Three Essays on Strategic Human Resource Management · Three Essays on Strategic Human Resource Management by Xiaoyu Huang ... Schuler & Jackson, 1987) suggests that a dynamic competitive

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


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



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



PERFORMANCE .................................................................................................................. 90


Differentiation ........................................................................................................................ 123

FIGURE 2a. The Moderation of HPWPs on the Relationship between HR Differentiation and

Strategic Performance of HR System .................................................................................... 123


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


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.


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


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


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.



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



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.


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


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.



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


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


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


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


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


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


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


ate

gic

per

form

an

ce o

f H

R s

yst

em

Low

Environmental

dynamism

High

Environmental

dynamism

Three Essays on Strategic Human Resource Management · Three Essays on Strategic Human Resource Management by Xiaoyu Huang ... Schuler & Jackson, 1987) suggests that a dynamic competitive

Documents