The Performance Benefits of Reward Choice: A Procedural Justice Perspective

Abstract

Reward choice – employees' ability to exercise control over the formal rewards they receive from work – is an important part of many HRM strategies. Reward choice is expected to increase employee performance, but conflicting findings highlight the need to better understand how and when it will do so. Based on fairness heuristic theory, we predicted that procedural justice mediates reward choice's influence on performance, and that choice attractiveness moderates that influence. A field study and an experiment both had similar results, supporting our predictions. Reward choice can increase performance by as much as 40 per cent, but only when the available choices are attractive to employees.

Full text

Reward choice p. 1
NOTE: This is the authors’ final submitted version (before editing by the publisher) of the
article published as
Caza, A., McCarter, M. W., & Northcraft, G. B. (2015). Performance benefits of reward
choice: A procedural justice perspective. Human Resource Management Journal, 25, 184–
199.
THE PERFORMANCE BENEFITS OF REWARD CHOICE:
A PROCEDURAL JUSTICE PERSPECTIVE
Arran Caza*
Griffith University
a.caza@griffith.edu.au; arran.caza@umanitoba.ca
Matthew W. McCarter
University of Texas at San Antonio
matthew.mccarter@utsa.edu
Gregory B. Northcraft
University of Illinois
northcra@illinois.edu
* Corresponding author
This research was supported by the Illinois Bureau of Economic and Business Research, the
University of Illinois Center for Human Resource Management, and the University of Illinois
Campus Research Board. The Centre for Work, Organisation and Wellbeing at Griffith
University supported the editing of this article. We benefited from the assistance of David
Collings and our anonymous reviewers, as well as Ariel Avgar, Brianna Caza, Robert Duncan,
Ashley Israel, and Lu Wang who commented on early drafts of this article.

Reward choice p. 2
Performance benefits of reward choice: A procedural justice perspective
ABSTRACT
Reward choice – employees’ ability to exercise control over the formal rewards they receive
from work – is an important part of many human resource management strategies. Reward
choice is expected to increase employee performance, but conflicting findings highlight the need
to better understand how and when it will do so. Based on fairness heuristic theory, we predicted
that procedural justice mediates reward choice’s influence on performance, and that choice
attractiveness moderates that influence. A field study and an experiment both had similar results,
supporting our predictions. Reward choice can increase performance by as much as 40%, but
only when the available choices are attractive to employees.
Keywords: choice; organisational justice; employee performance; rewards

Reward choice p. 3
An increasingly popular human resource management strategy is giving employees
reward choice, which provides employees with the ability to choose the level or type of rewards
they receive from work (IOMA, 2011; Miceli and Heneman, 2000; White, 2009). In the United
States, for example, fewer than 20 major employers allowed workers any control over how they
were rewarded in 1980 (Hewitt Associates, 1993), but almost all major employers were offering
some reward choice by 2007 (Employee Benefits, 2007). Adoption of reward choice in other
countries shows the same increasing trend (Koo, 2011; Rao, 2008). Allowing employees to
exercise some control over the nature of their rewards has become an important part of many
organisations’ human resource management strategy.
Organisations appear to have embraced reward choice for two reasons: to control costs by
only providing those rewards employees actually value, and to benefit from improvements in
workers’ attitudes and behaviour (Kelliher and Anderson, 2008; Lovewell, 2010). The adoption
of choice reflects the fact that the traditional approach of offering standardised rewards has not
succeeded in controlling costs or enhancing performance; such generic reward plans have tended
to produce poor results (Beer and Cannon, 2004; Chiang and Birtch, 2006). Reward choice
allows organisations to provide rewards that are customised to the individual, and is believed to
help when competing to recruit and retain the best employees, as well as contributing to their
subsequent performance (Fay and Thompson, 2001; Koo, 2011; Nazir et al., 2012). Consistent
with the belief that reward choice benefits the worker and the organisation, scholarship has
linked reward choice with both increased task performance and performance-related attitudes,
such as organisational commitment (Cole and Flint, 2004; Cooper et al., 1992; Lawler and
Hackman, 1969).

Reward choice p. 4
However, despite the widespread adoption of reward choice and research evidence of its
effect on performance, two important gaps remain in the literature. The first gap concerns the
mechanism linking reward choice to performance, which has remained unknown. The second
gap concerns anecdotal and empirical evidence suggesting that employers do not always get a
positive response from introducing reward choice (IOMA, 2011; Shreeram, 2012; Sullivan,
2009). While some studies have found a positive relationship between reward choice and
performance (Cooper et al., 1992; Lawler and Hackman, 1969), at least one study (Morgeson et
al., 2001) found that reward choice was not beneficial. It seems that reward choice can improve
employee performance, but does not always do so. As such, the crucial next step for theory and
practice lies in understanding how reward choice affects performance, including the mediating
mechanism and boundary conditions.
In service of this goal, we report two studies that answer the two research questions:
What is the mediator of the relationship between reward choice and work performance? And:
What are the boundary conditions of this relationship? Study 1 used a field survey to extend
previous research by showing that reward choice influences performance-related attitudes
through the mediating mechanism of procedural justice. Study 2 used an experimental design to
replicate the Study 1 findings and to examine boundary conditions for reward choice. Study 2
found that reward attractiveness is an important moderator of the effect of reward choice (i.e.
reward choice only increases justice and performance if the available choices are attractive).
THEORETICAL BACKGROUND AND HYPOTHESES
To study the relationship between reward choice and performance, we adopted the total
rewards perspective (Lin et al., 2014; Nazir et al., 2012). This perspective defines rewards to

Reward choice p. 5
include all of the valued outcomes that employees derive from their work, including base pay,
incentives, non-salary benefits and perquisites (Chiang and Birtch, 2006; Fay and Thompson,
2001; Gross and Friedman, 2004). Focusing on an employee’s total rewards is consistent with
larger trends in the study of HRM, which are recognising the need to move toward more holistic
views (Boxall, 2013). The total rewards perspective is based on the fact that different employees
value different types of rewards (Kinnie et al., 2005; Krausert, 2014), but their response to
having their desires satisfied will be similar. For example, imagine one worker who desires more
pay, and another who is most concerned with flexible work arrangements. Giving both workers
the same reward (e.g. more pay) would provoke differing responses, but if each worker is given
what s/he desires, they should have similarly positive responses (Cable and Edwards, 2004). The
total rewards perspective focuses on compensation practices in general, so as to accommodate
such diversity in the workforce.
Reward Choice and Performance
Organisations are increasingly offering customised total rewards by allowing workers to
control some aspects of the level or type of rewards they receive. The specifics of the reward
choice vary from simple forms such as choosing among different package options to situations
where an employee’s entire reward package is idiosyncratic (e.g. Anand et al., 2010). Providing
employees with reward choice is expected to reduce organisational costs, increase worker
satisfaction and improve worker performance (Kelliher and Anderson, 2008; Lovewell, 2010).
The first two effects of reward choice – reduced cost and increased reward satisfaction – are
straightforward. It has been documented that introducing reward choice decreases the cost of
reward provision while simultaneously increasing workers’ satisfaction (Dencker et al., 2007;
Karoly and Panis, 2004).

Reward choice p. 6
Beyond these two effects, however, there is some evidence that reward choice can also
improve performance. For example, one study found that workers who designed their own
reward system performed better than workers who had an identical reward system imposed on
them (Cooper et al., 1992). Cooper and colleagues (1992) speculated that reward choice will
increase performance when rewards are perceived as fair, and that having workers involved in
designing the reward system leads to perceptions of fairness.
Other findings also fit with the hypothesis that reward choice promotes perceptions of
fairness. One study showed that workers perceived a reward system as less fair, less responsive,
less motivating, and less satisfying when that system was created exclusively by management,
rather than through a joint management-union effort (Schwarz, 1989). Similarly, another study
found that allowing workers to design their own reward system made workers more trusting of
management (Jenkins and Lawler, 1981).
In contrast, however, one study failed to find any benefit from designing a new reward
system (Morgeson et al., 2001): workers were extensively involved in providing information
about reward preferences to management, but showed no positive response to the new rewards.
Although Morgeson and colleagues (2001) admitted surprise at this outcome, we submit that it
reflects the difference between choice and participation (Leana et al., 1990). In most studies of
reward choice, the comparison is between one group of workers who designed their own reward
system and another group of workers who had a system imposed on them. As such, these studies
confounded participation (i.e. having voice or giving input about preferred options) with choice
(i.e. making the actual reward selection). The studies that found positive responses compared
‘participation + choice’ to ‘no-participation + no-choice’. In contrast, Morgeson and colleagues
(2001) compared ‘participation + no-choice’ to ‘no-participation + no-choice’: some workers in

Reward choice p. 7
their study gave input, but the final system was chosen by management. That Morgeson and
colleagues (2001) found no positive outcomes suggests that workers respond to choosing their
rewards, rather than to participating in a process where someone else ultimately chooses. Choice
is the key factor.
Additional evidence for the importance of choice was provided by a field experiment
where some autonomous work groups were able to design their own reward plan, while control
groups had an identical plan imposed on them by management (Lawler and Hackman, 1969).
The initial results of Lawler and Hackman’s (1969) field experiment were consistent with other
work in this domain, showing that workers in the reward choice condition significantly increased
their performance-related behaviours while those in the control (no choice) condition did not.
While this design initially was like others in confounding participation and choice, a longitudinal
component of the study indirectly separated the two elements.
Follow-up data collected by Scheflen, Lawler and Hackman (1971) a year after the
reward redesign revealed two important findings. The first finding was that the behavioural
change lasted for a year; it was not a short-term result. The second finding was that losing the
chosen reward system reversed the benefits associated with it. After the first year, management
at the organisation discontinued the worker-designed reward plan. However, the removal of the
plan did not reduce workers’ rewards; in fact, management raised overall compensation levels,
so that workers were making more than they were under the employee-designed plan (Scheflen
et al., 1971). Nonetheless, when the reward system changed, performance behaviours reverted to
pre-experiment levels. The study did not measure the mechanism involved, but when considered
with the results of Morgeson and colleagues (2001), it seems likely that losing the feeling of
choice was crucial. The change was not caused by the level of reward, since the workers’

Reward choice p. 8
rewards actually increased when their plan was removed. The element that changed was that the
workers no longer had a reward plan they had chosen for themselves.
In summary, it is clear that allowing workers to design their own reward system can
improve performance. Furthermore, the evidence suggests that the choice exercised in system
design is the key to increasing worker performance (Fang and Gerhart, 2012). Changes in the
level of rewards do not account for observed performance gains, nor is it sufficient to participate
in the design if someone else chooses what rewards are received.
An important limitation of previous studies is that they examined choice in the context of
employees designing their own reward system. Obviously, such complete system design involves
a sweeping change, and of necessity cannot occur often. Nonetheless, the underlying mechanism
of positive responses to choice suggests that performance should be improved even if reward
choice is less sweeping than designing the entire system. We therefore propose the constructive
replication hypothesis that reward choice will increase performance, regardless of the scope of
change.
H1: Reward choice increases worker performance.
Reward Choice, Procedural Justice and Performance
Stating the hypothesis that reward choice will increase performance begs the question of
why. Drawing on fairness heuristic theory (Lind, 2001; Lind et al., 2001), we posit that
procedural justice is the mediating mechanism linking reward choice to worker performance. We
begin by defining procedural justice, and then use fairness heuristic theory to explain why
reward choice will increase procedural justice and why procedural justice will increase
performance.

Reward choice p. 9
Organisational justice is defined as an individual’s perception of fairness in an
organisation (Lind and Tyler, 1988), and while multiple types of justice have been found
(Colquitt et al., 2012), the most relevant to reward choice are distributive justice and procedural
justice. Distributive justice refers to workers’ evaluation of whether the results of a decision are
fair (Konovsky, 2000). For example, workers may judge whether the organisation is contributing
an appropriate amount toward their retirement savings. In contrast, procedural justice concerns
workers’ evaluation of the process used to reach an outcome, distinct from the outcome itself
(Thibaut and Walker, 1975). In the retirement contribution example, procedural justice would
involve workers’ views about the fairness of how contribution amounts were calculated (e.g.
more for employees with greater seniority, maximum amount limits).
Procedural justice is usually more important than distributive justice (Colquitt, 2001).
That is, workers will often accept a relatively poor outcome if they believe it was produced by a
fair process (van den Bos et al., 2001). Moreover, one of the key contributors to perceptions of
fair process is the power to choose (Korsgaard and Roberson, 1995; Lind et al., 1990). As a
result, procedural justice should be the most important issue in the context of reward choice.
Indeed, prior study has shown that reward choice contributes to perceptions of justice in the
reward system (Cole and Flint, 2004). In other words, if employees can choose their rewards,
they will perceive those rewards as more procedurally just.
Furthermore, fairness heuristic theory suggests that workers’ perceptions of the reward
system’s fairness will generalise to their perception of the organisation as a whole (Lind, 2001;
Lind et al., 2001; Mignonac and Richebe, 2013). Fairness heuristic theory explains that
particular occasions of fair treatment lead individuals to assume that those occasions reflect the
overall level of fairness in the organisation. If the organisation acts justly in important and salient

Reward choice p. 10
situations, then the individual will tend to assume that the organisation is fair in general. Because
of how important rewards are to workers, judgments about the fairness of the reward system
should be particularly influential in shaping overall perceptions of the organisation. As such,
reward choice should contribute to generalised perceptions of procedural justice among workers:
having reward choice will lead to perceptions of justice in the reward system, and evaluations of
the reward system will be generalised to the organisation as a whole (Thibaut and Walker, 1975;
Tyler, 1987).
Workers who believe their organisation is procedurally just subsequently believe that
there is greater interpersonal trust between themselves and the organisation, and that trust
motivates them to behave in a cooperative and supportive fashion (Cropanzano et al., 2001;
Lynch et al., 1999). Individuals who feel fairly treated take more action to support the common
good (Tyler and Blader, 2003). When workers believe their organisation is just, they are more
willing to contribute (Grover and Crooker, 1995; Lind, 2001) and their contributions improve
performance (Colquitt et al., 2012; Konovsky, 2000). As such, procedural justice should be the
mechanism by which reward choice increases performance.
H2: Procedural justice mediates the effect of reward choice on worker
performance, such that reward choice increases procedural justice and
procedural justice increases worker performance.
Choice Attractiveness: A Boundary Condition
Given the evidence that having choice increases perceptions of justice (Korsgaard and
Roberson, 1995; Lind et al., 1990), one might assume that any reward choice would increase
justice and subsequent performance. However, research in psychology shows that offering

Reward choice p. 11
choice does not always produce positive responses (Chua and Iyengar, 2006). For example, Botti
and Iyengar (2004) found that if individuals could only choose among meal options that they
found unappealing, the freedom to choose actually reduced their satisfaction with the meal.
Extending these results to the domain of work rewards suggests that not any choice will do; the
available options must be attractive to workers.
For example, imagine an organisation that offers only unsatisfactory reward choices to
employees. While the workers will be able to exercise reward choice, there is no reason to think
that doing so will improve their perception of the reward system in this hypothetical
organisation. Being compelled to choose among a series of unattractive options will not
contribute to perceptions of procedural justice, and is more likely to focus employee attention on
the poor level of their benefits (Colquitt, 2001). Since perceptions of an organisation’s overall
justice depend on perceptions in specific circumstances, the organisation that only offers choice
among unattractive options will not be perceived as more procedurally just in general. Since the
performance gains associated with reward choice derive from generalized perceptions of
procedural justice, offering reward choice among unattractive options will therefore fail to
improve performance. Consistent with this prediction, media reports have noted some overtly
negative responses to reward choice (Shreeram, 2012). Offering reward choice does not
guarantee positive results; it is necessary to offer attractive choices.
H3: Choice among unattractive rewards does not increase procedural justice.
H4: Choice among unattractive rewards does not increase work performance.
STUDY 1
Method

Reward choice p. 12
Sample and procedure. One hundred full-time workers were randomly selected from the
alumni database of a large Midwestern American university Master’s program and invited to
complete an online survey. Forty-six workers provided usable responses, representing a range of
industries and job functions. The respondents were primarily young professionals (mean age =
30.47 years, SD 6.02). All had graduate-level education, and most had recently been promoted
(mean job tenure = 2.30 years, SD 1.63; mean organisational tenure = 4.11 years, SD 3.79); 57%
of the participants were male. More than half (54%) has supervisory duties (mean number of
subordinates = 6.47, SD 17.07).
Variables. To measure reward choice, respondents were asked to describe the two most
important rewards they received from their employer. For each of those rewards, they reported
the extent of their reward choice in terms of degree of choice, satisfaction with degree of choice,
and quality of available choices on 5-point scales. We combined these six items (three items X
two rewards) into a reward choice measure that had good internal consistency (α = .80, 95% CI
[.59, 1.00]).
Because objective performance measures would not be comparable across our sample, we
used affective organisational commitment (Meyer et al., 1993) as a proxy measure for
performance (six items, e.g. ‘I really feel as if this organisation's problems are my own’).
Affective organisational commitment refers to a psychological state of positive attitudes toward
continuing one’s relationship with an organisation; workers are affectively committed to an
organisation to the extent that they identify themselves with that organisation and feel an
emotional attachment to their membership (Meyer and Allen, 1991). Affective organisational
commitment is not a measure of performance, but it is the type of commitment most strongly
associated with performance (Meyer et al., 2002) and meta-analysis has revealed a consistent

Reward choice p. 13
positive relationship between affective commitment and performance (Riketta, 2002). In
particular, when individuals are relatively new to their job – as was the sample used here – the
meta-analytic correlation between affective commitment and performance is .44 (Wright and
Bonett, 2002). Moreover, organisational commitment has been shown to mediate the relationship
between rewards and performance (Park and Kruse, 2014).
Participants also completed Colquitt’s (2001) scale measuring procedural justice (seven
items, e.g. ‘To what extent are the procedures use in the organisation free of bias?’), as well as
items about their demographic characteristics. In addition, to control for different levels of
rewards among participants, we had them report their overall satisfaction with the perceived
value of the rewards provided by their employer (benefit level, four items, e.g. ‘How satisfied are
you with the value of your benefits?’; Williams et al., 2008).
Results and Discussion
Based on responses to demographic questions, respondents did not appear to differ from
non-respondents. Demographic characteristics largely proved to be non-significant control
variables, and even when they were significant, their inclusion did not change any of the
substantive conclusions. As such, demographic variables were excluded for clarity and
parsimony (Aguinis and Vandenberg, 2014). Descriptive statistics are provided in Table 1.
Insert Table 1 here
We used ordinary least squares regression for our primary analyses. Consistent with H1,
reward choice was related to affective organisational commitment (Model 1 in Table 2).
Furthermore, consistent with H2, procedural justice mediated the relationship between reward
choice and affective organisational commitment. Reward choice predicted procedural justice

Reward choice p. 14
(Model 2), and when controlling for the relationship between procedural justice and affective
organisational commitment, there was no consistent relationship between reward choice and
affective commitment (Model 3). To further assess mediation, we conducted two additional tests:
a bootstrap estimation (Preacher and Hayes, 2008) revealed a standardised indirect effect of .23
(95% CI [.03, .46]); and a Sobel (1982) test found a significant effect (z = 2.24, p = .02, 95% CI
[.04, .40]). All of these results are consistent with H2, the prediction that procedural justice
mediates the relationship between reward choice and affective organisational commitment.
Insert Table 2 here
To provide a conservative test of our predictions, we also conducted several additional
analyses. First, we estimated a model that included the control variable for how satisfied
individuals were with their level of rewards (Model 4). This analysis revealed that the
relationship between procedural justice and performance remained after controlling for the level
of rewards. Second, we estimated alternative models using Huber, bi-square and sandwich
corrections, which can improve model estimates in the presence of outliers or heteroskedasticity
(Huber, 1981). In all cases, the results were substantively unchanged. Moreover, post hoc
diagnostics revealed that collinearity was not a concern; the largest variance inflation factor was
only 2.0 (O’Brien, 2007). Additionally, power analysis revealed an observed statistical power
greater than .9, which exceeds the traditional .8 guideline required for appropriate hypothesis
testing (Cohen, 1988). Taken together, all of these results suggest the robustness of the findings.
In sum, using a diverse sample of full-time workers, we found results consistent with
prior research: reward choice was associated with higher levels of the performance-related
attitude of affective organisational commitment. Also consistent with previous studies (e.g.
Tremblay et al., 1998), the effect of reward choice was independent of the level of the reward. In

Reward choice p. 15
addition to replicating previous results, these findings extended previous work in two important
ways. First, where previous studies were conducted primarily with first-level workers, the
majority of our respondents were managers or supervisors. The results therefore suggest that
reward choice has similar effects across hierarchical levels. Second, our results showed that the
effect of reward choice was mediated by procedural justice. This finding is consistent with the
longstanding, but previously untested, prediction that reward choice improves performance by
fostering perceptions of fairness.
Nonetheless, Study 1 had three limitations. First, it used affective organisational
commitment as a proxy, rather than measuring performance directly. Second, it used cross-
sectional data, which prevents us from ruling alternative explanations (e.g. committed employees
may be more likely to perceive the organisation as fair). And finally, the data did not allow a test
of the moderating effect of choice attractiveness. Moreover, the reward choices that the
respondents’ reported on were primarily non-salary benefits, rather than choice about pay. To
address these issues, we conducted a second study in a controlled experimental context.
STUDY 2
Method
Sample and procedure. Eighty-two undergraduate business majors at a large Midwestern
American university participated. Their mean age was 20.5 years (SD 1.61) and 47.5% were
male.
After providing consent, participants were told about the task and how they would be
rewarded (i.e. the experimental manipulation); they then completed the procedural justice
measure from Study 1, and finally completed the compound remote associates task (CRA). In the

Reward choice p. 16
CRA, participants try to find one word that can be combined with each of three different prompt
words to create meaningful compound terms. For example, the prompt ‘night/wrist/stop’ is
solved by the word ‘watch’, which can combine with each of the three prompts. Participants
were presented with 140 sets of prompts (Bowden and Jung-Beeman, 2003) and directed to
complete as many as they could. The CRA was originally developed as a measure of creativity
(Mednick, 1962), but recent research treats the task as an assessment of problem-solving ability
that does not require domain-specific knowledge (Bowden and Jung-Beeman, 2003). As such,
the CRA is considered a good measure of abstract knowledge-based work performance
(Ackerman and Zalmanov, 2012).
Reward choice was manipulated as a three-level, between-subjects factor (attractive-
choices, less-attractive-choices, no-choice), derived from a two-phase pre-test. In the first phase
of pre-testing, a sample of 43 participants was shown a list of 14 possible rewards for completing
an experiment, and asked to rate each reward’s attractiveness. In the second phase of pre-testing,
a different sample of 31 participants rated the overall attractiveness of the three highest-rated
rewards from phase 1 (Set A) and the three lowest-rated rewards from phase 1 (Set B). Set A was
rated as significantly more attractive than Set B, regardless of presentation order (M = 4.68 vs.
2.32 on a 7-point scale, paired t[31] = 8.56, p < .01).
The options in the attractive-choices condition were: (1) two points of extra credit +
$0.10 cash per correct answer; (2) two points of extra credit + $0.10 per correct answer donated
to a charity of their choice; (3) $0.30 cash per correct answer (no extra credit); or (4) two points
of extra credit (the university’s standard experimental reward). The options in the less-attractive-
choices condition were: (1) $0.20 cash per correct answer (no extra credit); (2) two points of
extra credit + $0.13 in fast food coupons per correct answer; (3) two points of extra credit +

Reward choice p. 17
$0.13 donated to the Humane Society per correct answer; or (4) two points of extra credit +
$0.10 cash per correct answer. The first three of the less attractive options were the set rated least
attractive in pre-testing (i.e. Set B); the fourth was the same as the first attractive-choices option
and was the single option judged most attractive in pre-testing. We offered this single, highly
attractive option with the three less attractive rewards to control for the effects of reward content,
as described below.
After learning about the task, participants in the attractive-choices and less-attractive-
choices conditions chose their reward from the appropriate list. All but three of the attractive-
choices participants chose the same reward, and all of the less-attractive-choices participants
chose that same reward. Therefore, participants in the no-choice condition were told that they
would receive that reward: two points of extra credit + $0.10 cash per correct answer. To prevent
any effect from reward content, the three participants who chose a different reward were not
included in the dataset (final N: attractive = 27, less-attractive = 25, no choice = 30). As such, all
participants operated under an identical reward structure; the only difference among groups was
whether they had it imposed on them or chose it from a list of otherwise attractive or less
attractive options.
Results and Discussion
As in Study 1, demographic variables were non-significant covariates. The mean CRA
performance was 30.8 correct responses (SD 17.88), the mean level of procedural justice was
3.92 on a 5-point scale (SD .74), and these two variables were significantly correlated (r = .25, p
= .02). Consistent with Study 1, reward choice increased performance (F[2, 79] = 5.22, p < .01,
η2 = .12) and procedural justice (F[2, 79] = 4.54, p = .01, η2 = .10).

Reward choice p. 18
Planned contrasts supported both H3 and H4 (see Table 3). Participants in the attractive-
choices condition reported greater procedural justice (M = 4.26) than those in the no-choice
condition (M = 3.74; t = 2.74, p < .01), and greater procedural justice than those in the less-
attractive-choices condition (M = 3.77; t = 2.47, p = .02). There was no significant difference in
reported procedural justice between the no-choice and less-attractive-choices conditions (t = .15,
p = .88). As one would expect, given the mediating role of procedural justice, the results were
similar for performance. Participants in the attractive-choices condition (M = 39.26) performed
39.6% better than participants in the no-choice condition (M = 28.13; t = 2.46, p = .02) and
57.8% better than participants in the less-attractive-choices condition (M = 24.88; t = 3.04, p <
.01). There was no statistically significant difference in performance between the no-choice and
less-attractive-choices conditions (t = .71, p =.48).
Insert Table 3
In sum, reward choice increased performance and perceptions of procedural justice, but
only when the choice was made from attractive options. These results extend our previous
study’s findings in three important ways. They replicated them in a different context, they
generalised them to a different criterion variable, which was an objective measure of task
performance, and they provided a clear demonstration of the causal effect of reward choice.
Providing one attractive option in the less-attractive-choices condition was necessary to
control for the effects of reward content, but it may have led participants to perceive this
condition as effectively offering no choice, since only one option was attractive. This could
explain the comparable results between the less-attractive-choices and no-choice groups. It may
be that if the less-attractive-choices condition had offered only unattractive options, performance
may actually have been reduced.

Reward choice p. 19
GENERAL DISCUSSION
In two studies, we examined the relationship between reward choice and performance.
Consistent with predictions, reward choice increased performance through the mediating effect
of procedural justice. Workers provided with reward choice perceived their organisation as more
fair, and that perception of fairness led to improved performance. Our results showed that the
effect of reward choice on performance is independent of the actual level of reward, but does
depend on choice attractiveness. Workers offered a choice among unattractive options did not
perceive greater procedural justice nor did their performance improve; only those choosing from
multiple attractive options had perceptions of greater procedural justice and subsequent
performance gains. Our data combined experimental and field observations, showing consistency
across diverse contexts, tasks, and measures. As a result, our research design provides strong
evidence for the causal effect of reward choice. These findings have important theoretical and
practical implications.
Theoretical Implications
The results make three theoretical contributions. The first is to demonstrate the
previously uncertain mediator of the effect of reward choice on performance. Human resource
management suffers from a lack of theoretical explanations for observed results (Monks et al.,
2013) and poor evidence of the causal nature of observed associations (Guest, 2011). Our data
address both of these issues. Prior to our research, there were no data on the mediated process by
which reward choice influences performance. Allowing workers to create their own reward
system had been shown to enhance performance (Cooper et al., 1992; Lawler and Hackman,
1969), but the explanation for this result remained unclear. We advanced this work by testing the
effects of reward choice in different contexts, with different outcomes and different types of

Reward choice p. 20
choice, while measuring the mediating mechanism. As a result, we extended and clarified those
previous studies, confirming that perceptions of fairness – specifically of procedural justice – are
the mechanism by which reward choice improves performance. Moreover, our use of both field
and experimental data supports inferences of cause: choice is the causal agent.
The second contribution arises from clearly distinguishing between choice and
participation (Barber et al., 1992). Previous studies confounded these two factors, which led to
some apparently anomalous findings (Morgeson et al., 2001). Our results appear to resolve the
anomaly, creating consensus in the literature. In our studies, workers had choice (i.e. they could
decide what type or level of reward they received), but they did not participate in creating those
choices. We found that choice improved performance, while previous studies found that the
combination of choice and participation improved performance (Cooper et al., 1992; Lawler and
Hackman, 1969). Combining these results with the observed lack of benefit among employees
who had participation without choice (Morgeson et al., 2001) suggests that choice, not
participation, is essential. As such, Morgeson and colleagues’ (2001) results are consistent with
previous work, so long as one distinguishes between choice and participation. Nonetheless,
opportunities remain for further investigation. While choosing among pre-defined options is
sufficient to improve performance, future scholarship can investigate whether the benefits are
greater when choice is combined with participation (Deci and Ryan, 2000).
The third contribution of our work is to highlight the complexities of choice, and thereby
identify a significant boundary condition to the benefits of reward choice. Research in other
contexts has shown that choice is not unequivocally regarded as good (Iyengar et al., 2004), and
we likewise found that in the case of work rewards, not any choice will do. Reward choice only
improved performance when individuals perceived their choice options as attractive ones. As

Reward choice p. 21
such, the preferences of workers will be an important consideration in future investigations of
reward systems and reward choice. All choice options will not have equivalent effects,
suggesting an important role for pre-choice participation to ensure that choice options are
attractive.
Practical Implications
Our findings suggest several related considerations for implementation. In discussing
these, it should be recalled that others found that the effects of reward choice persisted among
employees for more than a year (e.g. Scheflen et al., 1971). In the long run, the performance-
enhancing consequences of reward choice may be substantial, which makes this issue an
important one.
The most salient element of our findings is that choice is sufficient to improve
performance. Before our research, one might have inferred from the literature that performance
gains required allowing employees to design their own reward system, which would be
impractical in many contexts. Our results show that reward design is not necessary. It is enough
for management to have workers choose among attractive options. Doing so improved
performance in our study by almost 40%, compared to workers who had no choice. However,
reward attractiveness is an important contingency; there were no benefits from choice among
unattractive options. Organisations must ensure that the rewards offered meet not only their own
needs, but also those of the employees. It may be best to have workers participate in identifying
reward options to ensure that the results are sufficiently attractive. In addition, whatever system
is adopted, it will be crucial to evaluate it in terms of employee perception and response (Corby
et al., 2005).

Reward choice p. 22
In addition, organisational assistance may be required for the successful implementation
of reward choice. In practice, it will likely not be enough to simply provide choice. For example,
many organisations have moved from defined benefit pensions to defined contribution plans, and
while this move clearly gives workers more reward choice, it may not contribute to their
perceptions of organisational justice, particularly since prior work has shown that many
individuals make poor choices for their retirement (Benartzi and Thaler, 2007). Workers may
require help using their freedom to choose.
Conclusion
It seems obvious that the type of rewards provided would influence employees’ attitudes
and behaviour. Our findings support this intuition, but further demonstrate that rewards have an
effect beyond just their level. We integrated and extended previous studies by showing that being
able to choose how one is rewarded has potentially powerful implications for organisational
outcomes. Providing workers with choice over their rewards can lead them to feel more fairly
treated, and thus to provide better performance. This observation clarifies decades of prior
research, and, we hope, provides a foundation and stimulus for future investigation of the
interrelations among rewards, choice, and worker performance.

Reward choice p. 23
REFERENCES
Ackerman, R., and Zalmanov, H. (2012). ‘The persistence of the fluency–confidence association
in problem solving’. Psychonomic Bulletin & Review, 19: 1187–1192.
Aguinis, H. and Vandenberg, R.J. (2014). ‘An ounce of prevention is worth a pound of cure:
Improving research quality before data collection’. Annual Review of Organizational
Psychology and Organizational Behavior, 1: 569–595.
Anand, S., Vidyarth, P.R., Liden, R.C. and Rousseau, D.M. (2010). ‘Good citizens in poor-
quality relationships: idiosyncratic deals as substitutes for relationship quality’. Academy
of Management Journal, 53: 970–988.
Barber, A.E., Dunham, R.B. and Formisano, R.A. (1992). ‘The impact of flexible benefits on
employee satisfaction: A field study’. Personnel Psychology, 45: 55.
Beer, M. and Cannon, M.D. (2004). ‘Promise and peril in implementing pay-for-performance’.
Human Resource Management, 43: 3–48.
Benartzi, S. and Thaler, R.H. (2007). ‘Heuristics and biases in retirement savings behavior’.
Journal of Economic Perspectives, 21: 81–104.
Botti, S, and Iyengar, S.S. (2004). ‘The psychological pleasure and pain of choosing: When
people prefer choosing at the cost of subsequent satisfaction’. Journal of Personality and
Social Psychology, 87: 312–326.
Bowden, E.M. and Jung-Beeman, M. (2003). ‘Normative data for 144 compound remote
associate problems’. Behavior Research Methods, Instruments, and Computers, 35: 634–
639.
Boxall, P. (2013). ‘Mutuality in the management of human resources: Assessing the quality of
alignment in employment relationships’. Human Resource Management Journal, 23: 3–
17.
Cable, D.M. and Edwards, J.R. (2004). ‘Complementary and supplementary fit: A theoretical
and empirical integration’. Journal of Applied Psychology, 89: 822–834.
Chiang, F.F.T. and Birtch, T.A. (2006). ‘An empirical examination of reward preferences within
and across national settings’. Management International Review, 46: 573–596.
Chua, R.Y. and Iyengar, S.S. (2006). ‘Empowerment through choice? A critical analysis of the
effects of choice in organizations’. Research in Organizational Behavior, 27: 41–79.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.), Hillsdale, NJ:

Reward choice p. 24
Lawrence Erlbaum Associates.
Cole, N.D. and Flint, D.H. (2004). ‘Perceptions of distributive and procedural justice in
employee benefits: flexible versus traditional benefit plans’. Journal of Managerial
Psychology, 19: 19–40.
Colquitt, J.A. (2001). ‘On the dimensionality of organizational justice: a construct validation of a
measure’. Journal of Applied Psychology, 86: 386–400.
Colquitt, J.A., LePine, J.A., Piccolo, R.F., Zapata, C.P. and Rich, B.L. (2012). ‘Explaining the
justice–performance relationship: trust as exchange deepener or trust as uncertainty
reducer?’ Journal of Applied Psychology, 97: 1–15.
Cooper, C.L., Dyck, B. and Frohlich, N. (1992). ‘Improving the effectiveness of gainsharing: the
role of fairness and participation’. Administrative Science Quarterly, 37: 471–490.
Corby, S., White, G. and Stanworth, C. (2005). ‘No news is good news? Evaluating new pay
systems’. Human Resource Management Journal, 15: 4–24.
Cropanzano, R., Byrne, Z.S., Bobocel, D.R. and Rupp, D.E. (2001). ‘Moral virtues, fairness
heuristics, social entities, and other denizens of organizational justice’. Journal of
Vocational Behavior, 58: 164–209.
Deci, E.L. and Ryan, R.M. (2000). ‘The “what” and “why” of goal pursuits: Human needs and
the self-determination of behavior’. Psychological Inquiry, 11: 227–268.
Dencker, J.C., Joshi, A. and Martocchio, J.J. (2007). ‘Employee benefits as context for
intergenerational conflict’. Human Resource Management Review, 17, 208–220.
Employee Benefits. (2007, January). ‘Flexible benefits: Flexing up’. Employee Benefits, S.2.
Fang, M. and Gerhart, B. (2012). ‘Does pay for performance diminish intrinsic interest?’
International Journal of Human Resource Management, 23: 1176–1196.
Fay, C.H. and Thompson, M.A. (2001). ‘Contextual determinants of reward systems’ success:
An exploratory study’. Human Resource Management, 40: 213–226.
Gross, S.E. and Friedman, H.M. (2004). ‘Creating an effective total reward strategy: Holistic
approach better supports business success’. Benefits Quarterly, 20.
Grover, S.L. and Crooker, K.J. (1995). ‘Who appreciates family-responsive human resource
policies: the impact of family-friendly policies on the organizational attachment of
parents and non-parents’. Personnel Psychology, 48: 271–288.
Guest, D.E. (2011). ‘Human resource management and performance: still searching for some

Reward choice p. 25
answers’. Human Resource Management Journal, 21:, 3–13.
Hewitt Associates. (1993). Flexible Compensation Programs and Practices. Lincolnshire, IL:
Hewitt Associates.
Huber, P.J. (1981). Robust Statistics. San Francisco, CA: John Wiley & Sons.
IOMA. (2011). ‘Flexible benefits embraced as morale booster’. IOMA’s Report on Managing
Benefits Plans, 13: 4–6.
Iyengar, S.S., Huberman, G. and Jiang, W. (2004). ‘How much choice is too much?
Contributions to 401 (k) retirement plans’, in O.S. Mitchell and S.P. Utkus (eds), Pension
Design and Structure: New Lessons From Behavioral Finance, Oxford, UK: Oxford
University Press.
Jenkins, G.D. and Lawler, E.E. (1981). ‘Impact of employee participation in pay plan
development’. Organizational Behavior and Human Performance, 28: 111–128.
Karoly, L.A. and Panis, C.W.A. (2004). The 21st Century at Work: Forces Shaping the Future
Workforce and Workplace in the United States. RAND Corporation.
Kelliher, C. and Anderson, D. (2008). ‘For better or for worse? An analysis of how flexible
working practices influence employees’ perceptions of job quality’. International Journal
of Human Resource Management, 19: 419–431.
Kinnie, N., Hutchinson, S., Purcell, J., Rayton, B. and Swart, J. (2005). ‘Satisfaction with HR
practices and commitment to the organisation: why one size does not fit all’. Human
Resource Management Journal, 15: 9–29.
Konovsky, M.A. (2000). ‘Understanding procedural justice and its impact on business
organizations’. Journal of Management, 26: 489–511.
Koo, R.C. (2011). ‘The global added value of flexible benefits’. Benefits Quarterly, 27: 17–20.
Korsgaard, M.A. and Roberson, L. (1995). ‘Procedural justice in performance evaluation: the
role of instrumental and non-instrumental voice in performance appraisal discussions’.
Journal of Management, 21: 657–669.
Krausert, A. (2014). ‘HRM systems for knowledge workers: differences among top managers,
middle managers, and professional employees’. Human Resource Management, 53: 67–
87.
Lawler, E.E. and Hackman, J.R. (1969). ‘Impact of employee participation in the development of
pay incentive plans: A field experiment’. Journal of Applied Psychology, 53: 467–471.

Reward choice p. 26
Leana, C.R., Locke, E.A. and Schweiger, D.M. (1990). ‘Fact and fiction in analyzing research on
participative decision making: A critique of Cotton, Vollrath, Froggatt, Lengnick-Hall,
and Jennings’. Academy of Management Review, 15: 137–146.
Lin, Z., Yao, X. and Zhao, Z. (2014). ‘The direct and indirect impact of employee benefits on
firm performance in China’. Asia Pacific Journal of Human Resources, forthcoming.
doi:10.1111/1744-7941.12037
Lind, E.A. (2001). ‘Fairness heuristic theory: Justice judgments as pivotal cognitions in
organizational relations’, in J. Greenberg and R. Cropanzano (eds), Advances in
Organizational Justice, Stanford, CA: Stanford University Press.
Lind, E.A., Kanfer, R. and Earley, P.C. (1990). ‘Voice, control, and procedural justice:
Instrumental and noninstrumental concerns in fairness judgments’. Journal of Personality
and Social Psychology, 59: 952–959.
Lind, E.A., Kray, L. and Thompson, L. (2001). ‘Primacy effects in justice judgments: Testing
predictions from fairness heuristic theory’. Organizational Behavior and Human
Decision Processes, 85: 189–210.
Lind, E.A. and Tyler, T.R. (1988). The Social Psychology of Procedural Justice, New York, NY:
Plenum Press.
Lovewell, D. (2010, March). ‘Flexible benefits: structure’. Employee Benefits, S.10–S.12.
Lynch, P.D., Eisenberger, R. and Armeli, S. (1999). ‘Perceived organizational support: inferior
versus superior performance by wary employees’. Journal of Applied Psychology, 84:
467–483.
Mednick, S.A. (1962). ‘The associative basis of the creative process’. Psychological Review, 69:
220–232.
Meyer, J.P. and Allen, N.J. (1991). ‘A three-component conceptualization of organizational
commitment’. Human Resource Management Review, 1: 61–89.
Meyer, J.P., Allen, N.J. and Smith, C.A. (1993). ‘Commitment to organizations and occupations:
Extension and test of a three-component conceptualization’. Journal of Applied
Psychology, 78: 538–551.
Meyer, J.P., Stanley, D.J., Herscovitch, L. and Topolnytsky, L. (2002). ‘Affective, continuance,
and normative commitment to the organization: A meta-analysis of antecedents,
correlates, and consequences’. Journal of Vocational Behavior, 61: 20–52.

Reward choice p. 27
Miceli, M.P. and Heneman, R.L. (2000). ‘Contextual determinants of variable pay plan design: A
proposed research framework’. Human Resource Management Review, 10: 289–305.
Mignonac, K. and Richebe, N. (2013). ‘“No strings attached?”: How attribution of disinterested
support affects employee retention’. Human Resource Management Journal, 23: 72–90.
Monks, K., Kelly, G., Conway, E., Flood, P., Truss, K. and Hannon, E. (2013). ‘Understanding
how HR systems work: the role of HR philosophy and HR processes’. Human Resource
Management Journal, 23: 379–395.
Morgeson, F.P., Campion, M.A. and Maertz, C.P. (2001). ‘Understanding pay satisfaction: the
limits of a compensation system implementation’. Journal of Business and Psychology,
16, 133–149.
Nazir, T., Shah, S.F.H. and Zaman, K. (2012). ‘Literature review on total rewards: an
international perspective’. African Journal of Business Management, 6: 3046–3058.
O’Brien, R.M. (2007). ‘A caution regarding rules of thumb for variance inflation factors’.
Quality & Quantity, 41: 673–690.
Park, R. and Kruse, D. (2014). ‘Group incentives and financial performance: the moderating role
of innovation’. Human Resource Management Journal, 24: 77–94.
Preacher, K.J. and Hayes, A.F. (2008). ‘Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models’. Behavior Research Methods, 40:
879–891.
Rao, V.S.R. (2008, April 30). ‘Tailored employee benefits’. CiteHR. Retrieved September 27,
2012, from http://www.citeman.com/3078-tailored-employee-benefits.html
Riketta, M. (2002). ‘Attitudinal organizational commitment and job performance: a meta-
analysis’. Journal of Organizational Behavior, 23: 257–266.
Scheflen, K.C., Lawler, E.E. and Hackman, J.R. (1971). ‘Long-term impact of employee
participation in the development of pay incentive plans: A field experiment revisited’.
Journal of Applied Psychology, 55: 182–186.
Schwarz, R.M. (1989). ‘Participative decision making: Comparing union-management and
management-designed incentive pay programs’. Group & Organization Management, 14:
104–122.
Shreeram, J. (2012). ‘Employee satisfaction: Flexible benefits perks fail to woo’. CiteHR.
Retrieved September 28, 2012, from http://www.citehr.com/84293-employee-

Reward choice p. 28
satisfaction-flexible-benefits-perks-fail-woo.html
Sobel, M.E. (1982). ‘Asymptotic confidence intervals for indirect effects in structural equation
models’. Sociological methodology, 13: 290–312.
Sullivan, N. (2009, September). ‘Flexible benefits: a fair exchange’. Employee Benefits, S.3–S.4.
Thibaut, J.W. and Walker, L. (1975). Procedural Justice: A Psychological Analysis, Hillsdale,
NJ: Lawrence Erlbaum Associates.
Tremblay, M., Sire, B. and Pelchat, A. (1998). ‘A study of the determinants and of the impact of
flexibility on employee benefit satisfaction’. Human Relations, 51: 667–688.
Tyler, T.R. (1987). ‘Conditions leading to value-expressive effects in judgments of procedural
justice: A test of four models’. Journal of Personality and Social Psychology, 52: 333–
344.
Tyler, T. and Blader, S. (2003). ‘The group engagement model: Procedural justice, social
identity, and cooperative behavior’. Personality and Social Psychology Review, 7: 349–
361.
Van den Bos, K., Lind, E.A. and Wilke, H.A.M. (2001). ‘The psychology of procedural and
distributive justice viewed from the perspective of fairness heuristic theory’, in R.
Cropanzano (ed), Justice in the Workplace: From Theory to Practice, Mahwah, NJ:
Lawrence Erlbaum.
White, G. (2009). ‘Diversity in workplace causes rise in unique employee benefits and changes
in cafeteria plans’. Journal of Management & Marketing Research, 2: 1–9.
Williams, M.L., Brower, H.H., Ford, L.R., Williams, L.J. and Carraher, S.M. (2008). ‘A
comprehensive model and measure of compensation satisfaction’. Journal of
Occupational and Organizational Psychology, 81: 639–668.
Wright, T.A. and Bonett, D.G. (2002). ‘The moderating effects of employee tenure on the
relation between organizational commitment and job performance: A meta-analysis’.
Journal of Applied Psychology, 87: 1183–1190.

Reward choice p. 29
TABLE 1 Descriptive Statistics and Pairwise Correlations for Study 1A
Variable
Mean
s.d.
α
1
2
3
4
5
6
7
8
9
1
Reward choice
3.88
.92
.80
2
Affective organisational commitment
3.20
.93
.81
.33
3
Procedural justice
3.43
.95
.85
.40
.58
4
Reward level satisfaction
3.72
.91
.95
.69
.16
.34
5
Organisational size (thousands of employees)
24.40
53.78
.11
.22
.19
.23
6
Number of hierarchical levels from respondent
to the top of the organization
4.61
2.29
.14
-.17
-.14
.08
.34
7
Gender (1 = male; 0 = female)
.59
.50
-.26
-.26
-.32
-.26
.19
.32
8
Age (years)
30.48
6.08
.18
.09
.17
.21
.02
-.10
.13
9
Organisational tenure (years)
4.15
3.82
.19
.37
.36
.17
.54
-.03
.17
.40
10
Tenure in current position (years)
2.31
1.64
.02
.01
.00
.12
-.08
-.35
.19
.22
.32
A. n = 46
* Correlations with an absolute value greater than .30 were significant (p < .05).

Reward choice p. 30
TABLE 2 Regression Models for Study 1A
Independent
variables
Model 1
Affective organisational
commitment
Model 2
Procedural
justice
Model 3
Affective organisational
commitment
Model 4
Affective organisational
commitment
Reward choice
.33*
[.04, .62]
.41*
[.12, .70]
.11
[-.16, .39]
.25
[-.11, .60]
Procedural justice
.53*
[.26, .79]
.54*
[.28, .81]
Reward level
Satisfaction
-.20
[-.55, .15]
Constant
1.92*
1.83*
.96
1.15
R2
.11
.16
.35
.37
F
5.28* (df 1, 44)
8.30* (df 1, 44)
11.62* (df 2, 43)
8.28* (df 3, 42)
A. 95% CI in brackets; n = 46
* p < .05

Reward choice p. 31
TABLE 3 Group Means for Study 2A
No choice
Less-attractive choices
Attractive choices
Performance
(correct CRA answers)
28.13
24.88
39.26
Procedural justice
3.74
3.77
4.26
A. No choice N = 30, less-attractive choices N = 25, attractive choices N = 27.