International Journal of Hospitality Management 31 (2012) 10591066

Contents lists available at SciVerse ScienceDirect

International Journal of Hospitality Management

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

How leadermember exchange, work engagement and HRM consistency explain

Chinese luxury hotel employees job performance
Xiaobei Li a, , Karin Sanders b,c , Stephen Frenkel c
a
b
c

Guanghua Leadership Institute, Guanghua School of Management, Peking University, China

Organizational Psychology and Human Resource Development, Faculty of Behavioral Sciences, Twente University, Enschede, The Netherlands
School of Organisation and Management, Australian School of Business, University of New South Wales, Sydney, Australia

a r t i c l e

i n f o

Keywords:
Leadermember exchange (LMX)
HRM process
High performance work systems
Job performance
Hotels
China

a b s t r a c t
This paper provides insights into the relationship between leadermember exchange (LMX) and
employee job performance. An integrative model that includes work engagement and human resource
management (HRM) consistency, dened as the extent to which various HR practices are viewed as
consistent with one other, was developed to explain this relationship. Results from a hierarchical linear
model based on 298 employees (survey data) and 54 supervisors in a large luxury hotel in southern China
indicated that LMX was positively related to employee job performance. Moreover, as expected, work
engagement mediated this relationship and HRM consistency strengthened the inuence of LMX on work
engagement. Important research and practical implications are discussed.
2012 Elsevier Ltd. All rights reserved.

1. Introduction
Globalization has encouraged more intense competition among
service organizations and according to Schneider and Bowen (2010)
organizations need to go beyond customer expectations in order to
win the service game. Fullling customers variable and idiosyncratic needs is difcult and relies on the co-ordination of highly
motivated employees. Sub-optimal performance of any part of the
value chain is likely to reduce customer evaluations of service quality (Fantazy et al., 2010; Wang et al., 2012). Front-line employees
who frequently interact with customers need to be sensitive and
proactive regarding customer requirements. Back-stage employees
those with limited or no interaction with customers are expected
to provide effective support to front-line employees. Engaged and
well performing employees therefore become a key element in
maintaining service excellence.
Research shows that immediate managers (team leaders or
supervisors) can signicantly inuence employees work motivation and job performance (Gerstner and Day, 1997; Ilies et al., 2007;
Liden et al., 1993). In this study we use LMX theory to explore the
inuence of the supervisor, arguing that supportive supervision, as

The authors thank Min Li for facilitating data collection and Zhixue Zhang, Li Ma
and Xin Wei for comments on an earlier version. This research is supported by NFSC
and the number is 71032001.
Corresponding author.
E-mail addresses: Xiaobei.li@gsm.pku.edu.cn, xiaobei.li@gmail.com (X. Li).
0278-4319/$ see front matter 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhm.2012.01.002

perceived by employees, encourages employee engagement which

in turn contributes to job performance. LMX is dened as the quality of the relationship between supervisor and subordinate (Graen
and Scandura, 1987). LMX assumes that supervisors use a different
style for each of their subordinates. Based on dimensions such as
mutual trust, respect and obligation, differentiated relationships
between leaders and followers have been found across cultures
(see Graen and Uhl-Bien, 1995). Law et al. (2010) reported two
studies in China in which LMX was positively related to task performance and OCBs in the rst study while LMX was associated with
contextual performance in the second study. LMX inuences work
outcomes through different mechanisms including organizational
job embeddedness (Harris et al., 2009a,b), trust in the supervisor
(Wat and Shaffer, 2005) and feedback avoidance behavior (Moss
et al., 2009). Recent research has examined LMXs inuences on
service employees emotions. Using a Chinese call center sample,
Huang et al. (2010) found that LMX and burnout were negatively
associated. Emotional states are particularly relevant to service
employees whose performance, as we argue later, depends on their
willingness to go the extra mile In this study, we examine whether
work engagement a construct combining emotion and cognitive disposition acts as a mediator in the relationship between
LMX and employee job performance. Schaufeli et al. (2002) dene
work engagement as a positive, fullling, work-related state of
mind characterized by vigor, dedication and absorption (see also
Bakker and Demerouti (2007)). Modern organizations, especially
in the service sector, need employees who are engaged with their
work: they are more pro-active, take more initiatives and feel more

1060

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

responsible for delivering quality outcomes (Bakker and Leiter,

2010). In addition, our study integrates LMX with an additional
organizational inuence, namely the Human Resource Management (HRM) system. Recently attention has moved from a concern
with the content of HRM, e.g. recruitment and selection, and performance appraisal to HRM process (see Sanders and Frenkel,
2011), i.e. how HRM practices are implemented by line managers
and interpreted by employees, with consequences for organizational performance (Bowen and Ostroff, 2004). In this study we
propose that the more employees perceive consistent messages
concerning the purpose of HR practices (HRM consistency) the
stronger will be LMX inuences on work engagement and employee
job performance.
Prior studies have investigated LMX outcome relationship contingencies including leadership style (Harris et al., 2009a,b), task
characteristics (Dunegan et al., 1992, 2002); employee personality (Harris et al., 2009a,b), top management support (Erdogan and
Enders, 2007); and employee psychological climate (Tordera et al.,
2008). However, little is known about how the characteristics of an
HRM system inuence LMX-outcome relationships. This is surprising since scholars have for some time suggested more integration
between LMX and HRM studies (Lengnick-Hall and Lengnick-Hall,
2003; Uhl-Bien et al., 2000). While a high-quality supervisory
relationship is likely to be an important inuence on employees
performance, consistent HRM messages will tend to reinforce the
alignment between management expectations and employee goals,
fostering stronger engagement and ultimately leading to higher
individual performance.
The remainder of this paper is organized as follows. We begin
by noting key work requirements in luxury hotels followed by an
outline of our theoretical framework and hypotheses. The data and
methodology are then described followed by a presentation and
discussion of our results. We conclude by considering implications
for further research and practice and noting the studys major limitations.

2. Conceptual framework
2.1. LMX, work engagement and job performance
Contemporary hospitality organizations face a challenge of combining service efciency with service quality 24 h a day. This is
particularly the case for luxury (ve-star) hotels where high levels
of employee discipline and application of systematic routines are
required to execute tasks efciently while service quality requires
information and sensitivity to customer requirements and exibility in interacting with customers (Enz and Siguaw, 2000). It is
managements role to encourage and co-ordinate employees in
order to deliver seamless, quality service (Schneider and Bowen,
2010). In this paper we argue that it is the supervisors relations with employees that are crucial to maintaining employee
enthusiasm and that this enthusiasm is aimed at providing service
excellence by a set of supportive HR policies that are consistently
communicated to employees (Fig. 1).
LMX studies indicate the importance of the role supervisors
play in shaping employees work attitudes and performance.
High-quality exchange relationships between supervisors and
employees involve respect and mutual inuence while low-quality
exchange relationships are characterized by strictly contractual
exchanges and one-way, downward inuence. The supervisor is
the most immediate and salient superordinate (Janssen and Van
Yperen, 2004). In a high quality exchange relationship the supervisor as leader supports the subordinate by giving trust and emotional
support (in-group). In the low quality exchange relationship supervisor and subordinate contribute nothing extra than their job

description requirements (out-group) (Graen and Uhl-Bien, 1995;

Uhl-Bien et al., 2000).
LMX theory argues that positive actions by the leader can elicit
a sense of subordinate indebtedness in the form of a gift/favor
exchange (Liden et al., 1997). Thus, a favor induces an unspecied indebtedness with many content dimensions including trust,
competence, consideration, control of organizational resources, etc.
(Bernerth et al., 2007; Schriesheim et al., 1999). Prior research suggests that employees work motivation can be enhanced by high
quality supervisor relationships (Gerstner and Day, 1997; Klein and
Kim, 1998). According to reviews by Gerstner and Day (1997) and
Ilies et al. (2007), LMX has generally positive effects on employee
work outcomes including satisfaction with supervisors, overall job
satisfaction, employee job performance, and organizational citizenship behavior. However, some studies report non-signicant
relationships between LMX and performance (Wayne and Ferris,
1990; Liden et al., 1993; Scandura and Pellegrini, 2008) leading
to calls for exploration of potential mediators and moderators of
LMXoutcome relationships. This is a challenge we address in this
paper.
In a luxury hotel context, supportive leaders tend to provide various resources that assist in coordinating employees. These include
performance feedback and coaching that enhances employees professional skills and knowledge and permits supervisors to provide
more autonomy for employees to undertake their work. Supervisors may grant employees preferred work schedules and give
emotional and social support when difcult customers or onerous workloads place a heavier burden on employees (Vaux, 1988;
House, 1981). Social support provides opportunities for reappraisal
and adaptive responses to work stress; and facilitates well-being
(House, 1981; Cohen and Wills, 1985). Thus high-quality relationships with leaders offer distinct advantages for employees. In
addition, employees may perceive themselves as in-group members (Wayne and Green, 1993). These benets tend to generate
intrinsic motivation through identication with supervisors (Farh
et al., 2006). A willingness to repay supervisors and increased
engagement provide high levels of energy, mental resilience and
willingness to invest effort when encountering difculties, leading
to higher job performance (Bakker et al., 2007; Bakker and Leiter,
2010). A positive relationship between work engagement and job
performance has been supported by studies in a variety of industries (Gottschalg and Zollo, 2007; Leiter, 2010; Bakker and Leiter,
2010) including hotels (Salanova et al., 2005). Consequently, we
propose the following hypothesis:
H1. A positive relationship between leadermember exchange
(LMX) and job performance is mediated by work engagement.
2.2. The moderating role of HRM consistency
Employee reciprocity in high quality LMX relationships with
supervisors may not necessarily benet the organization. This is
because the norm of reciprocity (Blau, 1964) underlying social
interactions is interpersonal rather than organizational. Tekleab
and Taylor (2003) showed that, unlike LMX reported by supervisors,
LMX reported by employees is not signicantly associated with
their perceptions of obligations to the organization. Reciprocation
tends to reect what supervisors value (Hofmann et al., 2003; Liden
et al., 1993). This implies that a high-quality, leadermember relationship does not necessarily convey to employees behavior that is
valued by the organization.
The HRM system is a key organizational inuence that communicates the organizations values, goals and procedures to both
managers and employees (Nishii et al., 2008; Bowen and Ostroff,
2004). For example, HR practices like extensive training programs, performance-based reward systems and promotion criteria

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

usually make it clear to employees how high performance is dened

and the rewards that are attendant on achieving relevant standards or targets. HRM is an important addition to line management
relationships with employees. This is supported by evidence that
HR systems characterized by consistent HR practices are positively
related to organizational performance (Combs et al., 2006; Delery,
1998; Guest, 2007).
Regarding the hotel industry, Cho et al. (2006) found that rms
implementing HRM practices such as labor-management participation programs, incentive plans, and pre-employment tests are more
likely to experience lower employee turnover rates. Alleyne et al.
(2006) measured the effect of HRM on performance in the Barbados hotel industry and found no support for either external t (HR
is more effective when it is aligned with strategy) or internal t
(HR is more effective when different HR practices are aligned with
each other). Rather than focusing on the content of HRM practices,
Bowen and Ostroff (2004) emphasized implementation (process),
arguing that when employees receive consistent HR messages, they
are likely to better comprehend the managerial purposes underlying these practices and act accordingly. This is because HRM
messages are utilized by individuals to make sense of organizational goals and procedures, attributing causes and motives (Kelley,
1967), including their relationship with the organization (Li et al.,
2011; Sanders et al., 2008). Furthermore, managements objectives
are more clearly understood when employees receive consistent
messages over time and across different contexts (Bowen and
Ostroff, 2004; Kassin and Pryor, 1985). Conversely, if HRM messages are not communicated consistently across all HR practices,
employees may perceive a confusing or contradictory picture of
management objectives, making it difcult to connect individual
employee work roles and goals to a broader organizational purpose. Incentives attempting to motivate appropriate behavior may
therefore prove ineffective.
Given that HRM involves the salient motivational aspects of
employment (e.g. pay, working conditions, task attributes and work
relationships), the relationship between LMX and employee performance is likely to be strengthened by HRM consistency. In effect,
HRM messages can act as a lever that aligns employees responses
and interests with the organizations purposes (Gottschalg and
Zollo, 2007). A way of viewing this is that supervisors develop
differentiated relationships with subordinates (LMX) in order to
facilitate attainment of performance targets through employee
reciprocation. This provides obligation-based intrinsic motivation
for employees (Deci and Ryan, 1985). On the other hand, HRM
contributes to realizing organizational goals by offering a rewardsbased incentive (extrinsic motivation) and competency-enhancing
support (resources) to align employee behavior with the organizations goals (Brief and Aldag, 1977). When HR practices are
internally consistent, the organizational goals and the associated
incentives become salient to employees. Having been managed
in this way, employees are likely to recognize that, over-time
high-performance will be rewarded and increasingly value their
contributions to the organizations goals. In a strong situation
(where opportunities and incentives are aligned towards organizational goals), individual employees tend to reciprocate their
LMX in a way that contributes to job performance. This is because
higher performance means higher levels of utility to employees
including benets stemming from both ongoing interpersonal
relationships with supervisors and HR managers. Accordingly,
employees work motivation will be heightened (Deci and Ryan,
1985), ultimately leading to higher performance.
In contrast, in a weak situation where inconsistencies exist
between perceived organizational goals, performance objectives
and HRM messages, employees will see little connection between
their roles and goals and the organizations policies and objectives.
Therefore, when employees reciprocate in relation to supervisors,

1061

their behaviors tend to be in line with personal rather than organizational goals, weakening the relationships between LMX and work
outcomes (engagement and performance) valued by the organization. In addition, the inconsistency mentioned above may increase
employees work stress as a result of cognitive confusion (Bowen
and Ostroff, 2004) thereby reducing engagement.
So far, we have argued that HRM consistency strengthens the
relationship between LMX and two work outcomes (engagement
and job performance) and that work engagement predicts job
performance (part of H1). We further expect that the interaction
of LMX and HRM consistency affects job performance through
employee work engagement. This is because both supervisor support and HRM offer both intrinsic and extrinsic motivation to
engage employees in their work, while at the same time, high HRM
consistency helps employees reciprocate supervisor behavior in the
form of superior performance. Therefore, we offer the following
hypotheses:
H2. HRM consistency strengthens the relationship between LMX
and work engagement.
H3. The relationship between the interaction of HRM consistency
and LMX on the one hand and job performance on the other hand
is mediated by work engagement.

3. Methods and data

3.1. Sample and procedures
The study sample included 326 employees and 54 of their immediate supervisors working in several revenue centers in a luxury
hotel in 2009. This privately owned hotel is located in Guangdong
province in Southern China. The establishments main customers
include conference/leisure groups, business travelers, and local
customers. About 30 per cent of customers are foreign visitors. The
hotel has slightly more than two hundred guestrooms and employs
over 750 staff. The ve key revenue centers are: guestrooms, food
and beverage (three restaurants), health center (including swimming and tennis), entertainment (e.g. karaoke), and meeting space
rental. The hotel managements mission is to provide the best
service to customers. Emphasis is placed on people-oriented
management to create a fair, harmonious and creative work environment (Hotel employee handbook).
Questionnaires were constructed for both employees and supervisors. The detailed items and scales are presented in the following
section. Following agreement from management regarding the
research, two sealed questionnaire packages (employee and supervisor surveys) were administrated to the employees and the
supervisors across the revenue centers. Each employee received a
sealed package including a cover letter, which stated that the survey was voluntary and independent of management, and assured
them condentiality; a return envelope; and a numbered employee
survey. Supervisors were asked to rate the performance of each
employee against the names provided.
Two hundred and ninety eight employees and 54 supervisors
returned completed surveys in a sealed envelope directly to one of
the researchers. The high response rate of 91 per cent for employees and 100 per cent for supervisors respectively is not uncommon
in Chinese surveys, particularly when management supports the
research (Cooke, 2009). The sample includes 161 (54 per cent)
female employees. The average age of employees is 21.5 (SD = 4.1)
years old and average tenure in the hotel is 9.47 months (SD = 9.4).
Among the 54 supervisors, 30 (55 per cent) are female. The average age is 26.5 (SD = 5.9) years old and average hotel tenure is 35.6
months (SD = 17.7).

1062

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

Table 1
Means, standard deviations, reliabilities, and correlations between variables.

Age (in years).

Workplace tenure (in months)
Gender
Education level
Employment status
HPWS index
Leadermember exchange
HRM consistency
Work engagement
Job performance

Mean

S.D.

21.48
9.47
.54
2.82
.65
4.25
3.96
1.95
3.83
4.70

4.13
9.42
.49
.84
.47
.70
1.10
1.16
.97
.75

.21**
.01
.03
.21**
.08
.11*
.04
.12*
.06

.01
.05
.33**
.06
.08
.05
.09
.11

.15**
.16**
.09
.04
.07
.09
.01

.45**
.11
.05
.01
.02
.05

.13*
.12*
.04
.07
.01

.48*
.19*
.35*
.02

.13*
.35**
.12*

.09
.01

.19**

10

Note: Employee N = 298; Job performance was rated by the supervisor; S.D. means standard deviation; Gender (0 = male; 1 = female); Employment status (0 = temporary;
1 = formal).
**
p < .01.
*
p < .05.

3.2. Measures
Previously applied, validated scales were used to measure the
different concepts included in our survey. The questionnaires were
administrated in Mandarin after being developed in English. Two
bilingual Chinese persons back-translated the two survey independently (Brislin, 1980). In addition, a 15 employee pilot study
(excluded from the main survey) was undertaken resulting in revisions to the phrasing of some questions. Response items ranged
from 1 = strongly disagree to 6 = strongly agree. The six-point Likert
scales were chosen as a way of addressing Chinese peoples tendency to conceal positive emotions and hence select midpoints of
a range (Lee et al., 2002).
Leadermember exchange (LMX). LMX was measured by LMSX,
a shortened four-item scale developed and validated by Bernerth
and colleagues (2007). Two example items are Voluntary actions
on my part will be returned in some way by my supervisor and If
I do something for my supervisor, s/he will eventually repay me.
The scale had high reliability (Cronbachs = .82).
Following past research (Sun et al., 2007; Li et al., 2011), we
measured the content of HR practices using the 15-item high
performance work practices (HPWS) scale used in a study of Chinese hotel employees (Li et al., 2011). The scale covered four HR
practices: training (four items); internal promotion (three items);
employee participation (four items); and performance-oriented
rewards (four items). Two sample items include HR practices here
help me a great deal to develop my knowledge and skills. and This
company prefers to promote from within the company. Each HR
practice demonstrated high reliability (Cronbachs alphas are .86,
.77, .78 and .76 respectively). HRM consistency is measured by the
inverse of the average deviation (within-respondent agreement)
of the above-mentioned four HR practices (Sanders et al., 2008; Li
et al., 2011). A higher score denotes higher HRM consistency.
Work engagement was measured on a ve-item work vigor scale
(Schaufeli and Bakker, 2004). Example items included At my work,
I feel bursting with energy and When I get up in the morning, I feel
like going to work. The reliability value is acceptable (Cronbachs
= .76).
Job performance was measured by asking supervisors to rate
employees performance on the nine items from Grifn et al.s

(2007) role-based performance scale. Three dimensions were evaluated: task prociency (the extent to which an individual meets
role requirements that can be formalized), task adaptability (the
extent to which an individual adapts to changes in a work system
or work roles) and task proactivity (the extent to which the individual takes self-directed action to anticipate or initiate change in
the work system or work roles). The reliabilities of the different
sub-scales were high (Cronbachs alphas are .88, .75, and .83). Sample items included Carried out the core parts of his/her job well
and Come up with ideas to improve the way in which his/her core
tasks are done. Following other studies, we computed an overall
job performance score for each employee as an average of the three
dimensions (Salanova et al., 2005; Zacher et al., 2010).
Based on the proposition that employees personal and employment characteristics are likely to inuence their work outcomes,
the following characteristics were included as control variables:
age, gender, type of labor contract (permanent vs. temporary), educational level (from junior middle school to masters degree and
above), and workplace tenure (in months). Furthermore, following previous studies (Beugelsdijk, 2008; Doellgast, 2008; Ramsay
et al., 2000; Sanders et al., 2008), we calculated the HPWS index
(the average of the four HR practices).
3.3. Analysis
Data for this study is hierarchical in nature because employees
worked under different supervisors and each supervisor evaluates the job performance of ve to six employees. Therefore, we
used hierarchical linear modeling (HLM; Raudenbush and Bryk,
2002) when testing hypotheses related to job performance (H1 and
H3). HLM enabled simultaneous control of the effects of withinand between supervisor-level data (Raudenbush and Bryk, 2002).
The procedure has two steps. First, for each supervisor (level 2),
relationships at level 1 were estimated among level 1 variables,
analogous to ordinary least squares regression analysis at a single
level. Then, parameter estimates (intercepts for each supervisor)
generated from the rst step were used as outcome variables in
level 2 in order to examine supervisor-level effects. The Log likelihood statistics is a model t index similar to chi-square statistics.
Like R square in regression analysis, pseudo R squares are reported

HRM consistency

H2
Leader-member exchange

H1

Work engagement

H1 (H3)

Job performance

Fig. 1. Theoretical model showing the expected inuence of LMX, work engagement, and HRM consistency on job performance.

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

to demonstrate the variance explained in the dependent variable

by the predictors.
H1 suggests that LMX is related to job performance through
work engagement. To test this hypothesis we followed the procedure outlined by Baron and Kenny (1986).1 This includes the Sobel
test (Sobel, 1982) in order to conrm that the effect of LMX on job
performance is decreasing signicantly when work engagement is
added. H2 anticipates that the relationship between LMX and work
engagement is stronger when employees perceive HRM as consistent. To test this hypothesis and to ensure that the interaction term
is not related to one of the main effects, we mean-centered LMX
and work engagement and add the main and interaction effects
in model 2, Table 2 (Aiken and West, 1991). In H3 we expect that
the interaction between LMX and HRM consistency is related to
job performance through work engagement. To test this hypothesis
we again follow the procedure specied by Baron and Kenny (1986:
1179) and examine if the effect of the interaction between LMX and
HRM consistency on job performance is signicantly decreasing
when work engagement is added (see Liao and Chuang, 2007).
4. Results
Table 1 reports the means, standard deviations, and relations
between all variables at the employee level. As shown in Table 1,
leader member exchange (LMX) and work engagement were positively related to job performance (r = .12, p < .05 and r = .19, p < .01
respectively). LMX was also positively related to engagement
(r = .35, p < .01).
Table 2 reports the regression analysis predicting work engagement and the HLM results predicting job performance. First we
calculated an empty model in which no predictors or controls
were added in order to assess the variance in job performance at
the employee level and supervisor level respectively. The results
showed that 11 per cent of the variance in job performance can
be attributed to the supervisor, which we controlled for. Because
age and gender have trivial (<.02) and non-signicant effects on job
performance, these variables were excluded from the analysis.
H1 predicts that work engagement mediates the relationship
between LMX and job performance. Following Baron and Kenny
(1986)s approach outlined above, we found an effect of LMX on
work engagement (b = .18, p < .01) in Model 1. Moreover, Model 1
showed that workplace tenure (b = .01, p < .05) and the HPWS index
(b = .28, p < .01) had signicant effects. Models 3 and 4 indicated that
both LMX (b = .11, p < .05) and work engagement (b = .17, p < .01) had
positive effects on job performance. In Model 5, the effects of LMX
on job performance became non-signicant (b = .08, p > .10) after
entering work engagement, which remained a signicant inuence (b = .15, p < .05). The Sobel test demonstrated that the indirect
effect of LMX on job performance was signicant (z = 2.84, p < .01).
Therefore, H1 is supported.
H2 proposed that the LMX work engagement relationship is
strengthened by HRM consistency. As illustrated in Model 2, the
interaction of LMX and HRM consistency had a positive effect on
work engagement (b = .12, p < .05). This interaction effect between
LMX and HRM consistency is depicted in Fig. 2. This gure shows
that when employees perceive HRM practices as highly consistent
with one another, the relationship between LMX and work engagement is stronger than in the case that employees perceive HRM
practices as low consistent. These ndings support H2.

1
A mediation effect is said to exist when the independent variable (IV: e.g. LMX)
signicantly predicts the dependent variable (DV: e.g. job performance); the IV
signicantly predicts the mediator variable (e.g. work engagement); the mediator
signicantly predicts the DV; and the effects of the IV on the DV becomes smaller
or non-signicant when controlling for the mediator.

1063

Fig. 2. Effect of the interaction between LMX and HRM consistency on work engagement.

H3 was examined in Models 6 and 7. Model 6 showed that the

interaction of LMX and HRM consistency had positive effects on
job performance (b = .12, p < .05). This means that a higher HRM
consistency enhances the positive relationship between LMX and
employee performance. In Model 7, after entering work engagement, the effect of the interaction of LMX and HRM consistency
decreases. Having shown that the interaction effects had a positive effect on work engagement (Model 2) and on job performance
(Model 6) and that work engagement predicted LMX (Model 4),
these results support H3. In sum, the three hypotheses were supported by the results.
5. Discussion
In this paper we have taken up the call by some LMX researchers
(Bauer et al., 2006; Huang et al., 2010) to investigate the mechanisms through which LMX affects employees job performance. We
proposed a model involving LMX, HRM consistency and employee
performance in a service context. Our rst nding is that LMX
was positively associated with work engagement and, through
work engagement, further enhances employee job performance
as reported by supervisors (H1). This conrms that work engagement is an additional mechanism that links LMX and employee job
performance.
We then examined whether HRM consistency plays a role in
the LMX-outcome relationships. The results indicated that HRM
consistency strengthens two relationships LMX work engagement (H2) and LMX job performance through work engagement
(H3) respectively. In short, when an employee perceives HR practices are highly consistent with each other, they are more likely to
reciprocate high LMX with stronger work enthusiasm and higher
job performance.
These results highlight HRM as an important contingent factor in the LMX outcome relationship. When HRM messages are
consistent with each other, employees are able to detect the management purposes underlying HR practices and more condently
make sense of organizational goals (Bowen and Ostroff, 2004). This
helps align employees responses with the organizations objectives expectations thereby motivating employees to respond to
high LMX with appropriate attitudes and behavior. This is consistent with recent empirical studies (see Sanders et al., 2008; Li et al.,
2011), earlier HPWS research (Delery and Doty, 1996; Macdufe,
1995) and social psychology studies (for example, cognitive dissonance theory, Festinger, 1957; attribution theory, Kelley, 1967).
Cognitive dissonance theory (Festinger, 1957) assumes that people
have a motivational drive to reduce dissonance by changing their
attitudes, beliefs or behavior. In the context of HRM this means
that employees are motivated to balance their perceptions of the
consistency of the HRM system with their attitudes and behavior.
Attribution theory (Kelley, 1967) assumes that people are strongly
motivated to understand their environment, including their own

1064

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

Table 2
Results of hypotheses testing relating to work engagement and job performance.
Models

Workplace tenure (in months)

Education level
Employment status
HPWS index
Leadermember exchange (LMX)
HRM consistency
LMX HRM consistency
Work engagement
(Pseudo) R2
(Pseudo) R2 Change
F Change or 2 Restricted Log Likelihood

Work engagement

Job performance

.04
.10
.05
.28**
.18**

.17
.02
7.0**

2
.04
.09
.04
.27**
.20**
.02
.12*
.19
.02
4.1*

4
.01*
.09
.08
.02
.11*

.04
.03
596.5

5
.01*
.08
.08
.02

.17**
.04
.03
591.9

6
.01*
.06
.07
.06
.08

.15*
.05
.01
590.8

7
.01*
.08
.07
.01
.14*
.01
.12*

.08
.03
602.1

.01*
.08
.06
.05
.09
.00
.10
.14*
.10
.02
601.7

Note: Employee N = 298; Job performance was rated by the supervisors; Standardized parameter estimates were given. Employment status (0 = temporary; 1 = formal).
*
p < .05.
**
p < .01.

and others behaviors. The co-variation principle of attribution theory suggests that if employees perceive their environment (HRM
in this case) as distinctive (it is visible) and consistent, and if other
employees perceive HRM in the same way (consensus) employees
will clearly understand management expectations and associated
rewards. Consequently, their job performance will increase and the
effects of LMX on work engagement and job performance will be
enhanced.
After controlling for HRM content, our results conrmed that
highly consistent HRM messages, measured by within-respondent
agreement, helped employees reciprocate to leaders in ways that
are aligned with the organizations purpose. This may shed light
on why some studies have found a non-signicant relationship
between LMX and employee performance. In short, by including
HRM consistency as a moderator it is possible to see why under
certain conditions (low consistency) the relationship between
LMX and performance is unlikely to be signicant for reasons
outlined earlier. Perceptions of inconsistent HRM practices and
consequent confusion about alignment of employees roles and the
organizations objectives is likely to increase employee stress and
dissatisfaction leading to reduced performance. This is indicated by
Tordera and colleagues (2008)s nding that individual employees
goal orientation strengthened the negative relationship between
LMX and role overload in 33 health care service organizations. In
other words, lacking a clear goal, employees are likely to be more
stressed, perform at a lower level and be more likely to quit (Cho
et al., 2006).
Following this line of reasoning, further research may usefully focus on other aspects of HRM or managerial behavior that
are able to align employees behavior with organizational objectives in studying LMX-relationship outcomes. Candidates include
the extent of consensus between HRM and line management on
organizational values and goals (Bowen and Ostroff, 2004) and jobdesign and feedback mechanisms that align employees interests
with organizational purposes (Gottschalg and Zollo, 2007).
Also noteworthy is the nding shown in Fig. 2 that when LMX
is low, employees perform better when HRM consistency is low
than when it is high. A similar pattern exists when plotting the
joint effects of LMX and HRM consistency on job performance as
indicated in Model 6 of Table 2. One explanation is that the incongruence between the espoused values/goals of the organization
indicated by low HRM consistency, and the enacted values/goals
conveyed through LMX, is less salient when HRM consistency is
low. Employees are therefore less frustrated and their motivation to
work less impaired than if the organizations objectives were more
clearly understood and the incongruence between organizational
objectives and enacted goals more salient. Another possibility is

that when employees want to perform at a high level in response to

consistent HR practices and lack supervisor and co-worker support,
they are likely to feel unsatised and stressed. This may translate
into lower levels of engagement and performance. Clearly, further
research would benet by including these two variables as moderators in more complex models.
This study has two main limitations. First, although employee
performance was reported by supervisors, the relationship
between LMX and work engagement raises the issue of common
method variance. However, this is not a serious problem because
the two constructs are very different and the correlation between
them is acceptable (.35, p < .01). The second limitation is that our
results are based on a single hotel in China. This restricts generalizability and calls for further research in the hospitality sector, across
different hotel categories (3, 4 and 5-star) and in different cultural
contexts.
Our study has two important practical implications. First,
supervisors are encouraged to engage in social exchange
relationships with subordinates, particularly on-the-job coaching
and mentoring. This will enhance employees work motivation and
performance. Our ndings also suggest that supervisors communicate clearly and regularly to employees regarding organizational
strategies and objectives including the connection between individual roles and goals and these wider organizational features.
Employees need to be given opportunities to discuss and seek clarication. Such processes would help employees better understand
the organization and enable them to align personal goals with organizational objectives. Management expectations can be formalized
in periodic performance appraisals. The second implication is that
it is important for management to execute HR practices and initiatives in a consistent way so that desired attitudes and behavior are
clearly conveyed across the organization. High HRM consistency
relies on many factors including commitment to key values, unambiguous policies, and consistent HR practice implementation by HR
and line management (Sanders et al., 2008).

6. Conclusion
Our study addressed the challenge of extending LMX explanations of job performance by focusing on mediating and moderating
variables in a hospitality setting. LMX was found to inuence
job performance via work engagement while consistency in HRM
communications encouraged employees to respond to favorable LMX relations with increased work engagement and higher
performance.

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

This contribution suggests three directions for future research.

First, it is unclear what aspects of LMX are most important for
employees. Is supervisor contribution more important than respect
as a professional or loyalty to the person? Application of a multidimensional rather than a global measure of LMX would help to
answer this question. Second, many organizations seek to develop
a unique culture that is intended to encourage particular attitudes
and behaviors. Accordingly, future research should measure this
organizational attribute and examine how it strengthens or weakens employee engagement and performance. Finally, rather than
seeking performance assessments from supervisors, research in
service organizations should consider measuring individual performance from the standpoint of service recipients, i.e. from internal
or external customers depending on the individuals organizational
position.

References
Aiken, L.S., West, S.G., 1991. Multiple Regressions: Testing and Interpreting Interactions. Sage Publications, Newbury Park, CA.
Alleyne, P., Doherty, L., Greenidge, D., 2006. Human resource management and performance in the Barbados hotel industry. International Journal of Hospitality
Management 25, 623646.
Bakker, A.B., Demerouti, E., 2007. Using the job demands resources model to predict
burnout and performance. Human Resource Management 43, 83104.
Bakker, A.B., Hakanen, J.J., Demerouti, E., Xanthopoulou, D., 2007. Job resources
boost work engagement, particularly when job demands are high. Journal of
Educational Psychology 99, 274284.
Bakker, A.B., Leiter, M.P. (Eds.), 2010. Work Engagement: A Handbook of Essential
Theory and Research. Psychology Press, New York.
Baron, R.M., Kenny, D.A., 1986. The moderator-mediator variable distinction in social
psychological research: conceptual, strategic and statistical consideration. Journal of Personality and Social Psychology 51, 11731182.
Bauer, T.N., Erdogan, B., Liden, R.C., Wayne, S.J., 2006. A longitudinal study of the
moderating role of extraversion: leadermember exchange, performance, and
turnover during new executive development. Journal of Applied Psychology 91,
298310.
Bernerth, J.B., Armanakis, A.A., Feild, H.S., Giles, W.F., Walker, H.J., 2007.
Leadermember social exchange (LMSX): development and validation of a scale.
Journal of Organizational Behavior 28, 9791003.
Beugelsdijk, S., 2008. Strategic human resource practices and product innovation.
Organization Studies 29, 821847.
Blau, P., 1964. Exchange and Power in Social Life. Wiley, New York.
Bowen, D.E., Ostroff, C., 2004. Understanding HRM-rm performance linkages: the
role of the strength of the HRM system. Academy of Management Review 29,
203221.
Brief, A.P., Aldag, R.J., 1977. The intrinsicextrinsic dichotomy: toward conceptual
clarity. Academy of Management Review July, 496500.
Brislin, R.W., 1980. Translation and content analysis of oral and written material. In:
Triandis, H.C., Berry, J.W. (Eds.), Handbook of Cross-cultural Psychology. Allyn &
Bacon, Boston, pp. 349444.
Cho, S., Woods, R.W., Jang, S., Erdem, M., 2006. Measuring the impact of human
resource management practices on hospitality rms performances. International Journal of Hospitality Management 25, 262277.
Cohen, S., Wills, T.A., 1985. Stress, social support and the buffering hypothesis. Psychological Bulletin 98, 310357.
Combs, J., Liu, Y., Hall, A., Ketchen, D., 2006. How much do high-performance work
practices matter? A meta-analysis of their effects on organizational performance. Personnel Psychology 59, 501528.
Cooke, F.L., 2009. A decade of transformation of HRM in China. A review of literature
and suggestions for future studies. Asia Pacic Journal of Human Resources 47,
640.
Deci, E.L., Ryan, R.M., 1985. Intrinsic Motivation and Self-determination in Human
Behavior. Plenum, New York.
Delery, J.E., 1998. Issues of t in strategic human resource management: Implications
for research. Human Resource Management Review 8, 289309.
Delery, J.E., Doty, H.K., 1996. Modes of theorizing in strategic human resource management: test of universalistic, contingent and congurational performance.
Academy of Management Journal 39, 802835.
Doellgast, V., 2008. Collective bargaining and high-involvement management in
comparative perspective: evidence from U.S. and German call centers. Industrial
Relations 47, 284319.
Dunegan, K.J., Duchon, D., Uhl-Bien, M., 1992. Examining the link between leader
member exchange and subordinate performance. The role of task analyzability
and variety as moderators. Journal of Management 18, 5976.
Dunegan, K.J., Uhl-Bien, M., Duchon, D., 2002. LMX and subordinate performance:
the moderating effects of task characteristics. Journal of Business and Psychology
17, 275285.
Enz, C.A., Siguaw, J.A., 2000. Best practices in human resources. Cornell Hotel and
Restaurant Administration Quarterly 41, 4861.

1065

Erdogan, B., Enders, J., 2007. Support from the top: supervisors perceived organizational support as a moderator of leadermember exchange to satisfaction and
performance relationships. Journal of Applied Psychology 92, 321330.
Fantazy, K.A., Kumar, V., Kumar, U., 2010. Supply management practices and performance in the Canadian hospitality industry. International Journal of Hospitality
Management 29, 685693.
Farh, J.L., Cheng, B.S., Chou, L.F., Chu, X.P., 2006. Authority and benevolence: employees responses to paternalistic leadership in China. In: Tsui, A.S., Bian, Y., Cheng,
L. (Eds.), China Domestic Private Firms: Multidisciplinary Perspectives on Management and Performance. Sharpe, New York, pp. 230260.
Festinger, L., 1957. A Theory of Cognitive Dissonance. Stanford University Press,
Stanford, CA.
Gerstner, C.R., Day, D.V., 1997. Meta-analytic review of leadermember exchange
theory: correlates and construct issues. Journal of Applied Psychology 82,
827844.
Gottschalg, O., Zollo, M., 2007. Interest alignment and competitive advantage.
Academy of Management Review 32, 418437.
Graen, G.B., Scandura, T.A., 1987. Toward a psychology of dyadic organizing.
Research in Organizational Behavior 9, 175208.
Graen, G.B., Uhl-Bien, M., 1995. Relationship-based approach to leadership: development of leadermember exchange (LMX) theory of leadership over 25
years: applying a multi-level multi-domain perspective. Leadership Quarterly
6, 219247.
Grifn, M.A., Neal, A., Parker, S.K., 2007. A new model of work role performance:
positive behavior in uncertain and interdependent contexts. Academy of Management Journal 50, 327347.
Guest, D.E., 2007. HRM and the worker: towards a new psychological contract? In:
Boxall, P., Purcell, J., Wright, P. (Eds.), The Oxford Handbook of Human Resource
Management. Oxford University Press, New York, pp. 129146.
Harris, K.J., Harris, R.B., Brouer, R.L., 2009a. LMX and subordinate political skill: direct
and interactive effects on turnover intentions and job satisfaction. Journal of
Applied Social Psychology 39, 23732395.
Harris, K.J., Wheeler, A.R., Kacmar, K.M., 2009b. Leadermember exchange
and empowerment: direct and interactive effects on job satisfaction,
turnover intentions, and performance. The Leadership Quarterly 20,
371382.
Hofmann, D.A., Morgeson, F.P., Gerras, S.J., 2003. Climate as a moderator of the
relationship between leadermember exchange and content specic citizenship: safety climate as an exemplar. Journal of Applied Psychology 88, 170178.
House, J.S., 1981. Work, Stress, and Social Support. Addision-Wesley, MA.
Huang, X., Chan, S.C.H., Lam, W., Nan, X., 2010. The joint effect of leadermember
exchange and emotional intelligence on burnout and work performance in call
centers in China. The International Journal of Human Resource Management 21,
11241144.
Ilies, R., Nahrgang, J.D., Morgeson, F.P., 2007. Leadermember exchange and citizenship behaviors: a meta-analysis. Journal of Applied Psychology 92, 269277.
Janssen, O., Van Yperen, N., 2004. Employees goal orientations, the quality of
leadermember exchange, and the outcomes of job performance and job satisfaction. Academy of Management Journal 47, 368384.
Kassin, S.M., Pryor, J.B., 1985. The development of attribution processes. In: Pryor,
J.B., Day, J.D. (Eds.), The Development of Social Cognition. Springer-Verlag, New
York, pp. 334.
Kelley, H.H., 1967. Attribution theory in social psychology. In: Levine, D. (Ed.),
Nebraska Symposium on Motivation. University of Nebraska Press, Lincoln, pp.
192240.
Klein, H.J., Kim, J.S., 1998. A eld study of the Inuence of situational constraints,
leadermember exchange, and goal commitment on performance. The Academy
of Management Journal 41, 8895.
Law, K.S., Wang, H., Hui, C., 2010. Currencies of exchange and global LMX: how
they affect employee task performance and extra-role performance. Asia Pacic
Journal of Management 27, 625646.
Lengnick-Hall, L.M., Lengnick-Hall, A.C., 2003. HRs role in building relationship networks. The Academy of Management Executive 17, 5363.
Lee, J.W., Jones, P.S., Mineyama, Y., Zhang, X.E., 2002. Cultural differences in responses to a Likert scale. Research in Nursing & Health 25,
295306.
Leiter, M.P., 2010. A meta-analysis of work engagement: relationships with burnout,
demands, resources, and consequences. In: Halbesleben, J.R.B., Bakker, A.B.
(Eds.), Work Engagement: A Handbook of Essential Theory and Research. Psychology Press, New York, pp. 102117.
Li, X., Frenkel, S., Sanders, K., 2011. Strategic HRM as process: how HR system and
organizational climate strength inuence Chinese employee attitudes. International Journal of Human Resource Management 22, 18231840.
Liao, H., Chuang, A., 2007. Transforming service employees and climate: a multilevel,
multisource examination of transformational leadership in building long-term
service relationships. Journal of Applied Psychology 92, 10061019.
Liden, R.C., Sparrow, R.T., Wayne, S.J., 1997. LeaderMember Exchange Theory: The
Past and Potential for the Future. JAI Press.
Liden, R.C., Wayne, S.J., Stilwell, D., 1993. A longitudinal study on the early development of leadermember exchanges. Journal of Applied Psychology 78, 662674.
Macdufe, J.P., 1995. Human resource bundles and manufacturing performance:
organizational logic and exible production systems in the world auto industry.
Industrial & Labor Relations Review 48, 197221.
Moss, S.E., Sanchez, J.I., Brumbaugh, A.M., Borkowski, N., 2009. The mediating role
of feedback avoidance behavior in the LMX-performance relationship. Group &
Organization Management 34, 645664.

1066

X. Li et al. / International Journal of Hospitality Management 31 (2012) 10591066

Nishii, L.H., Lepak, D.P., Schneider, B., 2008. Employee attributions of the why of
HR practices: their effects on employee attitudes and behaviors, and customer
satisfaction. Personnel Psychology 61, 503545.
Ramsay, H., Scholarios, D., Harley, B., 2000. Employees and high-performance work
systems: testing inside the black box. British Journal of Industrial Relations 38,
501531.
Raudenbush, S.W., Bryk, A.S., 2002. Hierarchical Linear Models, 2nd ed. Sage Publications, Thousand Oaks.
Salanova, M., Agut, S., Peir, J.M., 2005. Linking organizational resources
and work engagement to employee performance and customer loyalty:
the mediation of service climate. Journal of Applied Psychology 90,
12171227.
Sanders, Dorenbosch, K., Reuver, L., R. de., 2008. The impact of individual and shared
employee perceptions of hrm on affective commitment: considering climate
strength. Personal Review 37, 412415.
Sanders, K., Frenkel, S., 2011. HR-line management relations: characteristics
and effects. International Journal of Human Resource Management 22 (8),
16111617.
Scandura, T.A., Pellegrini, E.P., 2008. Trust and leader member exchange: a closer
look at relational vulnerability. Journal of Leadership & Organizational Studies
15, 101111.
Schaufeli, W.B., Bakker, A.B., 2004. Job demands, job resources, and their relationship
with burnout and engagement: a multi-sample study. Journal of Organizational
Behavior 25, 293315.
Schaufeli, W.B., Salanova, M., Gonzalez-Roma, V., Bakker, A.B., 2002. The measurement of engagement and burn out: a conrmative analytic approach. Journal of
Happiness Studies 3, 7193.
Schneider, B., Bowen, D.E., 2010. Winning the Service Game. Harvard Business School
Press, Boston, MA.
Schriesheim, C.A., Castro, S.L., Cogliser, C.C., 1999. Leadermember exchange (LMX)
research: a comprehensive review of theory, measurement, and data-analytic
practices. The Leadership Quarterly 10, 63113.
Sobel, M.E., 1982. Asymptotic condence intervals for indirect effects in structural
equation models. In: Leinhardt, S. (Ed.), Sociological Methodology. San Francisco,
Jossey-Bass.

Sun, L.-Y., Aryee, S., Law, K.S., 2007. High Performance human resource practices,
citizenship behaviour, and organisational performance: a relational perspective.
Academy of Management Journal 50, 577588.
Tekleab, A.G., Taylor, M.S., 2003. Arent there two parties in an employment relationship? Antecedents and consequences of organization-employee agreement
on contract obligations and violations. Journal of Organizational Behavior 24,
585608.
Tordera, N., Gonzalez-Roma, V., Peiro, J.M., 2008. The moderator effect of psychological climate on the relationship between leadermember exchange (LMX) quality
and role overload. European Journal of Work and Organizational Psychology 17,
5572.
Uhl-Bien, M., Graen, G.B., Scandura, T., 2000. Implications of leadermember
exchange LMX for strategic human resource management systems: relationships as social capital for competitive advantage. In: Ferris, G. (Ed.), Research
in Personnel and Human Resource Management. JAI Press, Greenwich, CT, pp.
137185.
Vaux, A., 1988. Social Support: Theory, Research and Intervention. Praeger, New
York.
Wang, C.-H., Chen, K.-Y., Chen, S.-C., 2012. Total quality management, market orientation and hotel performance: the moderating effects of external environmental
factors. International Journal of Hospitality Management 31 (1), 119129.
Wat, D., Shaffer, M.A., 2005. Equity and relationship quality inuences on organizational citizenship behaviors: the mediating role of trust in the supervisor and
empowerment. Personnel Review 34, 406422.
Wayne, S.J., Ferris, G.R., 1990. Inuence tactics, affect, and exchange quality in
supervisor-subordinate interactions: a laboratory experiment and eld study.
Journal of Applied Psychology 75, 487499.
Wayne, S.J., Green, S.A., 1993. The effects of leadermember exchange on employee
citizenship and impression management behavior. Human Relations 46,
14311440.
Zacher, H., Heusner, S., Schmitz, M., Zwierzanska, M.M., Frese, M., 2010. Focus on
opportunities as a mediator of the relationships between age, job complexity,
and work performance. Journal of Vocational Behavior 76, 374386.