Upload
others
View
8
Download
0
Embed Size (px)
Citation preview
Three Essays on Strategic Human Resource Management
by
Xiaoyu Huang
A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy
Centre for Industrial Relations and Human Resources
University of Toronto
© Copyright by Xiaoyu Huang 2016
ii
Three Essays on Strategic Human Resource Management
Xiaoyu Huang
Doctor of Philosophy
Centre for Industrial Relations and Human Resources
University of Toronto
2016
Abstract
This thesis examines three key aspects of strategic human resource management (HRM).
Chapter 1 uses an eight-year longitudinal survey to investigate how patterns of change in high-
performance work systems (HPWS) relate to innovation and financial performance of
organizations. Results suggest that long-run consistency with continuous incremental change in
training and recruitment systems is positively related to organizations’ innovation and financial
performance. Conversely, short-run consistency with episodic change in compensation and
employee involvement systems is positively associated with both measures of performance.
Drawing on institutional theory and the contingency perspective of HRM, Chapter 2 uses
hierarchical linear model to study how multilevel environmental and organizational factors –
international competitive pressure, capital intensity, firm size, unionization and ownership –
influence firms’ use of short-term labor contracts, contract duration, training, and employee
involvement programs. Using a survey of 313 manufacturing plants in China, this paper finds that
iii
approximately 5 to 7 percent of the total variance in the four HRM programs are explained by
industry-level factor and provides empirical support for mimetic isomorphism.
Chapter 3 focuses on HR differentiation which refers to the practice of managing
individuals or groups of employees differently based on the value they deliver to an organization
using individual-based, workforce-based or job-based approaches. This paper suggests that HR
differentiation is a source of firm's competitive advantage. Findings supports that the hypothesized
causal chain linking HR differentiation, strategic performance of HR system, firm strategic
performance and financial performance.
iv
Acknowledgments
I would like to express my gratitude to Professor Anil Verma, my advisor and the chair of
my doctoral dissertation committee. His wisdom and commitment to my success as a researcher
and a teacher has been a most valued component of my graduate experience. It has been a great
honour to be his doctoral student. I would like to thank the rest of my thesis committee: Professor
Rafael Gomez and Professor Soo Min Toh for their valuable guidance. My sincere thanks goes to
Professor Morley Gunderson for his continuous support and to Professor Michele Campolieti for
his insightful suggestions. I also want to thank Professor Kaifeng Jiang for his help and valuable
advice. I sincerely thank Anil, Rafael, Soo Min, Kaifeng and Morley who have written numerous
recommendation letters for me. I could not have secured my doctoral scholarship from the Social
Sciences and Humanities Research Council (SSHRC) or my teaching opportunities without your
help.
I want to thank the capable administrative and library staff at the Centre for Industrial
Relations and Human Resources, Deborah Campbell, Carol Canzano-Hamala, Monica Hypher,
Vicki Skelton, and Michelle Petersen-Lee, as well as the Statistics Canada analysts at the Toronto
Research Data Center, Joanna Jacob and Carmina Ng, for their exceptional administrative and
research support. I thank my fellow students for their comments during seminars and most of all
for their friendship: Jing Wang, Lydia He, Tingting Zhang, Umar Boodoo, Bruce Curran, Alana
v
Arshoff, Rachel Aleks, Joanna Pitek, Tina Saksida, Elham Marzi, Amal Radie, Yao, Guenther
Lomas, Peter Bouris and Melissa Wawrzkiewicz.
I dedicate this thesis to my parents, Lihua Zhang and Xinzhong Huang, who encouraged
me to pursue this doctoral degree and offered their constant love and support. Mum, you have
showed me how learning and hard work can significantly transform one’s life path and help make
a contribution to society. I thank you for being the greatest inspiration in my life. To my
grandmother, Sufen Jiang, and my aunt, Chunmin Huang, thank you for your encouragement and
love. To my cousins, Rui Huang, Yushu Huang, Xigao Yuan, and Xuewei Chong, thank you for
believing in your big sister.
To my dear husband, Yu Hou, my thesis and my life would not be complete without you.
Lastly, I would like to thank the SSHRC for its generous financial support of this research.
vi
To my husband
and my parents,
...
for all their love and support over all these years.
vii
Table of Contents
ABSTRACT ................................................................................................................................... ii
ACKNOWLEDGMENTS ........................................................................................................... iv
TABLE OF CONTENTS ........................................................................................................... vii
LIST OF TABLES ....................................................................................................................... xi
LIST OF FIGURES .................................................................................................................... xii
CHAPTER 1 DO CHANGES IN HIGH-PERFORMANCE WORK SYSTEMS PAY
OFF? A LONGITUDINAL INVESTIGATION OF DYNAMIC FIT ................................ 1
ABSTRACT .................................................................................................................................... 1
1 INTRODUCTION...................................................................................................................... 2
2 THEORY AND HYPOTHESES ............................................................................................... 5
2.1 Conceptual Framework on Dynamic HR Fit ...................................................................... 5
2.2 Strategic HR Change as Beneficial Flexibility ................................................................... 7
2.3 Strategic HR Change as Deviation from Beneficial Stability ............................................. 9
2.4 Beneficial Strategic HR Change due to Increase in HPWS .............................................. 11
3 METHODS .............................................................................................................................. 13
3.1 Sample ............................................................................................................................... 13
3.2 Measures ........................................................................................................................... 14
3.2.1 Dependent Variables ............................................................................................. 14
3.2.2 Independent Variables .......................................................................................... 15
3.2.3 Control Variables .................................................................................................. 17
3.3 Empirical Analysis ............................................................................................................ 18
4 RESULTS ................................................................................................................................ 19
viii
4.1 Multivariate Analyses ....................................................................................................... 19
4.2 Additional Analyses and Robustness Check ..................................................................... 24
5 DISCUSSION .......................................................................................................................... 24
5.1 Overview and Contributions ............................................................................................. 24
5.2 Limitations and Suggestions for Future Research ............................................................ 28
6 CONCLUSIONS ...................................................................................................................... 30
REFERENCES ............................................................................................................................. 32
TABLE 1. A Dynamic Model of HR Fit and Misfit ..................................................................... 39
TABLE 2. Descriptive Statistics .................................................................................................. 40
TABLE 3. Patterns of Strategic HR Change as Predictors Long-run Average (LRA) Innovation
and Profitability Levels ........................................................................................................... 41
TABLE 4. AMO Dimensions of HPWS as Predictors Innovation and Profitability .................... 44
CHAPTER 2 INDUSTRY AND FIRM-LEVEL DETERMINANTS OF EMPLOYMENT
RELATIONS IN CHINA: A TWO-LEVEL ANALYSIS .................................................. 45
ABSTRACT .................................................................................................................................. 45
1 INTRODUCTION.................................................................................................................... 46
2 THEORY AND HYPOTHESES ............................................................................................. 52
2.1 International Competitive Pressure ................................................................................... 55
2.2 Capital Intensity ................................................................................................................ 57
2.3 Firm Size ........................................................................................................................... 59
2.4 Unionization ...................................................................................................................... 61
2.5 Ownership ......................................................................................................................... 63
3 METHODS .............................................................................................................................. 65
3.1 Data ................................................................................................................................... 65
ix
3.2 Measures ........................................................................................................................... 66
4 RESULTS ................................................................................................................................ 69
4.1 Descriptive Statistics and Bivariate Correlations ............................................................. 69
4.2 Hierarchical Linear Model Analyses ................................................................................ 70
5 LIMITATIONS ........................................................................................................................ 73
6 CONCLUSIONS ...................................................................................................................... 74
REFERENCES ............................................................................................................................. 79
TABLE 1. Descriptive Statistics and Bivariate Correlations for Variables .................................. 83
TABLE 2. Multilevel Predictors of Percentage of Short-Term Labor Contracts ......................... 85
TABLE 3. Multilevel Predictors of Weighted Average Labor Contract Duration ....................... 86
TABLE 4. Multilevel Predictors of Training Programs ............................................................... 87
TABLE 5. Multilevel Predictors of Employee Involvement Programs ........................................ 88
TABLE 6. Summary of Hypotheses and Findings ....................................................................... 89
CHAPTER 3 THE IMPACT OF HUMAN RESOURCE MANAGEMENT
DIFFERENTIATION ON CORPORATE STRATEGIC AND FINANCIAL
PERFORMANCE .................................................................................................................. 90
ABSTRACT .................................................................................................................................. 90
1 INTRODUCTION.................................................................................................................... 91
2 THEORY AND HYPOTHESES ............................................................................................. 94
2.1 Strategic Performance of HR System ............................................................................... 97
2.2 Moderators of HR Differentiation .................................................................................... 99
2.3 Firm Strategic Performance ............................................................................................ 102
2.4 Firm Financial Performance ........................................................................................... 102
3 METHODS ............................................................................................................................ 103
x
3.1 Sample ............................................................................................................................. 103
3.2 Measures ......................................................................................................................... 104
3.2.1 Independent Variables ........................................................................................ 104
3.2.2 Mediators and Dependent Variables ................................................................... 105
3.2.3 Moderators .......................................................................................................... 105
3.2.4 Control Variables ................................................................................................ 106
4 Results .................................................................................................................................... 107
4.1 Descriptive Statistics and Bivariate Correlations ........................................................... 107
4.2 Multivariate Analyses ..................................................................................................... 107
5 DISCUSSION AND CONCLUSIONS.................................................................................. 108
REFERENCES ........................................................................................................................... 112
TABLE 1. Descriptive Statistics and Bivariate Correlations for Variables ............................... 117
TABLE 2. Predictors of Strategic Performance of HR System .................................................. 119
TABLE 3. Predictors of Firm Strategic Performance ................................................................. 120
TABLE 4. Predictors of ROI ...................................................................................................... 121
TABLE 5. Summary of Hypotheses and Findings ..................................................................... 122
FIGURE 1. A Conceptual Framework on the Strategic and Financial Impact of HR
Differentiation ........................................................................................................................ 123
FIGURE 2a. The Moderation of HPWPs on the Relationship between HR Differentiation and
Strategic Performance of HR System .................................................................................... 123
FIGURE 2b. The Moderation of Firm Size on the Relationship between HR Differentiation
and Strategic Performance of HR System .............................................................................. 124
FIGURE 2c. The Moderation of Environmental Dynamism on the Relationship between HR
Differentiation and Strategic Performance of HR System ..................................................... 124
xi
List of Tables
CHAPTER 1 DO CHANGES IN HIGH-PERFORMANCE WORK SYSTEMS PAY
OFF? A LONGITUDINAL INVESTIGATION OF DYNAMIC FIT ................................ 1
TABLE 1. A Dynamic Model of HR Fit and Misfit ..................................................................... 39
TABLE 2. Descriptive Statistics .................................................................................................. 40
TABLE 3. Patterns of Strategic HR Change as Predictors Long-run Average (LRA) Innovation
and Profitability Levels ........................................................................................................... 41
TABLE 4. AMO Dimensions of HPWS as Predictors Innovation and Profitability .................... 44
CHAPTER 2 INDUSTRY AND FIRM-LEVEL DETERMINANTS OF EMPLOYMENT
RELATIONS IN CHINA: A TWO-LEVEL ANALYSIS .................................................. 45
TABLE 1. Descriptive Statistics and Bivariate Correlations for Variables .................................. 83
TABLE 2. Multilevel Predictors of Percentage of Short-Term Labor Contracts ......................... 85
TABLE 3. Multilevel Predictors of Weighted Average Labor Contract Duration ....................... 86
TABLE 4. Multilevel Predictors of Training Programs ............................................................... 87
TABLE 5. Multilevel Predictors of Employee Involvement Programs ........................................ 88
TABLE 6. Summary of Hypotheses and Findings ....................................................................... 89
CHAPTER 3 THE IMPACT OF HUMAN RESOURCE MANAGEMENT
DIFFERENTIATION ON CORPORATE STRATEGIC AND FINANCIAL
PERFORMANCE .................................................................................................................. 90
TABLE 1. Descriptive Statistics and Bivariate Correlations for Variables ............................... 117
TABLE 2. Predictors of Strategic Performance of HR System .................................................. 119
TABLE 3. Predictors of Firm Strategic Performance ................................................................. 120
TABLE 4. Predictors of ROI ...................................................................................................... 121
TABLE 5. Summary of Hypotheses and Findings ..................................................................... 122
xii
List of Figures
CHAPTER 3 THE IMPACT OF HUMAN RESOURCE MANAGEMENT
DIFFERENTIATION ON CORPORATE STRATEGIC AND FINANCIAL
PERFORMANCE .................................................................................................................. 90
FIGURE 1. A Conceptual Framework on the Strategic and Financial Impact of HR
Differentiation ........................................................................................................................ 123
FIGURE 2a. The Moderation of HPWPs on the Relationship between HR Differentiation and
Strategic Performance of HR System .................................................................................... 123
FIGURE 2b. The Moderation of Firm Size on the Relationship between HR Differentiation
and Strategic Performance of HR System .............................................................................. 124
FIGURE 2c. The Moderation of Environmental Dynamism on the Relationship between HR
Differentiation and Strategic Performance of HR System ..................................................... 124
1
CHAPTER 1 DO CHANGES IN HIGH-PERFORMANCE WORK SYSTEMS PAY
OFF? A LONGITUDINAL INVESTIGATION OF DYNAMIC FIT
ABSTRACT
Using an eight-year longitudinal survey, this study investigates the stability–change paradox in
human resource (HR) systems by examining how patterns of change in high-performance work
systems (HPWS) relate to innovation and financial performance of organizations. The contingency
perspective suggests that such change constitutes beneficial flexibility because changes in aspects
of HPWS are required to attain dynamic fit. By contrast, the universalistic perspective and
organizational ambidexterity suggest that HPWS provides both efficiency and flexibility, which
indicates beneficial stability. An exploratory analysis supports both theoretical perspectives and
reveals a positive relationship between two distinct patterns of change in the ability-motivation-
opportunity dimensions of HPWS and performance outcomes. Long-run consistency in the ability-
enhancing dimension (i.e., training and recruitment systems) with continuous incremental change
is positively associated with high performance. Conversely, short-run stability with episodic
change in the motivation- and opportunity-enhancing dimensions of HPWS (i.e., compensation
and employee involvement systems) is positively related to performance. The findings suggest that
organizations can benefit from both stability and flexibility in HR systems by appropriately
emphasizing long-run adaptability in the ability dimension and short-run adaptation in the
motivation and opportunity dimensions of HPWS.
2
1 INTRODUCTION
This paper re-examines Thompson’s (1967) “paradox of administration” between
efficiency and flexibility in the context of strategic human resource management (HRM). Two
competing theoretical perspectives suggest, paradoxically, that stability and change in human
resource (HR) systems can both be beneficial. On the one hand, organizational theory suggests that
bureaucracy is required to achieve efficiency but that it can hinder flexibility. Prior micro-level HR
studies demonstrate that there are meta-routines (i.e., routines for changing routines) that allow
organizations to achieve ambidexterity (Adler, Goldoftas, & Levine, 1999), supporting the
universalistic view (e.g., Huselid, 1995; Delery & Doty, 1996) that organizations can achieve both
efficiency and flexibility by adopting and maintaining high-performance work systems (HPWS)
(Patel, Messersmith, & Lepak, 2013). Therefore, stability in HPWS is sufficient for a positive HR
impact on performance. On the other hand, the contingency perspective (e.g., Miles & Snow, 1984;
Schuler & Jackson, 1987) suggests that a dynamic competitive environment demands sufficient
flexibility in HR systems; thus, organizations must strategically change aspects in their HR system
to achieve dynamic fit, which leads to superior performance (Wright & Snell, 1998). These two
competing theoretical predictions suggest a stability–change paradox in HR systems and raise
important questions – do changes in HR systems pay off, and if so, why? Additionally, if such
changes are beneficial, how should change be implemented to best enhance performance?
In an effort to reconcile the stability-change paradox in HR systems, the current paper
introduces a new construct, strategic HR change, and clarifies its influence on performance
3
outcomes. In particular, I define strategic HR change as change in the pattern of HR activities that
are intended to attain the dynamic goals of organizations. In contrast to the theoretical construct
known as flexibility of HR practices (hereinafter, HR flexibility), which refers to HR practices that
can promote organizational flexibility (Chang, Gong, Way, & Jia, 2013; Wright & Snell, 1998),
strategic HR change is an objective description of the actual changes in HR systems, which may
or may not have a positive impact on organizational performance. Unlike endogenous dynamic
organizational routines (Feldman & Pentland, 2003), changes in an HR system frequently require
a top-down approach that involves formal administrative interventions. Although some scholars
have noted that, HR systems are “intractable” (Cappelli & Neumark, 2001; Gerhart & Milkovich,
1990; Snell & Dean, 1994), and changes are rare events (Ichniowski, Shaw, & Prennushi, 1997)
due to substantial structural inertia (Astley & Van de Ven, 1983; Hannan & Freeman, 1984) and
organizational barriers (Wright & Snell, 1998), recent empirical research has shown the variability
in HR practices over time (e.g., Kim & Ployhart, 2014; Piening, Baluch, & Salge, 2013; Wang &
Shyu, 2009).
This study aims to provide insights into the change in HR systems from at least two aspects.
First, to extend the knowledge regarding how stability and change in HR systems relate to
performance in practice. The present study applies Zajac, Kraatz and Bresser’s (2000) model of
dynamic strategic fit to HR systems to derive a conceptual framework emphasizing that dynamic
HR fit can assume the form of both beneficial stability and beneficial flexibility. To test the
predictions of the dynamic HR fit model, I use an eight-year longitudinal survey to examine how
4
(1) the frequency of change in HPWS, (2) long-term variability (i.e., mean adjusted total amount
of change), and (3) the direction of change in terms of increases or reductions in the number of
high-performance work practices relate to innovation and financial performance. These three
dynamic aspects jointly delineate the beneficial patterns of change in HPWS.
Second, this paper addresses some methodological limitations in prior empirical studies.
As Luce (1997) insightfully noted, although theories are developed in a within-subject sense,
empirical studies are frequently based only on between-subject comparisons, which has largely
been the case in the field of strategic HRM. Despite the established HR systems–performance link
shown in recent meta-analyses (Combs, Liu, Hall, & Ketchen, 2006; Jiang, Lepak, Hu, & Baer,
2012), the weak evidence for causal inference due to various methodological limitations is an
important critique, particularly the lack of genuinely longitudinal studies (e.g., Boxall & Purcell,
2000; Guest, 2011; Jackson, Schuler, & Jiang, 2014; Wall & Wood, 2005; Wright, Gardner,
Moynihan & Allen, 2005). This weak evidence is partly caused by the intractable nature of HR
systems, which makes it difficult for cross-sectional studies to rule out the possibility that the HR
systems–performance relationship merely reflects the pre-existing variations among organizations
(Cappelli & Neumark, 2001; Wright et al., 2005). This eight-year longitudinal study enhances the
causal arguments in prior studies using evidence from within-estimates. By using the fixed effect
model, I offer stronger causal evidence than between-estimates from cross-sectional studies
because this approach captures individual firms’ actual changes in HR systems over time and
adjusts the confounded and stable unmeasured differences.
5
Furthermore, the inherent difficulty in capturing actual change in HR systems has caused
empirical research on HR practice flexibility to generally lag behind its theoretical developments.
Although empirical studies on HR practice flexibility offer evidence for the theoretical predictions
about beneficial flexibility, given the positive connotation of the theoretical construct and common
method variance, as discussed by Gerhart, Wright, McMahan, and Snell (2000), there remains a
research gap with respect to the actual impact of strategic HR change on performance outcomes.
This paper complements prior empirical studies and addresses some of their methodological
limitations by using objective measures of change derived from eight-year longitudinal data. This
paper also departs from prior studies on HR practice flexibility by considering the potential
negative impacts of change and by providing estimates from actual HR change.
In sum, as noted by many scholars (e.g. Guest, 2011; Jackson et al., 2014), strategic HRM
research to date has mainly focused on searching for the best HR systems in a given context, and
additional research is required to uncover how HR systems evolve in the dynamic competitive
environment. To address this research need, this research joins other recent research to explore the
dynamic nature of HR systems. In the following, I will first introduce a conceptual framework and
then elaborate on two competing perspectives regarding the forms of dynamic HR fit: beneficial
flexibility versus beneficial stability.
2 THEORY AND HYPOTHESES
2.1 Conceptual Framework on Dynamic HR Fit
This paper applies Zajac et al.’s (2000) dynamic strategic fit model to HR systems to
6
conceptualize different performance outcomes for strategic HR change. Table 1 presents the
conceptual framework for four scenarios of strategic HR change. The dynamic fit of the HR system
adds the temporal dimension to the horizontal and vertical fit and leads to a better performance
outcome than the misfit situations. A few assumptions are in play. Consistent with Zajac et al.’s
(2000) strategy model, dynamic HR fit is time-specific and multi-dimensional (i.e., there are
multiple environmental and organizational contingencies that determine HR fit); however, there
are three key distinctions. First, given the intractable nature of HR systems, strategic HR change
is generally more focused on the long run than a change in business strategies (Klarner & Raisch,
2013). Second, as Becker and Gerhart (1996) noted, the universalistic perspective tends to hold at
sufficiently high levels of abstraction, but the contingency perspective is essential at the lower HR
practice level. This paper focuses on the HR system level and thus assumes that dynamic HR fit is
not necessarily organization-specific in the short run because the adoption of HPWS may satisfy
the static HR fit. Third, the dynamic HR fit follows the “equifinality” assumption that different
combinations of HR practices can be equally effective (Delery & Doty, 1996). Therefore,
consistent with the configurational perspective, dynamic HR fit emphasizes both vertical fit and
horizontal fit among HR practices.
The vertical dimension of the dynamic HR fit model captures the prescriptive question of
whether an HR system should change to achieve a dynamic HR fit; the horizontal dimension asks
the descriptive question whether change in the HR system actually occurred. The following
sections elaborate further on the change-stability theoretical paradox in the vertical dimension and
7
present the hypotheses.
2.2 Strategic HR Change as Beneficial Flexibility
The contingency perspective suggests that organizations gain efficiency by aligning HR
systems with internal and external contingencies. Adapting to the dynamic competitive
environment also demands sufficient flexibility in HR systems (Tracey, 2012; Way et al., 2015).
Flexibility in HR systems is essential for both innovation and financial performance because many
organizations operate in highly dynamic competitive environments. Environmental dynamism
influences dynamic HR fit; thus, strategic HR change for organizations in stable environments is
more long run oriented than for those in dynamic environments (Datta, Guthrie, & Wright, 2005).
When aspects of HR systems no longer meet the changing needs of organizations, strategic HR
change helps organizations achieve dynamic fit and add value by improving the efficacy and
efficiency of HRM. From a capability perspective (Teece, Pisano, & Shuen, 1997), HR practice
flexibility is a particular type of dynamic capability that creates value for organizations through
fast and effective change. The resource-based view further suggests that organizations create
inimitable and non-substitutable capabilities through HR practice flexibility and that these
capabilities contribute to firms’ competitive advantages (Bhattacharya, Gibson, & Doty, 2005).
Strategic HR change can also lead to improvements in the overall management quality in
dynamic environments. As Becker, Huselid and Beatty (2009) noted, although the impact of
HPWS on performance has repeatedly been shown to be positive, there remains substantial
8
variation across organizations based on differences in the “quality” of HPWS across organizations.
One reason for poor HPWS quality is that a previously effective HPWS no longer meets the
changing demands of organizations, and no strategic HR change is implemented to improve the
quality of the HPWS. For example, when employees do not take advantage of a particular
employee involvement program, replacing this program with another empowerment initiative can
increase participation in—and the overall effectiveness of—HPWS.
In addition to the theoretical basis of HR practice flexibility as a desirable organizational
characteristic, empirical evidence affirms the beneficial impact of strategic HR change. Way et al.
(2015) noted that five empirical studies present evidence that HR flexibility is related to customer
effectiveness, market performance (Ngo & Loi, 2008), and financial performance (Ketkar & Sett,
2009, 2010; Bhattacharya et al., 2005). In sum, current theory on HR practice flexibility and
empirical evidence lead to the following beneficial flexibility hypothesis, which predicts a positive
relationship between the variability in HPWS (i.e., the amount of strategic HR change over time)
and performance:
Hypothesis 1. Variability in HPWS over time is positively related to long-run
innovation and financial performance.
The performance outcomes of strategic HR change also depend on the frequency or speed
of change. A high strategic speed of HR change indicates an organization’s dynamic capability to
swiftly adjust aspects of its HR system to achieve dynamic fit, which enhances performance—
9
particularly for innovators in dynamic environments. Wright and Snell (1998) emphasized this
point in defining coordination flexibility in HR practices as “how quickly the practices can be
resynthesized, reconfigured, and redeployed.” These authors argued that organizations should
develop a feedback mechanism to allow timely information on the implementation of change. The
faster organizations can implement strategic HR change and use feedback to make effective
adjustments, the better the performance outcome will be (Wright & Snell, 1998). In a validated
scale of HR flexibility, this construct measures firms’ ability to “quickly and effectively”
implement strategic HR change in HPWS. Frequent change enhances performance by allowing
organizations to swiftly adjust and refine HR practices, which thus achieves dynamic fit in fast-
changing environments.
Furthermore, organizations accumulate experience in changing routines (Amburgey, Kelly,
& Barnett, 1993); therefore, a relatively higher frequency of change in HR systems allows
organizations to develop a greater capability regarding the successful initiation and
implementation of strategic HR change than those who rarely change. Such theoretical
perspectives on beneficial flexibility jointly suggest the following:
Hypothesis 2. The frequency of change in HPWS is positively related to long-run
innovation and financial performance.
2.3 Strategic HR Change as Deviation from Beneficial Stability
Although strategic HR changes aim to achieve an organization’s dynamic goals, such
10
attempts may not always lead to higher performance compared with remaining unchanged. There
are several theoretical reasons why stability in HR systems can be more beneficial than change.
Earlier organizational theory on routines (Thompson, 1967) suggests that organizations can
achieve efficiencies by utilizing bureaucracies with a high level of formalization and
standardization. One of the most efficient HRM approaches for modern organizations is to adopt
HPWS. The universalistic perspective shows that HPWS leads to superior performance across all
firms and across all environmental conditions (Huselid, 1995). Thus, regardless of environmental
or organizational dynamism, once organizations achieve their optimal HR configurations,
remaining with their HPWS yields a consistent and positive impact on performance unless radical
change is required. Similarly, from an economics perspective (Kaufman & Miller, 2011),
consistency in the HR system over the long run indicates that long-run equilibrium has been
achieved and is likely to represent optimal bundles because organizations continuously pursue
efficiency in HRM. Beneficial consistency does not suggest that organizations should not change
their HR system at all; instead, it emphasizes the benefits of long-run stability in the HR system.
One example of such HR change is to reduce training investment and divert the resources into
other less efficient areas such that the overall impact of change on performance is negative.
Furthermore, following the argument that the level of analysis distinguishes the universalistic and
contingency predictions (Becker & Gerhart, 1996), beneficial consistency is likely to hold at the
HR system or HR-architecture level.
Notably, a number of HR practices within the HPWS inherently induce flexibility such that
11
organizations can achieve dynamic fit without changing their HR systems. For example,
continuous improvement and total quality management are meta-routines that change routines in
organization (Hackman & Wageman, 1995; Adler et al., 1999). Meta-routines serve as the micro-
foundation of the inherent flexibility of HPWS. More recently, Patel et al. (2013) showed that
HPWS is positively related to organizational ambidexterity, which mediates the impact of HPWS
on firm growth (2013). Ketkar and Sett (2009) summarized a number of “flexibility inducing HR
practices” from the prior literature, including training, variable pay, an open communication
system, and employee empowerment programs. Such HPWPs allow organizations to attain
ambidexterity (Gibson & Birkinshaw, 2004), which enables organizations to achieve adaptability
and fit without actual change in their HR system and is consistent with the theoretical view that fit
and flexibility can co-exist and complement one another (Wright & Snell, 1998).
To my knowledge, there is no empirical evidence at the HR system level showing a positive
impact for the stability of HR systems in the long run. Therefore, this paper follows the theoretical
predictions of beneficial flexibility to hypothesize a positive impact for strategic HR change, but
it recognizes beneficial stability as a possible and theoretically founded competing prediction on
the overall impact of change.
2.4 Beneficial Strategic HR Change due to Increase in HPWS
If the theoretical predictions of beneficial HR flexibility prevail in practice, the next
question is how to change HR systems to enhance performance. A substantial theoretical and
12
empirical body of literature has shown that an increase in the HPWS is positively related to
performance. Wright and McMahan (1992) provide a complete theoretical framework for
examining the causal impact of HR practices on organizational performance that involves six
theoretical perspectives. Different versions of the causal chain linking HPWS to financial
performance (Becker & Huselid, 1998; Jiang et al., 2012) generally follow the multilevel macro-
micro-macro approach (Hedstrom & Swedberg, 1998). In short, HPWS first improves employees’
abilities, skills, and motivation and gives them opportunities to contribute. Human capital theory
(e.g., Crook, Todd, Combs, Woehr, & Ketchen, 2011; Lepak & Snell, 1999, 2002) and the
behavioral perspective (Schuler & Jackson, 1987) serve as the primary theoretical bases for
deriving the causal impact of HPWS on employee ability, skill and motivation. Thus, desirable HR
outcomes, such as lower voluntary turnover, positively affect operational outcomes (e.g.,
innovation, customer service, and productivity) which consequently influence financial
performance. The resource-based view also helps explain the causal impact of HRM on
performance because HR is a “unique” source of sustained competitive advantage that is difficult
to imitate as a result of its causal ambiguity and path dependency (Becker & Gerhart, 1996). Thus,
HPWS can help improve innovation and financial performance.
There is substantial empirical support in the literature for a robust positive relationship
between HPWS and organizations’ financial performance, as has been shown in recent meta-
analyses (Combs et al., 2006; Jiang et al., 2012). Empirical evidence also reveals a positive impact
of HPWS on innovation (e.g. Delery & Doty, 1996). Particularly, studies have shown that the use
13
of appraisal systems, training, job autonomy, variable pay, employee involvement programs, and
teamwork are positively associated with product and process innovation (Beugelsdijk, 2008;
Lopez-Cabrales, Perez-Luño, & Cabrera, 2009; Walsworth & Verma, 2007). In addition, using
Danish data, Laursen and Foss (2003) found that HR systems had a positive impact on the
importance of the innovation, whether the innovation is new to the firm, the national market, or
the world. In sum, there is strong theoretical basis and empirical evidence that suggests that HPWS
have a positive impact on innovation and financial performance. To emphasize the causal
relationship, I further hypothesize that:
Hypothesis 3. An increase in HPWS is positively related to innovation and profitability.
3 METHODS
3.1 Sample
Despite the call for large-scale longitudinal surveys of HPWS, such surveys are costly to
perform (Boxall & Purcell, 2000; Huselid, 1995; Wall & Wood, 2005; Wright et al., 2005).
However, a few national surveys can partly serve this purpose. The data for this study come from
the Workplace and Employee Survey (WES), which was developed and collected by Statistics
Canada. In particular, this study uses the longitudinal workplace survey from 1999 to 2006. The
nationally representative workplace survey provides comprehensive information on HR practices,
innovation and financial outcomes. The unit of the survey is the workplace, which is ideal for
analyzing dynamic HR fit because there may be diverse HR and business strategies within an
organization’s HR architecture. Another advantage of this survey is its high response rate, which
14
ranges from 75 percent in 2006 to 94 percent in 1999 due to mandatory participation (Statistics
Canada, 2005, 2009; Haines, Jalette, & Larose, 2010).
The final sample excludes non-profit organizations. Only workplaces with at least five
years of observations (i.e., five to eight years) are included in the final sample to ensure a consistent
and meaningful interpretation of the frequency of change and long-run variability. I also restricted
the final sample to workplaces with at least five employees for all years. I chose five instead of a
larger cut-off number because of the panel structure, in which size fluctuates over time, and
because the survey unit is the workplace rather than the entire firm. The final sample includes 2302
workplaces in 14 industry categories and 15,679 workplace-year observations. The average years
surveyed for each workplace is 6.81 years.
3.2 Measures
In this section, I summarize the variables used in the study.
3.2.1 Dependent Variables
I measure innovation using product innovation. Product innovation refers to whether a
workplace has improved a product or introduced a new product during the survey period. The
product innovation measure is a dichotomous variable. Financial performance is measured by
Profitability, or operational margin, which is calculated as the ratio of profit over operating revenue
and expressed as a percentage, with profit equal to operating revenue minus operating cost.
15
3.2.2 Independent Variables
There is no consensus on what constitutes an HPWS (Gerhart, 2012). However, prior
studies have theorized and shown that HPWS consists of three dimensions that are based on their
distinct contributions to organizations—the ability-motivation-opportunity (AMO) dimensions—
and that each dimension has a positive impact on performance (Lepak, Liao, Chung, & Harden,
2006; Jiang et al., 2012). Therefore, to explore the unique patterns of change in different AMO
dimensions, this study adopts the AMO framework to operationalize HPWS rather than using a
single index. A confirmatory factor analysis of 13 HPWP variables affirms a three-factor AMO
structure in HPWS. Each index reflects multiple ways to combine HR practices under each AMO
dimension. I measured HPWS by four variables: training system, hiring system, compensation
system, and employee involvement system.1 This study further decomposes the ability dimension
into training and hiring systems for easier interpretation, and an additional factor analysis confirms
the two-factor structure of the ability dimension.
The index of the training system equals the percentage of employees who received training
times the average amount of training, which consists of cognitive and social skill training
(α=0.694). Cognitive skill training is a standardized and averaged index of 12 dichotomous items
1 Arthur and Boyles (2007) define the HR program as “the set of formal HR activities.” In this study, to distinguish
HR programs at the lower level of abstraction (e.g., a social skill training program), the term “system” is used to
represent the AMO HR sub-systems or a bundle of high performance work practices within each AMO dimension in
HPWS.
16
on classroom and on-the-job training (α=0.751). Social skill and leadership training consists of
eight items (α=0.805). Internal hiring system represents the extent to which vacant positions are
typically filled by internal applicants as opposed to external candidates; it is a standardized and
averaged index measuring whether vacant positions are typically staffed from within the workplace.
Compensation system is a standardized and averaged index of four dichotomous variables
(α=0.458): individual incentive plans, group incentive plans, merit pay, and profit sharing plans.
Employee involvement system equals the standardized and averaged index of five dichotomous
items (α=0.673): the employee’s suggestion program, flexible job design, information sharing with
employees, problem-solving teams, and self-directed work groups.
Variability in HR systems is operationalized as the coefficient of variation (i.e., the standard
deviation of HR systems divided by the mean of that HR system) and measures the dispersion or
total amount of change that occurred within the AMO dimensions of HPWS; hence, it does not
distinguish the direction of change. A value of zero refers to no change in HR systems during the
survey period, and high values indicate substantial amounts of change.
Frequency of change in HR systems is measured by the number of years a change occurred
in an AMO dimension over the number of years surveyed; hence, it is also non-directional.
Compared with the prior year, a difference in the hiring, compensation, and employee involvement
systems is counted once as change. All three frequency variables roughly follow a normal
distribution. Because of a much larger number of training programs (i.e., 20 programs) captured
in the training index, the frequency of change in the training system is calculated differently to
17
convert a highly skewed distribution to a normal distribution. Change in training is counted once
only if the difference between two succeeding years is equal to or more than two training programs.
The distribution of frequency of change in all four HR systems generally follows a normal
distribution. A value of 0 for frequency of change indicates that there is no change in the AMO
dimensions of HPWS, and a value of 1 indicates that change occurred every year during the survey
period.
3.2.3 Control Variables
To control for different organizational and environmental characteristics, I included a
number of control variables. In particular, this paper controls for business strategies, size, voluntary
turnover, percentage of permanent employees, union presence, 14 category industries, and year
variables. It is important to include these controls because they jointly indicate the environmental
and organizational dynamism of a particular workplace, which can significantly shape the form
and impact of strategic HR change.
This study controls for two broad generic business strategies conceptualized by Porter
(1980). Business strategy is operationalized as two continuous variables on cost leadership
strategy and differentiation strategy. The WES asked the respondents to rate the relative
importance of general business strategies on a 5-point Likert scale from 1 (not important) to 5
(crucial). I constructed two additive indexes for cost leadership and differentiation strategy using
the mean of three items for each index. The index of cost leadership (α=0.590) consists of reducing
labor cost, reducing operating cost, and using part-time, temporary, or contract workers. The index
18
of differentiation strategy (α=0.760) captures undertaking research and development, developing
new products or services, and developing new production/operating techniques. Size of workplace
is the logarithmic transformation of the total number of employees. Percentage of standard
employees equals the number of permanent full-time employees over the total of employees. Union
Presence is a dichotomous variable that shows whether a collective bargaining agreement covers
the workplace. Fourteen industry categories are used in this study.2 The reference category in the
regressions is retail trade and consumer services. Finally, I also included year dichotomous
variables in which 1999 is the reference category. Because of the significant mediating role of
voluntary turnover on the relationship between HPWS and performance (Batt & Colvin, 2011),
this paper controls for voluntary turnover to provide conservative estimates of the HR–
performance link, in which HPWS influences performance through other mechanisms (e.g., Jiang
et al., 2012). The voluntary turnover variable in this study equals the number of employees who
had resigned (with no special incentives) over the total number of employees.
3.3 Empirical Analysis
The data are analyzed using Stata 13. The effect of HR systems on performance in panel
data is derived from two sources of variation, which represent either a within-workplace effect
(i.e., a comparison among values in different years of the same workplace) or a between-workplace
2 Detailed industry definitions with mapping to the NAICS codes are found in the Workplace and Employee Survey
Compendium.
19
effect (i.e., differences among workplaces), while holding all other covariates constant. The
following paragraphs explain the two main multivariate analysis methods used in this study. First,
fixed effect models, or least square dummy variable estimators, estimate within-workplace effects
of HR change (i.e., how adding or removing a particular HR practice within the HR system affects
performance). Second, I use long-run average models to investigate how the variability of HR
systems over time influences performance after controlling for the main effects of HR systems.
For each model, I estimate the HR impact on product innovation and profitability. Because product
innovation is a dichotomous dependent variable, comparable logistics models are used to estimate
the probability of producing product innovation.
4 RESULTS
4.1 Multivariate Analyses
Table 2 presents the mean and standard deviation of all variables, including the 13 HPWP
variables used to construct the AMO indexes of HPWS.3 On average, organizations change their
ability-enhancing dimensions of HPWS (i.e., training and hiring systems) at least once every two
years, whereas adjustments in the other two dimensions (i.e., compensation and employee
involvement systems) are less frequent and occur approximately every three years. The variables
for long-run variability range from 0 to 2.828 for training and hiring systems and from 0 to 2.646
for compensation and employee involvement programs. The hiring system shows the highest
3 Due to page limits, two correlation matrices for long-run average variables in Table 3 and for workplace-year
variables in Table 4 are not presented but are available upon request.
20
amount of long-run variability, which indicates that organizations frequently adjust the extent of
internal and external recruitment. The other three indexes show similar variability over the survey
period, but the training system has a smaller standard deviation for long-run variability of 0.390,
which suggests incremental continuous change.
In summary, the descriptive statistics for the frequency and long-run variability variables
jointly suggest that the AMO dimensions of HPWS evolve in different patterns. The training
system has a high frequency of change in the form of smaller fluctuations. The hiring system shows
both high frequency of change and high long-run variability. The compensation and employee
involvement systems change less frequently but show similar long-run variability, which indicates
episodic major change. Therefore, change in the ability dimension of HPWS largely represents
Weick and Quinn’s (1999) metaphor of continuous change, whereas the motivation and
opportunity dimensions appear to resemble episodic change. Nevertheless, descriptive statistics
offer only a glimpse of the general trends. None of the means in frequency and variability are close
to the extreme values, which suggests that the trends are relative (not absolute) and that a
significant portion of change in the AMO dimension deviates from the two corresponding general
patterns.
21
Table 3 and 4 present the results of regression analyses.4 Table 3 shows results for the long-
run average models estimated using equation (5). Model a presents the base model with all the
control variables and the mean levels of HR systems. Long-run variability and frequency variables
are separately entered into models b and c. The full model is presented in model d. Unexpectedly,
the results reveal that the relationship between stability and change in HR systems is more than a
simple trade-off and suggests that organizations balance stability and change in different AMO
dimensions to maximize their gain from HPWS.
The findings with respect to profitability levels and the long-run innovation patterns
consistently show that long-run variability in compensation and employee involvement systems
are positively related to performance (models b and d in Table 3), which supports Hypothesis 1.
By contrast, variability in training system is negatively related to long-run product innovation and
profitability levels; in addition, variability in hiring system is negatively related to long-term
product innovation and profitability levels (models b and d in Table 3). These contrary findings
support the competing theoretical predictions that changes represent deviations from beneficial
consistency.
The frequency of change in training and hiring systems is positively related to long-run
product innovation patterns and profitability levels (models c and d in Table 3), which supports
4 Because of the stratified sampling design of the WES, I applied the bootstrapped sampling weights provided by
Statistics Canada. The bootstrapped weights ensure that any significance in results is not due to the sampling design.
In addition, for confidentiality precautions, only weighted results are allowed for release by Statistics Canada.
22
Hypothesis 2. However, the frequency of change in compensation systems is negatively related to
all four performance measures. The frequency of change in employee involvement systems is
negatively related to long-term product innovation patterns and profitability levels (models c and
d in Table 3). The conflicting findings support the theoretical predictions of beneficial stability in
these two dimensions.
Therefore, Hypotheses 1 and 2, which articulate beneficial flexibility, are only partially
supported, whereas the competing theoretical predictions of beneficial stability explain the other
half of the findings. The results jointly suggest that frequent incremental changes with long-run
stability in training and hiring systems, whereas episodic significant adjustments in compensation
and employee involvement systems, tend to be positively related to performance.
Hypothesis 3 re-states the extensively examined HR–performance relationship with an
emphasis on causality, where an increase in HPWS leads to higher product innovation and
financial performance. Table 4 presents the within-estimates from the fixed effect models using
equation (4) and pooled OLS estimates using equation (1). The fixed effects show that increases
in training and employee involvement systems are positively related to both measures of
performance in Table 4, which supports Hypothesis 3. These findings are consistent with the
between-estimates in Table 3, which show that the levels of training and employee involvement
systems are positively related to product innovation and financial performance.
23
Increases in the extent of external recruitment (as opposed to hiring internally) are
positively related to product innovation (models 1a and 1b in Table 4). However, such increases
show no significant impact on profitability (model 2a) and therefore only partially support
Hypothesis 3. The between-estimates in Table 3 confirm the positive association between external
recruitment and innovation. Consistent with the pooled OLS estimate (model 2b in Table 4),
models 2a and 2b in Table 3 show a positive association between the degree of internal hiring
system and profitability. However, after adding the frequency variables, the effect of the mean
level of internal hiring is no longer significant (models 2c and 2d in Table 3), suggesting that the
positive association is likely to be spurious. Results suggest that frequent adjustments in the
recruitment practices, rather than the actual extent of internal vs. external hiring, are positively
related to profitability.
Table 4 shows that an increase in the compensation system is not significantly related to
innovation (models 1a and 1b) but is negatively related to profitability (model 2a); therefore,
Hypothesis 3 with regard to compensation system is not supported. The negative impact on
profitability seems to contradict the positive between-estimates of compensation system on
profitability in Table 3. The results suggest the presence of confounding or omitted variable bias
in the between effect models, as suggested by Gerhart (1999). Another explanation is the lack of
internal fit among the four major compensation practices within a single workplace examined in
this study. Studies at the firm level may find a positive impact of complex compensation systems.
24
4.2 Additional Analyses and Robustness Check
I first conducted a series of lagged regression analyses with one- and two-year lags of HR
systems in predicting future innovation and financial performance. Results of fixed effect estimates
provide further temporal information about the impact of HPWS. In particular, increases in training
systems appear to have a more immediate effect because this impact mainly unfolds in the current
year (model a in Table 4). It is reasonable that the impact of change in hiring and employee
involvement systems would last longer than one year.
It is also important to note that strategic HR change can be driven by both internal and
external factors (Barnett & Carroll, 1995). To investigate the endogeneity of HR change, I
conducted further lagged fixed effect analyses to examine how change in profitability and
organizational factors is related to change in HR systems in the following year. The results show
that change in business strategies, size, union representation, and profitability in the last year
predict change in certain HPWS dimensions. Therefore, the findings from the long-run average
models should be interpreted as correctional rather than causal because profitability and other
factors partly influence future strategic HR change.
5 DISCUSSION
5.1 Overview and Contributions
This study reframes the “paradox of administration” that articulates a trade-off between
efficiency and flexibility (Thompson, 1967) from a strategic HRM perspective by asking whether
25
long-run stability or change in HPWS leads to superior organizational outcomes. Two competing
theoretical perspectives suggest that both stability and change in HR systems can be beneficial,
but how organizations change their HR systems and whether such attempts pay off are questions
that remain to be answered empirically. An exploratory analysis of the AMO dimensions of HPWS
reveals a more complicated picture than a simple stability–change trade-off in HR systems. The
results show that, in practice, organizations strive to maximize their gain from HR systems by
strategically balancing stability and change in the different aspects of HPWS. The pattern of
change is significantly related to performance. The collective evidence from different analytical
models suggests that an increase in HPWS positively impacts innovation and financial
performance, but effective strategic HR change unfolds in different forms in the AMO dimensions.
Continuous change with long-run stability in the ability-enhancing dimension (i.e., training and
hiring systems) is positively related to both innovation and financial performance, whereas
episodic change with short-run stability in the motivation- and opportunity-enhancing dimensions
(i.e., compensation and employee involvement systems) is positively related to both performance
outcomes.
The study bridges the strategic HRM and organizational change literature by showing that
the tempo of change—the “characteristic rate, rhythm, or pattern of work or activity” —is
significantly related to performance (Weick & Quinn, 1999). The beneficial tempos of change in
the AMO dimensions are largely consistent with Weick and Quinn’s (1999) conceptualization of
continuous and episodic change. Effective training programs and hiring systems focus on long-run
26
adaptability and involve micro-level adjustments. By contrast, compensation and employee
involvement systems—the most intractable features of HR systems—focus on short-run adaption.
Because the motivation- and opportunity-enhancing dimensions of HR systems inherently induce
flexibility, short-run stability can be sufficient to ensure high levels of performance. In the long
run, beneficial HR change tends to take the form of episodic HR system change to significantly
impact employees’ motivation and opportunities to contribute. Therefore, these findings support
Becker and Gerhart’s (1998) insights that the seemingly contradictory universalistic and
contingency views are complementary perspectives operating at different levels of analysis. As
shown in this paper, the two theoretical perspectives also operate in different AMO dimensions of
HPWS and time frames.
This study also contributes to the prior theoretical debates about the relationship between
HR fit and flexibility with empirical evidence. The orthogonal view argues that fit and flexibility
are two ends of a spectrum at a given point in time and that firms must choose a position
accordingly (Lengnick-Hall & Lengnick-Hall, 1988; Baird, L., & Meshoulam, I. 1988). Similarly,
Boxall and Purcell (2000) and Gerhart (2007) recognized the “strategic tension” between current
performance and adaptation to the future. By contrast, the complementary perspective suggests
that fit and flexibility complement one another because HR flexibility is necessary for
organizations to achieve fit in the dynamic environment (Wright & Snell, 1998). Organizational
ambidexterity also reflects the complementary perspective that HR systems simultaneously pursue
fit and adaptability (Adler et al., 1999; Patel et al., 2013). The findings in this paper support both
27
perspectives, which not only affirm the key distinction about the time frame but also emphasize
that the tempo of change in the AMO dimension is significantly related to performance. Dynamic
HR fit takes the form of both stability and flexibility but differs in the AMO dimensions. Consistent
with the orthogonal perspective, the results suggest that in the short run, fit (in the form of
consistency) in the motivation and opportunity dimensions positively impacts performance but that
flexibility is essential for the ability dimension of HPWS. In the long run, the results support the
complementary perspective that organizations balance the patterns of stability and change to
achieve high performance. This study demonstrates that both orthogonal and complimentary
perspectives inform the implementation of strategic HR change when matched with the appropriate
tempo of change in the corresponding AMO dimensions.
Finally, this paper complements prior empirical studies by addressing certain
methodological limitations. All the variables that relate to the dynamic aspects of HPWS (i.e.,
frequency and long-run variability) and performance are derived from the longitudinal data to
avoid potential common method bias in empirical studies on HR practice flexibility. This
longitudinal study also strengthens the causal arguments about the HR–performance relationship
by demonstrating that increases in the extent of training, external recruitment, and employee
involvement systems are positively related to performance. A lagged fixed effect analysis suggests
that the positive effect of more extensive training system is mainly manifested in the current year,
whereas the impact of external hiring and employee involvement programs largely manifests in
the subsequent year. In summary, this study strengthens the causal inference in prior research and
28
further demonstrates that the impact of HPWS on performance is more than merely the main effect;
the dynamic aspect, i.e., the tempo of strategic HR change, also matters.
5.2 Limitations and Suggestions for Future Research
There are several limitations to this study. This study captures only key HPWPs and may
not be generalizable to other aspects and measurements of HPWS. Similarly, the results of this
paper should be interpreted at the HR system level, and the findings may not be generalizable to
other levels of analysis. Nevertheless, the competing theoretical perspectives suggest that the
impact of change in more intractable structures of the HR system is likely to follow the findings
regarding the motivation and opportunity dimensions of HPWS. By contrast, the effect of change
in micro-level HR activities, such as the use of structured interviews or employment tests, is
expected to be consistent with the findings on the ability-enhancing dimension in which continuous
change is beneficial. By weighing the trade-off among accuracy, generality, and simplicity in
developing theory (Weick, 1979), this study shows that the AMO level of analysis shows
significant potential for informative future theoretical developments and empirical investigations.
Future studies might also examine the impact of moderators on the relationship between
strategic HR change and performance. As the contingency perspective suggests, the impact of
strategic HR change depends on organizational and environmental dynamism. For example, life
cycle stages influence the extent and outcome of change (Amburgey et al., 1993). Moreover, minor
and major dynamic HR misfits are mostly excluded in this study. As a result of annual
29
measurement, this study is unable to capture short-lived minor dynamic HR misfits, where
workplaces quickly adjust to regain alignment. To attain meaningful and consistent measures of
strategic HR change, this study only includes surviving workplaces with at least five years of
observations. However, poor fit may affect a firm’s survival (Gerhart, 2007). Future studies might
further explore dynamic HR misfit by investigating non-survivors.
A general limitation of this study is that the dynamic HR fit model is a parsimonious
depiction of macro-level relationships and does not fully capture complex multi-level causal
mechanisms. Analogous to Boxall, Purcell, and Wright’s (2007) distinction between strategic
HRM and micro-HRM, this strategic HRM research focuses on the macro-level patterns of change
in HR systems rather than on the effectiveness of particular HR interventions. The fixed effect
models in Table 4 capture the within workplace variation over time, and the lagged analysis
provides further support for the positive impact of HPWS on performance. However, since the
variables are only measured annually, this study would only address the issue of causality to a
limited extent. As shown in this paper, the AMO dimensions evolve in different patterns. Future
studies investigating the causal impact of HPWS may consider measuring the ability-enhancing
dimension with shorter time intervals. Another limitation is that the between-estimates in Table 3
are correlational, and there is no appropriate instrumental variable to test for the endogeneity of
strategic HR change. Thus, it is inadvisable to draw causal inferences about the patterns of HR
change and performance. Future studies should refine the research design to investigate the causal
impact of strategic HR change. One such approach advocated by Huselid and Becker (2011) is to
30
conduct multi-level studies, which can meaningfully enhance our understanding of the casual
mechanism of dynamic HR fit by bridging the macro-level findings on strategic HR change in this
paper with micro-level evidence, such as Leana and Barry’s (2000) theoretical development of
micro level stability and change.
6 CONCLUSIONS
This paper explores how organizations balance stability and change in HR systems to
ensure superior and sustained innovation and financial performance. The findings provide
empirical support for both competing theoretical perspectives; paradoxically, both stability and
change in HR systems can be beneficial. An exploratory analysis of the AMO dimensions of
HPWS reveals that the performance impact depends on the patterns of strategic HR change. The
results of different measures of innovation and financial performance consistently show that long-
run stability with a high frequency of change in the ability-enhancing dimension (i.e., training and
hiring systems) is related to high performance. By contrast, long-run variability with a low
frequency of change in the motivation- and opportunity-enhancing dimensions (i.e., compensation
and employee involvement programs) is associated with high performance. The findings suggest
that organizations can achieve dynamic HR fit by strategically balancing stability and change in
HPWS. Particularly, organizations should emphasize long-run adaptability in the ability-enhancing
dimension by implementing continuous incremental change, while stressing short-run adaptation
in the motivation- and opportunity-enhancing dimensions by providing short-run stability and
implementing episodic changes over the long run. Furthermore, there are three main
31
methodological advances in this paper. First, this paper complements prior studies on HR practice
flexibility by using eight-year longitudinal data to derive objective measures of HR change and by
showing that dynamic HR fit can also take the form of beneficial stability. Second, this study
introduces a between-estimate panel model to better represent the long-term oriented strategic
contribution of HPWS and to offer robust controls in examining the impact of variability and
frequency of strategic HR change. Third, this paper strengthens causal arguments about the
positive impact of HPWS on innovation and financial performance by using within-estimates. The
fixed effect models show that increased training and employee involvement programs are
positively related to performance. In sum, this paper provides initial evidence that patterns of
change in the AMO dimensions of HPWS are significantly related to innovation and financial
performance; thus, it can serve as the basis for future studies on the impacts of strategic HR change.
32
REFERENCES
Adler, P., Goldoftas, B., and Levine, D. 1999. Flexibility versus efficiency? A case study of
model changeovers in the Toyota production system. Organization Science, 10(1): 43-68.
Amburgey, T. L., Kelly, D., & Barnett, W, P, 1993. Resetting the clock: The dynamics of
organizational change and failure. Administrative Science Quarterly, 38(1): 51-73.
Arthur, J. B., & Boyles, T. 2007. Validating the human resource system structure: A levels-
based strategic HRM approach. Human Resource Management Review, 17(1):77−92.
Astley, W., & Van de Ven, A. 1983. Central perspectives and debates in organization theory.
Administrative Science Quarterly, 28(2): 245-273.
Baird, L., & Meshoulam, I. 1988. Managing two fits of strategic human resource management.
Academy of Management Review, 13(1): 116-128.
Barnett, W. P. & Carroll, G. R. 1995. Modeling internal organizational change. Annual Review of
Sociology, 21(1): 217-236.
Batt, R., & S. Colvin, A. J. 2011. An employment systems approach to turnover: human
resources practices, quits, dismissals, and performance. Academy of Management
Journal, 54(4): 695-717.
Becker, B. & Gerhart, B. 1996. The impact of human resource management on organizational
performance: progress and prospects. Academy of Management Journal, 39(4): 779-801.
Becker, B. E., & Huselid, M. A. 1998. High performance work systems and firm performance: A
synthesis of research and managerial implications. In G. R. Ferris (Ed.), Research in
Personnel and Human Resources Management: 53-101. Greenwich, CT: JAI Press.
Becker, B. E., & Huselid, M. A. 2006. Strategic human resources management: Where do we
go from here? Journal of Management, 32(6): 898-925.
Becker, B. E., Huselid, M. A., & Beatty, R. W. 2009. The differentiated workforce:
Transforming talent into strategic impact. Boston: Harvard Business Press.
Beugelsdijk, S. 2008. Strategic human resource practices and product innovation. Organization
Studies, 29(6): 821-847.
33
Bhattacharya, M., Gibson, D., & Doty, D. 2005. The effects of flexibility in employee skills,
employee behaviors, and human resource practices on firm performance. Journal of
Management, 31(4): 622-640.
Boxall, P. & Purcell, J. 2003. Strategy and human resource management. Hampshire,
England: Palgrave Macmillan.
Boxall, P. F., Purcell, J., & Wright, P. M. 2007. The Oxford handbook of human resource
management. Oxford University Press.
Cappelli, P. & Neumark, D. 2001. Do ‘high performance’ work practices improve
establishment level outcomes? Industrial and Labor Relations Review, 54(4): 737-775.
Chang, S., Gong, Y., Way, S. A., & Jia, L. 2013. Flexibility-oriented HRM systems, absorptive
capacity, and market responsiveness and firm innovativeness. Journal of Management,
39(7): 1924-1951.
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(3): 501-528.
Crook, T. R., Todd, S. Y., Combs, J. G., Woehr, D. J. & Ketchen, D. J. Jr., 2011. Does human
capital matter? A meta-analysis of the relationship between human capital and firm
performance. Journal of Applied Psychology, 96(3): 443-456.
Datta, D. K., Guthrie, J. P., & Wright, P. M. 2005. Human resource management and labor
productivity: Does industry matter? Academy of Management Journal, 48(1): 135-145.
Delery, J. E., & Doty, D. H. 1996. Modes of theorizing in strategic human resource
management: Tests of universalistic, contingency. Academy of Management Journal,
39(4): 802-835.
Feldman, M. S., & Pentland, B. T. 2003. Reconceptualizing organizational routines as a source of
flexibility and change. Administrative Science Quarterly, 48(1): 94-118.
Gerhart, B. 1999. Human resource management and firm performance: Measurement issues and
their effect on causal and policy inferences. Research in Personnel and Human Resources
management, Supplement, 4: 31-51.
34
Gerhart, B. 2007. Horizontal and vertical fit in human resource systems. Perspectives on
Organizational Fit, 317-348.
Gerhart, B. 2012. Construct validity, causality, and policy recommendations: The case of high
performance work practices systems. Human Resource Management Review, 22(2):
157-160.
Gerhart, B., & Milkovich, G. T. 1990. Organizational differences in managerial compensation
and financial performance. Academy of Management Journal, 33(4): 663-691.
Gerhart, B., Wright, P. M., McMahan, G. C., & Snell, S. A. 2000. How much error is there and
how does it influence effect size estimates? Personnel Psychology, 53(4): 803-834.
Gibson, C.B., & Birkinshaw, J. 2004. The antecedents, consequences, and mediating role of
organizational ambidexterity. Academy of Management Journal, 47(2): 209-226.
Guest, D. E. 2011. Human resource management and performance: still searching for some
answers. Human Resource Management Journal, 21(1): 3-13.
Hackman, J. R., & Wageman, R. 1995. Total quality management: Empirical, conceptual, and
practical issues. Administrative Science Quarterly, 40(2): 309-342.
Haines, V. Y., Jalette, P., & Larose, K. 2010. The influence of human resource management
practices on employee voluntary turnover rates in the Canadian non governmental sector.
Industrial & Labor Relations Review, 63(2): 228-246.
Hannan, M. T. & Freeman, J., 1984. Structural inertia and organizational change. American
Sociological Review, 49(2): 149-164.
Hedstrom, P., & Swedberg, R. 1998. Social mechanisms: An introductory essay, Social
Mechanisms: An Analytical Approach to Social Theory: 1-31. New York: Cambridge
University Press.
Huselid, M. A. 1995. The impact of human resource management practices on turnover,
productivity, and corporate financial performance. Academy of Management Journal,
38(3): 635-670.
Huselid, M.A., & Becker, B.E. 2011. Bridging micro and macro domains: workforce
differentiation and strategic human resource management. Journal of Management.
37(2): 421-428.
35
Ichniowski C, Shaw K, Prennushi G. 1997. The effects of human resource management
practices on productivity. American Economic Review, 87(3): 291-313.
Jackson, S. E., Schuler, R.S. & Jiang, K.F. 2014. An aspirational framework for strategic human
resource management. The Academy of Management Annals, 8(1): 1-56.
Jiang, K., Lepak, P. D., Hu, J., & Judith, C. B. 2012. How does human resource management
influence organizational outcomes? A meta-analytic investigation of mediating
mechanisms. Academy of Management Journal, 55(6): 1264-1294.
Kaufman, B. E., & Miller, B. I. 2011. The firm’s choice of HRM practices: economics meets
strategic human resource management. Industrial & Labor Relations Review, 64(3): 526-
557.
Ketkar, S. & Sett, P.K. 2009. HR flexibility and firm performance: Analysis of a multi-level
causal model. The International Journal of Human Resource Management, 20(5): 1009-
1038.
Ketkar, S. & Sett, P. K. 2010. Environmental dynamism, human resource flexibility, and firm
performance: analysis of a multi-level causal model. The International Journal of Human
Resource Management, 21(8): 1173-1206.
Kim, Y., & Ployhart, R. E. 2014. The effects of staffing and training on firm productivity and profit
growth before, during, and after the Great Recession. Journal of Applied Psychology, 99(3):
361.
Klarner, P., & Raisch, S. 2013. Move to the beat--rhythms of change and firm performance.
Academy of Management Journal, 56(1): 160-184.
Laursen, K., & Foss, N. 2003. New HRM practices, complementarities and the impact on
innovation performance. Cambridge Journal of Economics, 27(2): 243-263.
Leana, C. R., & Barry, B. 2000. Stability and change as simultaneous experiences in
organizational life. The Academy of Management Review, 25(4): 753-759.
Lengnick-Hall, C. A., & Lengnick-Hall, M. L. 1988. Strategic human resources management:
A review of the literature and a proposed typology. Academy of Management Review,
13(3): 454-470.
36
Lepak, D. P., Liao, H., Chung, Y., & Harden, E. E. 2006. A conceptual review of human
resource management systems in strategic human resource management research. In J. J.
Martocchio (Ed.), Research in Personnel and Human Resource Management, 25: 217-
271. Greenwich, CT: JAI Press.
Lepak, D. P., & Snell, S. A. 1999. The human resource architecture: Toward a theory of human
capital allocation and development. Academy of Management Review, 24(1): 31-48.
Lepak, D. P., & Snell, S. A. 2002. Examining the human resource architecture: The
relationships among human capital, employment, and human resource configurations.
Journal of Management, 28(4): 517-543.
Lopez-Cabrales, A., PÉRez-Luño, A., & Cabrera, R. V. 2009. Knowledge as a mediator
between HRM practices and innovative activity. Human Resource Management, 48(4):
485-503.
Luce, R. D. 1997. Several unresolved conceptual problems of mathematical psychology. Journal
of Mathematical Psychology, 41(1): 79-87.
Miles, R. E., & Snow, C. C. 1984. Designing strategic human resources systems.
Organizational Dynamics, 13(1): 36-52.
Ngo, H., & Loi, R. 2008. Human resource flexibility, organizational culture and firm
performance: An investigation of multinational firms in Hong Kong. International
Journal of Human Resource Management, 19(9): 1654-1666.
Patel, P. C., Messersmith, J. G., & Lepak, D. 2013. Walking the tightrope: An assessment of
the relationship between high-performance work systems and organizational
ambidexterity. Academy of Management Journal, 56(3): 1420-1442.
Piening, E. P., Baluch, A. M., & Salge, T. O. 2013. The relationship between employees’
perceptions of human resource systems and organizational performance: Examining
mediating mechanisms and temporal dynamics. Journal of Applied Psychology, 98(6): 926-
947.
Porter, M. E. 1980. Competitive strategy. New York, NY: Free Press.
Schuler, R. S., & Jackson, S. E. 1987. Linking competitive strategies with human resource
management practices. Academy of Management Executive, 1(3): 207-219.
37
Snell, S. A., & Dean, J. W., Jr. 1994. Strategic compensation for integrated manufacturing:
The moderating effects of jobs and organizational inertia. Academy of Management
Journal, 37(5): 1109-1140.
Statistics Canada. 2005. Workplace and Employee Survey Compendium 2005. Catalogue no.
71-585-X. http://www.statcan.gc.ca/pub/71-585-x/71-585-x2008001-eng.pdf
Statistics Canada. 2009. Workplace and Employee Survey.
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=2615&lang=en
%20&db=imdb&adm=8&dis=2
Teece, D., Pisano, G., & Shuen, A. 1997. Dynamic capabilities and strategic management.
Strategic Management Journal, 18(7): 509-533.
Thompson, J. D. 1967. Organizations in Action. McGraw Hill, New York.
Tracey, J. B. 2012. A contextual, flexibility-based model of the HR-firm performance
relationship. Management Decision, 50(5): 909-924.
Wall, T. D., & Wood, S. J. 2005. The romance of human resource management and business
performance, and the case for big science. Human Relations, 58(4): 429-462.
Walsworth, S., & Verma, A. 2007. Globalization, human resource practices and innovation:
recent evidence from the Canadian workplace and employee Survey. Industrial
Relations, 46(2): 222-240.
Wang, D. S., & Shyu, C. L. (2009). The longitudinal effect of HRM effectiveness and dynamic
innovation performance on organizational performance in Taiwan. The International
Journal of Human Resource Management, 20(8), 1790-1809.
Way, S. A., Tracey, J. B., Fay, C. H., Wright, P. M., Snell, S. A., Chang, S and Gong, Y. 2015.
Validation of a multidimensional HR flexibility measure. Journal of Management, 41(4):
1098-1131.
Weick, K. E., & Quinn, R. E. 1999. Organizational change and development. Annual Review
of Psychology, 50(1): 361-386.
Weick, K. 1979. The social psychology in organizing. Reading, MA: Addison-Wesley.
38
Wright, P. M., & McMahan, G. 1992. Theoretical perspectives for strategic human resource
management. Journal of Management, 18(2): 295-320.
Wright, P. M., Gardner, T. M., Moynihan, L. M., & Allen, M. R. 2005. The relationship
between HR practices and firm performance: Examining causal order. Personnel
Psychology, 58(2): 409-446.
Wright, P. M., & Snell, S. A. 1998. Toward a unifying framework for exploring fit and
flexibility in strategic human resource management. Academy of Management Review,
23(4): 756-772.
Zajac, E.J., Kraatz, M.S., & Bresser, R.K.F. 2000. Modeling the dynamics of strategic fit: a
Normative approach to strategic change. Strategic Management Journal, 21(4): 429-
453.
39
TABLE 1. A Dynamic Model of HR Fit and Misfit
Does Change in HR System Occur?
Is Change in HR
System Needed to
Establish Dynamic
HR Fit?
Yes No
Yes
HR flexibility is
desirable; HR
stability leads to
dynamic misfit.
I
Beneficial HR
Flexibility
(Dynamic Fit)
IV
Insufficient HR
Change
(Dynamic Misfit)
No
HR stability is
desirable; HR
change leads to
dynamic misfit.
II
Excessive HR
Change
(Dynamic Misfit)
III
Beneficial Consistency
(Dynamic Fit)
40
TABLE 2. Descriptive Statistics a
Mean s.d.
Product innovation 0.586 0.493
Profitability (%) 17.405 26.961
Ability-enhancing dimension of HPWS:
Training system 0.151 0.178
Average percentage of employees received classroom and on-
the-job training
0.443 0.290
Number of social skill training program 0.320 0.277
Number of cognitive skill training program 0.239 0.202
Internal hiring system 0.218 0.316
Motivation-enhancing dimension of HPWS:
Compensation system 0.310 0.284
Individual incentive systems 0.493 0.500
Group incentive systems 0.204 0.403
Merit pay and skill-based pay 0.342 0.474
Profit sharing plan 0.203 0.403
Opportunity-enhancing dimension of HPWS:
Employee involvement system 0.275 0.279
Employee suggestion program 0.380 0.485
Flexible job design 0.184 0.387
Information sharing with employees 0.470 0.499
Problem-solving teams 0.254 0.436
Self-directed work groups 0.089 0.285
Frequency of change in the AMO dimensions of HPWS b
Frequency of change in training system 0.575 0.205
Frequency of change in hiring system 0.507 0.251
Frequency of change in compensation system 0.344 0.128
Frequency of change in employee involvement system 0.382 0.130
Long-run variability in the AMO dimensions of HPWS b
Long-run variability in training system 0.854 0.390
Long-run variability in hiring system 1.172 0.848
Long-run variability in compensation system 0.749 0.577
Long-run variability in employee involvement system 0.855 0.581
Control Variables
Index of cost leadership strategy 2.956 0.772
Index of differentiation strategy 3.030 0.899
Total number of employees 57.856 100.642
Percentage of permanent full time employees (%) 78.601 26.307
Union representation 0.287 0.452
Voluntary turnover 0.159 0.301 a The number of workplace-year observations is 15,679 and the number of workplace is 2302. Each
workplace only has one value for frequency of change and long-run variability in the AMO dimensions.
41
TABLE 3. Patterns of Strategic HR Change as Predictors Long-run Average (LRA)
Innovation and Profitability Levels ab
Variables LRA Product Innovation
Model 1a Model 1b Model 1c Model 1d
Mean level of training system 0.240*** 0.137*** 0.208*** 0.098***
(0.013) (0.014) (0.014) (0.014)
Mean level of internal hiring
system -0.105*** -0.095*** -0.118*** -0.126***
(0.010) (0.009) (0.010) (0.010)
Mean level of compensation system 0.129*** 0.170*** 0.098*** 0.149***
(0.009) (0.010) (0.009) (0.010)
Mean level of employee
involvement 0.245*** 0.299*** 0.232*** 0.305***
system (0.009) (0.008) (0.010) (0.011)
Variability in training system -0.055*** -0.065***
(0.005) (0.005)
Variability in hiring system -0.003 -0.006**
(0.002) (0.002)
Variability in compensation system 0.035*** 0.032***
(0.003) (0.003)
Variability in employee
involvement 0.043*** 0.040***
system (0.003) (0.003)
Frequency of change in training
system 0.183*** 0.169***
(0.009) (0.009)
Frequency of change in hiring
system 0.025* 0.039***
(0.009) (0.008)
Frequency of change in
compensation system 0.008 -0.061***
(0.014) (0.015)
Frequency of change in employee
involvement system -0.001 -0.058***
(0.014) (0.014)
Control variables
Mean level of cost leadership
strategy -0.054*** -0.062*** -0.061*** -0.066***
(0.003) (0.003) (0.003) (0.003)
Mean level of differentiation
strategy 0.137*** 0.136*** 0.136*** 0.136***
(0.002) (0.002) (0.002) (0.002)
Mean level of size -0.008*** -0.008*** -0.008*** -0.008***
(0.002) (0.002) (0.002) (0.002)
Mean level of percentage of
permanent 0.000 0.000 0.000 0.000
full time employees (0.000) (0.000) (0.000) (0.000)
42
Mean level of union representation -0.019*** -0.015** -0.027*** -0.026***
(0.004) (0.004) (0.004) (0.004)
Mean level of voluntary turnover 0.004 0.018 -0.008 0.005
(0.008) (0.007) (0.007) (0.007)
Variability in voluntary turnover 0.022*** 0.019*** 0.021*** 0.023***
(0.003) (0.003) (0.003) (0.003)
Industry 5 Yes Yes Yes Yes
Constant 0.278*** 0.270*** 0.212*** 0.251***
0.015 0.016 0.016 0.017
R2 0.301 0.318 0.319 0.334 a These are OLS models estimated using equation (5). b Number of workplaces workplace is 2302 and the average number of years surveyed for each
workplace is 6.81 years.
* p <.01
** p < .001
*** p <.0001
5 The reference category is retail trade and consumer services. Due to space limit, coefficients of the 13 industry
dummy variables are not presented, but are available upon request.
43
TABLE 3 (Continued)
Variables LRA Profitability
Model 2a Model 2b Model 2c Model 2d
Mean level of training system 23.763*** 9.466*** 22.645*** 7.168***
(1.509) (1.400) (1.469) (1.393)
Mean level of hiring system 6.659*** 5.842*** 2.062 -0.98
(0.980) (0.911) (1.116) (1.106)
Mean level of compensation 4.376*** 5.015*** 3.452*** 5.755***
system (0.691) (0.612) (0.772) (0.694)
Mean level of employee -2.436* 2.415** 1.920 10.463***
involvement system (0.800) (0.667) (0.821) (0.815)
Variability in training system -8.966*** -9.267***
(0.419) (0.418)
Variability in hiring system -1.567*** -1.962***
(0.121) (0.120)
Variability in compensation 0.990* 1.782***
system (0.309) (0.338)
Variability in employee 3.371*** 4.684***
involvement system (0.253) (0.31)
Frequency of change in training 4.840*** 4.601***
system (1.038) (0.864)
Frequency of change in 5.950*** 8.532***
hiring system (0.854) (0.844)
Frequency of change in -3.209* -7.663***
compensation system (1.238) (1.486)
Frequency of change in -14.380*** -20.558***
employee involvement system (1.462) (1.593)
Control variables
Mean level of cost leadership -2.385*** -2.818*** -2.398*** -2.739***
strategy (0.260) (0.227) (0.250) (0.227)
Mean level of differentiation -2.922*** -2.595*** -2.817*** -2.548***
strategy (0.244) (0.214) (0.238) (0.205)
Mean level of size -1.812*** -1.898*** -2.052*** -2.203***
(0.190) (0.180) (0.190) (0.181)
Mean level of percentage of -0.045*** -0.037*** -0.044*** -0.035***
permanent full time employees (0.009) (0.008) (0.008) (0.007)
Mean level of union 0.898 1.103 0.024 -0.025
representation (0.532) (0.545) (0.449) (0.453)
Mean level of voluntary -0.821 0.254 -1.596** -0.382
turnover (0.520) (0.543) (0.480) (0.515)
Variability in voluntary turnover -1.715*** -1.118** -1.502** -0.905
(0.395) (0.337) (0.414) (0.366)
Industry Yes Yes Yes Yes
Constant 34.221*** 39.667*** 35.957*** 42.414***
1.213 1.354 1.271 1.345
R2 0.093 0.128 0.104 0.148
44
TABLE 4. AMO Dimensions of HPWS as Predictors Innovation and Profitability
Product Innovation Profitability
Model 1a Model 1b Model 2a Model 2b
Fixed effect OLS Fixed effect OLS
Training system 1.564*** 1.417*** 5.433*** 14.624***
(0.275) (0.042) (1.352) (0.799)
Internal hiring system -0.608*** -0.525*** -1.041 2.468***
(0.126) (0.026) (0.639) (0.462)
Compensation system -0.165 0.024 -6.868*** -1.636***
(0.163) (0.032) (0.847) (0.425)
Employee involvement 0.400* 0.754*** 4.392*** 3.741***
system (0.168) (0.028) (0.849) (0.41)
Control variables
Cost leadership strategy -0.196*** -0.147*** -0.629* -1.255***
(0.059) (0.01) (0.292) (0.15)
Differentiation strategy 0.16** 0.408*** -0.803** -1.445***
(0.051) (0.008) (0.249) (0.091)
Size 0.039 -0.0200* 1.602* -2.084***
(0.149) (0.008) (0.727) (0.182)
Percentage of permanent full 0.001 -0.005*** -0.056*** 0.029
time employees (0.003) (0.000) (0.014) (0.006)
Union representation -0.069 -0.28*** -1.018 -0.188
(0.15) (0.016) (0.803) (0.462)
Voluntary turnover -0.51** -0.398*** 1.759* 0.304***
(0.165) (0.033) (0.686) (0.222)
Year6 Yes Yes Yes Yes
Constant 0.566 27.61*** 34.918***
(0.056) (3.145) (0.822)
Number of observations 13030 15679 15679 15679
Number of workplaces 1889 2302 2302 2302
R2 0.07 0.061 0.047 0.042
* p <.05
** p < .01
*** p <.001
6 The reference category is 1999. Due to page limit, coefficients of the 7-year dummy variables (i.e. 2000 to 2006)
are not presented, but are available upon request.
45
CHAPTER 2 INDUSTRY AND FIRM-LEVEL DETERMINANTS OF
EMPLOYMENT RELATIONS IN CHINA: A TWO-LEVEL ANALYSIS
ABSTRACT
Factors influencing the adoption of human resource management (HRM) policies and practices are
nested within the multilevel contexts of firms and industries. Institutional theory focuses on
environmental pressure and suggests that an organization’s choice of an HRM system is partly
attributable to mimetic isomorphism. Drawing on different theoretical perspectives, this study
examines multilevel environmental and organizational contingencies as determinants of HRM
systems and tests their effects on the use of short-term labor contracts, contract duration, training,
and employee involvement programs using a survey of 313 manufacturing plants in China.
Utilizing a hierarchical linear model, our analysis shows that while most of the variance in HRM
systems occurred at the firm level, approximately 5 to 7 percent of the total variance in the four
HRM programs we studied are explained by industry-level factors. Findings suggest that
international competitive pressure, capital intensity, firm size, unionization, and ownership type
have significant effects on use of labor contracts in a manufacturing context. However, for training
programs, only capital intensity and firm size are significant positive predictors; for employee
involvement programs, only firm size and ownership are significant determinants.
46
1 INTRODUCTION
Convergences and divergences of human resource management (HRM) systems and, more
specifically, the antecedents of HRM systems are central issues in strategic HRM research. The
contingency perspective suggests that organizations gain efficiency by aligning their HRM
systems with environmental and organizational contingencies to achieve high performance.
Jackson and Schuler (1995) and Jackson, Schuler, and Jiang (2014) offer a comprehensive
conceptual framework in which HRM systems are the product of organizational and environmental
contingencies at multiple levels. Notably, few organizations face static environments. In today’s
dynamic business environment, most organizations need adequate flexibility to meet internal and
external challenges brought forth by change. As several studies have noted, human resource (HR)
flexibility is a key contributor to organizational financial success in dynamic environments (Ketkar
& Sett, 2009, 2010; Lepak, Takeuchi, & Snell, 2003). According to Wright and Snell’s (1998)
theoretical framework, there are two dimensions of flexibility—resource flexibility and
coordination flexibility—and each of these dimensions can be applied to three components of
flexibility: (1) HR practices, (2) employee skills, and (3) behavior. These are the three main ways
in which HRM systems can help organizations achieve flexibility. In the context of strategic HRM,
resource flexibility refers to the applicability of HR practices, employees’ variability of skills, and
their ability to learn and apply new skills and behaviors to a large range of uses; coordination
flexibility refers to the malleability of HR practices and the variety of employee skills and
behaviors that can be resynthesized, reconfigured, and redeployed. Based on Wright and Snell’s
47
(1998) theoretical framework, Way et al. (2015) further developed and validated a
multidimensional scale for HR flexibility; however, there remains a gap in the literature on the
antecedents of HR flexibility, specifically on organizational and external contingencies that are
related to the adoption of flexibility-inducing HR practices.
Some aspects of HR systems are arguably more crucial for HRM flexibility (Ketkar & Sett,
2009, 2010). For instance, training and employee involvement programs are particularly important
for flexibility in employee skills and behavior because these practices directly influence human
capital and the motivation of employees, which can be a source of firms’ sustained competitive
advantage (Campbell, Coff, & Kryscynski, 2012). In the China context, labor contract systems
give employers the ability to sign contracts of different durations with different employees. This
gives insight into the trade-offs that employers make between labor flexibility and the retention of
their core workforce. This study focuses on how employers’ use labor contract durations and
training and employee involvement programs, looking specifically at how internal and external
contingencies influence the adoption of flexibility-inducing HR programs such as these.
To fully understand the dynamic environment that propels in the adoption and assimilation
of HR systems requires multiple theoretical perspectives. Institutional theory and a resource-based
view help to explain the homogeneity and heterogeneity of HR systems across different
organizations. This theory predicts that certain factors contribute to an organization’s adoption of
certain structures and processes, including HRM practices. The state of resemblance of a particular
48
organization to other organizations in the field is known as “organizational isomorphism.”
According to DiMaggio and Powell (1983), three types of pressures——coercive, normative, and
mimetic pressures——cause organizational practices to diffuse to other organizations. The
institutional perspective offers insights into antecedents of HRM systems by suggesting that
through organizational isomorphism firms introduce similar HR practices resulting in a
convergence of HR systems among organizations. Paauwe and Boselie (2003) support this theory,
arguing that coercive, formalized, and mimetic mechanisms all contribute to isomorphism and the
resulting homogeneity in HR practices across organizations. While institutional theory stresses the
role of external pressure, a resource-based view (e.g., Barney, 1995) focuses on internal
considerations. Drawing on the resource-based view, Becker and Gerhart (1996) further argued
that HR can serve as a rare, valuable, and imitable source of an organization’s sustained
competitive advantage through causal ambiguity and path dependency.
Despite these theoretical advances in the field, empirical evidence of the multilevel impacts
of organizational and environmental contingencies on flexibility-inducing HR systems is limited,
specifically research into the extent to which industry-level isomorphism is a determinant of an
organization’s choice of HR systems. This paper aims to empirically test the relative importance
of internal organizational characteristics and external pressures at the industry level in predicting
an organization’s use of labor contracts and flexibility-inducing HR programs. I examine factors
influencing the use of short-term workers, labor contract duration, and training and employee
involvement programs. Briefly, the results suggest that international competitive pressure and
49
capital intensity at the industry level, as well as firm-level characteristics such as firm size,
unionization, and ownership, are significant determinants of flexibility-inducing HR systems. As
predicted by institutional theory, industry is important in influencing an organization’s adoption
of HR practices, and I found that industry-level factors account for about 5 to 7 percent of the total
variance in a manufacturer’s choice of HR system with a higher degree of isomorphism for labor
contracts than for the use of training and employee involvement programs.
As context is essential for understanding the use of HR systems, the next section introduces
economic and legal contexts in China. China’s economic reforms began in 1978 and economic
development in the decades since has led to remarkable growth in export-oriented enterprises and
increases in technological intensity in the manufacturing sector. China’s recent shift from low-cost
manufacturing towards a value-added economy has significantly influenced China’s employment
relations in the manufacturing sector. Indeed, industrial upgrading towards higher capital intensity
in manufacturing has required changes in HR strategies and practices to ensure workers are capable
and motivated to increase productivity, which is often associated with the use of high-performance
work systems (HPWSs). In addition to this, rising international competitive pressures, partly from
low-wage countries such as Vietnam, has not only profoundly shaped employment relations with
China’s exporters, but also requires HR systems to have considerable flexibility to gain and
maintain competitive advantage. Su and Wright (2012) found that in order to meet these
environmental challenges Chinese firms adopted hybrid HR systems that consisted of both
commitment- and control-based HR practices. However, it remains unclear how environmental
50
and organizational factors contribute in a manufacturing context to the use of labor contracts and
other HR practices to meet the dual goals of cost control through HR flexibility and high
productivity through the adoption of HPWSs.
In line with its national manufacturing strategy to move up the global value chain, the
Chinese government has also embarked on significant changes in labor legislation. China’s 2008
Labor Contract Law represents the most significant recent legislative change since the 1995 Labor
Law, which introduced a labor contract system and ended the 1978 socialist legislation of
guaranteed employment security known as the “iron rice bowl.” The current Labor Contract Law
of 2008 protects employees’ legal rights, generally places more burdens on employers, and
strengthens the employment relationship by clarifying the rights and responsibilities of both
parties. The new legislation shortens the probationary period for new hires to a maximum of two
months from the prior six months. It renders layoffs more difficult and expensive by mandating
that firing 20 or more employees, or 10 percent or more of the total workforce, would require a
30-day advance notice to unions or all employees and would be permissible only in cases of
bankruptcy, severe production difficulties, or significant changes in technology or market
conditions (Chen & Funke, 2009). The new legislation makes illegal certain retention practices
that prevent high-skilled employees from leaving, and it places an emphasis on equal pay for equal
work. The law also provides for fixed-term contracts to be converted to open-ended contracts after
the completion of two fixed-term contracts or 10 years of service (Lee & Liu, 2011). The
Economist Intelligence Unit reported that domestic enterprises have complained about the
51
overemphasis of employees’ rights in the law, and it estimates that the new Labor Contract Law
increases labor costs for employers by about 20 percent (Chen & Funke, 2009). In September 2008
the State Council of China issued a revision to clarify the conditions under which termination
would be legal.
A few empirical studies have examined the implementation of the new Labor Contract Law
and its impacts. Gallagher, Giles, Park, and Wang (2014) found that key determinants of a worker’s
likelihood of being employed under a fixed-term labor contract are education, sector, ownership
type, and location. They noted that migrant workers in the manufacturing sector are 13 to 36
percent more likely to have fixed-term labor contracts compared to workers in other sectors. Using
surveys in the Pearl River Delta area, Li and Freeman (2015) found that the new labor law led to
higher labor contract signing rates. Another study (Lee & Liu, 2011) discovered that Chinese firms
had increased the use of high-performance work practices (HPWP) between 2007 and 2008. They
further noted an increase in the usage of open-ended labor contracts and a decrease in short-term
contracts (i.e., labor contracts that are three years or less) since the introduction of the new Labor
Contract Law.
As the labor contract system has become increasingly a cornerstone of employment
relations in China, it is important to examine the trade-off between HR practices as cost control
measures, on the one hand, and measures to secure a stable and committed core workforce through
52
labor contracts, on the other hand. To do this, more must be known about how environmental and
organizational contingencies affect an employer’s choice in labor contracts in China.
In sum, China manufacturing sector faces both internal and external pressures to its hiring
practices. China’s Labor Contract Law and recent amendments offer greater employee protection
but restrict an employer’s flexibility in certain aspects of employment relations. At the same time,
international competition, especially from low-wage countries and China’s own industrial
upgrading in the manufacturing sector, pushes manufacturers to adopt more capital intensive
production and to increase labor productivity. Given these pressures, this paper examines HR
systems and key environmental and internal contingencies that affect the development of flexible
HR system models in response to the competitive challenges of dynamic international and
domestic markets.
2 THEORY AND HYPOTHESES
The contingency perspective argues that HR systems should align with environmental and
organizational factors in order to achieve high performances. Jackson and Schuler (1995) and later
Jackson et al. (2014) provided a comprehensive conceptual framework that underscores the
important role of the organizational context—technology, structure, size, life cycle stage, and
strategy—and the external context, including legal regulations, culture, politics, unions, labor
markets, and industry characteristics in shaping the HR systems of organizations. To further
explore to what extent context determines HR practices, Dewettinck and Remue (2011) analyzed
empirical studies using CRANET data and concluded that although certain universal best HR
53
practices do exist, context plays a key role in a firm’s choice of HR practices. They suggest that
empirical studies supported the contingency perspective that cultural, socio-political, economic,
and management-related contextual factors are key determinants of HR practices.
It is important to note that few organizations are in static environment. Many organizations
face dynamic environmental changes and demand HR system flexibility to ensure timely adaption
to important changes in the business environment. Wright and Snell (1998) proposed three types
of HRM flexibility: flexibility of HR systems for quick and timely adoption of HR practices,
flexibility of employee skills, and flexibility of employee behaviors. Drawing on their theoretical
framework of types of HR flexibility, I explored flexibility in labor usage, specifically predictors
of use of short-term labor contracts and predictors of contract duration. I also looked at training
and employee involvement programs that influenced a firm’s flexibility in terms of employee skills
and behaviors. Furthermore, I’d like to note that while training and employee involvement
programs mainly influence employees’ ability and opportunity to contribute respectively, these
two programs also significantly shape employees’ behaviors (Jiang, Lepak, Hu, & Baer, 2012).
Because labor contracts and training and employee involvement programs are important
determinants of HR flexibility in organizations, I focused on what influenced decisions to adopt
these HR practices.
While some HR practices are recognized as intractable features of HR systems (e.g.,
compensation systems), other aspects of HR systems, such as training programs, are more readily
54
adjustable to environment changes. Weick and Quinn (1999) conceptualized two types of
organizational change: episodic and continuous change. For instance, change of a compensation
system is likely to be episodic change in that once it is fixed, it is unlikely to change due to the
high cost of transition. By contrast, training and employee involvement programs are likely to
follow patterns of continuous change, as these practices directly influence flexibility in employee
skills and behavior. This paper focuses on HR practices that follow continuous change. Essentially,
flexibility-inducing practices, such as training, are more likely to be contingent on organizational
and environmental influences than more intractable aspects of HR practices, such as compensation
systems. In the China context, signing labor contracts of different durations with different
employees offers employers considerable flexibility in labor usage. The concept of organizational
ambidexterity, or an organization’s ability to achieve efficiency under one environment and to
continuously innovate to meet upcoming challenges (Gibson & Birkinshaw, 2004; Tushman &
O’Reilly, 1996), is an important consideration here, and research has shown that the use of HPWSs
(Patel, Messersmith, & Lepak, 2013) enhances organizational ambidexterity. Training and
employee involvement programs are key components of HPWSs, as they motivate employees with
a broad range of skills and behaviors to meet the demands of a dynamic business environment. In
the following sections, I also draw from contingency theory, analyzing not only an organization’s
choice of HR system, but also its use of labor contracts in a manufacturing context in China.
Subramony (2009) argued that an economic approach based on the rational choice model
is another important influence on an organization’s decision to adopt certain HR practices. Utility
55
analysis has suggested that firms will choose to adopt a certain HR practice if the expected benefits
exceed the costs of implementation. HR flexibility is one way that organizations can save on costs
by terminating or training excessive and less productive hires. Sanchez (1995) outlined two types
of employment flexibility: coordination flexibility and resource flexibility. Coordination flexibility
refers to the extent to which firms can resynthesize business strategy and reconfigure and redeploy
resources. Resource flexibility refers to extent to which a resource can be used in alternative ways,
the cost and difficulty of switching usage from one alternative to another, and the time required
for making the switch. The use of short-term contract duration is one way that employers can gain
greater coordination flexibility. However, manufacturers choosing to sign longer labor contracts
may benefit from the predictability in labor demand and develop HR resource flexibility in that
long-term workers will be capable of filling in positions as needed.
2.1 International Competitive Pressure
High-level institutional factors are significant determinants of an organization’s
implementation of HR systems. For instance, as shown by Lawler, Chen, Wu, Bae, and Bai (2011),
the rate of economic growth is positively related to the use of HPWSs. In China’s export-driven
manufacturing sector, firms face competitive pressure from multiple levels. At the industry level,
manufacturing firms need to compete internationally; at the firm level, they are often subject to
intensive internal competitions. Earlier research has shown that firms that engage in a high degree
of internationalization are more likely to adopt HR practices that enhance innovation than those
that do not (Walsworth & Verma, 2007). During the long period of growth in China’s economy
56
from the early 1990s up until 2013, enterprises that relied heavily on exports, experienced greater
competitive pressure compared to firms that targeted relatively more stable domestic markets.
Theoretically, firms experiencing high external pressures and uncertainties are more likely to adopt
HRM policies that allow flexibility than those that do not have these pressures (Sanchez, 1995;
Wright & Snell, 1998). Signing more short-term labor contracts of three or fewer years is one way
that export-oriented Chinese firms could gain flexibility. I hypothesize that firms in export-
oriented industries that are exposed to intense international competition are more likely to seek
HR flexibility through the use of labor contracts compared to firms focused on domestic markets.
The use of short-term contracts, then, could enhance an employer’s labor flexibility, but only in
the short term. In the long run, short-term contracts may reduce flexibility because an employee’s
completion of two fixed-term contracts will automatically lead to an open-ended labor contract.7
For this reason, I hypothesize the following:
Hypothesis 1(a): International competitive pressure (i) is positively related to the use of
short-term labor contracts and (ii) is negatively related to the weighted average of labor
contract duration.
Greater exposure to international competition means higher demand for product quality
and, thus, for higher production standards. Both conditions require a skilled and engaged
workforce to gain a competitive edge. Firms under international competitive pressures are likely
7 I thank one of the anonymous reviewers on this constructive comment.
57
to face mimetic and normative pressures to adopt HR practices that ensure product quality and
efficiency. To learn from the successes of competitors in the field, manufacturing firms tend to
model best practices, such as training and employee involvement programs, and this knowledge
and professionalization are sources of an organization’s normative isomorphism (Guler, Guillen,
& Macpherson, 2002). In other words, international pressures create incentives for manufacturing
firms to train their employees to acquire knowledge and skills based on global standards, and this
is likely to result in more extensive training and employee involvement programs.
Hypothesis 1(b): International competitive pressure is positively related to firm-level usage
of (i) training and (ii) employee involvement programs.
2.2 Capital Intensity
Capital intensity is a key organizational internal contingency affecting the adoption of HR
practices. According to Lepak and Snell’s (1999) conceptualization of HR architecture, capital
intensive firms are likely to develop human capital internally and to build commitment-based HR
systems because in general, capital intensive firms require less labor than less capital intensive
firms, but need higher quality human capital in operations. The high cost of investing in the training
and development of employees makes long-term labor contracts desirable for capital intensive
firms. Further, Koch and McGrath (1996) argue that employee training and participation are
essential for the successful operation of capital intensive businesses, as in firms such as chemicals,
metal production, and shipbuilding, poorly trained employees have the potential to be dangerous
to the production process. In China, one way to retain a stable workforce is to sign labor contracts
58
with longer durations with core knowledge employees. As capital intensive manufacturing firms
possess a large portion of well-trained employees, they are likely to sign long-term labor contracts.
Therefore, I hypothesize that
Hypothesis 2(a): Capital intensity (i) is negatively related to the incidence of short-term
labor contracts, and (ii) is positively related to the weighted average labor contract
duration.
Consistent with a commitment-based HR configuration, high capital intensity requires both
unique and strategically valuable human capital. Extensive training and employee engagement are
essential to ensure product quality. Firm-specific skills, such as operating particular machines, are
non-transferable from other employers, and firms often need to provide training and develop long-
term employment relationships to fulfill such skill demands. To maximize the return on investing
in employees, firms with high capital intensity are likely to adopt employee involvement programs,
so that employees can actively contribute to a firm’s productivity. There is also empirical support
for a positive interaction effect on labor productivity between HR practices and capital intensity
(Datta, Guthrie, & Wright, 2005; Koch & McGrath, 1996). Capital intensive firms are likely to use
HPWSs as these HR practices benefit them more than they do low capital intensive firms (Datta
et al., 2005). The next hypothesis is as follows:
Hypothesis 2(b): Capital intensity is positively related to the use of (i) training, and (ii)
employee involvement programs.
59
2.3 Firm Size
Firm size is a key determinant of HR practices (e.g., Guthrie, 2001). It can also influence
an employer’s choice of labor contracts. Small firms normally have limited resources to support a
large proportion of long-term employees. Start-up and small-to-medium-sized organizations thus
often adopt HR systems that enable considerable flexibility for survival in dynamic and
competitive environments. One way to achieve flexibility is to use more short-term employees
than long-term employees. In China’s export-oriented manufacturing sector, which faces
fluctuating global demand, short-term contingent workers allow for greater flexibility and better
control of labor costs than long-term contract workers.
While there are benefits to using short-term employees, it is important to weigh potential
drawbacks of employing short-term employees exclusively or predominantly. For example,
organizations may suffer from high voluntary turnover because short-term employees are known
to turn over frequently. The lack of a sustained pool of long-term permanent workers can also lead
to seasonal labor shortages. In China, millions of migrant workers go back to their hometowns for
Chinese New Year, and manufacturers face severe labor shortages after this long national holiday
because most of their workers have signed short-term labor contracts and do not return to work.
Labor shortages faced by manufacturers in the coastal region is further intensified by the fact that
fast economic development in China’s inner provinces has led to workers seeking employment
near their hometowns as opposed to travelling out to the coast. Consequently, manufacturers in the
coastal provinces have had to strike a balance between desirable levels of employee retention and
60
HR flexibility. Larger manufacturers are better off than smaller firms, as they use more long-term
employees, which helps reduce the costs and uncertainty associated with high turnovers and labor
shortages. For smaller manufacturers that have limited resources, flexibility in labor usage
arguably outweighs the costs of using short-term workers. The next hypothesis is:
Hypothesis 3(a): Firm size (i) is negatively related to the incidence of short-term labor
contracts, and (ii) is positively related to the weighted average labor contract duration.
Firm size is generally related to the adoption of HR practices (Van Eerde, Tang, & Talbot,
2008). At early stages of growth, organizations do not have adequate resources to build a complex
HR system. Larger organizations, however, possess the resources to develop and manage their
human resources. Compared to start-ups and smaller organizations, larger firms are more likely to
provide extensive training activities and to offer different employee involvement programs. Matlay
and Addis (2002) argue that larger firms have the resources to support HR departments and in-
house trainers to develop their employees whereas smaller organizations are less likely to have the
resources required for training programs. Similarly, larger manufacturing firms can support a
greater number of employee involvement programs to allow workers to actively contribute to their
organization. However, there is limited empirical support for the hypothesis that firm size is
positively related to training quantity (e.g., Colarelli & Montei, 1996; Van Eerde et al., 2008).
Further empirical examination is needed on the relationship between organizational size and the
61
use of training and employee involvement programs. Drawing on theoretical arguments, I
hypothesize that
Hypothesis 3(b): Firm size is positively related to the use of (i) training, and (ii) employee
involvement programs.
2.4 Unionization
Unions in the West have been characterized by two different faces: the “monopoly face,”
characterized by rent-seeking, and the more positive “voice face” that offers union members
channels to resolve workplace disputes (Freeman & Medoff, 1979), and both of these faces have
played an important role in influencing HRM policies. Unions increase workplace efficiency by
persuading management to rationalize and standardize HRM practices (Verma, 2005). Training
programs are often promoted by unions for the benefit of workers, but training programs also create
long-run benefits for the organization. The predominant form of corporate strategy tends to have
a short-term focus on firm performance and often fails to establish and achieve long-term
organizational goals.
Chinese unions differ from their Western counterparts in several important ways. There is
a significant difference in the ability of Chinese unions to influence an organization’s HR
decisions. Liu (2010) argued that although unions in China are under the umbrella of the official
All China Federation of Trade Unions (ACFTU), there are considerable internal variations,
including three major patterns, namely, (1) “the traditional ACFTU pattern,” (2) “the union
62
association pattern,” and (3) “the regional, industry-based bargaining pattern.” Liu (2010) found
that unions in medium to large enterprises that conformed to the traditional ACFTU pattern had
limited bargaining power. The union association pattern, which is found in small- and medium-
sized enterprises, also appears to have relatively low union bargaining power. Only the regional,
industry-based bargaining pattern in small- and medium-sized enterprises with strong local state
support and tight local labor markets possessed relatively high bargaining power. Because of these
internal variations in bargaining power, union effects found in empirical studies tend to be driven
by powerful and independent unions.
I expect Chinese unions to act similarly to Western unions by influencing employment
relations in several ways. Job security is one of the key priorities for unions. In the Chinese context,
unions ensure job security by pushing employers to sign longer-term labor contracts. Unions also
push management to adopt more training activities to invest in human capital because it helps not
only workers but also organizational interests in the long run. As noted by Verma (2005), there is
well-established empirical evidence that unions have a positive impact on training (e.g.,
Arulampalam & Booth, 1998; Heyes & Stuart, 1998; Osterman, 1995). Indeed, Lee and Liu’s
(2011) survey of service and manufacturing firms in China found that union elections and union
organizations are positively related to high investment in training. Unions are also positively
associated with employee involvement programs, as these programs provide opportunities for
employees to participate in decision making or to make suggestions for improvement and are
positively associated with training. Goll (1991), for instance, found that union presence is
63
positively related to the number of participative programs used by management. Despite the vast
internal variation in bargaining power of Chinese unions, I expect that the overall effect of unions,
which is likely driven by more powerful or autonomous unions in China, is positively associated
with the use of long-term labor contracts and training and employee involvement programs based
on union goals to improve productivity and job security.
Hypothesis 4(a): Unionization (i) is negatively related to the use of short-term labor
contract and (ii) is positively related to the weighted average labor contract duration.
Hypothesis 4(b): Unionization is positively related to the use of (i) training and (ii)
employee involvement programs.
2.5 Ownership
China’s rapid transition from a planned and stated-owned economy to a significant private
economy has given rise to a variety of ownership forms including wholly foreign-owned, Hong
Kong-, Macau-, or Taiwanese-owned, privately-owned, and state-owned enterprises, as well as
hybrid ownership forms through joint ventures. In general, today’s state-owned enterprises remain
significantly different from other types of ownership in several ways. Within state-owned
enterprises, there are two common types: central state-owned enterprises, which are owned by
central government and associated entities, such as the central state-owned assets supervision and
various ministries, and local state-owned enterprises controlled by local province, prefecture,
country, and township governments, such as local state-owned assets and finance bureaus (Wu,
64
Wu & Rui, 2010). State-owned enterprises, especially central state-owned enterprises, generally
concentrate in the monopolistic sectors. As state-owned enterprises often possess considerable
resources, they are able to afford a stable pool of employees.
The resource-based view predicts that state-owned enterprises will use their political
connections with governments to seek governmental support (Claessens, Feijien & Laeven, 2008).
As a result, state-owned enterprises enjoy preferential status in getting bank loans and gaining
access to key resources (Brandt & Li, 2003). According to agency theory, state-owned enterprises,
especially locally owned, have incentives to assist governments to achieve social and political
goals on top of their financial goals (Wu et al., 2010). One such goal is to reduce unemployment
(Jin, Qian & Weingast, 2005). In sum, state-owned enterprises have the resources to afford a
relatively large and stable workforce; they are less likely to use short-term labor contracts and tend
to adopt longer contract durations than non-state-owned enterprises.
State-owned enterprises also tend to have high power distance cultures and a hierarchical
organizational structure. Chen (1993) suggested that China’s state-owned enterprises lag behind
privately-owned firms in terms of management practices, as they tend to have deep bureaucracy
and rules accumulated over time have yet to be broken (Wei & Lau, 2008). For these reasons, I
expect state-owned enterprises to report less training and employee involvement activities than
other ownership types.
65
Hypothesis 5(a): State-owned enterprises (i) are negatively related to the use of short-term
labor contracts and (ii) are positively related to the weighted average labor contract
duration.
Hypothesis 5(b): State-owned enterprises are negatively related to the use (i) training and
(ii) employee involvement programs.
3 METHODS
3.1 Data
This study used matched enterprise and industry-level data on manufacturers in China from
two sources. The enterprise-level data come from an extensive International Labor Organization
(ILO) survey conducted by Lee and Liu (2011), which investigated labor relations and HRM
practices in 600 service and manufacturing enterprises in four major cities in China. This survey,
conducted in the third quarter of 2008, asked HR managers to provide data at two points of time
in 2007 and 2008. This study used a subsample of the 2008 manufacturing sector data. I first coded
the manufacturing enterprises into three-digit North American Industry Classification System
(NAICS) subsectors.8 I then matched the enterprise-level data with the corresponding industry
level data from the China Statistical Yearbook.
8 The 17 manufacturing subsectors are as follows with the number of firms in bracket: 313, Food (22); 312, Beverage
and tobacco product; 314, Textile product mills (39); 315, Apparel (39); 316, Leather and allied product (11); 322,
Paper product (10); 323, Printing and related support activities (5); 325, Chemical (39); 326, Plastics and rubber
products (18); 327, Non-metallic mineral product (4); 331, Primary metal (14); 332, Fabricated metal product (30);
66
3.2 Measures
The first two dependent variables were measured as, respectively, percentage of employees
on short-term labor contracts (i.e., three years or less) and weighted average labor contract
duration. Percentage of short-term labor contracts was computed as the number of employees
with three-year or less labor contracts divided by the total number of employees and expressed as
a percentage. Long-term labor contracts in this study included five-year, ten-year, and open-ended
labor contracts. Weighted average labor contract duration was computed as the average duration
of all labor contracts within a particular firm weighted by the number of employees.9
The other two dependent variables were indices of HR practices. The index of training is
the mean of four self-report items, measuring the use of extensive training, formal training systems,
cross-job/multi-skill training, and sufficient training funds based on a scale of 1 to 7. The reliability
coefficient as measured by Cronbach’s alpha for the training index is 0.788. The index of employee
involvement programs is the mean of five items, measuring the use of formal employee suggestion
systems, participatory teams, open communications between employees and supervisors on job-
333, Machinery (34); 334, Computer and electronic product (31); 335, Electrical equipment, appliance and component
(38); 336, Transportation equipment (2); 337, Furniture and related product (3).
9 Since there is no data on the average labor contract duration of employees with open-ended labor contracts, I assigned
15 years to these employees following the economics convention of assigning a value that is 50% more than the highest
value, which was 10 years.
67
related issues, formal information sharing systems, and formal complaint resolution systems based
on a self-report scale of 1 to 7. Cronbach’s alpha of this index is 0.901.
While these HR measures were collected in 2008, I used lagged industry level data to
ensure the causal temporal order. Specifically, I used end-of-2007 data for percent export and
capital intensity. International competitive pressure was operationalized as the percentage of
export of a particular industry. Percentage export equaled to exports over total sales in the
particular manufacturing industries expressed as a percentage. A higher percentage of exports in a
particular industry was used as a proxy for greater exposure to international competition. In
addition, theoretically high international exposure was generally associated with high international
competitive pressures unless firms established a dominant position in the international market. In
the case of China’s manufacturing sector, none of the subsectors was able to achieve a dominant
position. Therefore, I assumed that a high percentage of export was a proper measure for the degree
of international competitive pressures faced by Chinese manufactures. By this measure,
manufacturing subsectors most exposed to international competition were computer and electronic
products (68%), furniture (43%), leather and allied products (43%), and clothing (42%). Notably,
none of the Chinese firms are able to establish a dominant position in the global market that high
exposure to international competition could mean low competition pressure. Therefore, percentage
of exports is an appropriate proxy for the international competitive pressure faced by the
manufacturing firms.
68
Similar to Datta et al. (2005)’s measure of capital intensity, I operationalized capital
intensity as the fixed assets required to generate a certain level of sales. Because of a lack of
enterprise-level data, I used industry-level capital intensity as a proxy. I measured the capital
intensity of manufacturing firms as the ratio of total fixed assets to total sales of each industry
defined at the three-digit NAICS level in the manufacturing sector. As three-digit NAICS offers a
relatively detailed breakdown of manufacturing industries, I assumed that industry-level capital
intensity offered a proper proxy for a firm’s capital intensity.
Other variables were measured at the firm level. Union was a dichotomous variable
denoting union presence at the plant level. Business strategy was measured in three categories:
cost strategy, quality strategy, and innovation strategy. The survey asked the respondents to
indicate which one of the three factors — (1) labor cost, (2) product or service quality, and (3)
product or service innovation — was most important for the enterprise’s competitive advantage.
Size was measured as the total number of employees. The natural logarithm of size was used in
regressions. Ownership was a categorical variable capturing four different types of ownership,
namely, privately-owned enterprises (POEs), state- or collectively owned enterprises (SOEs),
Hong Kong-, Macau- or Taiwan-owned enterprises and joint ventures (HMTJV), and foreign-
owned enterprises (FOEs), which included wholly foreign-owned enterprises and joint ventures
with foreign investors.
69
4 RESULTS
4.1 Descriptive Statistics and Bivariate Correlations
Our final sample consisted of 313 enterprises. Table 1 presents the means, standard
deviations, and bivariate correlation for variables. The average size of the enterprises was 660
workers. The minimum and maximum size of the companies was 4 and 12,480 people respectively.
The average percentage of employees with short-term labor contracts was 73.8 percent. The
average contract duration of surveyed manufacturers was five years. Beginning with simple
bivariate correlations of variables, as shown in Table 1, capital intensity was associated with a low
percentage of exports (r = -0.739, p<0.01), suggesting that export-oriented manufacturing
subsectors are largely low in capital intensity.
Export percentage—an indicator of exposure to international competitive pressures—was
positively related to the percentage of short-term labor contracts (r = 0.251, p<0.01) and negatively
related to average labor contract duration (r = -0.251, p<0.01). This indicates initial support for
Hypothesis 1(a). In addition, higher capital intensity was associated with lower usage of short-term
labor contracts (r = -0.257, p<0.01), longer average contract durations (r = 0.263, p<0.01), and
more training activities (r = 0.152, p<0.01), providing initial support for hypotheses 2(a) and
2(b)(i).
Supporting hypotheses 3(a) and 3(b), firm size was negatively related to the use of short-
term labor contracts (r=-0.159, p<0.01) and positively related to contract durations (r=0.170,
p<0.01), training (r=0.269, p<0.01), and employee involvement programs (r=0.151, p<0.01).
70
Hypotheses 4(a), 4(b), 5(a), and 5(b) were also supported by correlational analysis. Union presence
was negatively correlated to percentage of exports (r=-0.216, r<0.01) and positively related to
average contract durations (r=0.248, p<0.01) and the training index (r=0.170, p<0.01). State-
owned enterprises were negatively related to the use of short-term contracts (r=-0.403, p<.01) and
positively related to average contract duration (r=0.429, p<.01). State-owned enterprises were
negatively related to employee involvement programs (r=-0.165, p<.01). To further examine if
relationships still hold after controlling for other characteristics, I proceeded to a multivariate
analysis.
4.2 Hierarchical Linear Model Analyses
I used hierarchical linear modelling (HLM; Raudenbush & Bryk, 2002) to estimate how
industry-level international competitive pressures and capital intensity affected manufacturers’ use
of labor contracts and HR systems, as individual firms are nested in their particular industry or
manufacturing subsector and a key advantage of HLM is to partition the variance explained at each
level of analysis (i.e., between- and within-industry components) (Cullen, Parboteeah, & Hoegl,
2004). Before proceeding to HLM analysis for hypotheses testing, I first examined whether
between-firm/within-industry variance existed for the four dependent variables by testing four
respective null models, following Hofmann, Griffin, and Gavin (2000). For each null model, I only
included industry identification; predictors in both firm and industry levels were not specified.
Results of the null models showed significant between-industry variance for all four dependent
variables: percentage of short-term labor contract (χ2 (17, 313) = 13.33, p<0.001), average contract
71
durations (χ2 (17, 313) = 11.80, p<0.001), training programs (χ2 (17, 313) = 8.61, p<0.01), and
employee involvement programs (χ2 (17, 313) = 5.84, p<0.01). I then calculated the intraclass
correlations (ICC [1]) to estimate the proportion of total variance explained by between-industry
factors. The ICC (1) for short-term labor contracts was 0.071, suggesting that 7.1 percent of the
total variance in the use of short-term labor contracts resided between industries and that 92.9
percent of the variance was accounted for by within-industry between-firm differences. The ICC
(2) values for average contract durations and training and employee involvement programs were
0.065, 0.054 and 0.045 respectively. Overall, the results for the four null hypotheses revealed that
HLM would be an appropriate analytical strategy to estimate the impacts of predictors on all four
dependent variables.
I present the results of the HLM analysis in Tables 2 through 5.10 Empirical findings on
predictors of short-term labor contracts are shown in Table 2 and for average contract durations in
Table 3. In both Tables 2 and 3, I first entered percent export (capital intensity) in Model 1a and
Model 2a and subsequently added training and employee involvement program indexes in Model
1b and Model 2b. The effect of percent export on labor contracts was robust even after controlling
for the effect of HR programs. The effects dropped only slightly from 0.317 (p<.01) to 0.303
10 Due to multicollinearity between percentage of export and capital intensity, they are entered individually to the
regression analyses.
72
(p<.01) for the percentage of short-term contracts in Table 2 and from -0.0391 (p<.01) to -0.0383
(p<.01) for contract durations in Table 3, but remained significant in both cases.
Table 4 shows regression estimates on training programs and Table 5 on employee
involvement programs. For both tables, I entered percent export in Model 1, then separately
entered capital intensity in Model 2, and lastly entered both factors in Model 3. In Table 4, the
employee involvement program index was entered as a control variable in Models 1b, 2b, and 3b.
Capital intensity was found to be a robust predictor of training programs; however, industry level
factors, percent export, and capital intensity did not have a significant effect on employee
involvement programs.
Table 6 summarizes the results in respect of all the hypotheses. There were some interesting
findings. Industry level factors—percentage export and capital intensity—significantly influenced
the use of labor contracts, but did not affect HR practices, except for the positive effect of capital
intensity on training. Organizational factors—size, unionization, ownership—all significantly
shaped the use of labor contracts, but not all affected the use of training and employee involvement
programs. Size was a significant determinant for both HR programs, supporting Hypothesis 3(b).
Union presence was not significantly related to the adoption of any of the HR programs measured
in this study; however, unionization was significantly associated with a smaller share of short-term
labor contracts and longer contract durations. Powerful and autonomous unions would be expected
to make exactly such demands. The results suggest that although Chinese unions have limited
73
influence on an employer’s choice of HR programs, they are capable of pushing for long-term
labor contracts in the interest of job security for their members. The results show that industry
level variance accounts for 7.1 percent, 6.5 percent, 5.4 percent, and 4.5 percent of the total
variance in manufacturers’ use of short-term contracts, average contract durations, training
programs, and employee involvement programs respectively. The findings provide empirical
evidence of industry-level convergence of HR systems and the presence of industry-level
isomorphism. However, results also show that most of the heterogeneity in HR systems
(approximately 93 to 95 percent) are due to organization-level differences.
5 LIMITATIONS
I acknowledge that this study is subjected to several limitations that require future
investigation. Although industry level data were collected prior to the dependent variables, which
were firm-level data to ensure causal precedence, firm-level independent variables were cross-
sectional and causality could not be established through the empirical analysis and are only inferred
from theoretical arguments. Future research can improve the rigor of the study by avoiding the use
single-source self-reporting surveys.
Our study focused on the 17 different industries within the manufacturing sector, and it
may not be generalizable to other services. With this said, I expect similar findings. As suggested
by institutional theory and shown in this study, the majority of the variance in HR systems occurs
at the firm level and only a small percent of the total variance happens at the industry level. I also
expect the percentage of total variance in labor contract durations explained by service industry-
74
level factors are likely to be higher than that in training and employee involvement programs, as
institutional influences from the industry level are more likely to occur in labor contracts than in
HR programs. Nonetheless, future studies may collect data from service firms and explore the
differences in terms of institutional influences. The results as tabulated in Table 6, with few
exceptions such as Hypothesis 5, are consistent with the theoretical predictions and empirical
results of studies using data collected in the West, confirming the generalizability of this empirical
study.
Our study also has several data limitations. First, I do not have data on firm-level capital
intensity and percent exports, although industry-level capital intensity shows robust effects on
most of the dependent variables. Second, findings from the 313 manufacturing firms in four major
cities in this study may not be generalizable to all parts of China given the substantial heterogeneity
in labor relations across China. Third, this study included only a few of the key variables that can
be measured. This two-level model is a simplified model of multilevel HR antecedents. Although
I demonstrated the important multilevel influences on labor contracts and HR systems, future
studies may include more relevant factors in different contexts to model the complexity of
employers’ HR decisions.
6 CONCLUSIONS
Building on Wright and Snell’s (1998) conceptualization of HR flexibility, this two-level
analysis focused on the effects of key environmental and organizational contingencies on
manufacturers’ use of flexibility-inducing HR practices in China. This paper focused on the use of
75
labor contracts, training programs, and employee involvement programs. Signing short-term labor
contracts offers flexibility in labor usage in the short run and training and employee involvement
programs are essential for firms to gain flexibility in employees’ skill and behavioral repertoires.
Focusing on these aspects of HR systems, I offered three main theoretical and empirical
contributions to the research literature.
Our main theoretical contribution to the field is to use both contingency theory and
institutional theory to explain and explore multilevel predictors of HR flexibility. I argue and find
support for the concept that internal competition and industry level factors are joint antecedents of
flexibility-inducing HR practices. This study is among the first few studies in the field to focus on
industry-level predictors of HR practices and offers a robust estimation of the percentages of
variance in the use of HR practices that is accounted for by industry and firm-level factors. I
conducted a multilevel analysis to examine the extent to which industry-level factors influence the
adoption of labor contracts and HR programs, and found empirical support for industry level
isomorphism as predicted by institutional theory. Specifically, I found that industry-level factors
account for approximately 5 to 7 percent of the total variances in an organization’s use labor
contracts, training programs, and employee involvement programs while the rest of the variances
are at the organization level. As Lawler and colleagues (2011) suggested, it is important to take
higher-level institutional factors into account when considering the antecedents of HPWSs. The
findings show that the effect of industry level isomorphism is stronger for the use of labor contracts
than for HR programs, which are influenced more by organizational characteristics. Overall, this
76
multilevel paper offers insights on the relative importance of firm and industry level antecedents
of flexibility-inducing HR systems. The results offer insights into the degrees of HR system
divergence and convergence in the manufacturing sector in China. Although industry isomorphism
plays a role in shaping an organization’s choice of HR practices, the effect is relatively small
compared to the effects of internal competition and organizational contingencies, which account
for 93 to 95 percent of the variance in the adoption of flexibility-inducing HR practices.
Our study also contributes to the literature on HR flexibility by examining the key
contributors of an organization’s choice of flexibility-inducing HR practices, and I are among the
first few studies to examine the important determinants of labor contracts in China. Given the
important yet differentiated role of the labor contract system in China compared to other Western
nations, it is important to examine the antecedents and rationale of a firm’s choice of HR practices,
a choice that is usually based on a calculated trade-off between short-run flexibility and the
retention of workers. I used a strategic HRM framework to analyze the key determinants of labor
contract durations in the manufacturing industry in China. I showed that international competitive
pressures, capital intensity, firm size, unionization, and ownership are all significant predictors of
a firm’s percentage of short-term labor contracts and average contract durations. However, not all
factors were significantly related to the use of training and employee involvement programs.
Training programs were mainly determined by firm characteristics, including capital intensity,
firm size, and ownership. Only firm size and ownership were significantly associated with
employee involvement programs. I also found that Chinese unions had limited power in
77
influencing an employer’s choice of HR programs but that powerful unions were capable of
fostering job security by signing labor contracts with longer durations. These results, which are
not all consistent with findings in Western countries, underscore the importance of conducting
indigenous research in China, and add to HR flexibility literature, specifically the key concept of
labor usage flexibility, which, in China, is reflected in a firm’s choice of labor contracts with
various durations.
Finally, this study contributes to the understanding of higher-level antecedents, namely
international competitive pressures, on an individual organization’s choice of HR practices. I found
that manufacturers in China that face high international competitive pressures tend to cluster in
industries with low capital intensity. Manufacturers under intense international competition often
face the dual HR priorities of high flexibility and high productivity. International competition
requires employers to gain flexibility in HR systems to meet fluctuating international demands and
also forces manufacturers to be more productive, which is often achieved through the use of
HPWSs. The results suggest that flexibility appears to be a higher HR priority than high
performance for manufacturing firms in China. The analysis further shows that manufacturers with
intense international competition tend to use more short-term contracts and have lower average
contract durations than manufacturers without this pressure. Notably, however, international
competitive pressure does not significantly affect the use of training and employee involvement
programs. This result underscores the importance of internal competition, as firms tends to
differentiate themselves from their competitors through their choice of HR practices as indicated
78
by a considerable divergence of HR practices among firms in the same three-digit NAICS industry
code. The results support the theoretical prediction of a resource-based view that an HR system
can serve as a unique source for an organization’s sustained competitive advantage (Becker &
Gerhart, 1996).
Although the analysis is based on a simplified model of the complex environmental and
organizational determinants of labor contracts and HR programs, this study advances the strategic
HRM literature in three main ways: estimating the relative effect of industry-level isomorphism,
examining the key determinants of an employer’s use of labor contracts in China, and analyzing
the impact of international competitive pressures. This study demonstrates that industry-level
factors, although smaller than the effect of organizational contingencies, are significant
determinants of HR systems and should be investigated further with a more comprehensive dataset.
79
REFERENCES
Arulampalam, W., & Booth, A. L. 1998. Training and labor market flexibility: is there a trade-
off. British Journal of Industrial Relations, 36(4): 521-536.
Barney, J. B. 1995. Looking inside for competitive advantage. The Academy of Management
Executive, 9(4): 49-61.
Becker, B., & Gerhart, B. 1996. The impact of human resource management on organizational
performance: progress and prospects. Academy of Management Journal, 39(4): 779-801.
Brandt, L., & Li, H. 2003. Bank discrimination in transition economies: ideology, information,
or incentives? Journal of Comparative Economics, 31(3): 387-413.
Campbell, B. A., Coff, R., & Kryscynski, D. (2012). Rethinking sustained competitive advantage
from human capital. Academy of Management Review, 37(3): 376-395.
Chen, Q. 1993. Enterprise management in china has entered a new stage. International
Business Administration, 3(1): 4-8.
Chen, Y. F., & Funke, M. 2009. China’s new labour contract law: no harm to employment?’
China Economic Review, 20(3): 558-572.
Claessens, S., Feijen, E., & Laeven, L. 2008. Political connections and preferential access to
finance: the role of campaign contributions. Journal of Financial Economics, 88(5): 554-
80.
Colarelli, S. M., & Montei, M. S. 1996. Some contextual influences on training utilization.
Journal of Applied Behavioral Science, 32(3): 306-323.
Cullen, J. B., Parboteeah, K. P., & Hoegl, M. 2004. Cross-national differences in managers’
willingness to justify ethically suspect behaviors: a test of institutional anomie theory.
Academy of Management Journal, 47(3): 411-421.
Datta. D. K., Guthrie, J. P., & Wright, P. M. 2005. Human resource management and labor
productivity: Does industry matter? Academy of Management Journal, 48(1): 135-145.
Dewettinck, K., & Remue, J. 2010. Contextualizing HR in comparative research: The role
of the Cranet network. Human Resource Management Review, 21(1): 37-49.
80
DiMaggio, P. J., & Powell, W. W. 1983. The iron cage revisited: Institutional isomorphism
and collective rationality in organizational fields. American Sociological Review, 48(2):
147-160.
Freeman, R. B., & Medoff, J. L. 1979. The two faces of unionism. Public Interest, 57: 69- 93.
Gallagher, M., Giles, J., Park, A., & Wang, M. (2014). China’s 2008 labor contract law:
Implementation and implications for China’s workers. Human Relations, 68(2): 197-235.
Gibson, C. B., & Birkinshaw, J. 2004. The antecedents, consequences, and mediating role
of organizational ambidexterity. Academy of Management Journal, 47(2): 209-226.
Goll, I. 1991. Environment, corporate ideology, and employee involvement programs.
Industrial Relations, 30(1): 138-149.
Guler, I., Guillen, M., & Macpherson, J. 2002. Global competition, institutions, and the
diffusion of organizational practices: the international spread of ISO 9000 quality
certificates. Administrative Science Quarterly, 47(2): 207-232.
Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity: Evidence from
New Zealand. Academy of Management Journal, 44(1): 180-190.
Heyes, J., & Stuart, M. 1998, Bargaining for skills: trade unions and training at the
workplace. British Journal of Industrial Relations, 36(3): 459-467.
Hofmann, D.A., Griffin, M.A., and Gavin, M.B. 2000, The application of hierarchical linear
modeling to organizational research, in Multilevel Theory, Research, and Methods in
Organizations, 11, eds. K.J. Klein and S.W.J. Kozlowski, San Francisco, CA: Jossey-
Bass, 467-511.
Jin, H., Qian, Y. & Weingast, B. R. 2005. Regional decentralization and fiscal incentives:
federalism Chinese style, Journal of Public Economics, 89(9): 1719-1742.
Ketkar, S., & Sett, P. K. 2009. HR flexibility and firm performance: Analysis of a multi-level
causal model. International Journal of Human Resource Management, 20(5): 1009-1038.
Ketkar, S., & Sett, P. K. 2010. Environmental dynamism, human resource flexibility, and firm
performance: Analysis of a multi-level causal model. International Journal of Human
Resource Management, 21(8): 1173-1206.
81
Koch, M. J., & McGrath, R. G. 1996. Improving labor productivity: human resource
management policies do matter. Strategic Management Journal, 17(5): 335-354.
Lawler, J. J., Chen, S. J., Wu, P. C., Bae, J., & Bai, B. 2011. High-performance work systems
in foreign subsidiaries of American multinationals: An institutional model. Journal of
International Business Studies, 42(2): 202-220.
Lee, C. H. & Liu, M. W. 2011, Measuring the Effects of the Collective Voice Mechanism and
the Labor Contract Law: A Survey of Labor Relations and Human Resources Management
Practices in 600 Enterprises in China. Beijing, ILO Regional Office for China and
Mongolia, 1-78.
Lepak, D. P., & Snell, S. A. 1999. The Human resource architecture: toward a theory of human
capital allocation and development. Academy of Management Review, 24(1): 31-48.
Lepak, D. P., Takeuchi, R., & Snell, S. A. 2003. Employment flexibility and firm performance:
Examining the interactive effects of employment mode, environmental dynamism and
technological intensity. Journal of Management, 29(5): 681-703.
Li, X., & Freeman, R. B. 2015. How does China's new labour contract law affect floating workers?
British Journal of Industrial Relations, 53(4): 711-735.
Liu, M. W. 2010. Union organizing in china: still a monolithic labor movement. Industrial
and Labor Relations Review, 64 (1): 30-52.
Matlay, H., & Addis, M. 2002. Competence-based training, vocational qualifications and learning
targets: some lessons for the Learning and Skills Council. Education+ Training, 44(6):
250-260.
Jackson, S. E., & Schuler, R. S. 1995. Understanding human resource management in the context
of organizations and their environment. Annual Review of Psychology, 46: 237-264.
Jackson, S. E., Schuler, R. S. & Jiang, K. F. 2014. An aspirational framework for strategic
human resource management. The Academy of Management Annals, 8(1): 1-56.
Jiang, K., Lepak, P. D., Hu, J., & Judith, C. B. 2012. How does human resource management
influence organizational outcomes? A meta-analytic investigation of mediating
mechanisms. Academy of Management Journal, 55(6): 1264-1294.
82
Osterman, P. 1995. Skill, training, and work organization in American establishments. Industrial
Relations, 34(2): 125-146.
Paauwe, J., & Boselie, P. 2003. Challenging 'strategic HRM' and the relevance of the
institutional setting. Human Resource Management Journal, 13(3): 56-70.
Patel, P. C., Messersmith, J. G., & Lepak, D. P. 2013. Walking the tightrope: an assessment
of the relationship between high-performance work systems and organizational
ambidexterity. Academy of Management Journal, 56(5): 1420-1442.
Raudenbush, S. W., & Bryk, A. S. 2002. Hierarchical Linear Models, Applications and Data
Analysis Methods, London: Sage Publications.
Verma, A. 2005. What do unions do to the workplace? Union effects on management and the
HRM policies. Journal of Labor Research, 26(3): 415-449.
Van Eerde, W., Tang, K. C. S., and Talbot, G. 2008. The mediating role of training utility in the
relationship between training needs assessment and organizational effectiveness.
International Journal of Human Resource Management, 19(1): 63-73.
Sanchez, R. 1995. Strategic flexibility in product competition. Strategic Management Journal,
16(1): 135-159.
Su, Z.-X., & Wright, P.M. 2012. The effective human resource management system in
transitional china: a hybrid of commitment and control practices. International
Journal of Human Resource Management, 23(10): 2065-2086.
Subramony, M. 2009. A meta-analytic investigation of the relationship between HRM bundles and
firm performance. Human Resource Management, 48(5): 745-768.
Tushman, M. L., & O’Reilly, C. A. III. 1996. Ambidextrous organizations: managing
evolutionary and revolutionary change. California Management Review, 38(4): 8-30.
Walsworth, S., & Verma, A. 2007. Globalization, human resource practices and innovation:
Recent evidence from the Canadian workplace and employee survey. Industrial Relations:
A Journal of Economy and Society, 46(2): 222-240.
Way, S. A., Tracey, J. B., Fay, C. H., Wright, P. M., Snell, S. A., Chang, S., & Gong, Y. 2015.
Validation of a multi-dimensional HR flexibility measure. Journal of Management, 41(4):
1098-1131.
83
Wei, L.Q., & Lau, C.M. 2008. The Impact of market orientation and strategic hrm on firm
performance: the case of Chinese enterprises. Journal of International Business Studies,
39(6): 980-995.
Weick, K. E., & Quinn, R. E. 1999. Organizational change and development. Annual Review
of Psychology, 50(1): 361-386.
Wright, P. M., & Snell, S. A. 1998. Toward a unifying framework for exploring fit and
flexibility in strategic human resource management. Academy of Management Review,
23(4): 756-772
Wu, W., Wu, C., & Rui, O. M. 2012. Ownership and the value of political connections: Evidence
from China. European Financial Management, 18(4): 695-729.LE 1. Descriptive
Statistics and Bivariate Correlations for Variables
84
TABLE 1. Descriptive Statistics and Bivariate Corrlations for Variables
85
TABLE 2. Multilevel Predictors of Percentage of Short-Term Labor Contracts
Model 1a Model 1b Model 2a Model 2b
Percent export 0.317**
(0.113)
0.303**
(0.114)
Capital intensity -51.032***
(12.892)
-50.299***
(13.032)
HR practices
Index of training -1.666
(1.980)
-1.312
(1.981)
Index of employee involvement 2.316
(2.153)
2.715
(2.140)
Union -8.198+
(4.828)
-8.068+
(4.832)
-8.197+
(4.815)
-8.159+
(4.820)
Business strategy
Ref: Cost strategy
Quality strategy 6.300
(4.588)
6.048
(4.596)
6.537
(4.569)
6.175
(4.574)
Innovation strategy 3.075
(4.983)
2.630
(4.999)
3.151
(4.939)
2.590
(4.952)
Log of firm size -3.526*
(1.372)
-3.403*
(1.409)
-3.343*
(1.353)
-3.332*
(1.391)
Ownership
Ref: POEs
FOEs -4.601
(4.602)
-4.829
(4.602)
-4.573
(4.583)
-4.901
(4.582)
SOEs -27.100***
(4.645)
-26.569***
(4.666)
-27.746***
(4.592)
-27.027***
(4.624)
HMTJVs 4.237
(5.022)
3.910
(5.026)
4.972
(5.003)
4.662
(5.008)
Constant
96.145***
(8.707)
91.875***
(10.444)
141.834***
(12.277)
135.534***
(13.372)
Random intercept variance 1.363*
(0.669)
1.378*
(0.658)
-14.211*
(5.613)
-11.250+
(5.989)
Variance for residuals 3.391***
(0.041)
3.388***
(0.041)
3.393***
(0.040)
3.390***
(0.040)
N=313; +p<.10; * p<0.05; ** p<0.01; *** p<0.001.
86
TABLE 3. Multilevel Predictors of Weighted Average Labor Contract Duration
Model 1a Model 1b Model 2a Model 2b Percentage export -0.0391**
(-2.875)
-0.0383**
(-2.754)
Capital intensity 6.520***
(4.119)
6.561***
(4.096)
HR practices
Index of training 0.070
(0.288)
0.0220
(0.092)
Index of employee
involvement
-0.199
(-0.750)
-0.208
(-0.790)
Union 1.344*
(2.260)
1.347*
(2.260)
1.343*
(2.273)
1.361*
(2.297)
Business strategy
Ref: Cost strategy
Quality strategy -0.867
(-1.535)
-0.838
(-1.479)
-0.906
(-1.615)
-0.864
(-1.537)
Innovation strategy -0.584
(-0.953) -0.543
(-0.882) -0.603
(-0.995) -0.546
(-0.897) Log of firm size 0.479**
(2.840)
0.481**
(2.770)
0.459**
(2.764)
0.475**
(2.777)
Ownership
Ref: POEs
FOEs 0.207
(0.366)
0.229
(0.404)
0.169
(0.301)
0.205
(0.365)
SOEs 3.426***
(5.988)
3.374***
(5.862)
3.469***
(6.155)
3.396***
(5.975)
HMTJVs -0.791
(-1.279)
-0.782
(-1.261)
-0.883
(-1.438)
-0.876
(1.423)
Constant 1.960+
(1.839)
2.445+
(1.905)
-3.848*
(-3.903)
-3.236*
(-1.969)
Random intercept
variance
-0.870
(-1.084)
-0.810
(-1.087)
-23.345***
(-3.903)
-27.701***
(-5.201)
Variance for residuals 1.297***
(31.856)
1.296***
(31.797)
1.295***
(32.412)
1.294***
(32.377)
N=313; +p<.10; * p<0.05; ** p<0.01; *** p<0.001.
87
TABLE 4. Multilevel Predictors of Training Programs
Model 1a Model 1b Model 2a Model 2a Model 3a Model 3b
Percentage Export -0.003 -0.005 0.009+ 0.002
(-0.573) (-1.400) (1.833) (0.619)
Capital Intensity 1.111* 1.022** 2.186** 1.265*
(1.984) (2.710) (3.241) (2.327)
Employee
Involvement
Programs
0.657*** 0.655*** 0.651***
(13.389) (13.470) (13.296)
Average Contract
Duration
-0.006 0.003 -0.009 0.001 -0.007 0.001
(-0.383) (0.249) (-0.524) (0.092) (-0.428) (0.108)
Union 0.249 0.171 0.258 0.184 0.268 0.184
(1.440) (1.230) (1.496) (1.328) (1.550) (1.330)
Business Strategy
Ref: Cost Strategy
Quality 0.199 0.062 0.178 0.039 0.19 0.044
Strategy (1.212) (0.475) (1.087) (0.299) (1.162) (0.333)
Innovation 0.192 0.008 0.177 -0.01 0.188 -0.006
Strategy (1.077) (0.056) (0.998) (-0.069) (1.070) (-0.041)
Log of Firm 0.210*** 0.150*** 0.212*** 0.147*** 0.206*** 0.145***
Size (4.239) (3.772) (4.316) (3.737) (4.205) (3.668)
Ownership
Ref: POEs
FOEs 0.146 0.026 0.138 0.015 0.13 0.013
(0.895) (0.196) (0.844) (0.118) (0.795) (0.102)
SOEs -0.165 0.011 -0.178 0.012 -0.169 0.02
(-0.947) (0.079) (-1.027) (0.087) (-0.971) (0.142)
HMTJVs -0.224 -0.205 -0.229 -0.216 -0.244 -0.224
(-1.251) (-1.433) (-1.284) (-1.511) (-1.359) (-1.561)
Constant 2.702*** 0.478 1.796*** -0.376 0.767 -0.601
(8.377) (1.594) (3.547) (-0.982) (1.184) (-1.139)
Random intercept
variance
-1.479*** -2.087*** -1.850** -17.259 -13.465* -24.306***
(-4.123) (-3.654) (-3.217) (-0.008) (-2.582) (-3.938)
Variance for
residuals
0.049 -0.169*** 0.051 -0.167*** 0.057 -0.167***
(1.197) (-4.156) (1.252) (-4.167) (1.419) (-4.183)
N=313; +p<.10; * p<0.05; ** p<0.01; *** p<0.001.
88
TABLE 5. Multilevel Predictors of Employee Involvement Programs
Model 1a Model 1b Model 2a Model 2a Model 3a Model 3b
Percentage
Export
0.003 0.004 0.010* 0.005
(0.814) (1.590) (2.190) (1.393)
Capital 0.263 -0.304 1.400* 0.203
Intensity (0.504) (-0.862) (2.231) (0.401)
Training 0.557*** 0.560*** 0.554***
(13.588) (13.470) (13.296)
Average
Contract
Duration
-0.012 -0.008 -0.015 -0.01 -0.013 -0.009
(-0.792) (-0.701) (-0.978) (-0.790) (-0.861) (-0.750)
Union 0.124 -0.021 0.128 -0.021 0.129 -0.02
(0.775) (-0.163) (0.801) (-0.162) (0.807) (-0.154)
Business Strategy
Ref: Cost Strategy
Quality 0.22 0.123 0.197 0.111 0.223 0.119
Strategy (1.456) (1.016) (1.302) (0.914) (1.482) (0.988)
Innovation 0.295* 0.197 0.277* 0.187 0.297* 0.194
Strategy (1.799) (1.516) (1.684) (1.434) (1.830) (1.490)
Log of Firm 0.091* -0.02 0.096* -0.016 0.094* -0.02
Size (1.994) (-0.542) (2.111) (-0.428) (2.077) (-0.538)
Ownership
Ref: POEs
FOEs 0.19 0.109 0.187 0.112 0.179 0.107
(1.257) (0.908) (1.235) (0.925) (1.186) (0.887)
SOEs -0.277* -0.198 -0.294* -0.214* -0.289* -0.196
(-1.723) (-1.538) (-1.828) (-1.668) (-1.800) (-1.528)
HMTJVs -0.02 0.109 -0.015 0.122 -0.031 0.104
(-0.123) (0.827) (-0.089) (0.920) (-0.188) (0.787)
Constant 3.359*** 1.840*** 3.243*** 2.156*** 2.116*** 1.676***
(11.555) (7.442) (6.899) (6.465) (3.513) (3.501)
Random
intercept
-1.841*** -17.248 -1.912** -17.826** -3.56 -22.160***
variance (-3.592) (-0.010) (-3.241) (-3.083) (-0.322) (-4.119)
Variance for -0.029 -0.247*** -0.027 -0.245*** -0.024 -0.248***
residuals (-0.706) (-6.189) (-0.661) (-6.118) (-0.588) (-6.195)
N=313; +p<.10; * p<0.05; ** p<0.01; *** p<0.001.
89
TABLE 6. Summary of Hypotheses and Findings
Hypothesis Hypothesized relationships Empirical
findings
supported?
1(a)(i) International competitive pressure
Percentage (short-term labor contract) + Yes
1(a)(ii) International competitive pressure Average
labor contract duration - Yes
1(b) (i) International competitive pressure Training
programs + No
1(b)(ii) International competitive pressure
Employee involvement programs + No
2(a)(i) Capital intensity Percentage (short-term
labor contract) - Yes
2(a)(ii) Capital intensity Average labor contract
duration + Yes
2(b) (i) Capital intensity Training programs + Yes
2(b)(ii) Capital intensity Employee involvement
programs + No
3(a)(ii) Firm size Percentage (short-term labor
contract) - Yes
3(a)(ii) Firm size Average labor contract duration + Yes
3(b) (i) Firm size Training programs + Yes
3(b)(ii) Firm size Employee involvement programs + Yes
4(a)(i) Union Short-term labor contract - Yes
4(a)(ii) Union Average labor contract duration + Yes
4(b)(i) Union Training programs + No
4(b)(ii) Union Employee involvement programs + No
5(a)(i) State-owned enterprises Percentage (short-
term labor contract) - Yes
5(a)(ii) State-owned enterprises Average labor
contract duration + Yes
5(b)(i) State-owned enterprises Training programs - No
5(b)(ii) State-owned enterprises Employee
involvement programs - Yes
90
CHAPTER 3 THE IMPACT OF HUMAN RESOURCE MANAGEMENT
DIFFERENTIATION ON CORPORATE STRATEGIC AND FINANCIAL
PERFORMANCE
ABSTRACT
Human resource (HR) differentiation refers to the practice of managing individuals or groups of
employees differently based on the value they deliver to an organization using individual-based,
workforce-based or job-based approaches. A differentiated HR architecture allows organizations
to prioritize their HR investments and gain cost advantages and flexibility through differential
treatments of strategic and nonstrategic employees. This study hypothesizes and tests a causal
model in which an organization’s HR differentiation practices enhance the strategic performance
of HR systems that then positively affects a firm’s strategic and financial performances. Results
revealed that after controlling for the positive effect of high-performance work practices (HPWPs),
firms with a greater degree of HR differentiation reported significantly higher strategic
performance of their HR systems. The positive relationship is moderated by the adoption of
HPWPs, firm size, and environmental dynamism. Notably, the effect of HR differentiation was
greater for firms with more HPWPs, a greater number of employees, or more dynamic
environments. These results support causal linkage between HR differentiation, strategic
performance of an HR system, and a firm’s strategic and financial performance.
91
1 INTRODUCTION
In his dystopian novella Animal Farm, George Orwell was thinking of the hidden
corruption of communism when he wrote, “All animals are equal, but some animals are more equal
than others.” His point was that total equality was impossible, as there are always those who will
hold more power or be more important to the machinery (whether the machinery is a cause or an
organization). In today’s dynamic business environment, this is an important lesson to remember.
While an organization stating that all employees are valuable is a nice sentiment, the truth is that
many organizations choose differentiated human resource (HR) practices to stay competitive. HR
differentiation refers to the practice of managing individuals or groups of employees differently
based on the value they deliver to an organization using individual-based, workforce-based or job-
based approaches within an organization (Becker & Huselid, 2006 & 2010; Becker, Huselid, &
Beatty, 2009; Zhou & Hong, 2008). As suggested by Huselid and Becker (2011), the essence of
HR differentiation is that “some jobs are more valuable (strategic) than others.” In other words,
strategic employees are more “equal” than others because they make disproportionally higher
contributions to their organizations. The concept of differentiated HR is based on earlier work on
employee–organization relationship framework (Tsui, Pearce, Porter, & Tripoli, 1997) and
concept of HR architecture (Lepak & Snell, 1999, 2002). Tsui and colleagues (1997) were among
the first to conceptualize different employee–organization relationships, and they found that
overinvestment or mutual investment relationships outperformed “quasi-spot contract” (i.e.,
temporary) or underinvestment relationships. Lepak and Snell (1999) later conceptualized HR
92
architecture as having two dimensions: the value and the uniqueness of human capital. When both
value and uniqueness of human capital are low, the employment mode is described as
“contracting” and the HR configuration as “compliance.” By contrast, when both value and
uniqueness of human capital are high, the employment mode is “internal development” and the
HR configuration is “commitment.” They note that these employment modes should not be
reduced to either/or distinctions, as a variety of approaches to manage human capital can exist
simultaneously. The value of differentiated HR architecture forms theoretical basis of using HR
differentiation.
Huselid and Becker (2011) built on this research and suggested that organizations should
develop an architecture that differentiates between strategic employees who make disproportional
strategic contributions to the organizations and nonstrategic employees who do not. Huselid and
Becker (2011) stressed that there is no clear line between strategic and nonstrategic jobs, and this
paper supports this idea by suggesting that the level of strategic contribution made by different
employees can be conceptualized as a continuum which is shaped by organizational and
environmental contingencies. Drawing on prior work, I argue that the degree of strategic
contribution depends on the uniqueness and value of human capital (Lepak & Snell, 1999, 2002).
HR differentiation measures the differential HR treatments within an organization. The
concept is multilevel since organizations can use HR differentiation at both individual and group
level. First, HR differentiation can occur to individual employees within a workgroup. For
example, HR system can differentiate between star and nonstar performers who hold the same job.
93
Aguinis and O'Boyle (2014) proposed that a compensation system that enhances the retention of
nonstar employees would lead to higher turnover of star performers. Therefore, by recognizing the
different needs of star and nonstar employees, individual level HR differentiation can create value
and make strategic contributions to the organization.
Second, this paper considers not only HR differentiations that occur between different
employees who are within the same job or workgroup, but also those that occur between different
jobs. Zhou and Hong (2008), for example, consider two types of HR differentiation: job-based
differentiation, which centres on the strategic contributions of jobs, and workforce-based
differentiation, which uses workforce characteristics as the criteria for HR differentiation such as
the value and uniqueness of human capital (Lepak & Snell, 1999). Organizations use jobs or
workforce characterises as criteria to engage in HR differentiation, such that people who hold the
same job or in the same workgroups are treated the same way. For job-based HR differentiation,
it is important to note here that HR differentiation processes do not all use the same criteria in
determining strategic and nonstrategic jobs. For example, Lepak, Taylor, Tekleab, Marrone, and
Cohen (2007) mark a division between core and support employees, and find that core employees’
exposure to high-investment HR systems is significantly greater than support employees’ in
nonmanufacturing contexts. These binary conceptualizations of HR differentiation, whatever their
terminology, remain highly simplified. Many organizations recognize that jobs vary in the degree
of their strategic contribution and there is a need to develop more complex HR architectures to
facilitate various types of HR differentiations. In sum, HR differentiation conceptualizes both
94
between-group (job-based or workforce-based) and within-group individual level differentiation
within an organization.
Whichever approach a firm chooses to use, HR differentiation aims to achieve the same
goals: improving strategic and financial performance. There are several mechanisms by which HR
differentiation can help create competitive advantages, including enabling firms to better allocate
resources to improve their attraction and retentions of strategic employees. It can also enhance HR
flexibility in terms of nonstrategic employees, which may provide a cost advantage to
organizations, and strengthens an organization’s strategic capability by more efficient investments
in human capital. Overall, differentiated HR is a complex social system, but one that can be a
competitive advantage. This study proposes and empirically tests a causal model of strategic HR
management to understand whether HR differentiation is positively related to an organization’s
strategic performance of an HR system and it will positively affect strategic and financial
performance.
2 THEORY AND HYPOTHESES
The contingency perspective of strategic human resource management (HRM) (e.g., Miles
& Snow, 1984; Schuler & Jackson, 1987) suggests that in order to be effective, organizations
should fit with both internal and external contingencies. Toh, Morgeson, and Campion (2008)
found that HR practices could be bundled into five categories: cost minimizers, contingent
motivators, competitive motivators, resource makers, and commitment maximizers. They argued
that these bundles would ensure a good fit between HR systems and the values and structures of
95
their organizations. By contrast, the universalistic perspective suggests that the adoption of a set
of best practices or high-performance work practices (HPWPs), such as selective recruitment,
extensive training, incentive compensation and employee participation, can enhance performance
across different situations (Huselid, 1995; Combs, Liu, Hall, & Ketchen, 2006). Generally, systems
of HPWPs are more likely to be a source of competitive advantage compared to individual HPWP
which could be imitated by competitors.
Although strategic HRM research (e.g. Huselid, 1995; Delery & Doty, 1996; Yanadori &
Van Jaarsveld, 2014) has established a significant positive HR–performance link, the effect of
HPWPs on financial performance varies considerable between organizations such that there are
still considerable differences in HRM “quality” across organizations (Becker, Huselid, & Beatty,
2009). One of the reasons for this may be the lack of studies that have looked into the differential
HR treatment across different groups of employees within organizations. Zhou and Hong (2008)
conceptualized two general approaches to differentiation: workforce-based and job-based. They
argue that workforce-based HR differentiation, which builds upon a resource-based view of firms
(e.g., Barney, 1991), is a bottom–up approach, as it is a result of workforce heterogeneity; and job-
based HR differentiation is a top–down approach that emphasizes the unique and inimitable
strategic process of an organization. However, few empirical studies have examined the impact of
HR differentiation. This is an important gap in the field, as internal HR differentiation can
positively influence the strategic performance of HR systems and create synergistic effect with
HPWPs.
96
This paper also looks at how HR differentiation contributes to financial performance.
Recent meta-analytic studies have shown that HPWPs significantly influence various HR,
operational, and financial outcomes (e.g., Combs et al., 2006; Jiang, Lepak, Hu, & Baer, 2012)
and that a causal chain can be found from HPWPs to human capital and motivation to HR outcomes
(e.g., employee satisfaction and turnover) to operational outcomes (e.g., productivity and
innovation) to financial performance (Becker & Huselid, 1998; Jiang et al., 2012; Subramony,
2009). Indeed, studies have repeatedly shown that the adoption of HPWPs is positively related to
financial performance by enhancing employee abilities, motivations, and opportunities to
contribute (Jiang et al., 2012), and scholars have called for increased focus on the strategic
contribution of HR systems (Lengnick-Hall, Lengnick-Hall, Andrad, & Drake, 2009). This paper
will control for the use of HPWPs and focus instead on the strategic processes by which HR
differentiation contributes to a firm’s financial success by creating strategic value for the
organization. In other words, I will address the HR–performance link from a strategic perspective.
In Figure 1, the hypothesized link between HR differentiation, the strategic performance of an HR
system, the strategic performance of a firm, and the financial performance of a firm is illustrated,
as well as the moderating influence of HPWPs, firm size, and environmental dynamism. The
following sections elaborate this conceptual framework.
97
2.1 Strategic Performance of HR System
Strategic performance of an HR system refers to the extent that a firm’s HR system can
serve as its competitive advantage. The resource-based view of firms suggests that physical
resources, financial resources, and human resources are three types of strategic resources that are
valuable, rare, and non-substitutable (Barney, 1991; Wernerfelt, 1984, 1995). Barney and Wright
(1998) moved further into this idea by suggesting that HR systems, as opposed to single HR
practices, can make strategic contributions, as they are able to create sustained competitive
advantages by working as complex, interdependent systems that are difficult to imitate. This paper
takes this a step forward by suggesting that HR differentiation can help organizations better
integrate and use their internal resources to enhance the strategic value to their HR systems.
Using a resource-based view of firms, I argue that developing a differentiated HR
architecture can serve as a key strategic capability for organizations because such value-creating,
complex social structures tend to be difficult to imitate. Also, HR differentiation helps
organizations to strategize and concentrate resources on the most valuable employees, and these
employees are likely to possess rare and unique skills that will then be difficult to acquire by
competitors. Moreover, Colbert (2004) extends the resource-based view to suggest that HR
systems are complex, interactive, living social systems that follow complexity principles and can
be integrated into HR architectures. As HR differentiation represents the strategic processes of an
HR architecture that provides multiple HR systems to different individual or groups of employees,
98
HR differentiation can also be explained by a complex, resource-based view (Colbert, 2014),
which will provide a competitive advantage to an HR system.
The positive impact of HR differentiation on the strategic performance of an HR system
can be created in several ways, including providing additional resources to better invest in strategic
jobs to enhance the attraction and retention of strategic employees. HR differentiation also helps
organizations create a cost advantage by treating nonstrategic employees differently than strategic
employees. The rationale behind this stance is that strategic employees create more value to
organizations and nonstrategic employees are peripheral, and providing different HR treatments
for these groups would better serve organizational goals by improving the efficiency and reward
of resource allocation. Matusik and Hill (1998), for example, suggest that contingent workers serve
as a source of competitive advantage because they lower costs, increase strategic flexibility, and
add valuable knowledge to organizations. By separating out contingent workers from standard
employees, HR differentiation enhances the strategic and financial performance of organizations.
Furthermore, HR differentiation improves the overall HR flexibility of organizations,
which can be a source of dynamic capability that is a necessary condition for competitive
advantage (e.g., Eisenhardt & Martin, 2000; Teece & Pisano, 1994; Teece, Pisano, & Shuen,
1997). Drawing on the Sanchez (1995), Wright and Snell (1998) distinguish between resource
flexibility (i.e. the ability to use a resource in different contexts) and coordination flexibility (i.e.
the ability to “resynthesize,” “reconfigure” and “redeploy” the resources) in three aspects:
flexibility in HR practices, flexibility in employee skills, and flexibility in employee behavior. In
99
other words, they see HR flexibility as a multidimensional construct (Way et al., 2012), one that
several studies have shown to have a positive link to financial performance (Bhattacharya, Gibson,
& Doty, 2005; Ketkar & Sett, 2009, 2010). I argue that HR differentiation is an important source
of flexibility in HR practices, as organizations with differentiated HR architecture are more agile
in their use of HR practices. Further, by strategically allocating resources to different employees,
organizations can gain greater flexibility in employee skills and behaviour, as a differentiated HR
structure better serves the various developmental and motivational needs of different employees.
Finally, I argue that HR differentiation can better serve the different needs of employees.
HR differentiation allows employees to select the HR practices, such as flexible benefit programs,
that best suit their interests, which leads to higher employee satisfaction and makes a positive
impact on the strategic performance of an HR system. Thus, HR differentiation enables
organizations to benefit from greater value that is added by strategic employees, as well as from
greater cost advantages and flexibility that is added by nonstrategic employees.
Hypothesis 1: HR differentiation is positively related to the strategic performance of an
HR system.
2.2 Moderators of HR Differentiation
The positive effect of HR differentiation can be influenced by a number of organizational
and environmental factors. It is well established in the field that HPWPs have a positive impact on
a variety of HR, operational, and financial outcomes. The adoption of HPWPs is related to higher
strategic performances of their HR system because these systems are difficult to imitate and more
100
likely to create strategic value (e.g., flexibility). Organizations with HPWPs are more likely to be
successful in their implementation of HR differentiation since such organizations are more likely
to create synergistic effect of HPWPs and HR differentiation. Thus, it is expected that the
relationship between HR differentiation and the strategic performance of an HR system will be
more positive when the level of HPWPs is high.
Hypothesis 2(a): HPWPs moderate the positive relationship between HR differentiation
and the strategic performance of an HR system such that the relationship is more positive
when the level of HPWPs is high.
The alignment of HR systems within organizational and environmental contingencies can
lead to superior performance (Jackson & Schuler, 1995; Jackson, Schuler, & Jiang, 2014). This
study focuses on firm size and environmental dynamism as two important contingencies. I propose
that firm size is another moderator of the HR–performance relationship because firm size is a key
determinant of an organization’s choice of HR practices (Guthrie, 2001; Huang & Verma, 2016)
and the development of HR architecture. It is expected that HR differentiation will be more
effective in larger organizations for several reasons. A greater number of employees creates an
economy of scale for implementing differential HR treatments and also allows for increased HR
differentiation alternatives, which means that firms have an increased chance of finding effective
HR differentiation. In addition, the negative aspects of HR differentiation, including potential
justice issues and administrative costs concerns, are more likely to be mitigated in larger
organizations
101
Hypothesis 2(b): Firm size moderates the positive relationship between HR differentiation
and strategic performance of an HR system such that the relationship is more positive when
firm size is high.
Environmental dynamism is a third factor that can significantly influence the impact of HR
systems. Datta, Guthrie, and Wright (2005) found that the impact of HPWPs on labor productivity
is stronger in more dynamic industries, possibly because firms face changing and uncertain
environments require greater flexibility than firms in stable and predictable environments. Lepak,
Takeuchi, and Snell (2003) argue that environmental dynamism moderates the relationship
between employment mode and financial performance because employees in dynamic
environments have greater task flexibility and thus may create more value to organizations.
Providing flexibility in an HR system is a central feature of HR differentiation. Its effect is likely
to be more pronounced in dynamic environments than in predictable and stable environments
because there is more need for flexibility. Thus, it is expected that the strategic impact of HR
differentiation will be more pronounced in more dynamic environments.
Hypothesis 2(c): Environmental dynamism moderates the positive relationship between HR
differentiation and the strategic performance of an HR system such that the relationship is
more positive when environmental dynamism is high.
102
2.3 Firm Strategic Performance
The strategic performance of organizations can be manifested in several ways, such as
gaining competitive advantages or greater market share, or being more successful than competitors
(Schilke, 2014). A resource-based view of the firm (Barney, 1991, 1995; Wernerfelt, 1984)
suggests that organizations can develop internal resources to create valuable, rare, and inimitable
competencies, and Becker and Gerhart (1996) argue that HR can serve as an organization’s unique
source of sustained competitive advantage because of causal ambiguity and path dependence. In
addition to the positive impact of HPWPs on a firm’s strategic performance, HR differentiation
can be a significant source of value creation, as it is able to allocate internal resources to enhance
the strategic performance of an HR system. Thus, it is expected that there is a positive relationship
between HR differentiation, the strategic performance of an HR system, and a firm’s strategic
performance.
Hypothesis 3(a): HR differentiation is positively related to firm strategic performance.
Hypothesis 3(b): Strategic performance of an HR system mediates the positive relationship
between HR differentiation and firm strategic performance.
2.4 Firm Financial Performance
In a resource-based view, the strategic performance of a firm’s HR system may offer a
competitive advantage and consequently a positively influence on the firm’s financial
performance. Crook, Ketchen, Combs, and Todd’s (2008) meta-analysis of more than 29,000
organizations in 125 studies found that 22 percent of the variance in a firm’s performance could
103
be explained by strategic resources. As an HR system serves as an integral part of a firm’s strategic
resources, I expect that the strategic performance of an HR system is positively related to a firm’s
strategic performance that will then influence a firm’s financial performance.
Hypothesis 4(a): Strategic performance of an HR system is positively related to a firm’s
financial performance.
Hypothesis 4(b): Strategic performance of a firm mediates the positive relationship
between the strategic performance of an HR system and a firm’s financial performance.
3 METHODS
3.1 Sample
My final sample consists of 240 enterprises in 27 cities in China. For each enterprise, one
manager is contacted to complete the survey. The survey respondents were general managers
(87%) and HR managers or directors (13%). The average tenure in their current job was 5.8 years.
Respondents were asked to complete a survey using one of three methods: (1) paper survey, (2)
online survey distributed through emails, and (3) mobile survey using the smartphone application
WeChat. In total, 250 surveys were distributed using these three methods, and 240 respondents
completed the survey (response rate of 96%). Specifically, 15 surveys were collected on paper, 8
online (through email), and 217 through WeChat mobile survey application.
104
3.2 Measures
3.2.1 Independent Variables
Building on the work of Huselid and Becker (2011) and Sun, Aryee, and Law (2007), I
developed a scale on HR differentiation using six items. Response options ranged from 1, “strongly
disagree,” to 7, “strongly agree.” The six questions include three items on group level HR
differentiation and three items on individual level HR differentiation. The items are as follows: “1.
We use different HR practices on groups of employees (e.g. administrative support, marketing,
sales and production workers),” “2. Our HR system differentiates between core and support
employees,” “3. Our HR system rewards different jobs based on their unique contributions to the
organizations,” “4. Within a particular job, we offer more training and development opportunities
to top performers than low performers,” “5. Within a particular job, our compensation system
rewards star or top performers much more than low performers,” and “6. Within a particular job,
our HR system offer more incentives for star or top performers to stay compared to low
performers.” To ensure the validity of scale, I conducted a pilot study of 359 enterprises. To
examine the underlying structure of the scale, I conducted an exploratory factor analysis and all
six items loaded on a single factor. The HR differentiation index (α=.79 in the pilot study) was the
average of the six items. In the final sample of 240 enterprises, confirmatory factor analysis
supported the one-factor solution and the Cronbach's alpha for the HR differentiation index was
.90.
105
3.2.2 Mediators and Dependent Variables
Strategic performance of HR system is the extent to which an HR system can serve as
firm’s competitive advantage. The four items to assess the strategic performance of HR systems
are as follows: “Our strategic employees have a rare skills or expertise which are difficult to access
by our competitors,” “Our nonstrategic employees give us a cost advantage over our competitors,”
“Our HR system can’t be easily imitated by our competitors,” and “Our HR system is a most
important reason that we gain competitive advantage over our competitors.” In the pilot study of
359 enterprises, I conducted an exploratory factor analysis of the scale and results supported a one-
factor structure. The index for strategic performance of HR system (α=.80 in the pilot study) is the
average of the four items. In the final sample of 240 enterprises, all four items load on one single
factor in the CFA and the reliability alpha for this scale was .86.
The firm strategic and financial performance measures were both adapted from Schilke
(2014). The strategic performance index was based on three items on a 1 to 7 scale (α =.89). Sample
items are “We have gained strategic advantages over our competitors,” and “We have a large
market share.” The firm financial performance measured return on investment (ROI), specifically
whether the firm’s ROI was continuously above the industry average. Response options ranged
from 1, “strongly disagree,” to 7, “strongly agree.”
3.2.3 Moderators
Sun et al. (2007) have developed a 27-item scale for HPWPs and the scale can be reduced
to eight dimensions using factor analysis. This study adapted Sun et al.’s (2007) scale on HPWPs
106
and used an eight-item abridged scale which captured all eight dimensions of HPWPs. Response
options ranged from 1, “strongly disagree,” to 7, “strongly agree.” Sample items used in this study
are as follows: “Very extensive efforts are made in selection,” “Extensive training programs are
provided for employees,” and “Close tie or matching of pay to individual/group performance.”
The HPWPs index was the average of the eight items (α =.90). Firm size was the natural logarithm
of the number of employees (e.g., Datta et al., 2005; Huselid, 1995; Koch & McGrath, 1996), and
the environmental dynamism index was adapted from Schilke (2014) with the average of five items
on a 1 to 7 scale (α =.82). Sample items for the environmental dynamism scale are, “The modes of
production/service change often and in a major way,” “Environmental changes in our industry are
unpredictable,” and “Marketing practices in our industry are constantly changing.”
3.2.4 Control Variables
I included several control variables to account for organizational characteristics. Union
presence was a dichotomous variable denoting whether a union was present in the firm. Firm age
was the number of years since the organization opened. Industry was one of three categories:
manufacturing, service, and other industries. Firm ownership was operationalized as a set of four
variables: private-owned enterprises (POE), state-owned enterprises (SOE), foreign-owned
enterprises or joint ventures (FOE), Hong Kong, Macau, or Taiwan-owned enterprises or joint
ventures (HMTOE). Four dummy variables were created to represent firm location, and were
categorized into four groups: Beijing and surrounding cities, Shandong province, Guangdong
province, Shanghai and surrounding cities.
107
4 Results
4.1 Descriptive Statistics and Bivariate Correlations
Table 1 presents the means, standard deviations, and bivariate correlation for the variables.
The average enterprise size was 16,158, and 85 percent of the enterprises had 100 or more
employees. The number of firms in manufacturing, service, and other industries were 78, 73, and
89, respectively. Of these, 173 enterprises were unionized.
4.2 Multivariate Analyses
Three sets of multivariate regressions were conducted to test the moderated mediation
model depicted in Figure 1. Table 2 presents the regression results for the strategic performances
of the HR systems. Model 1 includes all control variables and HPWPs, which shows a significant
positive impact on the strategic performance of HR systems (p<0.001). As shown in Model 2, HR
differentiation is positively related to the strategic performance of HR systems (p<.05). Hypothesis
1 is thus supported. The interaction terms between HR differentiation and three moderators—
HPWPs (p<.05), firm size (p<.001), and environmental dynamic (p<.10)—were added
independently in Models 3 to 5. Hypotheses 2a, 2b, and 2c are thus supported. See Figures 2.1 to
2.3 for plots of the interaction effects.
Table 3 shows the regression results for strategic performance. In Model 1, the adoption of
HPWPs is a significant predictor of a firm’s strategic performance (p<.001). HR differentiation
was added to the regression in Model 2, and it was found that HR differentiation was significantly
related to strategic performance (p<.10). Hypothesis 3a is thus supported. The mediator—the
108
strategic performance of an HR system—was added in Model 3. In this model, it was found that
the strategic performance of an HR system was a significant determinant of a firm’s strategic
performance (p<.001), and HR differentiation was no longer significant when compared to Model
2. Additional analysis based on Hayes’ (2013) bootstrapping procedures with 5,000 resamples was
conducted, and the results (95% confidence interval: .394 to .716) supported the mediation
suggested by Hypothesis 3b.
As shown in Table 4, the use of HPWPs (p<.001) is a significant predictor of ROI in Model
1. In Model 2, HR differentiation (p<.05) is positively related to ROI. Thus, Hypothesis 4a is
supported. Additional analysis using on Hayes’ (2013) bootstrapping procedures with 5,000
resamples shows that a firm’s strategic performance is a significant mediator (95% confidence
interval: .251 to .634) on the relationship between strategic performance of an HR system and ROI,
as suggested in Hypothesis 4b. Table 5 summarizes the hypothesized relationships and empirical
findings of this study.
5 DISCUSSION AND CONCLUSIONS
This paper suggests that HR differentiation can make significant strategic and financial
contributions to organizations and improve the strategic performance of HR systems on top of the
positive effect of HPWPs. Thus by developing differentiated HR architecture to provide multiple
HR treatment alternatives based on the strategic contribution of jobs and workforce, organizations
can better allocate resources to achieve their goals. There are several ways in which HR
differentiation enhances the overall strategic performance of HR systems. First, HR differentiation
109
means that more favourable and attractive HR treatments can be offered to employees who are
more valuable and strategic for the organization. Second, differential HR treatments for
nonstrategic employees allows organizations to gain cost advantages. Third, by developing
multiple HR systems within an organization, HR differentiation improves a firm’s overall HR
flexibility to better fit with organizational and environmental contingencies and be agile in
dynamic environments. Fourth, greater HR flexibility generated by HR differentiation can be a
source of dynamic capability. Fifth, following the guidelines of a resource-based view, HR
differentiation, as a manifestation of a firm’s complex HR architecture, makes it valuable, rare,
and non-substitutable (i.e., difficult to imitate by competitors). Sixth, HR differentiation may better
serve the unique needs of employees. Overall, differential HR treatments may improve the utility
of employees, which may positively influence employee behaviour and this may result in better
performance and HR outcomes.
Understanding HR differentiation helps to explain differences in the size of positive
impacts of HPWPs across different organizations. Decades of theoretical and empirical research
in the field of strategic HRM has established that the adoption of HPWPs is positively related to a
variety of HR, operational, and financial outcomes of organizations and can serve as a unique
source of competitive advantage; however, such positive effects vary as the quality of
implementation differs (Becker & Huselid, 2006). In this paper, I find that a firm’s ability to treat
their employees differently based on job or workforce heterogeneities creates strategic value to the
organization and significantly enhances a firm’s financial performance. HR differentiation helps
110
organizations to better attract and retain strategic employees and to gain greater flexibility and cost
advantages by differentiating nonstrategic employees. HR differentiation enhances organizations’
ability to develop unique HR systems that are socially complex and firm-specific and hence
enhance strategic performance, which in turn strengthens a firm’s strategic and financial
performance.
Although this study provides theoretical arguments and empirical support for the positive
impacts of HR differentiation on organizational strategic and financial performance, there are
several limitations that need to be acknowledged and some key aspects that need to be examined
by future studies. First, this study does not explore the determinants of HR differentiation. A firm’s
ability to implement differentiated HR architecture is often restricted by institutional and
environmental factors, including unionization. Future research may examine how such factors
influence the adoption of HR differentiation. Second, this study operationalizes HR differentiation
as an overall index of differential HR treatments between and within jobs. Future studies may
explore the specific types of criteria organizations use to conduct HR differentiation and
investigate the effects of different types of HR differentiation. Third, future research may study
the potential negative outcomes of HR differentiation, such as the cost of administration or
organizational injustice. Lastly, this paper is also subjected to the methodological limitations of
self-report questionnaires. Although this study aimed to survey the actual degree of HR
differentiation, the survey only captured managers’ perception of HR differentiation in their
111
organization. Future study may address this issue by adopting objective measures of the actual HR
differentiation practices.
In conclusion, this paper contributes to the strategic HRM literature in two main ways.
First, it shows that the adoption of HPWPs and HR differentiation significantly enhances the
strategic performance of HR systems. The positive impact of HR differentiation is greater for firms
with more HPWPs, more employees, and those working in more dynamic environments. Second,
drawing on a resource-based view of firms, this study shows a causal chain linking HR
differentiation to the strategic and financial performance of organizations. Overall, this empirical
study on HR differentiation offers guidance to business and HR managers in designing and
implementing more differentiated and strategic HR systems. The theoretical framework explaining
the causal mechanisms of HR differentiation and a firm’s financial performance helps to advance
strategic HRM research by showing how HR differentiation can make organizations more strategic
and profitable. Together, this paper offers an initial theoretical and empirical basis for future
research on HR differentiation.
112
REFERENCES
Aguinis, H. & O'Boyle, E. 2014. Star performers in twenty-first century organizations.
Personnel Psychology, 67(2): 313-350.
Barney, J. 1991. Firm resources and sustained competitive advantage. Journal of Management,
17(1): 99-129.
Barney, J. 1995. Looking inside for competitive advantage. Academy of Management
Executive, 9(4): 49-61.
Barney, J. B., & Wright, P. M. 1998. On becoming a strategic partner: The role of human resources
in gaining competitive advantage. Human Resource Management, 37(1): 31-46.
Becker, B. & Gerhart, B. 1996. The impact of human resource management on organizational
performance: progress and prospects. Academy of Management Journal, 39(4):779-801.
Becker, B. E., & Huselid, M. A. 1998. High performance work systems and firm performance: A
synthesis of research and managerial implications. In G. R. Ferris (Ed.), Research in
Personnel and Human Resources Management: 53–101. Greenwich, CT: JAI Press.
Becker, B. E., & Huselid, M. A. 2006. Strategic human resources management: Where do we go
from here? Journal of Management, 32(6): 898-925.
Becker, B. E., & Huselid, M. A. 2010. SHRM and job design: Narrowing the divide. Journal of
Organizational Behavior, 31(2-3): 379-388.
Becker, B. E., Huselid, M. A., & Beatty, R. W. 2009. The differentiated workforce: Transforming
talent into strategic impact. Boston: Harvard Business Press
Bhattacharya, M., Gibson, D., & Doty, D. 2005. The effects of flexibility in employee skills,
employee behaviors, and human resource practices on firm performance. Journal of
Management, 31(4): 622-640.
Colbert, B. 2004. The complex resource-based view: Implications for theory and practice in
strategic human resource management. Academy of Management Review, 29(3): 341−358.
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(3): 501-528.
113
Crook, T. R., Ketchen, D. J., Combs, J. G., & Todd, S. Y. 2008. Strategic resources and performance:
A meta-analysis. Strategic Management Journal, 29(11): 1141-1154.
Datta. D. K., Guthrie, J. P., & Wright, P. M. 2005. Human resource management and labor
productivity: Does industry matter? Academy of Management Journal, 48(1): 135-145.
Eisenhardt, K. M. & Martin, J. A. 2000, Dynamic capabilities: what are they? Strategic
Management Journal, 21(10-11): 1105-1121.
Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity: Evidence from
New Zealand. Academy of Management Journal, 44(1): 180-190.
Hayes, A. F. 2013. An introduction to mediation, moderation, and conditional process analysis:
A regression-based approach. New York, NY: Guilford Press.
Huang, X. Y. & Verma, A. 2016. Industry- and firm-level determinants of employment relations in
China: a two-level analysis. The International Journal of Human Resource Management,
3: 1-21.
Huselid, M. A. 1995. The impact of human resource management practices on turnover,
productivity, and corporate financial performance. Academy of Management Journal, 38(3):
635-672.
Huselid, M., & Becker, B. 2011. Bridging micro and macro domains: Workforce differentiation
and strategic human resource management. Journal of Management, 37(2): 421-428.
Jackson, S. E., & Schuler, R. S. 1995. Understanding human resource management in the context
of organizations and their environment. Annual Review of Psychology, 46: 237-264.
Jackson, S. E., Schuler, R. S., & Jiang, K.F. 2014. An Aspirational framework for strategic
human resource management. The Academy of Management Annals, 8(1): 1-56.
Jiang, K., Lepak, P. D., Hu, J., & Judith, C. B. 2012. How does human resource management
influence organizational outcomes? A meta-analytic investigation of mediating
mechanisms. Academy of Management Journal, 55 (6): 1264-1294.
Ketkar, S., & Sett, P. K. 2009. HR flexibility and firm performance: Analysis of a multi-level
causal model. International Journal of Human Resource Management, 20(5): 1009-1038.
114
Ketkar, S., & Sett, P. K. 2010. Environmental dynamism, human resource flexibility, and firm
performance: Analysis of a multi-level causal model. International Journal of Human
Resource Management, 21(8): 1173-1206.
Koch, M. J., & McGrath, R. G. 1996. Improving labor productivity: Human resource management
policies do matter. Strategic Management Journal, 17(5): 335-354.
Lengnick-Hall, M. L., Lengnick-Hall, C. A., Andrade, L. S., & Drake, B. 2009. Strategic human
resource management: The evolution of the field. Human Resource Management Review,
19(2): 64-85.
Lepak, D. P., & Snell, S. A. 1999. The human resource architecture: Toward a theory of human
capital allocation and development. The Academy of Management Review, 24(1): 31-48.
Lepak, D. P., & Snell, S. A. 2002. Examining the human resource architecture: The relationships
among human capital, employment, and human resource configurations. Journal of
Management, 28(4): 517−543.
Lepak, D. P., Takeuchi, R., & Snell, S. A. 2003. Employment flexibility and firm performance:
Examining the interaction effects of employment mode, environmental dynamism, and
technological intensity. Journal of Management, 29(5): 681−703.
Lepak, D. P., Taylor, M. S., Tekleab, A. G., Marrone, J. A., & Cohen, D. J. 2007. An examination
of the use of high-investment human resource systems for core and support employees.
Human Resource Management, 46(2): 223−246.
Matusik, S. F., & Hill, C. W. L. 1998. The utilization of contingent work, knowledge creation, and
competitive advantage. Academy of Management Review, 23(4): 680−697.
Miles, R. E., & Snow, C. C. 1984. Designing strategic human resources systems. Organizational
Dynamics, 13 (1): 36−52.
Sanchez, R. 1995. Strategic flexibility in product competition. Strategic Management Journal,
16(1): 135-159.
Schilke, O. 2014. On the contingent value of dynamic capabilities for competitive advantage: The
nonlinear moderating effect of environmental dynamism. Strategic Management Journal,
35(2): 179-203.
115
Schuler, R. S., & Jackson, S. E. 1987. Linking competitive strategies with human resource
management practices. Academy of Management Executive, 1(3): 207-219.
Subramony, M. 2009. A meta-analytic investigation of the relationship between HRM bundles and
firm performance. Human Resource Management, 48(5): 745-768.
Sun, L., Aryee, S., & Law, K. S. 2007. High-performance human resource practices, citizenship
behavior, and organizational performance: A relational perspective. The Academy of
Management Journal, 50(3): 558-577.
Teece D., & Pisano G. 1994. The dynamic capabilities of firms: an introduction. Industrial and
Corporate Change, 3(3): 537-556.
Teece D., & Pisano G, Shuen A. 1997. Dynamic capabilities and strategic management. Strategic
Management Journal, 18(7): 509-533.
Toh, S. M., Morgeson, F. P., & Campion, M. A. 2008. Human resource configurations:
Investigating fit with the organizational context. Journal of Applied Psychology, 93(4):
864-882.
Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. M. 1997. Alternative approaches to the
employee-organization relationship: Does investment in employees pay off? Academy of
Management Journal, 40(5): 1089-1121.
Way, S., Tracey, J., Fay, C., Wright, P., Snell, S., Chang, S., & Gong, Y. 2015. Validation of a
multidimensional HR flexibility measure. Journal of Management, 41(4): 1098-1131.
Wernerfelt, B. 1984. A resource-based view of the firm. Strategic Management Journal, 5(2): 171-
180.
Wernerfelt B. 1995. The resource-based view of the firm: ten years after. Strategic Management
Journal, 16(3):171–174.
Wright, P. M., & Snell, S. A. 1998. Toward a unifying framework for exploring fit and flexibility
in strategic human resource management. The Academy of Management Review, 23(4):
756-772.
Yanadori, Y., & Jaarsveld, D. D. 2014. The relationships of informal high performance work
practices to job satisfaction and workplace Profitability. Industrial Relations: A Journal
of Economy and Society, 53(3): 501-534.
116
Zhou, Y., & Hong, Y. 2008. A Differentiation Perspective of SHRM: Strategic Capabilities,
Configurations, and Complementarities. Academy of Management Annual Meeting
Proceedings, 8:1-6.
117
TABLE 1. Descriptive Statistics and Bivariate Correlations for Variables a
118
119
TABLE 2. Predictors of Strategic Performance of HR Systema
Model 1 Model 2 Model 3 Model 4 Model 5
HPWS 0.68*** 0.47*** 0.23 0.47*** 0.47***
(0.06) (0.09) (0.14) (0.08) (0.08)
HR differentiation (HRD) 0.25** 0 -0.2 -0.01
(0.08) (0.14) (0.15) (0.16)
Firm size 0.03 0.02 0.02 -0.30** 0.02
(0.03) (0.03) (0.03) (0.10) (0.03)
Environmental dynamism 0.16** 0.16** 0.13* 0.12* -0.1
(0.06) (0.06) (0.06) (0.06) (0.15)
Interaction terms
HPWS x HRD 0.06*
(0.03)
Size x HRD 0.07***
(0.02)
ED x HRD 0.06+
(0.03)
Control variables
Union presence -0.04 0.03 0.01 0.00 0.00
(0.17) (0.16) (0.16) (0.16) (0.16)
Firm age 0.00 0.00 0.00 0.00 0.00
(0.00) (0.00) (0.00) (0.00) (0.00)
Service -0.56** -0.47** -0.50** -0.42* -0.48**
(0.17) (0.17) (0.17) (0.17) (0.17)
Other industries -0.38* -0.27+ -0.29+ -0.26+ -0.25
(0.16) (0.16) (0.16) (0.16) (0.16)
FOE -0.02 0.10 0.07 0.08 0.10
(0.19) (0.19) (0.19) (0.19) (0.19)
HMTOE 0.06 0.12 0.13 0.24 0.12
(0.33) (0.32) (0.32) (0.31) (0.32)
SOE -0.13 -0.06 -0.05 -0.08 -0.07
(0.17) (0.17) (0.17) (0.16) (0.17)
Shandong -0.22 -0.27 -0.26 -0.29 -0.28
(0.23) (0.23) (0.23) (0.22) (0.23)
Guangdong -0.40** -0.43** -0.42** -0.35** -0.45**
(0.14) (0.14) (0.14) (0.14) (0.14)
Shanghai -0.28 -0.29 -0.24 -0.17 -0.31
(0.29) (0.28) (0.28) (0.28) (0.28)
Constant 0.57 0.29 1.40* 2.51*** 1.53*
(0.36) (0.36) (0.64) (0.73) (0.77)
R2 0.56 0.58 0.58 0.60 0.58
Adjusted R2 0.53 0.55 0.56 0.57 0.55 a Unstandardized coefficients are reported; the figures in parentheses are standard errors. N=240
for all models.
+p<.10; * p<0.05; ** p<0.01; *** p<0.001.
120
TABLE 3. Predictors of Firm Strategic Performance
Model 1 Model 2 Model 3
HPWS 0.66*** 0.52*** 0.26**
(0.07) (0.10) (0.10)
HR differentiation (HRD) 0.16+ 0.02
(0.09) (0.09)
Strategic performance of HR
system
0.56***
(0.07)
Control variables
Firm size 0.14*** 0.14*** 0.12***
(0.04) (0.04) (0.03)
Environmental dynamism 0.06 0.06 -0.03
(0.07) (0.07) (0.06)
Union presence -0.15 -0.11 -0.13
(0.20) (0.20) (0.18)
Firm age 0.00 0.00 0.00
(0.00) (0.00) (0.00)
Service -0.29 -0.23 0.03
(0.21) (0.21) (0.19)
Other industries -0.1 -0.04 0.11
(0.19) (0.20) (0.18)
FOE -0.46* -0.38 -0.43*
(0.23) (0.23) (0.21)
HMTOE 0.07 0.12 0.05
(0.39) (0.39) (0.35)
SOE -0.39+ -0.34 -0.31+
(0.20) (0.21) (0.18)
Shandong 0.25 0.22 0.37
(0.28) (0.28) (0.25)
Guangdong -0.08 -0.1 0.14
(0.17) (0.17) (0.15)
Shanghai 0.40 0.39 0.55+
(0.35) (0.35) (0.31)
Constant 0.69 0.51 0.35
(0.44) (0.45) (0.40)
R2 0.45 0.45 0.57
Adjusted R2 0.42 0.42 0.54 a Unstandardized coefficients are reported; the figures in parentheses are standard errors. N=240
for all models. +p<.10; * p<0.05; ** p<0.01; *** p<0.001.
121
TABLE 4. Predictors of ROI
Model 1 Model 2 Model 3 Model 4
HPWS 0.56*** 0.28* 0.12 0.01
(0.08) (0.12) (0.13) (0.12)
HR differentiation (HRD) 0.33** 0.25* 0.24*
(0.11) (0.11) (0.10)
Strategic performance of HR
system
0.33*** 0.09
(0.09) (0.10)
Firm strategic performance 0.44***
(0.08)
Control variables
Firm size 0.07 0.06 0.05 0.00
(0.04) (0.04) (0.04) (0.04)
Environmental dynamism 0.04 0.04 -0.02 0.00
(0.08) (0.08) (0.08) (0.07)
Union presence 0.15 0.24 0.23 0.28
(0.24) (0.24) (0.23) (0.22)
Firm age -0.01+ -0.01+ -0.01+ -0.01
(0.00) (0.00) (0.00) (0.00)
Service 0.13 0.25 0.41+ 0.40+
(0.24) (0.24) (0.24) (0.23)
Other industries 0.04 0.18 0.27 0.22
(0.23) (0.23) (0.23) (0.22)
FOE 0.06 0.22 0.19 0.38
(0.27) (0.27) (0.27) (0.25)
HMTOE -0.47 -0.39 -0.43 -0.45
(0.47) (0.46) (0.45) (0.42)
SOE -0.21 -0.11 -0.09 0.04
(0.24) (0.24) (0.24) (0.22)
Shandong -0.03 -0.08 0.01 -0.16
(0.33) (0.33) (0.32) (0.30)
Guangdong 0.10 0.06 0.20 0.14
(0.20) (0.20) (0.20) (0.19)
Shanghai 0.56 0.55 0.65 0.4
(0.42) (0.41) (0.40) (0.38)
Constant 1.22* 0.85 0.75 0.6
(0.52) (0.53) (0.51) (0.48)
R2 0.29 0.31 0.35 0.43
Adjusted R2 0.25 0.27 0.31 0.39 a Unstandardized coefficients are reported; the figures in parentheses are standard errors. N=240
for all models.
+p<.10; * p<0.05; ** p<0.01; *** p<0.001
122
TABLE 5. Summary of Hypotheses and Findings
Hypot
hesis
Hypothesized relationships Empirical
findings
supported?
1 HR differentiation Strategic performance of HR system + Yes
2 Moderators of the positive relationship in hypothesis 1
2a HPWP as a moderator Yes
2b Firm size as a moderator Yes
2c Environmental dynamism as a moderator Yes
3a HR differentiation Firm strategic performance + Yes
3b Strategic performance of HR system as a mediator Yes
4a Strategic performance of HR system Firm financial
performance
+ Yes
4b Firm strategic performance as a mediator Yes
123
FIGURE 1. A Conceptual Framework on the Strategic and Financial Impact of HR
Differentiation
FIGURE 2a. The Moderation of HPWPs on the Relationship between HR Differentiation
and Strategic Performance of HR System
2
2.5
3
3.5
4
4.5
5
Low HR Differentiation High HR DifferentiationStr
ate
gic
per
form
an
ce o
f H
R s
yst
em
Low HPWPs
High HPWPs
124
FIGURE 2b. The Moderation of Firm Size on the Relationship between HR Differentiation
and Strategic Performance of HR System
FIGURE 2c. The Moderation of Environmental Dynamism on the Relationship between
HR Differentiation and Strategic Performance of HR System
0
0.5
1
1.5
2
2.5
3
Low HR Differentiation High HR DifferentiationStr
ate
gic
per
form
an
ce o
f H
R s
yst
em
Low Firm Size
High Firm Size
0
0.5
1
1.5
2
2.5
3
Low HR Differentiation High HR DifferentiationStr
ate
gic
per
form
an
ce o
f H
R s
yst
em
Low
Environmental
dynamism
High
Environmental
dynamism