance pay and top-management
incentives. Journal of Political Economy. 98: 225-263.

Kahneman, D, Knetsch, J. L. and Thaler, R. (1986). Experimental Tests of the
Endowment Effect and the Coase Theorem. Journal of Political Economy.
98(6):]325-1348.

Kahneman, D. and Tversky, A. (1979). Prospect theory: an analysis of decision under
risk. Econometrﬂa 47: 263-290.

Kameda T, Davis J .H. (1990). The function of the reference point in individual and

group risk decision making.” Organizational Behavior and Human Decision
Processes, 46: 55-76.

Kﬁhberger, A., Schulte-Mecklenbeck, M. and Pemer, J. (1999). The effects of ﬂaming,
reﬂection, probability and payoff on risk preference in choice tasks. Organizational
Behavio_r_ and Humaa Decision Processes. 78: 204-231.

Lambert, R.A, Larcker, D. F. and Baker, G. (1987). An analysis of the use of accounting
and market measures of performance in executive compensation contracts. Journal
of Accounting Resear_9_h, 25: 85-129.

Lambert, Richard. (1993). Long Term Contract and Moral Hazard. Bell Journal of
Economics, 14, 441-452.

Lant, T.K. (1992). Aspiration level adaptation: An empirical exploration. Management
Science 38: 623-645.

 

Larraza-Kintana, M., Wiseman, R.M., Gomez-Mejia, LR. and Welboume, T.M. (1999).
Distinguising between employment and compensation risk inﬂuences on perceived
risk taking. Universidad Publica De Navarra working paper.

Loehman E. (1998). Testing risk aversion and nonexpected utility theories. Journal of
Economic Behavior and Organization, 33: 285-302.

Loewenstein, M.J. (2000). The conundrum of executive compensation. Wake Forest Law
Review 35: 1-30.

Lopes, L. L. (1987). Between hope and fear: The psychology of risk. In L. Berkowitz

(Ed). Advances in exm’mental social psychology, Vol. 20. (pp. 255-295). San
Diego, CA, US: Academic Press, Inc. vii, 351pp.

71

MacCrimmon, K.R. and Wehrung, DA. (1986). Taking risks: The management of
uncertaing, New York: The Free Press.

MacNeil, LR. (1978). Contracts: Adjustments of long-term economic relationship under
classical, neoclassical, and relational law. Northwestern Law Review. 72: 854-906.

March, J .G. and Shapira, Z. (1992). Variable risk preferences and the focus of attention.
Psychological Review, 99: 172-183.

March, J. M. (1988). Variable risk preferences and adaptive aspirations. Journal of
Economic Behavior and Organization. 9: 5-24.

March, J .G. and Shapira, Z. (1987). Managerial perspectives on risk and risk taking.
Management Science. 33: 1404-1418.

McNamara G. and Bromiley P. (1997). Decision making in an organizational setting:
Cognitive and organizational inﬂuences on risk assessment in commercial lending.
Academy of Management Journal, 40: 1063-1088.

McNamara, G. and Bromiley, P. (1999). Risk and return in organizational decision
making. Academy of Management Journal, 42: 330-339.

Mehran, H. (1995). Executive compensation structure, ownership, and ﬁrm performance.
Joumg of Financial Economics. 38: 163-184.

Murphy, K. J. (1999). Executive compensation. In 0. Ashenfelter and D. Card (Eds),
H_andbool_( of Labor Economics. Vol. III. North Holland.

Murphy, K. J. (2000). Performance standards in incentive contracts. Journal of
Accounting and Economics, 30: 245-278.

O’Reilly, C.A., Main, B.G., and Crystal, GS. (1988). CEO compensation as a
tournament and social comparison: A tale of two theories. Administrative Science

Mm, 20: 559-569.
Perry, T. and Zenner, M. (2000). Wake Forest Law Review 35: 123-152.

Sanders, W. G. (2001). Behavioral responses of CEOs to stock ownership and stock
option pay. Academy of Management Jom 44: 477-492

Seymour, J. (2002). Super portfolio near misses: More risk, more return. TheStreet.com
[WWW document]. URL:
http:www.thestreet.com/ funds/ j imseymour_up/ 1 0006840.html.

Shapria, Z. (1995). Risk taking: A managerial perspective. New York: Russell Sage
Foundation.

72

Shelley, MK. (1993). Outcome signs, question ﬂames, and discount rates. Management
Science, 39, 806-815.

 

 

Shelley, MK. and Omer, T.C. (1996). Intertemporal ﬂaming issues in management
compensation. Organizational Behaviorﬁ and Human Decision Processg 66: 42-58.

Shavell, S. (1979). Risk sharing and incentives in the principal and agent relationship.
Bell Economic Journal, 10: 55-73.

Sheﬂin, HM. and Thaler, RH. (1988). The behavioral life-cycle hypothesis. Economic
I_n_gpi_ry, 26: 609-643.

Sitkin, SB. and Weingart, LE. (1995). Determinants of risky decision-making behavior:
A test of the mediating role of risk perceptions and propensity. Academy of
Management Review. vol. 38, no. 6, pp. 1573-1592.

Sitkin, SB. and Pablo, AL. (1992). Reconceptulizing the determinants of risk behavior.
Academy of Management Review, 17: 9-38.

Sullivan, K. and Kida, T. (1995). The effects of multiple reference points and prior gains

and losses on managers’ risky decision making. Organizational Behavior and
Human Decision Procesgs. 64: 76-83.

Staw, B.M., McKechnie, RI. and Puffer, SM. (1983). The justiﬁcation of organizational
performance. Administrative Science Quarterly. 28: 582-600.

Stevenson, M. (1986). A discounting model for decisions with delayed positive or
negative outcomes. Journal of Expe_ri_raental Psychology Geneﬂ, 115: 131-154.

Thaler, RH. (1980). Toward a positve theory of consumer choice. Journal of Economic
Behavior and Organization, 1: 39-60.

Thaler, R. H. and Johnson, E. J. (1990). Gambling with the house money and trying to

break even: The effects of prior outcomes on risky choice. Management Science.
36: 643-660.

Tosi, H.L. and Gomez-Mejia, L. (1989). The decoupling of CEO pay and performance:
an agency theory perspective. Administrative Science Ougerly. 34, 169-189.

Tversky, A., and Kahneman, D. (1986). Rational choice and the ﬂaming of decisions.
Journg of Bu_siness 59: 251-78.

Weber, E. and Milliman, R (1997). Perceived risk attitudes: Relating risk perception to
risky choice. Management Science. 43, 122-143.

73

Wiseman, R. M.and Gomez-Mejia, L. R. (1998). A behavioral agency model of
managerial risk taking. Academy of Management Review, 23: 133-153.

Wiseman, R.M., Gomez-Mejia, LR, and F ugate, M. (2000). Rethinking compensation
risk. In B. Gerhart (Ed.) Compenaation in organizations: 311-347. San Francisco:
Jossey-Bass, Inc.

Zall, M. (2002). Are stock options getting unbearable? Perm] Business [WWW

document]. URL:
http://pubs.acs.org/subscribe/joumals/tcaw/O9/i12/html/120ers.htm.

74

APPENDIX A

COMPENSATION SCENARIO COVER PAGE FOR ALL SURVEYS

Please complete the following information regarding your current or previous job(s) :

I. The largest budget that I have been responsible for managing (in US. dollars) is
$

11. Thebrgestexpenditumlhavehadmeauﬂnntymmakewmutsupewisowappmval
(in US. dollars) is $

11. Please rate the current performance of the ﬁrm in which you work:

 

 

 

 

III. Please rate your willingness to undertake risky business propositions as compared to

others at or near your level in your firm:

 

 

Muchleswilling Mudtmorewililng
toacceptrisks toacceptrisls
1 2 4 5 7

 

IV. Please rate your feelings aboutthe riskofstnckoptionsasa form of compensation:

 

riskyformof
compensation
1

 

Stodtoptionsarea very...

safeformof
compensation

7

 

75

 

 

V. Please indicate which of the following compensation forms have been a part of your

current or past compensation packages (circle yes or no).

a. Salary .................................................................. Yes No
b. Annual bonus ...................................................... Yes No
c. Stockoptions ...................................................... Yes No
d. Long-terrn performance plan/bonus ................... Yes No
e. Restricted stock .................................................. Yes No

The following pages contain background information and a set of scenarios. Please read and answer all
questions.

***IMPORTANT***

Once you have turned the page, please do not go back and change
any of your previous answers.

We are investigating responses to compensation forms. There are no
right or wrong answers.

76

APPENDIX B

NO IMPLICIT RISK - 18 MONTH CONDITION

18NIR-1

W
You are the CEO of a small public ﬁrm. Since the ﬁrm is small, your actions directly
impact the stock price.

In December 2002 you were awarded the option to purchase 10,000 shares of
your ﬁrm’s stock for $29.00; however, you cannot exercise (purchase, sell, or
trade these shares) for 18 months. If you do not exercise the option on June
30, 2004 it will expire. This is an option to purchase shares - you are not

obligated in any way to make the purchase.

W
Your goal, set by the board of directors, is to increase your ﬁrm’s stock price to $35.00

a share (by June 30, 2004).

In 18 months (on June 30, 2004) you may purchase any or all of the 10,000 shares for
$29.00 per share. At that time you can immediately sell the stock at the current market
price for a proﬁt or keep the stock. Your ﬁrm will pay all of yourfees and axes
asadated with any gain- The following is a hypothetical situation to illustrate:

Example

On June 30, 2004 the market price of your ﬁrm is stock is $35. 00 per share. At that time
you can purdrase your shares at $29.00 and immediately resell the stodr and earn a
$6. 00 per share proﬁt ($60, 000).

Your ﬁrm manufactures critical components for your customers. As a result, your ﬁnn's
stock price is sheltered from any negative effects of the economy. Current conditions
are quite stable. In addition, your stock price and others in your industry have
generally continued to rise over the last decade. This pattern is projected to continue
over the next few years. As such, industry analysts consider your stock to be a very
safe investment.

77

High Implicit Risk - 18 month condition

18HIR-3

WED
You are the CEO of a small public ﬁrm. Since the ﬁrm is small, your actions directly
impact the stock price.

In December 2002 you were awarded the option to purchase 10,000 shares of
your ﬁrm’s stock for $29.00; however, you cannot exerc'me (purchase, sell, or
trade those shares) for 18 months. Ifyou do not exerc'ne the option on June
30, 2004 it will expire. This is an option to purchase shares - you are not

obligated in any way to make the purchase.

IQIIBEQAI.
Your goal, set by the board of directors, is to increase your ﬁrm's stock price to $35.00
a share (by June 30, 2004).

In 18 months (on June 30, 2004) you may purchase any or all of the 10,000 shares for
$29.00 per share. At that time you can immediately sell the stock at the current market
price for a proﬁt or keep the stock. Your ﬁrm will pay all of yourfax and axe
associated with any pain- The following is a hypothetical situation to illustrate:

Example

On June 30, 2004 the market price of your ﬁrm ’5 stock is $35. 00 per share. At that time
you can purchase your shares at $29.00 and immediately resell the stock and earn a
$6. 00 per share proﬁt ($60,000).

Your ﬁrm manufactures nonessential components for your customers. As a result,
economic conditions directly affect your ﬁrm's stock price. Current economic conditions
are quite unstable. In addition, your stock price and others in your industry have been
quite volatile (rising and falling dramatically) over the last decade. This pattern is
projected to continue over the next few years. As such, industry analysts consider your
stock to be a high risk investment.

78

APPENDIX C

SURVEY - 18 MONTH CONDITIONS (NO IMPLICIT RISK & HIGH IMPLICIT
RISK)

Please answer the following questions:
1. What is your proﬁt per share goal at exercise (when you are able to purchase and resell
these shares)?: $ per share.

2. How much of an impact do your actions have on the ﬁrm's stod< price?

 

No impact Direct impact
1 2 3 4 S 6 7

 

 

 

3. On the scale below please indicate your perception of the likelihood of making your proﬁt

goal:

 

Low likelihood High likelihood
1 2 3 4 5 6 7

 

 

 

DECISION 1 - It is now April 15, 2003
Current market price = $33. 00.
The exercise price in ONE YEAR AND TWO 1/2 MONTHS (JUNE 30, 2004) = $29.00

 

 

 

 

 

 

 

4J1 5I03
1/15/03 29.00
2/1/03 29.00
2115/03 29.50
311/02 32.00
3/15/03 32.00
4/1/03 32.50
4115/03 33.00

 

 

 

 

 

 

 

 

300091“
3000913
WM:
0300910
2000915
3000910
300091”.
WHO
ZOO091I6
30009101
. 300091111
30009121
800091“
80009117.
€00091l€
900091”
8000915
80009119

 

 

79

Atthis point in time, whatistheproﬂtpershareyou expecttoeam atexercise (whenyou
are able to purchase and resell these shares) in 14 and 1/2 months?: $

per share.

Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of personal

wealth?: $ per share.

At this point in time I feel as though the shares are performing ......

 

Cirdeonenumber
Behindofwhere Aheadofwhere
I'dlikelhemtobe I'dlikethemtobe
1 2 3 4 5 6 7

 

 

 

At this point in time I am more concerned about ......

 

Circle one number
Maintaining the Increasing the
arrrent stodt price and: price
1 2 3 4 S 6 7

 

 

 

At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure

ﬁnancing on a purchase you really want to make?: $ per share.

DECISION 2 - It is now June 1, 2003
Current market price = $42. 00.
The exercise price in 1 YEAR AND 29 DAYS (JUNE 30, 2004) = $29.00

 

l

 

 

 

 

 

6/01/03
1/15/03 29.00
211/03 29.00
2/15/03 29.50
3/1/02 32.00
3/15/03 32.00
411/03 32.50
4415/03 33.00
5/1l03 33.50
5/15/03 36.00
611/03 42.00

 

80

9. Atthis point in time, whatistheprol‘ltpershareyouexpecttoeam atexercise (when you
are able to purchase and resell these shares)?: $ per share.

10. Defining personal wealth to include both current income and fully anticipated future income,

at this point in time, what proﬁt per share would you include in your calculations of personal

 

 

 

 

 

 

wealth?: $ Jet share.
11. At this point in time I feel as though the shares are performing ......
Circle one number
Behind of where Ahead of where
I'dlikethemtobe I'dlikethemtobe
1 2 3 4 5 6 7
12. At this point in time I am more ccncemed about ......
Clrde one number
Maintaining the Increasing the
cunent stodr price stodt price
1 2 3 4 5 6 7

 

 

 

13. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure

ﬁnancing on a purchase you really want to make?: $ Jer- share.

Decision 3 - It is now June 15, 2003

 

 

 

 

 

 

 

 

 

Current market price = $42. 00
The exercise price in ONE YEAR AND TWO WEEKS (JUNE 30, 2004) = $29.00
1/15/03 29.00
“‘5’” 2/1/03 2900
2/15/03 29.50
’5°-°° 3/1/02 32.00
347.50 -— ——~ - ---—-——_,~- -_ “W -_——
mo, __ ._ . __._______. 3/15/03 32.00
$42.50 - -- -- « -—-— ,_____,,____,__ _,,,,,,, 4 4/1/03 3250
340130 ”AW -...-.________.__ _ . _ _ 2, M ..
,,,_,,, __ _ __ ,_ , 4/15/03 33.00
$35.00- ~~— --——-—»--————~ W -.__,. 5/1/03 33.50
3:33 ‘ “L ‘;,::;::,‘:ii: 5’15’03 360°
”7,01 __ _____ ,n f, 6/1/03 42.00
325.00 .— Wm. _ .- . . _..-_-..- 2 6/15/03 42.00
322.50 - —-----~—w—- —--———~—-- ——-- ———-—--~--——-—~———— --~—_———-. —— ____._______ _
820.00 r.— . . e , , , , . . . . .4
§§§§§§§§§§§§§§§§§§
§§§§§§§§§§§§§§§§§§

 

 

 

 

 

81

14.

15.

16.

17.

18.

Atthis pointintime,whatistheproﬁtpershareyouexpecttoeamatexercise(whenyou
are able to purchase and resell these shares)?: 3 pershare.

Deﬁning personal wealth to include boﬂi cunent income and fully anticipated future income,

at this point in time, what proﬁt per share would you include in your calculations of personal

 

 

 

 

 

 

 

 

 

wealth?: $ per share.
At this point in time I feel as though the shares are performing ......
Clrde one number
Behind of where Ahead of where
I'd like them to be I’d like them to be
1 2 3 4 5 6 7
At this point in time I am more ccncemed about ......
Circle one number
Maintaining the Increasing the
current stod( price stock price
1 2 3 4 5 6 7
At this point in time, what is the profit per share you are conﬁdent enough to use to secure

ﬁnancing on a purchase you really want to make?: $ per share.

DECISION 4 - It is now January 1, 2004
Current market price = $42. 00.
The exercise price in SD( MONTHS (JUNE 30, 2004) = $29.00

 

 

 

 

 

 

 

 

 

 

 

 

 

-: 7
' l
E 111104 1011103 42.00
" 10115103 4075
l 39:$..._.. .--.___-_._:_::__,__ , ..___,,,,:: 11/1103 43.00
4500 ~—- ----- -— -- ~ 11115/03 4100
23:3 2-: - " _ ,_ _ ___,_.,_,‘_jfi—‘:“'i:j 1211/03 4300
3:053 -—— - -» . -- 12115103 41.00
32:50 gun/L-” _ 1:“— “1 :: 111104 42.00
30.“) - . .. ..._1_-___.-_.....- .. _ ________
27,50 —-————-—————-» . — .-...-..--..__ -. .. . . ,,___,,_ , 1
25.“) _i.....r__..._. M.,. ..._. ﬁ....__.. ...-_____._.._ .. . .--_._._______._. . - _ _4 l
22.50 ~ —— — “.._—“_— _ _e l
20.00 + .
3§3533333§§§3§35S
gmgéggggggémél

 

82

19. Atthis pointintime,whatisd'ieproﬂtpershareyouexpecttioeamatexercise(whenyou
are able to purchase and resell these shares)?: $ per share.

20. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what profit per share would you include in your calculations of personal

wealth?: $ per share.

21. Atthispointintimelfeelasthoughthesharesareperforming ......

 

Ordeonenumber
Behindofwhere Aheadofwhere
i'dlikethemtobe I’dlikethemtobe
1 2 3 4 S 6 7

 

 

 

22.Atthispointintime1ammoreconcemedabout .....

 

 

 

 

0rde one number
Maintaining the Increasing the
current stock price stock price
1 2 3 4 5 6 7
23. Has your initial profit per share goal changed? Circle one: yes no

If yes please indicate the new ﬁgure: $ per share

 

24. At this point in time, what is the profit per share you are conﬁdent enough to use to secure
ﬁnancing on a purchase you really want to make?: $ per share.

Decision 5 - It is now February 15, 2004
Current market = $36.00.
The exercise price in FOUR 1/2 MONTHS (JUNE 30, 2004) = $29.00

 

2’15’04 l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l 1011/03 42.00

10115103 40.75

333%... H. _ __. I 1111/03 43.00

23.053 _ 1 11115103 41.00

40200 12I1I03 43.00

gg-g 12115103 41.00

33: _ 111/334 1:33

$2.33 --._....“ "i:;1:i.__-_ __, _ :1: _ “ ‘31 2,1104 39:00
, . ...- .. ___»--- M _.

m'msssssssségggésssé 2,15,“ 3600
§§§§§§§§§§§§§§§§§

 

83

25. At this point in time, what is the proﬁt per share you expect to earn at exercise (when you
are able to purchase and resell these shares)?: $ per share.

26. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of personal
wealth?: $ per share.

27. At this point in time I feel as though the shares are performing ......

 

Cirde one number
Behind of where Ahead of where
I'dllkethemtnbe I'dlikethemtobe
1 2 3 4 5 6 7

 

 

 

28. At this point in dme I am more concerned about ......

 

Circle one number
Maintaining the Increasing the
current stock price stodt price
1 2 3 4 5 6 7

 

 

 

29. Please indicatetheprol'ltpershareyou would needtoeam atexercise in ordertofeelthat
you have not lost any wealth you felt you had previously gained?: $ per
share.

30. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure
ﬁnancing on a purchase you really want to make?: $ per share.

DECISION 6 - It is now March 15, 2004
Current market price = $33. 00.
The exercise price in THREE 1/2 MONTHS (JUNE 30, 2004) = $29.00

r
!

 

 

 

 

 

 

 

 

31504 1011103 42.00

10115103 4075

1111103 43.00

11115103 41.00

1211103 43.00

12115103 41.00

111104 42.00

1115104 41.00

211104 39.00

2115104 36.00

ssssessssasesssss 3"“ 360°

- .. - 3115104 33.00
33333333333333333
N N N N N N N § 0

 

 

 

84

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

31. At this point in time, what is the proﬁt per share you expect to eam at exercise (when you
are able to purchase and resell these shares)?: $ per share.

32. Deﬁning personal wealth to include both current income and fully anddpatied future income,
at this point in time, what proﬁt per share would you include in your ulculatlons of personal
wealth?: $ Jet share.

33. At this point in time I feel as though the shares are performing ......

9&8 one My
Behind of where Ahead of where
I'dlikethemtobe I'dlikethemtobe
1 2 3 4 5 6 7
34. At this point in time I am more concerned about ......
Circle one number
Maintaining the Increasing the
current stodt price stodr price
1 2 3 4 5 6 7

35. Please lndlcatetheprofltpershareyou would needtoeamatexerciseinordertiofeelthat
youhavenotlcstanyweald1youfeltyouhadpreviouslygained?:$ per
share

36. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure
ﬁnancingonapurchaseyoureallywanttomake?: $ Jet-share.

DECISION 7 - It is now June 1, 2004

Current market price = $25. 00.

The exercise price in ONE MONTH (JUNE 30, 2004) = $29.00

“1’“ 1011103 42.00
10115103 40.75

1111103 43.00

11115103 41.00

1211103 43.00

12115103 41.00

111104 42.00

1115104 41.00

211104 39.00

2115104 30.00

311104 36.00

3115104 33.00

g g S E S S E 411104 32.50
4115104 20.00

E g g g g g E 511104 27.00
'° 5115104 27.00
011104 25.00

85

37.

39.

41.

42.

At this point in time, what is die proﬁt per share you expect to earn at exercise (when you
are able to purchase and resell these shares)?: $ per share.

. Deﬁning personal wealth to include both current income and fully anticipated future income,

at this point in time, what proﬁt per share would you include in your calculations of personal
weald'i?: $ pershare.

Atthispointintimelfeelasdioughthesharesareperforming ......

 

Orde one number
Behind of where Ahead of where
I'd like them to be I'd like them 00 be
1 2 3 4 S 6 7

 

 

 

. At this point in time I am more concerned about ......

 

Cirde one number
Maintaining the Inaeadng the
current stodt price stodt price
1 2 3 4 5 6 7

 

 

 

Pleaseindicatetheprofltpershareyouwouldneedtaoeamatexerciselnordertofeeithat
youhavenotlostanywealthyoufeltyouhadprevlouslygained?: $ per
share.
Atthispointintime,whatistheproﬁtpershareyouareconﬁdentenoughtousetosecure

ﬁnancing on a purchase you really want to make?: $ per share.

86

APPENDIX D
No Implicit Risk - 3-year condition

3NIR-2

You are the CEO of a small public ﬁrm. Since the ﬁrm is small, your actions direcﬂy
impact the stock price.

In December 2002 you were awarded the option to purchase 10,000 shares of
your firm’s stock for $29.00; however, you cannot exercise (purchase, sell, or
trade these shares) for three years. Ifyou do not exercise the option on
December 31, 2005 it will expire. This is an option to purchase shares - you

are not obligated in any way to make the purchase.

101135021.
Your goal, set by the board of directors, is to increase your ﬁnn's stock price to $35.00
a share (by December 31, 2005).

In three years (on December 31, 2005) you may purchase any or all of the 10,000
shares for $29.00 per share. At that time you can immediately sell the stock at the
current market price for a proﬁt or keep the stock. Your ﬁrm will pay all of your
fees and taxes associated with any gain- The following is a hypothetical situation
to illustrate:

Example

On December 31, 2005 the market price of your ﬁrm’s stout is $35.00 per share. At
that time you can purchase your shares at $29. 00 and immediately resell the stock and
earn a $6. 00 per share proﬁt ($60,000).

1821518!
Your ﬁrm manufactures critical components for your customers. As a result, your ﬁrm's

stock price is sheltered from any negative effects of the economy. Current conditions
are quite stable. In addition, your stock price and others in your industry have
generally continued to rise over the last decade. This pattern is projected to continue
over the next few years. As such, industry analysts consider your stock to be a very
safe investment.

87

High Implicit Risk — 3-year condition
3HIR-4

W
You are the CEO of a small public ﬁrm. Since the ﬁrm is small, your actions directiy

impact the stock price.
In December 2002 you were awarded the option to purchase 10,000 shares of
your ﬁrm’s stock for $29.00; however, you cannot exercise (purchase, sell, or
trade these shares) for three years. If you do not exercise the option on
December 31, 2005 it will expire. This is an option to purchme shares - you

are not obligated in any way to make the purchase.

W
Your goal, set by the board of directors, is to increase your ﬁrm's stock price to $35.00
a share (by December 31, 2005).

In three years (on December 31, 2005) you may purchase any or all of the 10,000
shares for $29.00 per share. At that time you can immediately sell the stock at the
current market price for a proﬁt or keep the stock. Your firm will pay all of your
fees and taxes asodated with any gain— The following is a hypothetical situation
to illustrate:

Example

On December 31, 2005 the market price of your firmfs stock is $135.00 per share. At
that time you can purdiase your shares at $29.00 and immediately resell the stock and
earn a $6. 00 per share proﬁt ($60, 000).

Your ﬁrm manufactures nonessential components for your customers. As a result,
economic conditions directly affect your ﬁrm's stock price. Current economic conditions
are quite unstable. In addition, your stock price and others in your industry have been
quite volatile (rising and falling dramatically) over the last decade. This pattern is
projected to continue over the next few years. As such, industry analysts consider your
stock to be a high risk investment.

88

APPENDIX E

Survey - 3—year conditions (no implicit risk & high implicit risk)

43. What is your proﬁt per share goal at exercise (when you are able to purchase and resell
these shares)?: $ per share.

44. How much of an impact do your actions have on the ﬁrm’s stock price?

 

No impact Direct impact
1 2 3 4 5 6 7

 

 

 

45. On the scale below please indicate your perception of the likelihood of making your proﬁt
goal:

 

Low likelihood High likelihood
1 2 3 4 S 6 7

 

 

 

DECISION l - It is now August 1, 2002

Current market price = $33.00.

The exercise price in TWO YEARS AND FIVE MONTHS (DECEMBER 31, 2004) =
$29.00

 

Feb. 2002 29.00
Mar. 2002 29.00
Apr. 2002 29.50
May. 2002 32.00
Jun. 2002 32.00
Jul. 2002 32.50
Aug. 2002 33.00

 

 

 

 

 

 

 

 

89

46. Atthis point in time, whatlstheproﬁtpershareyouexpecttoeam atexercise (whenyou
are able to purchase and resell these shares) in two years and ﬁve months?:
$ per share.

47. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of
personal wealth?: $ per share.

48. At this point in time I feel as though the shares are performing ......

 

0rde one number
Behind of where Ahead of where
I'd like them to be I'd like them to be
1 2 3 4 5 6 7

 

 

 

49. At this point in time I am more ccncemed about ......

 

Circle one number
Maintaining the Increasing the
uncut stodt price St0d( price
1 2 3 4 S 6 7

 

 

 

50. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure
ﬁnancing on a purchase you really want to make?: $ Jer share.
DECISION 2 - It is now November 1, 2002

Current market price = $42.00.
The exercise price in TWO YEARS AND TWO MONTHS (DECEMBER 31, 2004) = $29.00

 

 

 

 

 

November 2002 Feb. 2002 29.00
1 Mar. 2002 29.00
$50.00 Apr. 2002 29.50
$47-50 M . 2002 32.00
545.00 ay
542.50 Jun. 2002 32.00
33:23 .....sz 32.50
$35.00 Aug. 2002 33.00
33:33 Sep. 2002 33.50
$27.50 Oct. 2002 36.00
33% Nov. 2002 42.00
52000
Le - _.

 

 

 

90

51. Atthis point in time, whatistheproﬁtpershareyou expecttoeam atexercise(when you
are able to purchase and resell these shares)?: $ per share.

52. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of

personal wealth?: $ per share.

53. At this point in time I feel as tough the shares are performing ......

 

Clrdeone number
Behind ofwhere Aheadofwhere
I'dlikethemtobe I’dlikethemtobe
1 2 3 4 5 6 7

 

 

 

54. Atthispointintimelamrnoreconcemedabout ......

 

Circle one number
Maintaining the Increasing the
cunent StOdK price shod: price
1 2 3 4 5 6 7

 

 

 

55. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure

ﬁnancing on a purchase you really want to make?: $ per share.

DECISION 3 - It is now December 1, 2002
Current market price = $42.00
The exercise price in TWO YEARS AND 30 DAYS (DECEMBER 31, 2004) = $29.00

 

 

 

 

 

 

 

December 2002 Feb- 2002 29.00

Mar. 2002 29.00

Apt. 2002 29.50

550.00 May. 2002 32.00

gig-g Jun. 2002 32.00

“2'50 Jul. 2002 32.50

$37.50 Sep. 2002 33.50

- NOV. 2002 42.00

gag-g3 Dee 2002 42.00
$25.00
$22.50
$20.00

L _m

 

 

 

91

56. Atthis point in time, whatistheproﬁtpershareyouexpecttoeam atexercise (whenyou
are able to purchase and resell these shares)?: $ per share.

57. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of

personal wealth?: $ per share.

58. At this point in time I feel as though the shares are performing ......

 

Grde one number
Behind of where Ahead of where
I’d like them to be I’d like them to be
1 2 3 4 5 6 7

 

 

 

59. At this point in time I am more concerned about ......

 

Clrde one number
Maintaining the Increasing the
cunent stodt price stock price
1 2 3 4 5 6 7

 

 

 

60. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure

ﬁnancing on a purchase you really want to make?: $ per share.

DECISION 4 - It is now January 1, 2004
Current market price = $42.00.
The exercise price in ONE YEAR (DECEMBER 31, 2004) = $29.00

 

January 2004 FED. 2W3 42.00
Mar. 2003 40.75
Apr. 2003 43.00
May. 2003 41.00
Jun. 2003 42.00
Jul. 2003 42.00
Aug. 2003 40 75
Sep. 2003 43.00
Oct. 2003 41.00
Nov. 2003 43.00
Dec. 2003 41.00
Jan. 2004 42. 00

 

 

 

 

 

 

 

 

383838333581333
8888888888888

ES iiiéiiiiiiiiiiﬁiiiiiiﬁ H858?

 

 

 

92

61. At this point in time, what is the proﬁt per share you expect to earn at exercise (when you
are able to purchase and resell these shares)?: $ per share.

62. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of

personal wealth?: $ per share.

63. At this point in time I feel as though the shares are performing ......

 

Circle one number
Behind of where Ahead of where
I’d like them to be I'd like them to be
1 2 3 4 5 6 7

 

 

 

64. At this point in time I am more ccncemed about ......

 

 

 

 

Orde one number
Maintaining the Increasing the
current stock price stock price
1 2 3 4 5 6 7
65. Has your initial proﬁt per share goal dianged? Circle one: yes no

If yes please indicate the new ﬁgure: $ per share

 

66. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure
ﬁnancing on a purchase you really want to make?: $ oer share.

Decision 5 - It is now April 1, 2004
Current market = $36.00.
The exercise price in NINE MONTHS (DECEMBER 31, 2004) = $29.00

April 2004 Apr. 2003 43.00
May. 2003 41.00

Jun. 2003 42.00
Jul. 2003 42.00
Aug. 2003 4075
Sep. 2003 43.00
Oct. 2003 41.00
Nov. 2003 43.00
Dec. 2003 41.00
Jan. 2004 42.00
Feb. 2004 41.00
Mar. 2004 39.00
Apr. 2004 36.00

 

 

 

 

 

 

 

 

93

67. At this point in time, what is the proﬁt per share you expect to earn at exercise (when you
are able to purchase and resell these shares)?: 5 per share.

68. Deﬁning personal wealth to include both cunent income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of

personal wealth?: $ ﬁner share.

69. At this point in time I feel as though the shares are performing ......

 

Cirde one number
Behind of where Ahead of where
I’dlikethemtobe I'dlikelhemtobe
1 2 3 4 5 6 7

 

 

 

70. At this point in time I am more concerned about ......

 

Circle one number
Maintaining the Inc'easlng the
current stodt price St0d( price
1 2 3 4 5 6 7

 

 

 

71. Please indicate the proﬁt per share you would need to earn at exercise in order to feel that
you have not lost any wealth you felt you had previously gained?: $ per
share.

72. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure
ﬁnancing on a purchase you really want to make?: $ oer share.

DECISION 6 - It is now June 1, 2004

Current market price = $33.00.
The exercise price in SEVEN MONTHS (DECEMBER 31, 2004) = $29.00

June 2004

 

May. 2003 41 .00
Jun. 2003 42.00
Jul. 2003 42.00

 

 

 

 

 

50.00 Aug. 2003 40.75
47.50 Sep. 2003 43.00
:g-gg Oct. 2003 41 .00
40:00 Nov. 2003 43.00
37.50 Dec. 2003 41.00
3500 Jan. 2004 42.00
33:33 Feb. 2004 41.00
2 7 _5 0 Mar. 2W4 39.00
25.00 Apr. 2004 36.00
3338 May. 2004 36.00

Jun. 2004 33.00

 

 

 

 

94

73. At this point in time, what is the proﬁt per share you expect to earn at exercise (when you
are able to purdiase and resell these shares)?: $ per share.

74. Deﬁning personal wealth to include both current income and fully anticipated future income,
at this point in time, what proﬁt per share would you include in your calculations of
personal wealth?: $ per share.

75. At this point in time I feel as though the shares are performing ......

Clrde one number

 

Behind of where Ahead of where
I'd like them to be I'd like them to be
1 2 3 4 5 6 7

 

 

 

76. At this point in time I am more concerned about ......

 

Grde one number
Maintaining the Ina'easing the
cunent shod: price stodt price
1 2 3 4 5 6 7

 

 

 

77. Pleaseindlcatetheproﬁtpershareyou would needtoearn atexerciseinordertofeelthat
youhavenotlostanywealth youfeltyou hadpreviously gained?: $ per
share.

78. At this point in time, what is the proﬁt per share you are conﬁdent enough to use to secure
ﬁnancing on a purchase you really want to make?: $ per share.

DECISION 7 - It is now November 1, 2004
Current market price = $25.00.
The exercise price in TWO MONTHS (DECEMBER 31, 2004) = $29.00

 

November 2004 Jul. 2003 42.00

 

 

50.00 Oct 2003 41.00
47.50 Nov. 2003 43.00
45.00 Dec. 2003 41.00
4250 Jan. 2004 42.00
4000 Feb. 2004 41.00
37.50 Mar. 2004 39.00
33% Apr. 2004 36.00
”'00 May. 2004 36.00
27°50 Jun. 2004 33.00
25:00 Jul. 2004 32.50
2250 Aug. 2004 28.00
20_00 Sep. 2004 27.00

Oct. 2004 27.00

Nov. 2004 25.00

 

 

 

 

95

79.

81.

82.

83.

Atthis point in time, whatistheproﬁtpershareyouexpecttoeamatexercise (when you
are able to purchase and resell these shares)?: $ per share.

. Deﬁning personal wealth to include both current income and fully anticipated future income,

at this point in time, what proﬁt per share would you include in your calculations of

personal wealth?: $ per share.

At this point in time I feel as though the shares are performing ......

 

Clrde one number
Behind of where Ahead of where
I’d like them to be I'd like them to be
1 2 3 4 S 6 7

 

 

 

At this point in time I am more concerned about ......

 

Circle one number
Maintaining the Increasing the
current stodt price stodt price
1 2 3 4 5 6 7

 

 

 

Pleaseindicatetheproﬁtpershareyouwouldneedtoeamatexerciseinordertofeelthat
you havenotlostany wealth you feltyou had previously gained?: $ per
share.

. Atthis pointintime,whatistheproﬁtpershareyouareconﬁdentenoughtousetosecure

ﬁnancing on a purchase you really want to make?: $ per share.

96

APPENDIX F

Non-flexible condition (Internal and External)

InternallNon-flexible condition

IMNF-Z

BACKGROUND
You are the vice president Of sales Of a small ﬁrm. You have complete authority over all sales

activities, including, accounting, pricing, collection, and delivery policies. In slow periods this
ﬂexibility allows you to adjust policy to increase or manipulate the timing Of sales (shift sales
from one period to another), to meet your internal sales goals. Although industry-wide sales
have been ﬂat over recent years, analysts expect this ﬁgure to begin to increase over the next ﬁve
years. However, analysts caution some ﬁrms may be better positioned to take advantage of this

growth than others.

In December 2001 a long-term bonus plan was added to your compensation package. The length
of the plan is three years. The plan works as follows:

On December 31, 2004 your ﬁrm’s three-year unit sales ﬁgure will be
compared to the unit sales ﬁgure of the last three-year period. The plan
rewards you for achieving percentage increases in three-year unit sales

according to the following schedule:

 

 

 

 

 

 

 

 

 

 

Percentage points Bonus amount
Increase in unit sales
3% - 3.99% $45,000
4% - 4.99% $60,000
5% - 5.99% $75,000
6% - 6.99% $90,000
7% - 7.99% $105,000
8% - 8.99% $120,000
9% - 9.99% $135,000

 

 

MEDAL
TheCEOsetagoaltoexceedlauperlod’sunitsalesbyolllofortheplan period.

97

External/Non flexible condition

EMNF-4

BACKGROUND

You are the vice president of sales Of a small ﬁrm. You have complete authority over all sales
activities, including, accounting, pricing, collection, and delivery policies. In slow periods this
ﬂexibility allows you to adjust policy to increase or manipulate the timing Of sales (shiﬁ sales
from one period to another), to meet your internal sales goals. Although industry-wide sales have
been ﬂat over recent years, analysts expect this ﬁgure to begin to increase over the next ﬁve
years. However, analysts caution some ﬁrms may be better positioned to take advantage of this
growth than others.

In December 2001 a long-term bonus plan was added to your compensation package. The
lengthofthe plan isthreeyears. The planworksasfollows:

On December 31, 2004 your ﬁrm’s three-year unit sales will be compared
with the average industry unit sales for the same period. Most of your
competitors abo have bonus plans that are abo based on exceeding the
industry average unit sales. Consequently, you are unsure what the period
industry average unit sales ﬁgure will be on December 31, 2004. The plan
rewards you for achieving percentage increases in over the three-year

industry average according to the following schedule:

 

 

 

 

 

 

 

 

Percentage increase Bonus amount
Over industry avegle
3% - 3.99% $45,000
4% - 4.99% $60,000
5% - 5.99% $75,000
6% - 6.99% $90,000
7% - 7.99% $105,000
8% - 8.99% $120,000
9% - 9.99% $135,000

 

 

 

 

mecca:
The CEO set a goal to exceed last period’s unit sales by 6% for the plan period.

98

APPENDIX G

Survey - Non-ﬂexible conditions (Internal and External measures)

Please answer the following questions:

85. What is your bonus 90a": $

86.

Onthescalebelowpleaseindlutethelevelofcontrolyoufeelyouhaveoverthestructureof

this bonus (how much control do you feel you have over your ﬁrm exceeding last period’s
sales in the allotted time frame?:

 

 

Lowcontrol

I-lighconuol

 

 

87.

On the scale below please indicate your perception of the likelihood of receiving this bonus:

 

 

Low likelihood

1

Hid! likelihood

7

 

 

Decision 1 - It is now August 1, 2002

Bonus in TWO YEARS AND FIVE MONTHS (DECEMBER 31, 2004) would be

$0.00

MONTH

% INCREASE

BONUS

LEVEL

 

Feb. M2

0.00%

_

0.00%

Jun. 2002
0.00%

0.00%

 

 

 

0

 

 

0

 

 

 

 

Please answer the following questions:

88. Atthispointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:

$

89. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to

use to secure ﬁnancing on a purchase you really want to make?:

$

99

 

90. At this point in time I feel as though my bonus is ......

 

Circle one number
Behind of where Ahead of where
I’d like it to be I'd like it to be
1 2 3 4 5 6 7

 

 

 

91. At this point in time I am more concerned about ......

 

Clrde one number
Maintaining 100835509 mY
my current current bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

Decision 2 - It is now November 1, 2002

Bonus in TWO YEARS AND NO MONTHS (DECEMBER 31, 2004) would be
$15.18!!

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

92. Atthis pointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:
$

93. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnandng on a purchase you really want to make?:

$

94. Atthispolntintimelfeelasthoughmybonusis ......

 

Cirdeonenumber
Behindofwhere Aheadofwhere
I'd like ltto be I'd like ltto be
1 2 3 4 S 6 7

 

 

 

100

95. At this point in time I am more concerned about. .....

 

Clrde one number
Maintaining Increasing my
my current cunent bonus
bonus level level
1 2 3 4 S 6 7

 

 

 

Decision 3 - It is now December 1, 2002
Bonus in TWO YEARS AND 30 DAYS (DECEMBER 31, 2004) would be M

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

96. Atthispointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:
$

97. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to

use to secure ﬁnancing on a purchase you really want to make?:

$

98. At this point in time I feel as though my bonus is ......

 

Cirdeonenumber
Behindofwhere Aheadofwhere
I’dlikelttobe I'dlikeittobe
1 2 3 4 5 6 7

 

 

 

99. Atthispointintimelammoreconcernedabout ......

 

Clrde one number
Maintaining Ina'easing my
my lament axrent bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

101

Decision 4 - It is now December 1, 2003

Bonus in ONE YEAR & ONE MONTH (DECEMBER 31, 2004) would be $129M
Optional extension to December 31, 2005

 

 

 

 

 

 

 

 

 

 

 

 

MONTH mm ”.2002 Aphm “.2002 M2002 um um 800.2%2
% INCREASE 0.%% 0.00% 0.%% 0.%% 0.00% 0.00% 1.00% 1.75%
BONUS LEVEL 0 O 0 0 0 0 0 0
MONTH 061.2%2 ”.2002 Ram JUL2%3 Fob.2%3 MU.2%3 Aw.2%3 Mly.2%3
% INCREASE 2.00% 3.%% 3.%% 4.50% 7.25% 725% 6.88% 7.55%
BONUS LEVEL 0 45.0% 45.0% 60.%0 105.000 105.0% 90.0% 105.%0
MONTH mm MM mm 302.2%: 061.2%3 Nomm Doom

% INCREASE 6.90% 8.%% 7.%% 7.00% 7.%% 7.00% 8.%%

BONUS LEVEL 90.0% 120.0% 105.0% 105.0% 105.0% 105.11%) 120.11!)

 

 

 

 

 

 

 

 

 

100. Atthis pointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:
$

101. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purchase you really want to make?:

$

102. At this point in time I feel as though my bonus is ......

 

Circleonenumber
Behindofwhere Aheadofwhere
I'dlikeittobe I'dlikeittobe
1 2 3 4 5 6 7

 

 

 

103. At this point in time I am more ccncemed about ......

 

C'lrde one number
Maintaining Indexing my
my current orient bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

102

Decision 5 - It is now January 1, 2004

Bonus in ONE YEAR (DECEMBER 31, 2004) would be $129,099

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

104. Atthispointintime,whatisthebonusyouexpecttoeamattheendofthlsplah?:

105.

106.

107.

108.

$

At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purchase you really want to make?:
$

Atthispointintimelfeelasthoughmybonusis ......

 

Cirdeone number
Behindofwhere Aheadofwhere
I'd like it to be I’d like it to be
1 2 ' 3 4 5 6 7

 

 

 

Atthis pointintimeIamrnoreooncemedabout. .....

 

 

 

 

Cirde one number
Maintaining Increasing my
my current cunent bonus
bonus level level
1 2 3 4 5 6 7
Has your initial bonus goal changed? Circle one: yes no

If yes please indicate the new ﬁgure $

 

103

Decision 6 - It is now April 1, 2004
Bonus in NINE MONTHS (DECEMBER 31, 2004) would be 51%

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

109. Atthis pointlntime,whatisthebonusyouexpecttoeamattheendofthlsplan?:
$

110. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purdiase you really want to make?:
$

111. Atthispointintimelfeelastl'loughmybonusis ......

 

Ordeonenumber
Behindofwhere Aheadofwhere
I'dlikeittobe I'dlikelttobe
1 2 3 4 S 6 7

 

 

 

112. Atthispointintimelamrnoreconcernedabout ......

 

Clrdeonenumber
Maintaining Inaeasngmy
myalrrent currentbonus
bontslevel level
1 2 3 4 S 6 7

 

 

 

104

113. Pleaseindicatethebonusamountyouwouldneedtoeamattheendoftheplanlnorder
tofeel that you havenotlostany wealthyoufeltyou had previously gained?:
4

Decision 7 - It is now June 1, 2004
Bonus in SEVEN MONTHS (DECEMBER 31, 2004) would be m

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

114. Atthis pointintime,whatisthebonusyouexpecttoeamattheendofthis plan?:
$

115. At this point in time, how much ofthe ﬁgure you listed above are you conﬁdent enough to

use to secure ﬁnancing on a purdiase you really want to make?:

$

116. At this point in time I feel as though my bonus is ......

Cirdeonenumber

 

Behindofwhere Aheadofwhere
l’dliltelttobe I'dliltelttobe
1 2 3 4 5 6 7

 

 

 

105

117. At this point in time I am more concerned about ......

 

Circle one number
Maintaining 100835509 mt
my current current bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

118. Pleaseindicatethebonusamountyouwould needtoearn attheendofthe plan in order
to feel that you have not lost any wealth you felt you had previously gained?:
$

106

APPENDIX H

Flexible condition (Internal and External)

Internal/Flexible condition

IMF-1

W
You are the vice president of sales of a small ﬁrm. You have complete authority over all sales

activities, including, accounting, pricing, collection, and delivery policies. In slow periods this
ﬂexibility allows you to adjust policy to increase or manipulate the timing of sala (shiﬁ sales
from one period to another), to meet your internal sales goals. Although industry-wide sales
have been ﬂat over recent years, analyss expect this ﬁgure to begin to increase over the next
ﬁve years. However, analysts caution some ﬁrms may be better positioned to take advantage of
this growth than others.

In December 2001 a long—term bonus plan was added to your compensation package. The
minimum length of the plan is three years, but you have the option to extend it to four years.
The plan works as follows:
On December 31, 2004 your ﬁrm’s three-year unit sales ﬁgure will be
compared to the unit sales ﬁgure of the last three—year period. The plan

rewards you for achieving percentage increases in three-year unit sales

according to the following schedule:

 

 

 

 

 

 

 

 

 

 

 

Percentage points Bonus amount
Increase in unit sales
3% - 3.99% $45,000
4% - 4.99% $60,000
5% - 5.99% $75,000
6% - 6.99% $90,000
7% - 7.99% $105,000
8% - 8.99% $120,000
9% - 9.99% $135,000

 

OPTION

If you wish, you can extend your plan for up to one additional year, in the event sales are not
performing to plan on December 31, 2004. This allows you to the opportunity to cash out your
bonus anytime between December 31, 2004 and December 31, 2005 - at the time most

advantageous to you.

10.08.6961.
‘lheCEOsetagoaltoexoeedlastperiod'sunitsalesby6%fordleplan period.

107

 

External/Flexible condition
EMF-3

You are the vice president of sales of a small ﬁrm. You have complete authority over all
sales activities, including, accounting, pricing, collection, and delivery policies. In slow
periods this ﬂexibility allows you to adjust policy to increase or manipulate the timing of
sales (shift sales from one period to another), to meet your internal sales goals.
Although industry-wide sales have been ﬁat over recent years, analysts expect this
ﬁgure to begin to increase over the next ﬁve years. However, analysts caution some
ﬁrms may be better positioned to take advantage of this growth than others.

In December 2001 a long-term bonus plan was added to your compensation package.
The minimum length of the plan is three years, but you have the option to extend it
to four years. The plan works as follows:

On December 31, 2004 your ﬁrm’s three-year unit sales will be compared with the
average industry unit sales for the same period. Most of your competitors also have
bonus plans that are also based on exceeding the industry average unit sales.
Consequently, you are unsure what the period industry average unit sales ﬁgure will
be on December 31, 2004. The plan rewards you for achieving percentage increases

in over the three-year industry average according to the following schedule:

 

 

 

 

 

 

 

 

Percentage increase Bonus amount
Over industry average
3% - 3.99% $45,000
4% - 4.99% $60,000
5% - 5.99% $75,000
6% - 6.99% $90,000
7% - 7.99% $105,000
8% - 8.99% $120,000
9% - 9.99% $135,000

 

 

 

 

m

If you wish, you can extend your plan for up to one additional year, in the event sales are not
performing to plan on December 31, 2004. This allows you to the Opportunity to cash out your
bonus anytime between December 31, 2004 and December 31, 2005 - at the time most
advantageous to you.

museum.

TheCEOhassetagoaltoexcaedtheaveragelndustryunltsalesbyoqbfortheplan
period.

108

APPENDIX I

Survey - Flexible conditions (Internal and External measures)
Please answer the following questions:
119. What is your bonus goal?: $
120. Onthescalebelowpleaseindicatethelevelofcontrolyoufeelyou haveoverthestructure

ofthis bonus (how muchcontroldoyoufeelyou haveoveryourﬁnn exceeding last
period’s sales in the allotth time frame?:

 

Lowcontrol Highcontrol
1 2 3 4 5 6 7

 

 

 

121. On the scale below please indicate your perception of the likelihood of receiving this
bonus:

 

Low likelihood High likelihood
1 2 3 4 5 6 7

 

 

 

Decision 1 - It is now August 1, 2002

Bonus in TWO YEARS AND FIVE MONTHS (DECEMBER 31, 2004) would be

$0.99
Optional extension to December 31, 2005

MONTH Feb.2%2 Ilium .2002 ._01.m2| .1li 301.2002 I manual
% INCREASE 0.%% 0.%% 0.%% 0.%% 0.%% 0.00% 1.00%

BONUSLEVELIOIOIO olololoJ

 

 

122. Atthispointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:
$

123. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to

use to secure ﬁnancing on a purdiase you really want to make?:
$

109

124. Atthispointintimelfeelasthoughmybonusis ......

 

Cirdeonenumber
Behindofwhere Aheadofwhere
I'dlikeittobe I'dlikeittobe
1 2 3 4 5 6 7

 

 

 

125. At this point in time I am more concerned about ......

 

Orde one number
Maintaining lnaeasing my
my client alrrent bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

Decision 2 - It is now November 1, 2002

Bonus in TWO YEARS AND TWO MONTHS (DECEMBER 31, 2004) would be

5.15.9182
Optional extension to December 31, 2005

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

126. Atthis pointintime,whatisthebonusyouexpecttoeamattheendoftl'llsplan?:
$

127. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purchase you really want to make?:
$

128. Atthispolntintimelfeelasthoughmybonusis ......

Ordeonenumber
Behindofwhere Aheadofwhere
I'dlikeittobe I'dlikeittobe

1 2 3 4 5 6 7

 

 

 

 

110

129. At this point in time I am more concerned about ......

 

Circle one number
Maintaining Increasing my
my current alrrent bonus
bonus level level
1 2 3 4 S 6 7

 

 

 

Decision 3 - It is now December 1, 2002

Bonus in TWO YEARS AND 30 DAYS (DECEMBER 31, 2004) would be w
Optional extension to December 31, 2005

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

 

130. Atthis point in time, whatisthebonusyouexpecttoeamattheendofthis plan?:
15

131. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purchase you really want to make?:

$

132. At this point in time I feel as though my bonus is ......

 

Clrcleonenumber
Behindofwhere Aheadofwhere
I'dlikeittobe I’dlikeittobe
1 2 3 4 5 6 7

 

 

 

133. At this point in time I am more ccncemed about ......

 

Cirdeonenumber
Maintaining Increasing my
mycurrent currentbonus
bontslevel level
1 2 3 4 5 6 7

 

 

 

111

Decision 4 - It is now December 1, 2003

Bonus in ONE YEAR & ONE MONTH (DECEMBER 31, 2004) would be $1.2m
Optional extension to December 31, 2005

 

 

 

 

 

 

 

 

 

 

 

 

MONTH 5.52002 ”.2002 113.2002 “Jim mm .lul.2002 ELM 55pm
% INCREASE 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 1.00% 1.75%
BONUS LEVEL 0 0 0 0 0 0 0 0
MONTH 0.51.2002 Nov.2002 oer-.3002 31.2003 5.52003 “.2003 Apr.2%3 Haynes
% INCREASE 200% 3.00% 3.00% 4.50% 7.25% 725% 6.88% 7.55%
BONUS LEVEL 0 45.000 45.000 60,000 105.000 105.000 90.000 105.000
MONTH Jun. 2003 311 2003 riﬂe: Sep. 2003 Oct. 2003 Nov. 2003 Dec. 2003

% INCREASE 5.90% 3.00% 7.00% 7.00% 7.00% 7.00% 3.00%

BONUS LEVEL 90.000 120.000 105.000 105.000 105.000 105.000 120.000

 

 

 

 

 

 

 

 

 

134. Atthispointintime,whatistl'lebonusyouexpecttoeamattheendofthisplan?:
$

135. At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purchase you really want to make?:
$

136. At this point in b'me I feel as though my bonus is ......

 

Cirdeonenumber
Behindofwhere Aheadofwhere
I'd llkeittobe I'dlikeittobe
1 2 3 4 5 6 7

 

 

 

137. At this point in time I am more concerned about ......

 

Clrdeone number
Maintaining Increas‘ng my
myalrrent amelitbonus
bonuslevel level
1 2 3 4 5 6 7

 

 

 

112

Decision 5 - It is now January 1, 2004

Bonus in ONE YEAR (DECEMBER 31, 2004) would be 51,29,909
Optional extension to December 31, 2005

 

138. Atthispointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:

139.

140.

141.

142.

$

At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to
use to secure ﬁnancing on a purdiase you really want to make?:
$

Atthispointintimelfeelasthoughmybonusis ......

 

Cirdeonenumber
Behindofwhere Aheadofwhere
I’d like it to be I'd like it to be
1 2 3 4 5 6 7

 

 

 

At this point in time I am more concerned about ......

 

 

 

 

Circle one number
Maintaining Inaeasng my
my current errant bonus
bonus level level
1 2 3 4 5 6 7
Has your initial bonus goal changed? Clrde one: yes no

Ifyespleaseindicatethenewﬁgures

113

Decision 6 - It Is now April 1, 2004

Bonus in NINE MONTHS (DECEMBER 31, 2004) would be M
Optional extension to December 31, 2005

MONTH

% INCREASE
BONUS LEVEL
MONTH

% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

MONTH
% INCREASE
BONUS LEVEL

143.

144.

145.

146.

147.

 

Atthispointintime,whatisthebonusyouexpecttoeamattheendofthisplan?:
$
At this point in tlrne, how much oftheﬁgureyou listed above are you conﬁdent enough to
usetosecure ﬁnancingona purdiaseyou really wanttomake?:

$ .
Atthispointintimelfeelastl'loughmybonusis ......

Clrcleonenumber

 

Behindofwhere Aheadofwhere
I'dlikeittobe I'dlikeittobe
1 2 3 4 5 6 7

 

 

 

Atthispointintimelammoreconcemedabout ......

 

Orde one number
Maintaining “10835109 my
my current alrrent bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

Pleaseindicatethebonusamountyouwouldneedtoeamattheendoftheplaninorder
tofeelthatyou havenotlostanywealthyoufeltyouhadpreviouslygained?:
$ .

114

Decision 7 - It is now June 1, 2004
Bonus in SEVEN MONTHS (DECEMBER 31, 2004) would be w
Optional extension to December 31, 2005

MONTH

% INCREASE
BONUS LEVEL
MONTH

% INCREASE
BONUS LEVEL

MONTH

% INCREASE
BONUS LEVEL
MONTH

% INCREASE
BONUS LEVEL

148.

149.

150.

151.

152.

 

Atthispointintime,whatisthebonusyouexpecttoeamattheendofthis plan?:

6
At this point in time, how much of the ﬁgure you listed above are you conﬁdent enough to

use to secure ﬁnancing on a purchase you really want to make?:
4

Atthispointintimelfeelasthoughmybonusis ......

 

Ordeonenumber
Behindofwhere Aheadofwhere
I'dlikeittobe I'dlikeittobe
1 2 3 4 5 6 7

 

 

 

Atthispointintimelammoreconcemedabout ......

 

Clrde one number
Maintaining Increasing my
my current anrent bonus
bonus level level
1 2 3 4 5 6 7

 

 

 

Pleaseindicatethebonusamountyouwouldneedtoeamattheendoftheplaninorder
tofeelthatyou havenotlostanywealthyoufeltyouhadpreviouslygalned?:
S

 

115

1 .
J

illliiil

‘lli'iiliiliilliili[1H
93 02460 4823

 

41 ~4-F._