This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
 Author's personal copy

Organizational Behavior and Human Decision Processes 123 (2014) 79–89

Contents lists available at ScienceDirect

Organizational Behavior and Human Decision Processes

journal homepage: www.elsevier.com/locate/obhdp

The dark side of consecutive high performance goals: Linking goal

setting, depletion, and unethical behavior
David T. Welsh a,⇑, Lisa D. Ordóñez b,1
a
The University of Washington, Michael G. Foster School of Business, 574 Paccar Hall, Seattle, WA 98195-3226, United States
b
The University of Arizona, The Eller College of Management, 405 McClelland Hall, Tucson, AZ 85721-0108, United States

a r t i c l e i n f o a b s t r a c t

Article history: Over 40 years of research on the effects of goal setting has demonstrated that high goals can increase per-
Received 23 June 2012 formance by motivating people, directing their attention to a target, and increasing their persistence
Accepted 31 July 2013 (Locke & Latham, 2002). However, recent research has introduced a dark side of goal setting by linking
high performance goals to unethical behavior (e.g., Schweitzer, Ordóñez, & Douma, 2004). In this paper,
Accepted by: Maureen Ambrose,
Scott Reynolds and Marshall Schminke
we integrate self-regulatory resource theories with behavioral ethics research exploring the dark side of
goal setting to suggest that the very mechanisms through which goals are theorized to increase perfor-
mance can lead to unethical behavior by depleting self-regulatory resources across consecutive goal peri-
Keywords:
Behavioral ethics
ods. Results of a laboratory experiment utilizing high, low, increasing, decreasing, and ‘‘do your best’’ goal
Unethical behavior structures across multiple rounds provide evidence that depletion mediates the relationship between
Goal setting goal structures and unethical behavior, and that this effect is moderated by the number of consecutive
Self-regulation goals assigned.
Depletion Ó 2013 Elsevier Inc. All rights reserved.

Introduction behavioral ethics research on the dark side of goal setting by

A growing body of research suggests that there is a dark side to
the widely used practice of organizational goal setting (e.g., Barsky,
2008; Jensen, 2003; Ordóñez, Schweitzer, Galinsky, & Bazerman,
2009; Schweitzer et al., 2004). Over 40 years of empirical research
has documented the potency of goal setting as a means of increas-
ing effort, persistence, and performance on a task (Locke & Latham,
2002, 2006), leading some to promote goal-setting theory as the
most important theory in the organizational behavior literature
(Miner, 2003). Yet in some cases, high performance goals may lead
to undesirable outcomes. For example, goals have been linked to
increased unethical behavior, particularly when individuals are
close to meeting difficult performance targets (Schweitzer et al.,
2004).
Demanding performance goals have become a common aspect
of organizational life. In many organizations, after employees meet
their goal for a particular period, they find that management has
ratcheted up future performance targets (Ackman, 2002). Although
the process through which a particular goal is set, pursued, and
achieved has been well documented in the academic literature, rel-
atively little attention has been paid to the impact of consecutive
goals set within a period of time. In this paper, we seek to advance
⇑ Corresponding author. Fax: +1 (520) 621 4171.
E-mail addresses: davidtwelsh@hotmail.com (D.T. Welsh), lordonez@email.
arizona.edu (L.D. Ordóñez).
1
Fax: +1 (520) 621 4171.
exploring the impact of various goal structures on unethical behav-
ior over time. We provide a novel perspective by linking self-regu-
latory resource theories with the mechanisms underlying the
pursuit of consecutive performance goals that may exacerbate
unethical behavior. Empirical evidence suggests that individuals
become depleted when they engage in tasks that require them to
direct their attention, overcome their impulses, and persevere over
time (Baumeister, 2002; Muraven & Baumeister, 2000). Surpris-
ingly, the connection between goal setting and depletion has not
been explored to date even though the process through which
goals are theorized to operate includes many of these same mech-
anisms. Thus, self-regulatory theories may provide new theoretical
insight into the connection between consecutive high performance
goals and unethical behavior. This integration extends behavioral
ethics research by considering how an important contextual factor,
the use of high performance goals, may diminish self-regulatory
resources thereby facilitating unethical behavior in the workplace.
The remainder of this article unfolds as follows. First, we review
recent research on the dark side of goal setting and consider how
various goal structures may contribute to negative outcomes. Sec-
ond, we integrate research on the depletion of self-regulatory re-
sources to theorize how various goal structures may increase
unethical behavior over time. Specifically, we predict that the rela-
tionship between goal structures and unethical behavior will be
mediated by depletion. Third, we explore the importance of con-
secutive goal periods as a moderator of the mediated relationship
between goal structures, depletion, and unethical behavior at

0749-5978/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.obhdp.2013.07.006
 Author's personal copy

80 D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89

two points along the causal sequence. Specifically, we predict that
the relationship between goal structures and depletion will be
strengthened across initial consecutive goal periods as individuals
become increasingly depleted. We further predict that the relation-
ship between depletion and unethical behavior will be weakened
during later consecutive goal periods as some individuals experi-
ence goal disengagement thereby reducing the temptation to cheat
in order to achieve the goal. We test our hypotheses using a variety
of goal structures in a multi-round laboratory experiment.
The dark side of organizational goal setting
Over 400 goal-setting studies conducted over a 40-year period
have demonstrated the benefits of performance goals (Locke & La-
tham, 2006). These studies have repeatedly shown that setting
specific, difficult goals leads to higher levels of task performance
than low or ‘‘do your best’’ goals across both cognitive and physical
tasks. Goal-setting theory has become an important aspect of man-
agement education (Ambrose & Kulik, 1999) and is now a widely
used organizational practice (Locke & Latham, 2002).
However, recent research has suggested that high performance
goals sometimes lead to undesirable outcomes. Goal-setting schol-
ars have suggested that in order to optimize performance, high
goals should be set at the 90th percentile (e.g., Latham & Seijts,
1999). Thus, although a small group of people may experience
heightened self-efficacy and satisfaction upon reaching the goal,
a much larger group will not achieve the goal and consequently
may experience negative effects such as stress, lowered self-es-
teem, and demotivation (Cochran & Tesser, 1996; King & Burton,
2003; Soman & Cheema, 2004). For example, Soman and Cheema
(2004) demonstrated that not meeting a goal led to further poor
performance related to both personal savings and meeting dead-
lines. For many participants, missing a goal led to worse behavior
than having no goal at all. From this perspective, organizations
must employ a balancing act between motivating employees and
pushing them beyond their capacity to perform.
Additional research has suggested that goals can create ‘‘hyper-
motivation’’ by inducing a visceral state that leads individuals to
act in ways that normally would be perceived as unacceptable
(Rick & Loewenstein, 2008). Hypermotivation to perform can also
lead to hypercompetitive behaviors. For example, Poortvliet and
Darnon (2010) found that performance goals led to a competitive,
individualistic mindset coupled with maladaptive and adversarial
social behaviors. In contrast, mastery goals designed to improve
one’s own performance led to positive interactions with others.
In sum, an emerging body of research has connected the imple-
mentation of performance goals to a variety of negative outcomes.
In the next section, we consider how various aspects of goal struc-
ture may influence the relationship between performance goals
and unethical behavior.
Goal structure and unethical behavior
Schweitzer et al. (2004) found that specific, difficult, and unmet
performance goals motivated unethical behavior, both for goals
associated with monetary rewards as well as ‘‘mere goals’’ with
no external incentives. In the study, individuals who were closest
to meeting their goal were the most likely to behave dishonestly.
Similarly, Cadsby, Song, and Tapon (2010) found that goal-based
compensation systems increased unethical behavior more than lin-
ear piece-rate and tournament-based compensation. One explana-
tion for these findings is that individuals’ motivational calculus
may shift depending on a goal’s perceived achievability. Barsky
(2008) suggests that performance goals may also increase unethi-
cal behavior by creating high cognitive loads that focus mental re-
sources on goal attainment rather than moral standards.
Consequently, individuals with high performance goals may have
a narrowed focus and may experience lower levels of moral aware-
ness than those with lower goals. Recent research by Welsh and
Ordóñez (in press) provides empirical evidence supporting Bar-
sky’s (2008) theoretical arguments. Welsh and Ordóñez (in press)
first replicated Schweitzer et al.’s (2004) finding that high goals
can increase unethical behavior and then extinguished this effect
by using subconscious ethical priming to subtly trigger moral
awareness within individuals who had high performance goals.
Barsky (2008) suggests that characteristics of various goal
structures, including the presence, content, and attributes of goals,
will influence the extent to which they increase unethical behavior.
For example, the mere presence of a low goal may be insufficient to
increase unethical behavior. In contrast, a specific, difficult goal
may increase dishonesty (e.g., Schweitzer et al., 2004). The meth-
ods through which goals are set and the rewards associated with
goal attainment may also influence subsequent ethicality. For
example, assigned ‘‘all-or-nothing’’ goals may increase unethical
behavior more than goals with incremental rewards (e.g., Jensen,
2003). Thus, various aspects associated with the structure of a sin-
gle goal have been shown to influence unethical behavior.
Although the process through which a particular goal is set,
pursued, and achieved has been well documented, negative effects
associated with the cycle through which this process is repeated
over time have not been empirically tested in a controlled environ-
ment. In most goal-setting studies, all participants are given a task
to complete; one group receives a specific, high performance goal
and another group is told simply to ‘‘do your best’’ (Locke & La-
tham, 1990). At the end of either the designated time period or
the task itself, performance is assessed and the study is concluded.
However, in an organizational context, goals are often used consec-
utively over time rather than in isolated circumstances. For exam-
ple, an organization may set a performance goal for the quarter. At
the end of the quarter, supervisors evaluate this goal and set a new
goal. Similarly, an organization may set the goal of completing a
specific project. When the project is done, a new project begins,
and a new goal is set. According to goal-setting theory, high goals
induce individuals to devote extra energy and align their behavior
toward a specific target. However, some of the negative effects of
overprescribed goals may have been obscured by a lack of research
examining consecutive organizational goals over an extended time
period.
In practice, organizations often use consecutive goals, in which
the assignment of a new goal immediately follows the completion
of an existing goal; however, goal-setting research has rarely stud-
ied consecutive goal periods (cf. Locke, 1982; Vancouver, Putka, &
Sherbaum, 2005). Consequently, there appears to be a gap between
empirical research focusing on the impact of a single goal as com-
pared to a series of goals. In the next section, we propose a novel
theoretical account of the negative effects of consecutive goals by
suggesting that the mechanisms associated with goal-directed
behavior may lead to diminished self-regulatory capacity when
individuals are presented with a series of high performance goals.
We then explore how this depletion of self-regulatory resources
may exacerbate unethical behavior over time when individuals
pursue consecutive goals.
Applying a self-regulatory perspective to goal-setting theory
According to goal-setting theory (Locke & Latham, 2002), there
are four mechanisms through which goals operate. First, goals di-
rect attention towards goal-relevant activities. Second, goals moti-
vate and energize individuals to put forth greater effort. Third,
goals increase persistence and prolong the effort devoted to a task.
 Author's personal copy
D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
Fourth, goals lead individuals to discover and use new task-rele-
vant knowledge in order to improve performance. These mecha-
nisms increase the productivity of employees with high goals.
For example, loggers with high goals cut more trees (Latham &
Kinne, 1974), truck drivers increased their truck loads (Latham &
Baldes, 1975), and word-processing operators raised their typing
output (Latham & Yukl, 1976). Ultimately, goal-setting scholars ar-
gue that high goals create a cycle of high performance that leads to
rewards such as promotion (Latham, Locke, & Fassina, 2002). Re-
wards increase satisfaction, thereby enhancing self-efficacy and
improving one’s ability to complete future goals.
Given the importance of goal setting as a motivational tool and
the rapidly growing literature on the depleting effects of self-regu-
latory behavior, it is somewhat surprising that the connection be-
tween goal setting and self-regulation has not yet been explored.
Self-regulatory behavior enables individuals to discipline them-
selves by bypassing immediate desires in order to engage in behav-
iors aimed at long-term satisfaction (Baumeister, 2002; Muraven &
Baumeister, 2000). This capacity for volition allows people to exer-
cise self-control by overriding impulses through conscious choices.
Through self-regulation, people have the ability to control cogni-
tive processes that govern thoughts, emotions, and behaviors
(Bandura, 1999; Muraven & Baumeister, 2000). In contrast, lack
of self-control has been linked to behaviors such as substance
abuse, excessive debt, lack of exercise, poor nutrition, unwanted
pregnancy, and dishonesty (Baumeister, 2002; Mead, Baumeister,
Gino, Schweitzer, & Ariely, 2009).
Depletion can be caused by a variety of conditions that require
self-control, regulation of emotions, or high information processing
demands (Hagger, Wood, Stiff, & Chatzisarantis, 2010). However,
despite the many domains that may require self-control, research
has suggested that the ability to self-regulate stems from a single,
common, psychological resource (Baumeister, 2002; Baumeister,
Bratslavsky, Muraven, & Tice, 1998). Baumeister’s strength model
of self-control posits that ego depletion occurs when regulatory
behaviors draw from a limited reservoir of self-control and that re-
straint reduces capacity for further self-regulation (Baumeister &
Heatherton, 1996; Baumeister et al., 1998; Muraven & Baumeister,
2000). Thus, individuals who put forth significant effort and persis-
tence on a particular task may experience depletion and reduced
self-regulatory capacity on future tasks.
Overall, a large body of empirical evidence indicates that indi-
viduals have a limited capacity for self-regulation (Hagger et al.,
2010). For example, participants assigned to exercise self-control
by eating a bowl of radishes while leaving a bowl of chocolates un-
touched were less persistent on a subsequent task involving an
unsolvable geometric puzzle (Baumeister, Bratslavsky, Muraven,
& Tice, 1998). Similarly, participants required to suppress their
emotions while watching an evocative video clip performed worse
on a subsequent self-control task than did participants who had
not engaged in self-regulatory behavior (Muraven, Tice, & Baumei-
ster, 1998). Other self-regulatory tasks, such as completing math
problems under high stress (Webb & Sheeran,2003), giving a coun-
ter-attitudinal speech (Baumeister et al., 1998), and regulating
one’s thoughts (Baumeister, 2002), have all reduced performance
on later tasks requiring self-control. In sum, these studies demon-
strate that tasks requiring motivation, focus, and persistence can
drain self-regulatory resources.
Interestingly, motivation, focus, and persistence are the very
mechanisms through which goal-setting scholars theorize that
high goals are predicted to increase performance (Locke & Latham,
2002). Unlike the vast majority of goal-setting studies, self-regula-
tory studies generally use a sequential-task paradigm in which
individuals complete consecutive tasks. Reduced self-regulation
becomes apparent over time as depletion induced by the first task
diminishes self-regulatory capacity on a subsequent task. Given
81
this close theoretical link between the mechanisms associated
with goal attainment and the mechanisms associated with deple-
tion, we predict that the use of consecutive high performance goals
will have a depleting effect over time in comparison to low and ‘‘do
your best’’ (DYB) goals. The similarity between low goals and DYB
goals may further indicate that it is not merely having a specific
goal that induces depletion, but rather the motivation, focus, and
persistence required to meet the goal that consumes self-regula-
tory resources. In some circumstances, goals that are low enough
to require only minimal motivation, focus, and persistence may
be so easy that they deplete self-regulatory resources even less
than the instruction to ‘‘do your best.’’ In sum, consecutive high
goals may require recurring motivation, focus, and persistence that
may induce depletion in order to meet the target, whereas low and
DYB goals, requiring lower levels of motivation, focus, and persis-
tence, may lead to lower levels of depletion. We hypothesize:
Hypothesis 1. Consecutive high goals will increase depletion more
than consecutive low and DYB goals.
Increasing and decreasing goals
Goal structure involves not only the level of a particular goal
(e.g., high versus low goals) but also changes in this level over time
(e.g. increasing versus decreasing goals). Thus, an additional means
of testing the relationship between goals and depletion involves
comparing goal structures that start high and then decrease over
consecutive goal periods (decreasing goals) with goal structures
that start low and then increase (increasing goals). Decreasing
goals that initially start high may immediately demand high levels
of motivation, focus, and perseverance that quickly diminish self-
regulatory capacity with little opportunity for recovery between
goal periods. Once depleted, self-control must be replenished over
time through a period of rest (Muraven & Baumeister, 2000). Pre-
vious research has shown that engaging in depleting activities
without a break between tasks can have negative implications
for how employees feel and perform at work (Trougakos, Beal,
Green, & Weiss, 2008). However, when individuals are faced with
consecutive depleting tasks, they may not be able to restore de-
pleted resources and may thus have a diminished capacity to reg-
ulate their behavior. Although in a decreasing goal structure, the
rate of depletion may slow in later rounds as the goals decrease,
self-regulatory resources may have already been mostly consumed
during the pursuit of initial goals with no opportunity to recover
these resources during later consecutive goal periods.
In contrast, increasing goals that start low and then gradually
build with each consecutive goal period may allow individuals to
preserve their self-regulatory capacity longer and to experience
lower overall levels of depletion. Initial goals that are not too diffi-
cult will not be overly taxing to self-regulatory resources, thus
allowing individuals to maintain much of their self-regulatory
capacity for later goal periods. In sum, we predict that across var-
ious goal structures, both starting goal level (high versus low) and
changes in goal level across consecutive goal periods (stable goals
versus increasing or decreasing goals) will be associated with
depletion. However, we expect the starting level of the goal (e.g.,
high versus low) will impact depletion more than subsequent
changes in goal level because goals that start high will consume
self-regulatory resources that cannot be recovered during consec-
utive goal periods even when later goals decrease. If the starting le-
vel of a goal has more influence on depletion than changes in goal
level across consecutive goal periods, we would expect the highest
levels of depletion to be associated with high goals, followed clo-
sely by decreasing goals and more distantly by increasing goals,
DYB goals, and low goals. Thus, depletion associated with decreas-
 Author's personal copy
82 D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
ing goals should be somewhat similar to high goals but signifi-
cantly different than low and DYB goals. Depletion associated with
increasing goals should be somewhat similar to low and DYB goals
but significantly different than high and decreasing goals. We
hypothesize:
Hypothesis 2. Consecutive goals that start high and decrease over
time (decreasing goals) will increase depletion more than consec-
utive low and DYB goals.
Hypothesis 3. Consecutive goals that start low and increase over
time (increasing goals) will increase depletion less than consecu-
tive high and decreasing goals.
Linking goals, depletion, and unethical behavior
Two emerging areas within behavioral ethics research involve
the relationship between goals and unethical behavior (e.g., Bar-
sky, 2008; Ordóñez et al., 2009) and the relationship between
depletion and unethical behavior (e.g., Christian & Ellis, 2011;
Mead et al., 2009). However, to date these emerging streams of re-
search have not been connected. Integrating recent research on the
dark side of goal setting with the self-regulation literature may
provide additional theoretical insight into the process through
which high performance goals can increase unethical behavior. If
high performance goals lead to diminished self-regulatory capac-
ity, many of the negative effects associated with depletion may
be applicable to high-goal contexts.
In many cases, compliance with moral standards requires
individuals to overcome the temptation to engage in questionable
conduct. Under the strength model of self-regulation, acts of self-
control draw from a common, global resource that allows execu-
tive control over thoughts, emotions, and behaviors (Baumeister
et al., 1998). However, as this resource is diminished, non-optimal
motivational tendencies may exert a greater influence on behavior
(Christian & Ellis, 2011). Thus, as cognitive resources are drained, it
may become increasingly difficult to conform one’s behavior to
comply with personal standards or organizational policies. For
example, Mead et al. (2009) found that having participants engage
in a self-regulatory task in a laboratory setting increased cheating
on later tasks. Subsequent research has yielded similar findings
regarding the connection between depletion and unethical behav-
ior (Gino, Schweitzer, Mead, & Ariely, 2011). Additionally, Christian
and Ellis (2011) demonstrated that sleep deprivation increased
unethical behavior through the mediating mechanism of depletion.
If consecutive high goals lead to depletion, as we have theorized, it
follows that depletion may act as a mediator between goals and
unethical behavior. Thus, we predict that consecutive high perfor-
mance goals will increase unethical behavior as compared to low
and DYB goals and that this relationship will be mediated by deple-
tion, as high goals will both require and consume more self-regu-
latory resources than low and DYB goals. Additionally, if the
starting level of a goal has more influence on depletion than
changes in goal level across consecutive goal periods, then we
would expect the highest levels of unethical behavior to be associ-
ated with high goals, followed closely by decreasing goals and
more distantly by increasing goals, DYB goals, and low goals. Thus,
we hypothesize:
Hypothesis 4. Depletion of self-regulatory resources will increase
unethical behavior.
Hypothesis 5. Depletion will mediate the relationship between
consecutive goal structures and unethical behavior.
The moderating effect of consecutive goal periods
The strength of the mediating effect of depletion on the rela-
tionship between consecutive goal structures and unethical
behavior may vary based on the number of consecutive goal
periods assigned. The number of consecutive goal periods refers
to how many goals an individual is sequentially required to
complete. We expect that the number of consecutive goal peri-
ods will moderate the mediated relationship between goal struc-
tures, depletion, and unethical behavior at two points along the
causal sequence. First, we predict that the relationship between
goal structures and depletion will be strengthened during initial
consecutive goal periods as individuals become increasingly de-
pleted. At the start, reported levels of depletion may primarily
be the product of a variety of preexisting factors, such as one’s
blood-glucose level (Gailliot & Baumeister, 2007), how many
hours of sleep one has been getting (Christian & Ellis, 2011),
and whether one has engaged in any self-regulatory behaviors
prior to the task at hand. During each consecutive goal period,
high goals will require motivation, focus, and persistence, and
individuals may become increasingly depleted over time as they
must continually engage in goal-directed behavior. After a num-
ber of consecutive goal periods, the effects of depletion may
emerge as self-regulatory capacity is diminished. Thus, the rela-
tionship between high goal structures and depletion will be
strengthened during initial consecutive goal periods.
Second, during later consecutive goal periods the repeated
failure to achieve high goals may lead some individuals to expe-
rience goal disengagement in which the motivation to achieve
the goal is reduced (Wrosch, Scheier, Carver, & Schulz, 2003). Re-
search suggests that although individuals differ widely in their
reactions to seemingly unachievable goals, the desire to meet
the goal is often reduced if the goal appears unobtainable
(Wrosch, Scheier, Miller, Schulz, & Carver, 2003). In the hypoth-
esized mediated model, we predicted that depletion mediates
the relationship between goal structure and unethical behavior
because the temptation to cheat to reach the goal becomes hard-
er to resist as individuals become increasingly depleted. How-
ever, after repeated failures to attain a high goal, some
individuals may experience a diminished desire to achieve the
goal thereby attenuating the link between depletion and uneth-
ical behavior as the temptation to cheat is also reduced. Because
depletion is cumulative across consecutive goal periods and the
additional energy expended to pursue high goals in later rounds
will increase depletion even further relative to easily obtainable
low goals, we expect that there will continue to be a significant
relationship between high versus low goal structures and re-
ported depletion levels during later consecutive goal periods.
Yet despite high levels of depletion in later consecutive goal
periods, as individuals begin to experience goal disengagement
the temptation to cheat in order to attain the goal may decline
leading to a reduction in unethical behavior.
In sum, we predict that the number of consecutive goal peri-
ods will moderate the mediated relationship between goal struc-
ture, depletion, and unethical behavior at two points along the
causal sequence. Specifically, we expect the mediating effect to
initially grow stronger across consecutive goal periods as the
relationship between goal structures and depletion is strength-
ened as individuals’ self-regulatory resources are drained by high
goals. We also expect that the mediation effect will begin to
attenuate in later consecutive goal periods as some individuals
experience goal disengagement thereby weakening the relation-
ship between depletion and unethical behavior as the temptation
to cheat in order to achieve the goal is reduced. Thus, we
hypothesize:
 Author's personal copy
D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
Hypothesis 6. The number of consecutive goal periods will
moderate the mediated relationship between goal structure,
depletion, and unethical behavior.
Method
Participants and design
This study was conducted in a laboratory setting using 159
undergraduate students from a large public U.S. university with
an average of 32 subjects participating in one of five randomly as-
signed goal conditions: high, low, increasing, decreasing, or DYB
goal. The median age of participants was 21, and 40% percent were
female. Participants received class credit for their participation and
were also instructed that they would be paid $1 for each round in
which they achieved the performance goal (except for the DYB
condition, in which no goals were assigned and participants were
told they would be paid a flat fee of $5). Participants completed five
rounds of a problem-solving task and responded to a depletion
scale prior to each round. In accordance with previous experiments
that have used this problem-solving task (e.g., Gino, Norton, & Ari-
ely, 2010; Mazar, Amir, & Ariely, 2008), participants were in-
structed that they would throw away their worksheets for each
round at the end of the experiment and would turn in only the task
packet containing their self-reported performance as well as the
depletion measures. Thus, participants’ work was not checked,
and they could cheat by overstating their performance if they de-
sired. Unbeknown to participants, a unique coded number ap-
peared on both the worksheets and task packets that enabled the
two to be linked after participants had disposed of their materials
and left the room (e.g., Gino et al., 2010; Schweitzer et al., 2004).
Manipulated factors and measures
Goal structure. Participants completed five rounds of a problem-
solving task (Mazar et al., 2008). This task included a series of 20
matrices per round, each containing a set of 12 three-digit num-
bers (e.g., 4.73). Participants had three minutes to circle the two
numbers in each matrix that added up to exactly 10. In the high-
goal condition, participants were given the specific and difficult
goal of solving at least 12 matrices. As in prior goal-setting studies
(e.g., Latham & Seijts, 1999; Schweitzer et al., 2004), the goal in the
high-goal condition was set equal to the 90th percentile of perfor-
mance based on pretesting. In the low-goal condition, participants
were given the goal of solving at least three matrices. Pretesting
indicated that approximately 90% of participants in the low-goal
condition were able to meet the goal. In the increasing condition,
participants were given the initial goal of solving three matrices
in goal period 1, which was increased by an additional three matri-
ces each goal period to establish goals of 6, 9, 12, and 15 matrices.
In the decreasing condition, the pattern was reversed; participants
were given an initial goal of solving 15 matrices in goal period 1,
followed by goals of 12, 9, 6, and 3 matrices in subsequent goal
periods. In the DYB condition, participants did not receive a goal
for any of the rounds and instead were instructed to do their best
to solve as many matrices as possible. To ensure that participants
knew the goal for each round, the goal was stated aloud by the
experimenter and appeared at the top of the task sheet for each
round, and the number of problems needed to meet the goal was
outlined with a dashed line on the task sheet.
Depletion. After participants saw the goal for each round, depletion
was measured using four items from the State Ego Depletion Scale
(Ciarocco, Twenge, Muraven, & Tice, 2010). A sample item is ‘‘I feel
mentally exhausted.’’ Participants responded to these items on a
seven-point scale ranging from strongly disagree to strongly agree.
The reliability of the scale was .96.
83
Unethical behavior. Similar to other studies (Gino et al., 2010; Ma-
zar et al., 2008), the over-reporting of correctly solved matrices in
the problem-solving task was used to assess unethical behavior.
Matrix over-reporting is a good proxy for unethical behavior be-
cause correct solutions are unambiguous, legitimate mistakes are
infrequent, and participants rarely underreport the number of
matrices solved correctly (Mazar et al., 2008). Given the straight-
forward nature of the reporting and the systematic pattern of over-
statements, it appears that most overstatements in our study were
intentional.
Demographics. At the end of the experiment, participants com-
pleted a three-item demographic questionnaire that included
questions about age, gender, and ethnicity.
Results
Table 1 provides means, standard deviations, and correlations be-
tween measured variables and manipulated goal structures. As ex-
pected, goal structures were significantly correlated with
depletion and unethical behavior. Across all rounds and conditions,
participants met the goal approximately 38% of the time. The overall
mean level of depletion (3.43) was slightly below the midpoint of the
seven-point scale, suggesting that, on average, participants were
moderately depleted. As in previous goal-setting studies, goals gen-
erally improved performance on the task. Mean actual performance
was better in the high (5.63), increasing (5.48), and DYB (5.74) goal
conditions than in the low (4.73) and decreasing (4.53) goal condi-
tions, where participants performed relatively poorly.
Table 2 presents mean depletion and unethical behavior by goal
structure condition (in addition to t-values comparing means
across conditions). As depicted in Table 2, levels of depletion across
conditions were generally as predicted, with the highest average
level of depletion in the high-goal-structure condition
(mean = 4.07) followed by the decreasing (mean = 3.97), DYB
(mean = 3.62), increasing (3.09), and low (2.43) conditions. On
average, participants engaged in .51 instances of unethical behav-
ior in each goal period (approximately 2.5 instances total). Levels
of unethical behavior across conditions were as predicted, with
the highest average level of unethical behavior in the high-goal-
structure condition (mean = .94) followed by the decreasing (mean
= .85), increasing (mean = .45), DYB (mean = .30), and low (mean =
.11) conditions.
The correlations in Table 1, means in Table 2, and patterns de-
picted in Figs. 1–3 (shown later) are as predicted. However, in or-
der to account for the within-subjects design, to test interactions
between variables, and to assess the indirect effect of goal struc-
ture on unethical behavior through depletion, we used a combina-
tion of Linear Mixed Modeling (LMM) and regression techniques in
SPSS 19 to analyze the data. LMM (also commonly referred to as
multilevel modeling or hierarchical linear modeling, see Garson,
2012) offers additional flexibility beyond General Linear Modeling
(GLM) based Analysis of Variance (ANOVA) methods because LMM
allows for the simultaneous estimation of fixed and random ef-
fects, as well as the examination of time-varying covariates (Coo-
per, Keage, & Hermens, 2005; Cudeck & Klebe, 2002; Yeates
et al., 2002). Additionally, LMM utilizes model fit indices, such as
the Akaike Information Criterion (AIC) (Akaike, 1973), to assess
the improvement of fit as parameters are added or removed from
the model. The AIC is based on a ‘‘smaller is better’’ heuristic in
which a smaller number is indicative of a better model fit (Marcou-
lides & Hershberger, 1997). In the remaining section, results will be
summarized with respect to each hypothesis in terms of the mean
plots and correlations. Then, the specific LMM will be explained
that fully tests each hypothesis while controlling for subject and
goal period variation.
 Author's personal copy

84 D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89

Table 1
Means, standard deviations, and correlations among variables.a.

Variable Mean S.D. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

b
1. High vs. Low goals .54 .50 –
2. High vs. Low & DYBc .34 .48 – –
3. Decreasing vs. Low & DYBd .25 .44 – – –
4. High & Decreasing vs. Increasinge .60 .49 – – – –
5. High & Decreasing vs. Low & DYBf .46 .50 – – – – –
6. High & Decreasing vs. Low, DYB, & Increasingg .35 .48 – – – – – –
7. Met goalh .38 .36 .94** .94** .84** .35** .87** .58** –
8. Depletion 3.43 1.51 .53** .32** .25* .30** .31** .30** .37** –
9. Unethical behavior .51 1.02 .30* 30** .43** .18  .31** .29** .28** .15  –
10. Age 21.4 2.96 .02 .06 .08 .05 .06 .06 .01 .07 .10 –
11. Genderj .40 .49 .03 .07 .08 .02 .08 .04 .09 .12 .13 .08 –
a
n = 159 Unless otherwise noted.
b
n = 63, Goal structure coded as 1 = High goal, 0 = Low goal.
c
n = 100, Goal structure coded as 1 = High goal, 0 = Low & DYB goal.
d
n = 88, Goal structure coded as 1 = Decreasing goal, 0 = Low & DYB goal.
e
n = 93, Goal structure coded as 1 = High & Decreasing goal, 0 = Increasing goal.
f
n = 122, Goal structure coded as 1 = High & Decreasing goal, 0 = Low & DYB goal.
g
Goal structure coded as 1 = High & Decreasing goal, 0 = Increasing, low & DYB goal.
h
Met goal coded a mean of the number of times a participant met the goal across all goal periods.
j
Gender coded as 0 = Male, 1 = Female.
 
p < .10.
*
p < .05.
**
p < .01.

Table 2
Means, standard deviations, and contrasts for depletion and unethical behavior across goal structures.a.

Goal structureb Variablec Mean S.D. High Decreasing Increasing DYB

High Depletion 4.07 1.32
Unethical 0.94 1.82
Decreasing Depletion 3.97 1.60 0.26
Unethical 0.85 1.01 0.20
Increasing Depletion 3.09 1.53 2.89** 2.10*
Unethical 0.45 0.45 1.57 2.02*
DYB Depletion 3.62 1.29 1.45 0.91 1.63
Unethical 0.30 0.45 2.08* 2.92** 1.32
Low Depletion 2.43 1.36 4.85** 3.70** 1.81  3.63**
Unethical 0.11 0.14 2.45* 3.93** 3.32** 2.16*
a
t Value and significance level displayed for all contrasts.
b
Goal structures in this table are theoretically ordered from highest expected levels of depletion and unethical behavior to lowest expected levels.
c
Mean depletion level and instances of unethical behavior across all rounds.
 
p < .10.
*
p < .05.
**
p < .01.

Fig. 1. Theoretical diagram.

Hypothesis 1 predicted that consecutive high goals would in- each). Fig. 2 indicates that consecutive high goals increased deple-
crease depletion more than consecutive low and DYB goals. Fig. 2 tion more than low and DYB goals. As predicted, we also found a
depicts mean depletion levels (y-axis) over consecutive goal peri- significant positive correlation between goal structure (high versus
ods (x-axis) across goal structure conditions (separate curve for low and DYB) and depletion (r = .32, p < .01). However, to account
 Author's personal copy
D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
Fig. 2. Mean depletion ratings plotted as a function of goal period, with a separate
curve for each goal structure condition.
Fig. 3. Cumulative number of times that completed matrices were overstated
plotted as a function of goal period, with a separate curve for each goal structure
condition.
for within-subjects variation across goal periods, we used LMM to
test Hypothesis 1. A random subject intercept model was specified
with goal structure (high versus low and DYB) entered as a fixed
effect covariate with depletion as the dependent variable. Goal
structure was significant (t(61) = 3.52, p < .001), revealing that con-
secutive high goals increased depletion more than low and DYB
goals. Both the residual (Wald Z = 15.78, p < .001) and subject
intercept (Wald Z = 6.73, p < .001) covariance parameters were sig-
nificant, indicating that the variability in depletion could be ex-
plained by a combination of within- and between-subject effects.
Additionally, a linear mixed model fit the data better
(AIC = 1347.66) than a between-subjects model (AIC = 1858.03,
DAIC = 510.37). Thus, Hypothesis 1 was supported.
Hypothesis 2 predicted that consecutive decreasing goals that
started high and decreased over consecutive goal periods would in-
duce higher levels of depletion than consecutive low and DYB
goals. Fig. 2 indicates that levels of depletion were generally higher
in the decreasing condition than in the low and DYB conditions.
We also found the predicted significant positive correlation be-
tween goal structure (decreasing versus low and DYB) and deple-
85
tion (r = .25, p < .05). To test Hypothesis 2 using LMM, a random
subject intercept model was specified with goal structure (decreas-
ing versus low and DYB) entered as a fixed effect covariate with
depletion as the dependent variable. Goal structure was significant
(t(85) = 2.75, p < .01), revealing that consecutive decreasing goals
increased depletion more than low and DYB goals. Both the resid-
ual (Wald Z = 14.80, p < .001) and subject intercept (Wald Z = 6.32,
p < .001) covariance parameters were significant, indicating that
the variability in depletion could be explained by a combination
of within- and between-subject effects. Additionally, a linear
mixed model fit the data better (AIC = 1187.70) than a between-
subjects model (AIC = 1671.34, DAIC = 483.64). Thus, Hypothesis
2 was supported.
Hypothesis 3 predicted that consecutive increasing goals that
started low and increased over consecutive goal periods would in-
crease depletion less than consecutive high and decreasing goals.
Fig. 2 indicates that levels of depletion were generally lower in
the increasing condition than in the high and decreasing condi-
tions. As predicted, we also found a significant positive correlation
between goal structure (high and decreasing versus increasing)
and depletion (r = .30, p < .01). To test Hypothesis 3 using LMM,
a random subject intercept model was specified with goal struc-
ture (high and decreasing versus increasing) entered as a fixed ef-
fect covariate with depletion as the dependent variable. Goal
structure was significant (t(90) = 3.20, p < .01), revealing that con-
secutive increasing goals increased depletion less than high and
decreasing goals. Both the residual (Wald Z = 15.22, p < .001) and
subject intercept (Wald Z = 6.42, p < .001) covariance parameters
were significant, indicating that the variability in depletion could
be explained by a combination of within- and between-subject ef-
fects. Additionally, a linear mixed model fit the data better
(AIC = 1313.03) than a between-subjects model (AIC = 1767.38,
DAIC = 454.35). We also compared depletion in the increasing
and decreasing conditions at the crossover point in round 3, in
which participants in both conditions had the goal of solving nine
matrices. As predicted, participants in the decreasing-goal condi-
tion (mean = 4.18) were significantly more depleted than partici-
pants in the increasing-goal condition (mean = 3.00, (t(57) = 2.64,
p < .05). Thus, Hypothesis 3 was supported.
Hypothesis 4 predicted that depletion of self-regulatory re-
sources would increase unethical behavior. Consistent with this
prediction, we found a marginally significant correlation between
depletion and unethical behavior across all rounds and goal struc-
tures (r = .15, p < .10). Additionally, this finding is more fully expli-
cated by Hypotheses 5 and 6, which predicted a more complex
moderated mediation model of the relationship between goal
structures, depletion, and unethical behavior over multiple goal
periods. Hypothesis 5 predicted that depletion would mediate
the relationship between consecutive goal structures and unethical
behavior. Additionally, Hypothesis 6 predicted that the number of
consecutive goal periods would moderate the mediated relation-
ship between goal structure, depletion, and unethical behavior.
To test for evidence of moderated mediation, we used Preacher
and Hayes’ (2008) approach to examining moderated mediation
models. This procedure is an extension of the Sobel test (Sobel,
1982) and is recommended over Baron and Kenny (1986) because
it does not assume a normal sampling distribution of indirect ef-
fects (Preacher & Hayes, 2008; Williams & MacKinnon, 2008). As
Preacher and Hayes (2008) have recommended, we estimated the
indirect effect using unstandardized coefficients and utilized boot-
strapping procedures with 1,000 resamples to place a 95% confi-
dence interval around the estimate of the indirect effect.
Bootstrapping provides evidence of mediation if the bias-corrected
95% confidence interval excludes zero for indirect effects. Because
we predicted that the magnitude of the indirect effects would vary
across consecutive goal periods in a non-linear manner (increasing
 Author's personal copy
86 D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
in early rounds and diminishing in later rounds), we ran the medi-
ation analysis separately for each goal period in order to provide a
more detailed analysis of potential conditional mediation effects.
We used round-specific measures of depletion and unethical
behavior rather than cumulative measures. Additionally, this
goal-period-specific analysis eliminated potential within-subjects
problems that cannot be controlled for with this procedure across
a simultaneous examination of multiple rounds.
Supporting Hypotheses 5, we found a significant indirect effect
of high versus low and DYB goals on unethical behavior through
depletion. Supporting Hypothesis 6, the strength of this indirect ef-
fect was moderated by the number of consecutive goal periods.
Specifically, as predicted, effect sizes were small and non-signifi-
cant in goal period 1 (coefficient = .01, 95% CI = .04, .14) and goal
period 2 (coefficient = .03, 95% CI = .30, .15). In goal period 3, the
indirect effect was significant (coefficient = .22, 95% CI = .07, .53)
because the bias-corrected confidence interval did not include
zero. This effect diminished in goal period 4 (coefficient = .08,
95% CI = -.04, .14) and goal period 5 (coefficient = .09, 95% CI =
.01, .30), with the indirect effect approaching significance in the
former and a marginally significant indirect effect in the latter.
We also ran the mediation analysis using the same procedure
but with the increasing and decreasing conditions added to the
model. A similar pattern emerged with a significant indirect effect
of goal structure on unethical behavior through depletion in goal
period 3 (coefficient = .15, 95% CI = .04, .34) in which the bias-cor-
rected confidence interval did not include zero.
We also conducted a supplemental analysis in which we
decomposed the indirect effects of high goal structure on unethical
behavior via depletion across consecutive goal periods. As pre-
dicted, we found that the relationship between high goal structure
and depletion increased from goal period 1 (r = .15) to goal period 2
(r = .27) to goal period 3 (r = .35) suggesting that the indirect effect
grew stronger as high goals increased depletion during initial
rounds. Consistent with our hypotheses, we also found a decrease
in the relationship between depletion and unethical behavior in la-
ter rounds with the correlation dropping from goal period 3
(r = .28) to goal period 4 (r = .13) to goal period 5 (r = .05) suggest-
ing that depleted participants’ motivation to cheat was reduced in
later goal periods. In sum, results indicated that both the relation-
ship between goal structures and depletion and the relationship
between depletion and unethical behavior varied across consecu-
tive goal periods as hypothesized.
Discussion
The purpose of this study is to extend research on the dark side
of goal setting though an integration of behavioral ethics research
with self-regulatory resource theories. Specifically, we provide evi-
dence of process through which consecutive performance goals can
exacerbate unethical behavior over time. Results suggest that con-
secutive high goals can diminish self-regulatory resources and in-
crease unethical behavior. Depletion was found to mediate the
relationship between goal structure and unethical behavior and
the number of consecutive goal periods was found to moderate this
mediating effect, thereby providing new insight into the mecha-
nisms through which high performance goals can increase dishon-
esty over time. The results of this integration of behavioral ethics,
goal setting, and self-regulatory theories suggest a variety of theo-
retical and practical implications.
Theoretical implications
Our results extend current theory in four ways. First, we inte-
grate behavioral ethics research on the dark side of goal setting
with self-regulatory resource theories to provide new evidence of
the process through which goals can increase unethical behavior.
Although goal setting has been rated as the most important theory
in the organizational behavior literature (Miner, 2003) and uses
many of the same mechanisms as self-regulatory resource theories,
these literatures had not yet been integrated. This integration pro-
vides new insight into the heated debate regarding the ethical con-
sequences of goal setting. Some have argued that prior articles on
the negative effects of overprescribed goals have relied largely on
anecdotal evidence (Locke & Latham, 2009). Our results provide
empirical evidence suggesting that overuse of high goals can lead
to diminished self-regulatory capacity and unethical behavior. Be-
cause few goal-setting studies have considered the effects of con-
secutive goals over time, we suggest a new framework through
which behavioral ethics can inform goal-setting theory and the
unethical consequences of high performance goals should be con-
sidered. Our results provide evidence of moderated mediation, in
which consecutive goals can increase unethical behavior through
depletion, an effect moderated by the number of consecutive goal
periods. Although some previous research has demonstrated a con-
nection between goals and unethical behavior, to our knowledge
this is the first study to find empirical evidence of both the medi-
ated process through which negative outcomes associated with
high goals are manifested and the moderating influence of consec-
utive goal periods.
Second, our results extend behavioral ethics research by
emphasizing the importance of performance goals as a contextual
factor that can influence unethical behavior over time. Recent
meta-analytic evidence has demonstrated that unethical behavior
is often more closely associated with one’s organizational environ-
ment than individual differences (Kish-Gephart, Harrison, & Tre-
viño, 2010). For example, the organizational environment at the
accounting firm Arthur Anderson changed over time as manage-
ment’s focus shifted from delivering an honest audit to maximizing
return business in order to meet profitability goals (Toffler & Rein-
gold, 2003). Similarly, our results suggest that when organizations
consistently set aggressive performance targets for employees and
appear to care more that the numbers were achieved than how the
numbers were achieved, they may be creating an environment in
which employees will both be highly depleted and highly tempted
to cheat in order to reach the goal. Our findings contribute to a
growing body of research suggesting that other antecedents of
depletion, such as sleep deprivation, can increase the unethical
behavior of employees (e.g., Christian & Ellis, 2011). However,
one important difference is that sleep deprivation generally occurs
outside of the workplace, whereas the depleting effects of high per-
formance goals are actually caused by the organization itself. In
sum, we advance behavioral ethics research by demonstrating that
the use of consecutive high performance goal structures may rep-
resent a key contextual factor that can shape the ethicality of the
environment in which task performance takes place.
Third, our results indicate that behavioral ethics researchers
should give greater attention to the ethical consequences associ-
ated with particular goal structures. Specifically, we found that cer-
tain types of goal structures are especially likely to lead to high
levels of unethical behavior over time. Across conditions, unethical
behavior was influenced not only by the difficulty of the goal but
also by the availability of self-regulatory resources to resist the
temptation to cheat. A comparison of the increasing and decreasing
goal conditions highlights the ethical implications associated with
different goal structures. Participants in the decreasing condition
continued to experience high levels of depletion and committed
more unethical acts than their counterparts in the increasing con-
dition even as goal levels dropped. At the crossover point during
round 3, participants in both the increasing and decreasing condi-
tions had the goal of solving nine matrices. Yet participants in the
 Author's personal copy
D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
decreasing condition were significantly more depleted and en-
gaged in significantly more unethical behavior during this round.
As hypothesized, it appears that high goals quickly drained self-
regulatory resources that could not easily be recovered, even when
goal levels later decreased. In fact, mean levels of depletion were
always higher in the decreasing condition than in the increasing
condition, even during the final round, where participants in the
decreasing condition had a goal that was five times higher than
that in the increasing condition. These findings suggest that it is
not only absolute goal level that influences unethical behavior,
but also changes in the goal level over time.
Fourth, previous research has suggested that high performance
goals should be set at the 90th percentile in order to maximize per-
formance (e.g., Latham & Seijts, 1999). However, our results indi-
cate that when setting goals, it is important to consider not only
what will maximize performance on a specific goal, but how a par-
ticular goal structure will influence depletion and unethical behav-
ior related both to this goal and subsequent goals. During initial
goal periods, we found that the relationship between high goal
structure and depletion was strengthened as participants grew
increasingly depleted. In later goal periods, the relationship be-
tween depletion and unethical behavior was attenuated as partic-
ipants started to disengage from the goal and thus the temptation
to cheat in order to achieve the goal was diminished. However, this
disengagement also hurt performance with participants in the high
goal condition performing worse than participants in the increas-
ing goal condition during the last two rounds and only slightly bet-
ter than participants in the DYB goal condition. Future research is
needed to more fully explore factors that can influence the rela-
tionship between goal structure, depletion, and unethical behavior
across consecutive goal periods. However, our results suggest that
it may be time to consider replacing goal-setting theory’s 90th per-
centile framework with a more nuanced approach that considers
ethical, self-regulatory, and temporal factors when setting a goal.
Managerial implications
Our findings regarding the ethical consequences of consecutive
high goals have important implications for managers because goal
setting has become an important part of management education
(Ambrose & Kulik, 1999) and is a frequently used organizational
practice (Locke & Latham, 2002). However, a recent debate in the
goal-setting literature has left many unanswered questions for
managers regarding the degree to which goals facilitate unethical-
ity (Locke & Latham, 2009; Ordóñez et al., 2009). Given that indi-
vidual instances of unethical behavior in the workplace are
estimated to cost organizations more than $600 billion per year
in the United States alone (Ariely, 2008), many managers are inter-
ested in discovering factors that may beassociated with employee
dishonesty. One of the limitations of much of the goal-setting liter-
ature is that the consecutive use of goals, which is common in
many organizations, has been overlooked by research using a sin-
gle-round, single-goal paradigm. Our results integrate goal-setting
theory with recent behavioral ethics research on self-regulation to
demonstrate that some negative effects associated with goal set-
ting may manifest themselves more strongly over time as individ-
uals become depleted. Thus, this study may have implications for
organizations that regularly use consecutive goals to increase the
performance of their employees.
A common business ethics myth is that unethical behavior in
the workplace is simply the result of a few bad apples (Treviño &
Brown, 2004). However behavioral ethics scholars have argued
that situational factors in organizations can create pressures that
influence the (un)ethical behavior of ordinary employees (e.g., Tre-
viño & Brown, 2004). Because managers often set the performance
87
goals for their employees, some managers may actually be creating
an environment that facilitates unethicality. For example, over
time automotive technicians at Sears began completing unneces-
sary repairs and overcharging customers in order to meet sales
goals that had been ratcheted up to $147 per hour by management
(Dishneau, 1992). Similarly, energy traders at Enron responded to
increased performance objectives by figuring out creative ways
of moving power around the grid, such as by creating artificial
power emergencies to drive up the price of electricity (McLean &
Elkind, 2003). In each of these examples, high goals may have led
to the depletion of self-regulatory resources as performance expec-
tations were increased across a series of consecutive goals. The
dark side of organizational goal setting appears to frequently
emerge when employees are worn down by consecutive high per-
formance goals set by management and begin to cheat in order to
meet the target.
The over-reporting behavior in this study has organizational
equivalents, such as self-reporting the number of hours worked
in an accounting firm, the number of billable tasks charged to a cli-
ent in a law firm, or the number of customers contacted in a sales
agency. In many organizations, employees are given high perfor-
mance goals that are linked to significant financial incentives.
Widespread dishonesty in the workplace has led many firms to de-
vote significant resources to screening applicants through integrity
testing and monitoring employees through surveillance technolo-
gies (Sackett, Burris, & Callahan, 1989; Young & Case, 2004). Yet
our results introduce the possibility that managers could unwit-
tingly be increasing the unethical behavior of employees through
the overuse of consecutive high performance goals. Like previous
goal-setting studies, we found that high goals can increase perfor-
mance, yet the corresponding increase in depletion and unethical
behavior that we identified suggests that the overuse of goals
may also generate negative consequences for organizations.
Although further research is needed to explore the extent to which
these findings will generalize to the process through which goals
are implemented over time in an organizational environment,
managers may want to consider how the frequency and difficulty
of the goals that they assign may influence their employees. We
hope that, in practice, this study will lead managers to more care-
fully consider the process through which they assign employee
goals and the potential ethical consequences associated with the
overuse of high goals.
Limitations and directions for future research
This study is not without limitations. For example, although
goal-setting studies are often conducted in a laboratory setting, it
is unclear the extent to which these findings can be generalized
across different types of organizations and performance goals.
Nevertheless, the laboratory setting allowed us to conduct a mul-
ti-round study in which we measured depletion and unethical
behavior at multiple times and inferred causality between these
variables. We believe that this research design strengthened our
theoretical contribution, given our objective of extending current
theory by integrating two previously unconnected literatures (goal
setting and self-regulation), the complexity of the phenomena in-
volved (manipulating multiple goal structures across multiple
rounds with multiple repeated measurements, including the
assessment of actual unethical behavior), and the importance of
causality in testing the mediated process through which goals
can influence unethical behavior through depletion.
The relatively short time frame of each goal-setting round in
our study is comparable to previous research (e.g., Mazar et al.,
2008; Schweitzer et al., 2004) but may not reflect the duration of
time required to complete many organizational goals. Yet because
 Author's personal copy
88 D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
high goals were found to increase depletion over time, longer
rounds would likely increase depletion and unethical behavior
even further, potentially strengthening the effects that we found.
Relatively short rounds also allowed us to reduce potential learn-
ing effects from one round to the next. However, future research
could consider how time horizons influence the relationship be-
tween consecutive organizational goals on a daily, weekly,
monthly, or yearly basis. Future research could also consider the
effects of goals during periods when employees may be particu-
larly susceptible to the effects of depletion, such as right before a
meal, at the end of the day, or near the end of the work week. It
is also possible that self-regulatory depletion caused by certain
goal structures could spill over into other domains requiring self-
regulation, such as overeating, impulsive purchasing, or emotional
outbursts.
One potentially interesting finding that was not hypothesized
or discussed previously was that participants who started with
challenging goals in the high and decreasing conditions reported
feeling more depleted after seeing the goal but before even begin-
ning performance. This may imply that high goals have the poten-
tial to make individuals feel depleted even before they have begun
a task and is consistent with recent research suggesting a psycho-
logical component of depletion (Job, Dweck, & Walton, 2010).
Some research has even argued that, in some cases, depletion ef-
fects may be better predicted by perceptions of depletion than by
the actual depletion experience itself (Clarkson, Hirt, Jia, & Alexan-
der, 2010). Our results suggest both a psychological component of
depletion merely through exposure to a high goal as well as task-
induced depletion that increased over time when participants had
high goals. Future research is needed to examine multiple types of
depletion relating not only to the task itself but also to how the
task connects to underlying psychological processes. For example,
the psychological component of depletion may be reduced when
employees participate in setting goals or have high levels of intrin-
sic motivation. Additionally, whereas our analyses used mean lev-
els of depletion for each round, future research could more
explicitly consider the effects of changes in depletion across
rounds. For example, modest changes in depletion may have little
impact for individuals with very high or very low levels of self-reg-
ulatory resources, but may have significant impact for those with
moderate levels of self-regulatory resources. Similarly, future re-
search could examine whether there are specific self-regulatory
thresholds at which the effects of depletion become particularly
salient.
The depleting effects of high goals may suggest that in some sit-
uations goals may be associated with other negative outcomes in
addition to unethical behavior. For example, other antecedents of
depletion, such as sleep deprivation, have been linked to negative
emotional states such as hostility (Christian & Ellis, 2011). Thus, fu-
ture research could consider whether the depletion induced by
high performance goals leads to other types of negative outcomes.
Additionally, individual personality traits may strengthen or atten-
uate the connection between goals and depletion. For example,
conscientious individuals may have greater self-regulatory capac-
ity and thus may be less susceptible to the effects of depletion.
Feelings of self-efficacy or the perception of social resources might
also provide a buffer against depleting events (Beckes & Coan,
2011; Schnall, Zadra, & Proffitt, 2010). Finally, interventions such
as rest breaks could also be tested as a means of reversing the
depleting effects of performance goals.
Conclusion
Although goal setting is one of the most important theories
in the field of organizational behavior and is a widely used
management practice, researchers had not explored the negative
effects of consecutive goals over time. By integrating recent re-
search related to behavioral ethics and goal setting with self-regu-
latory resource theories, we found that high goals can increase
unethical behavior by depleting self-regulatory resources. The re-
sults also provide evidence of moderated mediation through which
the effects of goal structure on depletion and unethical behavior
are influenced by the number of consecutive goal periods. Given
the growing interest in understanding the drivers of unethical
behavior and the ubiquity of goal setting as an organizational prac-
tice, these results extend behavioral ethics research and have
implications for managers that regularly use consecutive goals as
a motivational tool. We hope these findings will encourage future
research exploring additional mechanisms through which goals
may influence unethical behavior over time.
Acknowledgments
This research was supported by funding from EthicsPoint. We
would like to thank Stephen Gilliland, Nathan Podsakoff, Aleksan-
der Ellis, and Sam Birk for their valuable insights and suggestions.
Additionally, we would like to thank Andrea Pittarello, Daren An-
sher, and Zach Wojtowicz for their assistance in data collection.
References
Ackman, D. (2002, March 22). Pay madness at Enron. Forbes.com. <http://
www.forbes.com/2002/03/22/0322enronpay.html> (Retrieved 01.06.12).
Akaike, H. (1973). Information theory and an extension of the maximum likelihood
principle. In B. Petrov & F. Csaki (Eds.), Second international symposium on
information theory (pp. 267–281), Budapest.
Ambrose, M. L., & Kulik, C. T. (1999). Old friends, new faces: Motivation research in
the 1990s. Journal of Management, 25, 231–292.
Ariely, D. (2008). Predictably irrational: The hidden forces that shape our decisions.
HarperCollins
Bandura, A. (1999). Social cognitive theory of personality. Handbook of personality:
Theory and Research, 2, 154–196.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical
considerations. Journal of Personality and Social Psychology, 51, 1173–1182.
Barsky, A. (2008). Understanding the ethical cost of organizational goal-setting: A
review and theory development. Journal of Business Ethics, 81, 63–81.
Baumeister, R. (2002). Ego depletion and self-control failure: An energy model and
the self’s executive function. Self and Identity, 1, 129–136.
Baumeister, R., Bratslavsky, E., Muraven, M., & Tice, D. (1998). Ego depletion: Is the
active self a limited resource? Journal of Personality and Social Psychology, 74,
1252–1265.
Baumeister, R., & Heatherton, T. (1996). Self-regulation failure: An overview.
Psychological Inquiry, 7, 1–15.
Beckes, L., & Coan, J. A. (2011). Social baseline theory: The role of social proximity in
emotion and economy of action. Social and Personality Psychology Compass, 5,
976–988.
Cadsby, C. B., Song, F., & Tapon, F. (2010). Are you paying your employees to cheat?
An experimental investigation. The B.E. Journal of Economic Analysis & Policy,
10(1).
Christian, M., & Ellis, A. (2011). Examining the effects of sleep deprivation on
workplace deviance: A self-regulatory perspective. Academy of Management
Journal, 54, 913–934.
Ciarocco, N. J., Twenge, J. M., Muraven, M., & Tice, D. M. (2010). The state self-control
capacity scale: Reliability, validity, and correlations with physical and psychological
stress. Monmouth University. (Submitted for publication).
Clarkson, J. J., Hirt, E. R., Jia, L., & Alexander, M. B. (2010). When perception is more
than reality: The effects of perceived versus actual resource depletion on self-
regulatory behavior. Journal of Personality and Social Psychology, 98, 29–46.
Cochran, W., & Tesser, A. (1996). The ‘‘what the hell’’ effect: Some effects of goal
proximity and goal framing on performance. In L. L. Martin (Ed.), Striving and
feeling: Interactions among goals, affect, and self-regulation (pp. 99–120).
Hillsdale, NJ: Lawrence Erlbaum Associates, Inc..
Cooper, N. J., Keage, H., & Hermens, D. (2005). The dose-dependent effect of
methylphenidate on performance, cognition, and psychophysiology. Journal of
Integrative Neuroscience, 4, 123–144.
Cudeck, R., & Klebe, K. J. (2002). Multiphase mixed-effects models for repeated
measures data. Psychological Methods, 7, 41–63.
Dishneau, D. (1992, June 22). Sears admits mistakes, takes workers off commission.
The Associated Press.
Gailliot, M. T., & Baumeister, R. F. (2007). The physiology of willpower: Linking
blood glucose to self-control. Personality and Social Psychology Review, 11,
303–327.
 Author's personal copy
D.T. Welsh, L.D. Ordóñez / Organizational Behavior and Human Decision Processes 123 (2014) 79–89
Garson, D. (2012). Hierarchical linear modeling: Guide and applications. Thousand
Oaks, CA: Sage Publications, Inc..
Gino, F., Norton, M., & Ariely, D. (2010). The counterfeit self: The deceptive costs of
faking it. Psychological Science, 21, 712–720.
Gino, F., Schweitzer, M., Mead, N., & Ariely, D. (2011). Unable to resist temptation:
How self-control depletion promotes unethical behavior. Organizational
Behavior and Human Decision Processes, 115, 191–203.
Hagger, M., Wood, C., Stiff, C., & Chatzisarantis, N. (2010). Ego depletion and the
strength model of self-control: A meta-analysis. Psychological Bulletin, 136,
495–525.
Jensen, M. C. (2003). Paying people to lie: The truth about the budgeting process..
European Financial Management, 9(3), 379–406.
Job, V., Dweck, C. S., & Walton, G. M. (2010). Ego depletion – Is it all in your head?:
Implicit theories about willpower affect self-regulation. Psychological Science,
21, 1686–1693.
King, L. A., & Burton, C. M. (2003). The hazards of goal pursuit. Virtue, Vice, and
Personality: The Complexity of Behavior, 53–69.
Kish-Gephart, J. J., Harrison, D. A., & Treviño, L. K. (2010). Bad apples, bad cases, and
bad barrels: meta-analytic evidence about sources of unethical decisions at
work. Journal of Applied Psychology, 95(1), 1–31.
Latham, G. P., & Baldes, J. (1975). The ‘‘practical significance’’ of Locke’s theory of
goal setting. Journal of Applied Psychology, 60, 122–124.
Latham, G. P., & Kinne, S. B. (1974). Improving job performance through training in
goal setting. Journal of Applied Psychology, 59, 187–191.
Latham, G. P., Locke, E. A., & Fassina, N. E. (2002). The high performance cycle:
Standing the test of time. In S. Sonnentag (Ed.), Psychological management of
individual performance (pp. 201–228). Chichester, England: Wiley.
Latham, G. P., & Seijts, G. H. (1999). The effects of proximal and distal goals on
performance on a moderately complex task. Journal of Organizational Behavior,
20, 421–429.
Latham, G. P., & Yukl, G. (1976). Effects of assigned and participative goal setting on
performance and job satisfaction. Journal of Applied Psychology, 61, 166–171.
Locke, E. A. (1982). Relation of goal level to performance with a short work period
and multiple goal levels. Journal of Applied Psychology, 67, 512–514.
Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance.
Englewood Cliffs, NJ: Prentice Hall.
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal
setting and task motivation: A 35-year odyssey. American Psychologist, 57,
705–717.
Locke, E. A., & Latham, G. P. (2006). New directions in goal setting theory. Current
Directions in Psychological Science, 15, 265–268.
Locke, E. A., & Latham, G. P. (2009). Has goal setting gone wild, or have its attackers
abandoned good scholarship? Academy of Management Perspectives, 23, 17–23.
Marcoulides, G. A., & Hershberger, S. L. (1997). Multivariate statistical methods: A first
course. Mahwah, NJ: Lawrence Erlbaum.
Mazar, N., Amir, O., & Ariely, D. (2008). The dishonesty of honest people: A theory of
self-concept maintenance. Journal of Marketing Research, 45, 633–644.
McLean, B., & Elkind, P. (2003). The smartest guys in the room: The amazing rise and
scandalous fall of Enron. Penguin.
Mead, N., Baumeister, R., Gino, F., Schweitzer, M., & Ariely, D. (2009). Too tired to tell
the truth: Self-control, resource depletion, and dishonesty. Journal of
Experimental Social Psychology, 45, 594–597.
Miner, J. B. (2003). The rated importance, scientific validity, and practical usefulness
of organizational behavior theories: A quantitative review. AOM Learning and
Education, 2, 250–268.
Muraven, M., & Baumeister, R. (2000). Self-regulation and depletion of limited
resources: Does self-control resemble a muscle? Psychological Bulletin, 126,
247–259.
89
Muraven, M., Tice, D. M., & Baumeister, R. F. (1998). Self-control as a limited
resource: Regulatory depletion patterns. Journal of Personality and Social
Psychology, 74(3), 774.
Ordóñez, L., Schweitzer, M. E., Galinsky, A. D., & Bazerman, M. H. (2009). Goals gone
wild: How goals systematically harm individuals and organizations. Academy of
Management Perspectives, 23, 6–16.
Poortvliet, P. M., & Darnon, C. (2010). Toward a more social understanding of
achievement goals. The interpersonal effects of mastery and performance goals.
Current Directions in Psychological Science, 19(5), 324–328.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for
assessing and comparing indirect effects in multiple mediator models. Behavior
Research Methods, 40, 879–891.
Rick, S., & Loewenstein, G. (2008). Hypermotivation. Journal of Marketing Research,
45, 645–653.
Sackett, P. R., Burris, L. R., & Callahan, C. (1989). Integrity testing for personnel
selection: An update. Personnel Psychology, 42, 491–529.
Schnall, S., Zadra, J. R., & Proffitt, D. R. (2010). Direct evidence for the economy of
action: Glucose and the perception of geographical slant. Perception, 39,
464–482.
Schweitzer, M. E., Ordóñez, L., & Douma, B. (2004). Goal setting as a motivator of
unethical behavior. Academy of Management Journal, 47, 422–432.
Sobel, M. (1982). Asymptotic confidence intervals for indirect effects in structural
equation models. In S. Leinhart (Ed.), Sociological methodology (Vol. 13, pp. 290–
312). San Francisco: Jossey-Bass.
Soman, D., & Cheema, A. (2004). When goals are counterproductive: The effects of
violation of a behavioral goal on subsequent performance. Journal of Consumer
Research, 31, 52–62.
Toffler, B. L., & Reingold, J. (2003). Final accounting: Ambition, greed, and the fall of
Arthur Andersen. New York: Broadway Books.
Treviño, L. K., & Brown, M. E. (2004). Managing to be ethical: Debunking five
business ethics myths. Academy of Management Executive, 19(2), 69–81.
Trougakos, J., Beal, D., Green, S., & Weiss, H. (2008). Making the break count: An
episodic examination of recovery activities, emotional experiences, and
performance of positive affective displays. Academy of Management Journal,
51, 131–146.
Vancouver, J. B., Putka, D. J., & Sherbaum, C. A. (2005). Testing a computational
model of the goal-level effect: An example of a neglected methodology.
Organizational Research Methods, 8, 100–127.
Webb, T., & Sheeran, P. (2003). Can implementation intentions help to overcome
ego-depletion? Journal of Experimental Social Psychology, 39, 279–286.
Welsh, D. T., & Ordóñez, L. D. (in press). Conscience without cognition: The effects of
subconscious priming on ethical behavior. Academy of Management Journal.
Williams, J., & MacKinnon, D. P. (2008). Resampling and distribution of the product
methods for testing indirect effects in complex models. Structural Equation
Modeling: A Multidisciplinary Journal, 15, 23–51.
Wrosch, C., Scheier, M. F., Carver, C. S., & Schulz, R. (2003a). The importance of goal
disengagement in adaptive self-regulation: When giving up is beneficial. Self
and Identity, 2(1), 1–20.
Wrosch, C., Scheier, M. F., Miller, G. E., Schulz, R., & Carver, C. S. (2003b). Adaptive
self-regulation of unattainable goals: Goal disengagement, goal reengagement,
and subjective well-being. Personality and Social Psychology Bulletin, 29(12),
1494–1508.
Yeates, K. O., Taylor, H. G., Wade, S. L., Drotar, D., Stancin, T., & Minich, N. (2002). A
prospective study of short- and long-term neuropsychological outcomes after
traumatic brain injury in children. Neuropsychology, 16, 514–523.
Young, K. S., & Case, C. J. (2004). Internet abuse in the workplace: New trends in risk
management. CyberPsychology and Behavior, 7, 105–111.