out formatting
Transcript of out formatting
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 1/147
HUMAN RESOURCE MANAGEMENT, HIGH PERFORMER TURNOVER, AND FIRM
PERFORMANCE
BY
KI-WOOK KWON
B.S., Hanyang University, 1999
M.A., University of Minnesota, Twin Cities, 2002
DISSERTATION
Submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy in Human Resources and Industrial Relations
in the Graduate Co llege of the
University of Illinois at Urbana-Champaign, 2009
Urbana, Illinois
Doctoral Com mittee:
Professor John Lawler, Chair
Professor Craig Olson
Associate Professor Deborah Rupp
Assistant Professor Gerald Ericksen
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 2/147
UMI Number: 3363137
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
subm itted. Broken or indistinct print, colore d or poor quality illustrations
and photographs, print b leed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also , if unauthorized
copyright material had to be removed, a note will indicate the deletion.
UMI
UMI Microform 3363137
Copyright 2009 by ProQuest LLC
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 3/147
ABSTRACT
This d issertation includes three separate but interrelated pape rs. I attempt to examine
the relationships among human resource systems, high performers' organizational
com mitm ent and turnov er, and firm performance at the individual as well as organizational
levels. First, I show that high performance (comm itment) work systems influence
organizational c om mitm ent of high performers at the individual level. Second, I report that
high performer turnover mediates the relationship between high performance work systems
and firm p erformance at the organizational level. Third, I argue and em pirically show that
high performer turnover could be detrimental for firm performance, but the degree of human
resource investment in employee training and development and firm reputation could
moderate the relationship between high performer turnover and firm performance.
11
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 4/147
To my parents,
to my w ife,
an d to my son and daughter
i i i
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 5/147
ACKNOWLEDGEMENTS
I would like to thank the mem bers of my dissertation com mittee, Professor John
Lawler, Craig Olson, Deborah Rupp, and Gerald Ericksen for their invaluable suggestions
and contributions. I would especially like to thank Professor Joh n Law ler, my advisor, for
his advice and encouragem ent. I am grateful to my parents, Sang W oon Kw on and Jung Ok
Oh, my brother, Jung II Kwon, my wife, Jeeseon Song, my son, Seahoon Kwon, and my
daughter, Yeaseo Kw on. W ithout their endless support, this dissertation could not have been
possible.
IV
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 6/147
TABLE OF CONTENTS
CHAPTER 1: INTRODU CTION 1
References 4
CHAPTER 2: HIGH COMMITMENT HR PRACTICES AND TOP PERFORMERS:
IMPACTS ON ORGANIZATIONAL COM MITMENT AND TURNOVER
INTENTION 7
Theoretical Backg round and Hyp otheses 10
Methods 18
Results 24
Discussion and Conclusion 26
References 33
Tables 45
Figures 49
CHAPTER 3: HIGH PERFORMANCE WORK SYSTEMS, TURNOVER, AND
FIRM PERFORMANC E: WHOSE TURNOVER MATTERS? 51
Previou s Literature and Hyp otheses 54
Methods 65
Results 71
Discussion and Conclusion 74
References 79
Tables 91
Figure 95
CHAPTER 4: A RELATIONSHIP BETWEEN HIGH PERFORMER TURNOVER
AND FIRM PERFOR MA NCE: A FIRM-LEVEL FIELD STUDY 96
Research Framewo rk 98
Theoretical Backgro und and Hypo theses 99
Methods 110
Results 115
Discussion and Conclusion 117
References 124
Tables 132
Figures 135
CHAPTER 5: CONCLU SION 138
APPENDIX 140
AUTH OR'S BIOGRAPHY 141
v
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 7/147
CHAPTER 1
INTRODUCTION
In a current knowledge-based econom y, organizational value creation activities
are highly influenced b y the quality of a firm's hum an resources (e.g., Batt, 2002; Collins
& Sm ith, 200 6; Huselid, 1995; Pfeffer, 1994; Wright, McM ahan, & McW illiams, 1994).
In particular, the retention/turnover of
a
firm's high performers tends to significantly
influence its performance (Cappelli, 2000; Lepak & Snell, 1999; Randall, 1987; Sturman,
Trevor, Boudreau, & Gerhart, 2003). According to a recent survey of 410 corporate
officers, M cKinsey & Com pany reported that the top 20 % of performers, relative to
average performers, generate 40 % increased productivity in operational roles, 49 %
increased profit in general man agemen t roles, and 67 % increased revenue in sales roles
(referred to Axelrod , Handfield-Jones, & Welsh, 2001).
Acad emic research argues that high performers' turnover negatively affects firm
performance because it impo ses high organizational costs related to replacing them and
socializing new comers, and also decreases organizational morale (Cascio, 1995;
Hollenback & Williams, 1986; Mobley, 1982; Randall, 1987; Staw, 1980). Furthermore,
in a current knowledge-driven economy, the negative impact of high performers'
turnover on firm performance appears to be amplified due to the potential threat of
valuable knowledge transfer by high-performing leavers to competing organizations
(Cappelli, 2000).
Despite the im portance o f retaining high performers, the inter-organizational
com petition for highly talented em ployees has intensified, as organizations have
1
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 8/147
increasingly recognized th e importance of high performing employees in organizational
value creation activities (Gardner, 2 005 ; Cappelli, 2000 ; Rao & Drazin, 2002). For
example, Gardner (2005) observed that more than 20 percent of organizations
experienced 'purposive talent raiding' from their competitors, which implies that high
performers m ay leave o rganizations more frequently than other levels of performers. In
fact, existing theoretical and empirical w ork indicates that high performers are more
likely to leave their org anizations than relatively lower performers (Jackofsky, 1984;
Trevor, Gerhart, & Bou dreau, 1997). Therefore, existing studies suggest that although
high performer retention is important, it has been increasingly difficult for firms to retain
high performers.
From the p erspectives of aforementioned theoretical and empirical studies, it is
evident that scholars need to pay significant attention to the implications of the
retention/turnover of high performers for firm performance and to how firms can increase
their retention. H ow ever, only a few studies investigated this issue, focusing on
relationships between one or different bundles of huma n resource (HR) practices and
high performer turnove r (Salam in & Horn 200 5; Shaw, Dineen, Fang, & Vellella, in
press; Shaw & Gup ta, 200 7; Trevor et al., 1997). These studies, however, did not fully
investigate complex relationships among HR practices, turnover between high performers
and relatively low performers, and firm performance.
In this dissertation, I attempt to fill this research gap. M y dissertation includes
three separate but highly interrelated essays (chap ters 2, 3, and 4), which exam ine the
relationships among high performance work systems, high performers' organizational
comm itment and turnov er, and firm performance at the individual as well as
2
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 9/147
organizational leve ls. In chapter 2, building upon un iversal and conditional aspects of
social exchange theory, I hypothesize and analyze how a bundle of high-commitment
human resource practices (HCHRPs) enables organizations to increase organizational
comm itment of both ordinary and high performers, bu t this effect w ill be stronger for the
latter than the former. The results support the hypotheses and some im plications are
derived in the context of managing high performers and HCHRPs.
In chapter 3,1 argue that most prior work in human resource management (HRM)
and firm perform ance literature has focused on investigating the relationships between
high performance work systems (HPWS), total turnover, and firm performance without
considering the performance level of leavers. Then, I simultaneously exam ine the
relationships among HPWS, both total turnover and high performer turnover, and firm
performance at the org anizational level. Chapter 3 reports that (1) HPW S reduces not
only total workforce turnover b ut also turnover of high performers and (2) high performer
turnover mediates the relationship between HPWS and firm performance with a control
for rest-of-workforce turnover.
In chapter 4 ,1 argue that high performer turnover has been viewed as a critical
organizational problem in the emp loyee turnover and firm performance literature as well
as in practice, but there h ave bee n surprisingly few studies that theoretically and
empirically address h ow high performer turnover is related to firm performance. I
investigate (1) the overall relationship between high performer turnover and firm
performance and (2) mod erating effects of a firm's investment in employee training and
development, and firm reputation. This chapter reports that, whereas high performer
3
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 10/147
turnover is negatively related to firm performance, a firm's investment in employee
training and development and firm reputation moderate this relationship.
REFERENCES
Axelrod, E. L., Handfield-Jones, H., & W elsh, T. A. 20 01 . War for talent, part two.
McKinsey Quarterly, 2: 9-12.
Batt, R. 2002. Man aging customer services: Human resource practices, quit rates, and
sales growth. Academy of Management Journal. 45 : 587-597.
Cappelli, P. 2000. A m arket-driven approach to retaining talent. Harvard Business
Review, 7 8: 103-113.
Cascio, W.F. 1995. Man aging Huma n Resources: P roductivity, quality of work life,
profits. New York: McGraw-Hill.
Collins, C. J., & Smith, K. G. 2006. Knowledge exchange and combination: The role of
human resource practices in the performance of high-technology firms. Academy
of Management
Journal,
4 9: 544-560.
Gardner, T. M. 2005. Interfirm competition for human resources: Evidence from the
software industry. Academy of Management Journal. 48: 237-256.
Hollenbeck, J. R., & W illiams, C. R. 1986. Turnover functionality versus turnover
frequency: A note on work attitudes and organizational effectiveness. Journal of
Applied Psychology,
71:
606-611.
Jackofsky, E. F. 1984. Turnover and job performance: An integrated process model.
Academy of Management Review, 9: 74-83.
Lepak, D. P., & Snell, S. A. 1999. The human resou rce architecture: toward a theory of
4
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 11/147
human capital allocation and development. Academy of Management Review, 24:
31-48.
Mobley, W. H. 1982. Some unanswered questions in turnover and withdrawal behavior.
Academy of Management Review, 7: 111-116.
Pfeffer, J. 1994. Competitive advantage through people. Boston: Harvard Business
School Press.
Randall, D.M. 1987. Commitment and the organization: The organization man revisited.
Academy of Management Review, 12 : 460-471.
Rao, H., & Drazin, R. 2002. Overcoming resource constraints on product innovation by
recruiting talent from rivals: A study of the mutual fund industry, 1986-94.
Academy of Management Journal,
4 5: 491-507.
Salamin, A., & Horn, P. W . 2005. In search of the elusive U-shaped Performance-
Turnov er R elationship: Are high performing Swiss bankers more liable to quit?
Journal of Applied Psychology, 90: 1204-1216.
Shaw, J. D., Dineen, B . R., Fang, R., & Vellella, R. F. in press. Em ployee-organization
exchang e relationship s, HRM practices, and quit rates of good and poor performer.
Academy of Management Journal.
Shaw, J. D., & G upta, N. 20 07. Pay system characteristics and quit patterns of good,
average, and poor performers. Personnel Psychology, 60: 903-928.
Staw, B. M. 1980. The consequence of turnover. Journal of Occupational Behavior, 1:
253-273.
Sturman, M. C , Trevor, C. O., Boud reau, J. W., & Gerhart, B. 2003. Is it worth it to win
the talent war? Evaluating the utility of performance-based pay. Personnel
5
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 12/147
Psychology, 56: 997-1035.
Trevor, C O ., G erhart, B ., & Boudreau, J.W. 1997. Voluntary turnover and job
performance: Curvilinearity and the moderating influences of salary growth and
promotions. Journal of Applied Psychology, 82: 44-61.
Wright, P. M., McMah an, G. C , & McW illiams, A. 1994. Human resources and
sustained competitive advantage: a resource-based perspective. International
Journal of Human Resource Management, 5: 301-326.
6
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 13/147
CHAPTER 2
HIGH COMMITMENT HR PRACTICES AND TOP PERFORMERS: IMPACTS
ON ORGANIZATIONAL COMMITMENT AND TURNOVER INTENTION
Human resources have been increasingly recognized as a key source for value
creation (Batt, 2002 ; Collins & S mith, 2006 ; Huselid, 1995; Pfeffer, 1994, 1998; Wright,
McMahan, & McWilliams, 1994). In particular, attracting and retaining top performing
know ledge wo rkers wh ose levels of performance are extremely high may significantly
influence organizational effectiveness (Cappelli, 2000; Lepak & S nell, 1999; Randall,
1987; Sturman, Trevor, Boudreau, & Gerhart, 2003). Scholars have argued that the
retention of top perform ers impacts organizational effectiveness because their turnover
not only imposes high replacement costs but also decreases organizational morale
(Cascio, 1995; Hollenback & Williams, 1986; Lucas, 1999; Mobley, 1982; Randall,
1987;
Staw, 1980). Given knowledge-based global competition, the importance of
retaining top pe rformers appears to be amplified due to the potential threat of valuable
know ledge transfer by top performers to competing organizations (Cap pelli, 2000). Also,
in a recent survey o f 410 corpo rate officers, McK insey & Comp any reported that the top
20 % of performers, relative to average performers, g enerate 40 % increased productivity
in operational ro les, 49 % increased profit in general managem ent roles, and 67 %
increased revenu e in sales roles (referred to Axelrod, Hand field-Jones, & W elsh, 2001).
W hile the retention of top performers is recognized as a key organizational
initiative (Cappelli, 2000; Martel, 2003; Michaels, Handfield-Jones, & Axelrod, 2001),
existing studies indicate that organizations may find it difficult to achieve this goal
7
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 14/147
because top performers tend to leave organizations m ore frequently than other levels of
performers (Jackofsky , 1984; Trevor, Gerhart, & Boudreau, 1997 ). Mo reover, recent
studies suggest that retaining top performers beco mes even m ore challenging due to the
inter-organizational market competition for valuable human capital (Cappelli, 2000;
Gardner, 2005; Rao & Drazin, 2002). Gardner (2005) argues that organizations compete
for valuable human capital not only to enhance their own competencies but also to
deplete their competitors' advantage. He observed that more than 20% of organizations
experienced 'purposive' talent raiding from their competitors. Recognizing the impact of
the retention of top performers on organizational performance, human resource
management (HRM) literature has recently argued that it is necessary to understand how
human resource practices influence the retention of high performing valuable employees,
beyond addressing their effects on organizational com mitment (OC) and overall retention
(Delery & Shaw, 200 1; Lepak & Snell, 1999).
From the theoretical perspectives and empirical evidence mentioned above, it
seems evident that both scholars and practitioners agree on the impo rtance of retaining
valuable hum an resource s for organizations to enhance organizational effectiveness.
How ever, few em pirical studies describe practices that can be used to retain top
performers and explain the reasons w hy the practices influence top perform ers' retention.
This paper proposes and tests the hypotheses that a bundle of high commitment human
resource practices (HCHRPs) may enable organizations to retain not only ordinary
employees but also top performers through their positive impact on O C, which is a strong
predictor of emp loyee turnov er (e.g., Meyer & Allen, 1997; Mathieu & Zajac, 1990). In
doing so, this pape r builds up on social exchange and perceived organizational support
8
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 15/147
theories that suggest that people tend to reciprocate benefits from other social entities, an
individual or organization, but the degree of reciprocation may vary depending on
individuals' perceived benefits (Blau, 1964; Eisenberger, Huntington, Hutchison, &
Sowa, 19 86; Gou ldner, 1960; Homa ns; 1974). This paper takes into account the
conditional aspect of social exchang e theory (Gouldner, 1960; Hom ans, 1974), and
argues that different work preferences (S mits, McLean, & Tanner, 1993; Trank, Rynes, &
Bretz, 2002) and degree of self-efficacy (Ackerman, Kanfer, & Goff, 1995; Wood &
Bandura, 1989) between top performers and other levels of performers m ay lead top
performers to perce ive H CH RPs m ore favorably. T his perception results in a higher
increase of OC of top performers comp ared to others. In addition, this paper tests that the
relationship between HCHRPs and turnover intention is mediated by employee OC.
This study enab les us to understand how o rganizations can effectively manage the
retention of top performers, which h as been recognized as a key research question but has
not been explored yet. Second, it shows that HCHR Ps may help organizations to retain
not only top performers but also other levels of performers w hose cooperation is
indispensable for an organization's success in a knowledge-based economy (Collins &
Sm ith, 20 06 ; Pfeffer, 1994 , 1998, 200 1). Third, this study prov ides a fresh insight for
strategic human resource management literature by suggesting that HCHRPs may
influence an org aniza tion's perform ance by reducing dysfunctional turnover and overall
workforce turnover. Lastly, practical implications are discussed based on the results of
this study.
9
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 16/147
THEORETICAL BACKGROUND AND HYPOTHESES
The literature in human resource management and organizational behavior has
identified several key factors and m echanisms that help retain employees in organizations.
First, existing studies have consistently reported that one of the strongest predictors of
employee turn over intention and actual turnover is OC , which dev elops through favorable
and fair exchanges between organizations and individual employees (Meyer & Allen,
1997;
Steers, 1977; Wallace, 1997). Second, employees' diverse work experiences
related to hu ma n resource practices have been identified as a significant antecedent of
OC among others (Meyer & Allen, 1997). Several studies in HRJV1 also indicate that
HCH RPs positively influence em ployees' O C (Agawalar, 2003; Appelbaum, Bailey,
Berg, & K alleberg, 20 00; G uest, 2001) or play a significant m oderating role in predicting
employees' OC (Whitener, 2001). Lastly, a few recent studies reported that
organizations' compensation practices or frequency of promotion also influence top
perform ers' turn over (A llen & Griffeth, 20 01 ; Lucas, 1999; Trevor et al., 1997).
Although existing studies help understand the determinants of turnover, few
studies exist which systematically investigate the factors and reasons that influence the
retention of top performers compared to the plethora of studies that explore how
individual demographics, work experiences, and organizational practices influence the
retention of employ ees in general (Griffeth, H orn, & Gaertner, 2000 ; Mathuie & Zajac,
1990; M eyer & Allen, 1997). This study considers both universal and contingent
perspectives of social exchang e theory and explores (1) the general relationship between
HCHRPs and employees' affective OC, (2) the stronger relationship between HCHRPs
and the affective OC of top performers relative to others, and (3) the mediation by
10
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 17/147
affective OC in the relationship between HCHRPs and employees' turnover intention.
Figure 1 briefly illustrates our research framework and hypotheses.
High Comm itment Human Resource Practices and Em ployees Affective OC
Research ers have con ceptualized H CH RPs in various ways, labeling a different
combination of human resource practices as high commitment, high performance, or high
involvement human resource practices. Despite varying concepts and labels of HCHRPs,
researchers share the perspective that organizations need to adopt HCHRPs which allow
organizations to develop skillful, motivated, and committed employees, who can
ultimately create superior value for o rganizations. In particular, the positive impact of
HCHRPs on employee OC and turnover has been a central premise in empirical studies in
the HRM literature (Arthur, 1992, 1994; Batt, 2002; Guest, 2002; Huselid, 1995).
Although the comp onents of HC HRPs vary in current HRM literature, HCHR Ps
generally constitute the following comm on practices: enriched job design, team-based
approach, participation in decision making, pay for performance
1
, high level of pay,
accurate performance appraisal, extensive training and developmen t, and selective
staffing (Ap pelbaum , et al., 2000 ; Arthur, 1994; Delery & Sh aw,
2001;
Guest, 1997;
Huselid, 1995; Pfeffer, 1998; Whitner, 20 01 ; W right, Gardner, & Moynihan , 2003).
OC refers to a strong belief in and acceptance of the organiza tion's goals and
values; a willingness to exert considerable effort on behalf of the organization; and a
definite desire to ma intain organizational mem bership (Porter, Steers, Mo wday, &
1
There is an ongoing debate on designing compensation practices of HCHRPs. Whereas some studies are
more likely to emph asize that organ izations need to pay for performance according to individual as well as
group and organization performance for motivating employees (e.g., Agarwala , 2003; Appelbaum et al.,
2000; G uthrie, 200 1; Huselid, 1995), others appear to suggest that organizations need to pay for
performance based on grou p or organization performance rather than individual performance in order to
enhance coop eration am ong organizational me mbers (e.g., Collins & Sm ith, 2006 ; Pfeffer, 1994, 1998).
How ever, both schools of scholars seem to agree that performance differences should be recognized.
11
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 18/147
Boulian, 1974: 604). The overarching theoretical rationale of the development of OC
draws upon social exchange theory and Gouldner's (1960)
'norm of reciprocity''
that
people should help those who have helped them (p. 171). Previous research suggests
that when em ployees perceive that organizational reward s and treatment are favorable
and fair to them , they develop OC as reciprocation toward organizations (Meyer & Allen,
1997; Steers, 1977; Wallace, 1997). W ork experiences and w ithdrawal behavior, such as
turnover, have been consistently found to be a strong antecedent and consequence of OC
(Angel & P erry, 198 1; Mathieu & Zajac, 1990; Meyer & Allen, 1997; Porter et al., 1974;
Steers, 1977).
The organ izational support theory put forward by Eisenberger and his colleagues
(1986,
1990) provides a theoretical rationale regarding why HCHRPs influence
employees' OC. Building upon social exchange theory as mentioned in the previous
section, Eisenberger et al. argued that employees tend to regard organizations as a person
and form a global belief regarding how this personified organization recognizes their
contribution and cares for their well-being. They prop osed that, if employees perceive
that their organ izations are supp ortive, they tend to reciprocate the organizations'
supportiveness w ith increased O C. In addition, they argued that perceived organizational
support is strongly influenced by organizational practices and policies such as pay, job
enrichment, and p articipation, wh ich signal the degree to which organizations value and
commit to their employees. A recent study by Rhoades, Eisenberger, and Armeli (2001)
showed that factors such as organizational rew ards, procedural justice, and supervisor
support influence affective OC by the m ediation of perceived organizational support.
This perspective points out that employees can perceive HCH RPs as evidence that their
12
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 19/147
organizations sup port and care for their well-being. The adoption of HC HRP s sends out
positive signals and communication, which may lead employees to reciprocate with OC
in exchange for the organizations' support for them.
Several studies have found a positive relationship betw een individual com ponents
of HCHRPs and em ployees' OC . Appelbaum et al. (2000) provided comprehensive
results in the relationships between the individual components of HCHRPs and
employees' OC. They found that pay for performance, employment security, fairness of
pay, flexible work schedu les, and prom otion opportunities positively influence
employees' OC. Meyer and Smith (2000) observed that HR practices such as
performance appraisal, benefits, training, and career developmen t influence perceived
organizational support and procedural justice which eventually enhance employees'
affective OC. Using a sample of
67
managers in an agricultural production company and
controlling supervision and work environment, Ogilvie (1986) found that employees'
perceptions to merit accuracy and promotion influence their OC. Furthermore, Agarwala
(2003) used a sample of 422 executives and manag ers in seven organizations and found
that a bundle of innova tive H R practices constituting 14 different practices influences
em ployee s' O C. Con sidering existing theories and empirical evidence, I predict that a
bundle of HCHRPs will influence employees' affective OC .
Hypothesis 1: A bun dle of high comm itment human resource practices is positively
related to employees' affective organizational commitment.
2
Researchers suggest that an organizational commitment q uestionnaire (OCQ) may com pound different
meanings of
OC
such as affective, continuance, and normative (M eyer & Allen, 1997; Meyer & Sm ith,
2000). In particular, it has been a rgued that, whereas affective OC which stems from emp loyees'
identification with the value s and goals of organizations may lead to beneficial ou tcomes, continuance
commitment base d on calcu lation on the costs of leaving may negatively influence organizations'
performance (lies, Mab ey, & R obertson, 1990; Meyer & Allen, 1997; Randall, 1987). In fact, most
discussion in HRM literature implicitly refers to affective OC rather than continuance or norm ative
commitment.
13
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 20/147
HCH RPs and O C of Top performers and Others
Social exchange theory and the norm of reciprocity, which provide overarching
theoretical explanations on the development of
OC,
argue that social and economic
exchange are inv olved in huma n interactions and people repay the benefits or debts
received from the other party (Blau, 1964; Gouldner, 1960). Social exchange theory, on
the other hand, sugg ests that the pattern and strength of reciprocation in exchange
relationships may vary de pending on different individu als' needs, preferences, and
situations, wh ich influence the degree to w hich the individuals perceive the value given
by the other party (Gouldn er, 1960; Hom an; 1974). Gouldner (1960) mentioned that a
norm of reciprocity is universal is not, of course, to assert that it is uncond itional.. .The
value of the benefit and hence the debt is in proportion to and varies with - among other
things - the intensity of the recipie nt's need at the time the benefit was bestowed (p.
171). Ho man s (1974) suggested that the one unit increase of value may be more
positively perceived by people who have less. Meyer and Allen (1997) also recognized
the impact of individual differences on OC. They stated that Given that individuals
differ in various w ays (e.g., personality, valu es, needs, expectations), it seems likely that
individual differences will have implication for which workplace experiences em ployees
would find particularly rew arding or fulfilling (p. 53). These conditional perspectives in
social exchange theo ry and its implication on OC suggest that the value of benefits and
debts bestowed by the other party m ay be perceived differently depending on individuals'
need and preferences, w hich m ay have a different degree of impact on individuals'
reciprocation processes.
14
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 21/147
By considering the conditional aspect of social exchange theory on the
development of
OC,
it is critical to identify if top performers and others have meaningful
individual differences to predict the differential impact of HCH RPs on OC of top
performers
and
other level of perform ers.
A few
studies indicate that top performers
and
others may have different p atterns of work preferences and self-efficacy. First, Trank,
Rynes, and Bretz (2002) conducted a study with a sample of college students to explore if
high achievers
and
lower ac hievers have different w ork preferences. Based
on
mo tivational research, they argued that high achievers tend to seek environments that
provide challenge
and
test themselves compared
to
others. They found that high
achievers prefer ch allenging and interesting w ork, pay for performance compensation,
and opportunity
for
promotion
and
training com pared
to low
achievers. Similarly, Smits,
McLean, and Tanner (1993) ob served that high achieving students prefer jobs with
challenge and autonomy. Second, top performers tend to have a higher level of self-
efficacy relative
to
others. Wood
and
Bandura (1989) argued that the degree
of
individuals' self-efficacy is most strongly influenced by mastery experience (previous
performance success). Ackerman, Kanfer, and Goff (1995) also reported that the previous
performance success is the most significant predictor of the level of individuals' self-
efficacy.
The different levels of self-efficacy and work preferences between top
performers
and
others m ay play
a
mod erating role
in
the relationship between HCHRPs
and OC. For example, Schaubroeck and Merritt (1997) ob served that, in the co ntext of
high job dem ands, people w ith high self-efficacy favor high levels of job control, but
3
Wood and Bandura (1989) defined self-efficacy as belief in one 's capabilities to mobilize the motivation,
cognitive resource, and course of action needed to meet given situational dema nds (p. 408).
15
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 22/147
people w ith low self-efficacy prefer low levels of job control. Litt (1988) argued that
control may b e beneficial only for people who hav e high levels of self-efficacy. In
addition, Godard (2001) argued that excessive team autonomy and responsibility may
decrease employees' OC due to their impact on work stress. However, Defrank and
Ivancevich (1998) po inted out that whether these practices are perceived as stress or not
depends on em plo yee s' levels of self-efficacy. These studies suggest that although
participation and job autonomy practices of HCHRPs may positively influence employee
OC,
top performers may perceive these practices more favorably than other levels of
performers, resulting in higher levels of OC of top performers relative to others, because
of their higher level of self-efficacy and matched work preference.
In addition, some theoretical and empirical work helps explain why top
performers are more likely to prefer H CH RPs to control-based traditional HR practices.
Therefore, the O C of top performers will be more likely to be affected by the
motivational components of HCHRPs. Research in organizational justice argues that
distributive and procedu ral justice strongly influence em ploye es' O C (e.g., Colquitt et al.,
2001; Greenberg, 2001). A few studies demonstrated that HR practices such
compensation and performance appraisal may influence individuals' (injustice
perceptions (e.g., Folger & G reenberg, 1985). Considering the theoretical perspective of
organizational justice, if em ployees are paid less than other comp anies on the basis of
their seniority, withou t the benefit of performance appraisal and rew ards, top performers
are less likely to pe rceive organizations fair. Top p erformers are likely to perceive that
their organizations are not supportive and, as a consequence, reciprocate perceived lower
organizational supportiveness with reduced OC.
16
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 23/147
In sum, the aforementioned theoretical perspectives and empirical works suggest
that a bundle of HC HR Ps will be mo re likely to increase the affective OC of top
performers than that of others because HC HR Ps tend to better meet top performers w ork
preferences and provide them with appropriate working conditions.
Hypothesis 2: A bundle of high commitment human resource practices is positively
related to the affective organ izational co mm itment of top performers more than that of
lower performers.
Affective OC as a Mediating Mechanism between HCHRPs and Turnover Intention
Research shows that employees' turnover intention, the most proximal predictor
of actual turnov er, is influenced b y their job attitudes (Griffeth & Horn, 1995; Steers &
Mo wday , 1981). In particular, OC h as been identified as a strong predictor of turnover
intention and the plethora studies have reported that employees com mitted to their
organizations are less likely to intend to leave their organizations (M athuie & Zajac,
1990; Mey er & A llen, 1997; Tett & Meyer, 1983). In the previous sections, I addressed
that HCH RPs m ay significantly influence OC , which im plies that OC may mediate the
relationship between HCHRPs and turnover intention. Additional evidence in this
relationship is given by a recent study of Batt and Valcour (2 003). They conducted a
study to explore how incentive and work design practices influence employees' turnover
intention. They found that incentive practices such as relatively high pay, em ployment
security, and career develop men t significantly influenced em ploye es' turnover intention,
and a work practice designed to increase decision-making autonomy also influenced,
though weak, their turnover intention.
17
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 24/147
Based on the resu lts of existing emp irical studies, I develop a hypothesis that the
relationship between HCHRPs and turnover intention will be mediated by their impact on
OC.
This hypo thesis is not new but attempts to confirm the validity of the assumption in
most previous studies in HRM literature that HCHRPs will influence employees'
turnover through their impact on employees' attitude toward their organization (Guest,
2002).
Hypothesis 3: The relationship between a bundle of high commitment human resource
practices and turnover intention will be mediated by affective organizational commitment.
METHODS
This study uses d ata which w ere collected by a research institute from 11
subsidiaries of a multinational conglomerate located in East Asia. Each participant
organization produces different material goods or provides financial services for diverse
custom ers, except for an organization that condu cts the research and development for this
whole conglomerate. Each organization consists of 1,000 to 50,000 employees and
operates its own bu siness in a fairly autonom ous w ay.
Sample and Procedures
The survey was conducted in the middle of 2002 with the cooperation of
researchers and the human resource headquarters of this conglomerate. The participants
of this study were ran dom ly selected through its computerized database system. After
sampling participants, a staff member in the HR headquarters of the conglomerate
contacted a senior HR man ager of each subsidiary of the organization and asked him or
her to distribute questionnaires to the participants, who w ere working in his or her
18
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 25/147
organization. Each participant received a survey questionnaire in which, in the front page,
researchers guaranteed the confidentiality of the survey. Self-addressed stamped
envelopes were provided to participants, and the survey was conducted anonymously.
638 out of total 800 targeted subjects, which were composed of 500 top performers and
300 relatively lower level performers, responded to the questionnaire resulting in a total
response rate of 80%. A total of 589 cases out of 638 were used for the analysis after
excluding a few invalid responses, which consisted of
363
top performers and 226 other
levels of performers. All participants were full-time managerial or professional workers
whose job levels comp rised the m iddle level of the organizational hierarchy. Table 1
presents the detailed dem ographic characteristics of the subjects used in this study.
Measures
The survey asked em ployees a wide variety of questions to measure job attitudes,
human resource practices, and turnover intention. Based on the survey questions,
independent and dependent variables were measured.
High co mm itment h uman resource practices HCH RPs were measured by
employees' perceptions of human resource practices. Measuring human resource
practices by em ploy ees' perception is not unusual in the current HRM literature,
especially for studies exploring the relationship between HR practices and an individual
employee's OC (Agawala, 2003; Applebaum et al., 2000; Gould-Williams, 2004; Guest,
2002; Smith & Meyer, 2000). The perceptual measures appear appropriate from the
theoretical perspective that HR practices influence employees' OC or other individual
and organizational outcomes, when the meaning of the HR practices are well
comm unicated to employees (Bowen &
Ostroff,
2 004; Eaton, 2003; G aertner & N ollen,
19
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 26/147
1989;
Guest, 2002 ; Guzzo & N oonan, 1994). In addition, Meyer and Allen (1997) argued
that perceptual measures would be more relevant in exploring the relationship between
HR practices and em ploy ees' affective OC because emp loyees will react to conditions
as they perceive them (p.88).
HCHRPs were measured by employees' perception of six different HR practices,
which are frequently men tioned in previous research: job design, participation in decision
making, training and development practice, pay for performance, high level of pay, and
performance appraisal. All individual q uestions used a 1-5 Likert-type response format,
where 1 denotes strongly disagree and 5 denotes strongly agree . Job design was
measured by asking employees three questions that assessed the perceived degree to
which organ izations allow them to use their discretion in performing their job
(Appelbaum et al, 200 0; Bae & Lawler, 2000 ). The three items (a = 0.70) are as follows:
This organization provides me with job autonom y, This organization allows me to set
rules and goals of my w ork , and This organization provides me with a challenging job .
Participation in decision making was measured by two items (a = 0.79), which are
consistent with questions used by Delery and Doty (1996) and Wright, McCormick,
Sherman, and McM ahan (1999). The two items were, In this organization, I can give
opinions on various business activities, and This organization provides me with
opportunities to be involved in mak ing important decisions. Pay for performance was
measured by a single item by asking emp loyees, This organization pays me according to
my performance . Performance appraisal was measured by asking employees two items
(a = 0.84) adapted from Snell and Dean (1992): This organization conducts performance
appraisal with clear performance criteria and proced ures, and This organization
20
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 27/147
evaluates my performance accurately . Training and development was measured by a
single item, This organization provides training and development opportunities for me to
acquire necessary skills and know ledge. Lastly, high level of pay was measured by a
single item, This org anization pays me a high level of salary . Then, I aggregated scores
of each practice ( a-0 .86 ) to test the impact of HCH RPs as a bundle on employees' OC.
Top and other levels of performers Employees were classified into two groups,
top performers v ersus other performers, by the orga nizations' archival data of each
employee's performance appraisal. Using the organizations' performance measurement
scheme, a group of top performers were classified if either one of the two following
conditions was m et: first, the employees h ave received all 'A ' ratings (A is given to the
top 20% performers in each year) in the past three years ' annual performance appraisal;
or second, the employees received 'Super' ratings (extraordinary high performance) in
the previous one or two y ears ' annual performance appraisal. Less than 5% of total
employees belonged to the group of top performers. Employees who received B, C, or D
in the last one year's perform ance appraisal, which w ere given to the other 80% of
emp loyees, were classified into a group of Others. A dum my coding was 1 for to p
performers and 0 for others. Organizations measured employees' performance by two
dimensions: task performance on the basis of MBO and employees' competencies
including several sub -dimen sions such as task specific and general co mpetencies. The
exemplary items for competencies were leadership, communication, teamwork and
collaboration. The weights for task performance and competencies were approximately
60% and 40%, respectively, although the weights vary somewhat among organizations.
The performance rating of each emp loyee was measured b y three steps: first, employee
21
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 28/147
self-appraisal; second, supervisor evaluation; and third, final evaluation in the human
resource department. During this evaluation process, frequent interviews between
emp loyees and their supervisors w ere utilized. All of these organizations used the results
of performance appraisal as a key decision factor in promotion and pay raise. They used a
forced distribution method to measure employees' performance, which appears to suggest
the validity of the measurement of employees' performance.
Dependent variables Affective OC w as measured by matching seven items of the
nine items scale me asure used by Tsui, Pearce, Porter, and Tripoli (1997) who developed
scales adapted from Ang el and P erry 's (1981) OC scales. The reliability of the affective
OC in this study was .79. Responses were recorded on a five-point scale, where
1
means
strongly disag ree and 5 mea ns strongly agree . Sample items were I find that my
values and the orga nizatio n's values are very similar , I really care about the fate of this
organization , and This organization inspires the very best in the way of job
perform ance. Turno ver intention was measured with a single item, If you have
intention to leave this organization, when are you going to leave? Response options
were 1 (one year ), 2 (two y ears), 3(three to five years), 4 (five to ten years), 5 (will not
leave). The responses to this question w ere treated as a continuous variable, 1 (highest
turnover intention) to 5 (lowest turnover intention).
Control variables Dem ographic variables such as age, organizational tenure,
education, and profession were con trolled because previous studies consistently indicated
that these variables influence OC (M athuie & Zajac, 1990; Meyer & Allen, 1997). Each
personal characteristic was measured using self-report measures. Age was measured
using a one-item self-report mea sure, which consisted of three categories (1 = 30 to 34, 2
22
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 29/147
= 35 to 39, 3 = over 40 y ears). Organizational tenure was measured by the year served in
the current organization from one self-report item. Professions w ere measured by one
item self-report measu re (0 = scientists in R& D, 1 = managers). Education was measured
by one item self-report measu re (1 = high school, 2
=
junior college (2 years), 3 =
bachelor (4 ye ars), 4 = master, 5 = doctorate) and I converted this final educational
degree into the years o f education. In addition, I controlled the deg ree of supervisor
support, which has be en identified as a strong predictor of OC (Mathuie & Zajac, 1990;
Meyer & Allen, 1997; Tusi et al., 1997). How ever, previous studies did not control this
variable in exploring the relationship between HR practices and employees' OC (except
for Ogilvie, 19 86). Supervisor support was m easured by five items (a = 0.89) adapted
from Tsui et al. (1997 ). Finally, ten organizational dumm ies were included in the analysis,
to capture organizational differences (Tsui et al., 1997).
Analysis. I tested hypothesis 1 and 2, using a hierarchical regression method
(Cohen & Cohen, 1983) that was designed to assess whether a single or sets of variables
explain additional variances explained by a set of control variables. I first loaded control
variables in step 1, including age, tenure, profession, education, organization dummies,
and supervisor support. In the following steps, I added independent variables representing
HC HR Ps, a perform ance group dum my (top or other levels of performers), and an
interaction variable between HCHRPs and the performance group dummy. I observed the
significance of coefficients of variables as well as R
2
change to assess whether the
hypotheses were supported or not. In order to reduce any multi-co-linearity problem, the
score of HCHRPs was mean centered in testing hypotheses (Aiken & West, 1991). In
order to test the mediation effect of OC in the relationship between HCHRPs and
23
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 30/147
turnover intention (hypothesis 3), I followed three steps procedures to test the existence
of mediation effect suggested by Baron and Kenny (1986).
RESULTS
Table 2 contains the m eans, standard deviations, and a correlation matrix of
variables in this study. First, consistent with the findings of previous studies on OC
(Mathuie & Z ajac, 1990; Steers, 1977), most of the demographic variables are
significantly correlated with affective OC. For example, affective OC is positively
correlated w ith tenure, age, and profession (m anagerial), and affective OC is negatively
correlated with education. However, as Meyer and Allen (1997) suggested, the magnitude
of the correlation betw een H CH RPs index and affective OC is much stronger than those
between perso nal ch aracteristics and affective OC . Second, as expected, a significant
correlation exists betw een H CH RPs index and affective OC . Third, interestingly,
affective OC and performan ce are not significantly correlated, which is in line with
previous studies (e.g., Keller, 1997; Mathuie & Zajac, 1990; Somers & Birnbaum, 1998;
Steers, 1977).
HCH RPs and Affective OC Hypothesis 1 predicted that a bundle of HCHRPs
would po sitively influence e m ploy ees' affective OC . The results of the two regressions,
mod el 1 and mode l 2 in Table 3, support this hypothesis. Model 1 (F=2 3.8, p < 0.001), a
reduced m odel in the hierarchical regression ana lyses, shows that a relational variable -
supervisor support (p < 0.001) and demographic variables
4
- tenure (p < 0.01) and age (p
4
1 regressed affective OC on only demog raphic variables, tenure, age, education, and profession. That
regression model was significant (F = 19.249, p < 0.001; if = 0.14) and tenure (p < 0.01), profession (p <
0.05), and age (p < 0.001). However, as supervisor support and HCHRPs were added in subsequent
regression mod els, the significance of demographic variables such as tenure and profession disappeared.
24
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 31/147
< 0.001) - significantly influence affective OC . Mod el 2, where the HPH RP index is
added to model 1, indicates that a bundle of HC HR Ps (p < 0.001) significantly explains
additional variance (A
R
2
=
0.118,
F
for
R
2
=139.6, p < 0.001). Th is result supports
Hypothesis 1 even after variables such as demographics and supervisor support are
controlled suggesting a strong positive impact of a bundle of HCHRPs on employees'
affective OC.
HC HR Ps a nd Affective OC between top performers and others Hypothesis 2
predicted that a bund le of HC HR Ps wou ld more positively influence the affective OC of
top performers than that of other levels of performers. In order to test hypothesis 2,1 first
ran model 3, which includes all control variables, HCHRPs index, and a performance
group dum my . Mod el 3 was significant (F=3 4.5, p < 0.001), indicating that both the
HC HR Ps index (p < 0.001) and the performance group dum my (p < 0.05) significantly
influence affective OC. Interestingly, model 3 points out that the coefficient of the
performance group du mm y is significant but negative, which m eans that top performers
are less likely to be affectively comm itted to their organizations. Second, I ran model 4,
where an interaction term (HCHRPs index x Performance dummy) was added to model 3.
This showed that the coefficient of the interaction term is significant (p < 0.05) and
accounts for significant portions of additional variance (A R
2
= 0.004, F for A R
z
=5.0, p <
0.05). These results support hyp othesis 2 .
Mediation by Affective OC in the relationship between HCHR Ps and Turnover
Intention Hypothesis 3 predicted that the relationship between HCHRPs and turnover
intention will be me diated by affective OC . In order to test the mediation effect of
These results are consistent with arguments by Meyer and Allen (1997) that relational and work practices
variables generally have a strong impact on affective OC but demographic variables have, at best, a weak
impact on affective OC.
25
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 32/147
affective OC in this relationship, I ran three different regressions with all subjects (Baron
& Kenny, 1986): first, model 5 regressing turnover intention on HCHRPs index; second,
mod el 6 regressing turnov er intention on affective OC ; and third, model 7 regressing
turnover intention on both H CH RPs and affective OC . The results of this series of
regressions are reported in Table 4 . In addition, I ran three different regressions, models 8,
9, and 10, with only top perform ers in order to investigate whe ther the relationship holds
for top performers.
The significant impact of both HCH RPs (in model 5, p < 0.01; in model 8, p <
0.01) and affective OC (in mo del 6, p <
0.001;
in model 9, p < 0.01) on turnover intention
were found. W hen I added both H CH RPs and affective OC to the equation (models 7 and
10), the significance of the coefficient of affective OC (in model 7, p < 0.001; in model
10, p < 0.001) remains but that of HCHRPs disappears in model 7 and model 10.
Therefore, the results of the series of regressions support hyp othesis 3 and also suggest
that this relationship ho lds for top perform ers.
DISCUSSION AND CONCLUSION
The results of this paper show that HCHRPs positively influence employees'
affective OC an d is con sistent with the findings of previous studies. Furthermore, I tested
this relationship using the construct of affective OC and by controlling the effect of
supervisor support on affective OC . The supportive result of hypothesis 1, therefore,
strengthens the argum ents and findings of previous studies in this relationship (e.g.,
Agawala, 2003; A ppelbau m et al., 2000 ). At the same time, the results of this study
indicate that researchers w ho are interested in investigating this relationship may need to
26
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 33/147
take into account the q uality of the relationship between supervisor and subordinate or
between coworkers (cf., Ogilvie, 1986; Tsui et al., 1997).
Second, I found that the relationship betw een HC HR Ps and turnover intention is
mediated by affective OC . This result contributes to understanding the relationship
between HRM and firm performance. The HRM-firm performance literature argued that
HCHRPs influence firm performance through their impact on turnover, drawing upon the
assumption o n the positive impact of HC HR Ps on employees OC (Arther, 1992, 1994;
Batt, 2002; G utherie, 20 01 ; Huselid, 1995). Recent studies in fact explored and reported
that HCHRPs influence either employees' OC or turnover intention, but these studies
have not explored the mediation effect of em ployee s' OC in the relationship between
HCHRPs and turnover intention or actual turnover (Agawalar, 2003; Appelbaum et al.,
2000; Batt & Valcour, 2003; G ould-W illiams, 2004). The results of this study provide
strong evidence that HCHRPs not only influence employees' OC but also reduce
em ploye es' turnove r intention through increased affective OC . Consequently, this result
provides evidence that HCHRPs influence employees' OC and turnover intention, which
results in lower organizational turno ver and higher firm performance.
Third, the results of this study indicate that HCHR Ps have a stronger impact on
top perform ers' affective OC than o thers. Figure 2 illustrates this relationship showing
that top performers tend to hav e a lower affective OC relative to others when the level of
HC HR Ps is low. Howe ver, as the level of HCHR Ps increases, the affective OC of top
performers increases more rapidly than o thers. When the practical significance was
calculated based on one standard deviation increase of HCHRPs from the mean (Aiken &
W est, 199 0), the affective OC of both top performers and others increased but the
27
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 34/147
magnitudes were
10.1%
for top performers and 7.3 % for others. This result can be
explained from the conditional perspective of social exchange and perceived
organizational support theories that the same exchange relationship between
organizations and e mp loyees can be interpreted differently.
Fourth, this study p rovides interesting theoretical q uestions for the OC and
turnover literature. O C literature has tried to find a positive relationship between O C and
job performance (Somers & Birnbaum, 1998). This study implies that OC may be
influenced by the interaction betw een perceived or actual practices and job performance
rather than job perform ance being influenced by OC influence. The direct or moderating
impact of job performance on OC is not new but relatively unexplored in the OC
literature. For exam ple, Steers and Porter (1981) argued that performance may influence
OC, and Lee and Mowday (1987) empirically observed the positive impact of job
performance on OC . Furthermore, Law ler and Porter (1967) emphasized that job attitude
may not influence job perform ance but performance can be influenced by the interaction
of employees' performance and organizational treatment. The results of this study
indicate that it m ay be fruitful for researchers to take into accou nt this interactional
perspective while exploring how OC is developed and how OC is related to job
performance.
Lastly, this paper sheds light on why hum an resource practices, particularly
HCHRPs, influence an organization's performance. Diverging from previous studies in
HRM focusing on the impact of HCHRPs on employees' OC, this study explores the
different impact of HC HR Ps on affective OC between top performers and others using
professional knowledge workers who can be a significant source of competitive
28
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 35/147
advantage for the firm (Lepak & Snell, 1999). The supporting results imply that a
mechanism through which HCHRPs influence firm performance is they may enable
organizations to increase the affective OC o f top performers and, thus, retain them. In
addition, the positive impact of HCH RPs on affective OC of other levels of performers
implies that they may enhance the cooperation among organizational members, which is
another key factor for organizational effectiveness (O'Reilly & Pfeffer, 2000; Pfeffer,
1994, 1998). By considering both these positive im pacts at the same time , this paper
concludes that HCH RP s may enhance o rganizational effectiveness because they not only
enable an organization to take advantage of superior hum an capital but also increase
cooperation am ong organizational m embers. Consequently, HCH RPs may help
organizations to create and maintain competitive advantage through people.
Practical Implications
Man y firms have increasingly paid attention to the retention of top performers,
but they often fail to establish a more systematic approach. As Gardner (2005) pointed
out, firms a ttempted to change their HR M practices only after top performers left the
organization. A s show n in the results (mod el 3 in Table 3), top performers revealed lower
OC (and higher tu rnover intention) when firms do not provide fair treatment for their
contributions.
For attracting and retaining the best peop le, engaging in a 'war for talent' and
provision of special monetary incentives for them are suggested (Cappelli, 2000;
Micha els et al., 2001). H owev er, Pfeffer
(2001:
249) argued that firms should not
necessarily try to win the war for talent, even adopting this image as a management
metaphor can be quite haza rdou s to organization 's health. This is because engaging in a
29
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 36/147
'war for talent' creates internal competition and disharmony deemphasizing cooperation
and team wo rk and g enerates a self-fulfilling prop hecy for two g roups (i.e., top talent vs.
other people). Therefore, fair treatment does not necessarily mea n providing higher
monetary rewards based on individual performance. Our results suggest that HCHRPs
enhance O C of top performers more rapidly and can eventually help to retain them.
To further investigate the specific effects of HRM functions on affective OC, an
exploratory factor analysis was conducted and arrived at two comm on factors -
inducement and involvement from six HCHRPs. While the inducement factor
(eigenvalue of 3.12 and explained
51%
of total variance) includes pay for performance,
high level of pay, and training and developmen t, the involvement factor (eigenvalue of
1.17 and explained 20 % of total variance) covers job design and participation in decision
mak ing. The results of a series of hierarchical regressions using these two factors
demon strated that the interaction effect between the inducem ent factor of HC HRPs and
the performance dum my (top versus other performers) is significant and explained
additional variance (A R
2
=
0.004, F for A R
2
=4.57, p < 0.05); and the interaction effect
between the involvement factor and the performance dummy is also marginally
significant and explained additional variance (A R
2
= 0.003, Ffo r
A
R
2
=3.33, p < 0.1).
From th e results of additional analyses, it was observed that basically both factors
have significant interaction effects with top performers on affective OC . Therefore, both
involvement and inducement are important. Enhancing the involvement factor through
job autonom y and participation in decision m aking is critical to gain high affective OC of
employees (Appelbaum et al., 2000; Meyer & Allen, 1997; Wright et al., 2003). However,
their impact on the affective OC of top performers and others may not be necessarily
30
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 37/147
homogeneous. Several studies pointed out that job autonomy and participation in decision
making may not be universally rewarding practices because they may increase work
stress (Defrank & Ivance vich, 1998; Truss, 2001). However, a well-designed
involvem ent program is critical to enhance affective OC . The involvement of human
resources training and development, as a high commitment HRM system suggests,
generates better results. Furthermore, high involvement creates much better outcomes in
terms of affective OC for top performers.
The results of additional analyses also suggest that inducem ent is a critical factor
for enhan cing affective OC , especially for top performers. Ry nes, Gerhart, and Minette
(2004) emp hasized that the impo rtance of monetary rew ards for top performers is
significantly under-eva luated by researchers and practitioners. Recent studies suggest that
monetary reward, which has been regarded as a lower needs or an extrinsic motivator,
can be one of the highest levels of needs for top performers because m onetary rewards to
top performers may indicate their
worth
in organizations and
social status
(Frank, 1999;
Rynes et al., 200 4; Trank et al., 2002). Therefore, mo netary incentives can have an
informational effect rather than a controlling effect (cf, Shalley & Perry-Smith, 2001).
How ever, this does no t necessarily m ean that firms should have strong incentive
intensity and w ide pay differential. As discussed, the inducem ent factor in this study also
includes relatively high pay and training and developm ent beyond pay for performance.
In addition, though our m easuremen t on pay for performance contains both individual
and group perform ances, th e ratio of task performance versus com petence is 6 to 4. Also,
competence includes leadership, communication, teamwork and collaboration, which are
closely related to the high commitment HRM approach.
31
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 38/147
Therefore, a bundle of HCH RPs - both inducement and involvement factors -
generate differences in terms of affective OC and turnover intention of top performers
and other groups of people . The results of this study suggest that current changes toward
performance-based system s (i.e., individual performance-based pay and wide pay
differentials amon g em ployees) in Asia are not the only answer. It may solve some
existing problems such as rigidity and complacence but at the cost of generating another
set of problem s such as internal com petition, high turnover rate, and loss of trust and
loyalty. Personal interview s with m anagers of firms participated in this survey revealed
that they were concerned about possible disharmony when they provide special monetary
treatment to top perform ers. T he results of this study suggest that HC HR Ps for all
employees may solve this problem by providing justice and perceived organizational
support which are critical influences of OC.
Much academic and practical attention has been paid to understanding how
organizations can retain top performers. The purpose of this paper is to address this issue
focusing on the relationship betw een H CH RPs and the affective OC of top and other
levels of perform ers. The pape r also attempted to show that HCH RPs influence
em ploy ees' turnov er intention through their impact on affective OC . I mainly explored
these relationships on the basis of the social exchange and perceived organizational
support theories, which suggest that when individuals receive a favor from other parties,
the individuals return the favor to them but the degree of reciprocation m ay vary
depending on their perceived value of benefits and debts. In particular, I concluded that
the degree of the relationship between HC HR Ps and affective OC m ay be different
between top perform ers and others due to different work preferences, levels of self
32
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 39/147
efficacy, and perform ance. Based o n these theoretical perspectives, this paper predicted
and established that HC HR Ps not only increase the affective OC of emp loyees but are
also more likely to influence the affective OC of top performers relative to that of other
levels of performers.
Althoug h this paper provide s promising results and future research directions
about how organizations may effectively manag e the affective OC and turnover of top
performers, more studies are required to broaden scientific know ledge in this topic. In
particular, future research ne eds to explore the interrelated relationships among actual
and perceived HCHRPs, turnover, and firm performance while examining employees' job
performance at both individual and organizational levels. This study has limitations
because it measures several components of HCHRPs with single-item measures as the
survey was designed not for this study but for organ izations' needs. Also, this study lacks
the ability to address the cau sality inherent in cross-sectional data. D espite these
limitations, the study sheds light on how organizations can create and m aintain
competitive advantage through people by addressing the relationship among HCHRPs,
affective OC , and turnover intention of top and other levels of performers using a sample
of middle level managerial and R &D knowledge w orkers.
REFERENCES
Ackerman, P. L., Kanfer, R., &
Goff,
M. 1995. Cognitive and noncognitive determinants
and consequences of complex skill acquisition. Journal of Experimental
Psychology: Applied,
1: 270-304.
Adler, P. S. 20 01 . M arket, hierarchy, and trust: The know ledge econom y and the future
33
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 40/147
of capitalism. Organization Science, 12: 215-234.
Adler, P.S. & Heckscher, C. 2006. Towards collaborative community. In C. Heckscher &
P.S. Adler (Eds.), Th e firm as a collaborate com munity: Reconstructing trust in
the knowledge economy (pp, 11-105). Oxford: Oxford University Press.
Agarwala, T. 2003. Innovative human resource practices and organizational commitment:
An empirical investigation. International Journal of Human Resource
Management, 14: 175-197.
Aiken, L. S., & West. S. G. 1991. Multiple regression: testing and interpreting
interactions. Newbury Park: Sage Publications.
Allen, D. G., Griffeth, R. W . 200 1. Test of mediated performance-turnover relationship
highlighting the moderating roles of visibility and reward contingency. Journal of
Applied Psychology, 86: 1014-1021.
Ang le, H. L., & Perry, J. L. 19 81. Organizational com mitment and organizational
effectiveness: An empirical assessment. Administrative Science Quarterly, 26: 1-
14.
Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. 2000. Manufacturing Advantage:
Why high-performance work systems pay off Ithaca, NY : Cornell University
Press.
Arther, J. B. 1992. The link between business strategy and industrial relations systems in
Am erican steel minimills. Industrial and Labor Relations Review, 45 : 488-506.
Arther, J. B. 1994. Effects of human resource systems on manufacturing performance and
turnover. Academy of Management Journal, 37: 670-687.
Axelrod , E. L., Han dfield-Jones, H., & W elsh, T. A. 2001 . War for talent, part two .
34
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 41/147
McKinsey Quarterly, 2: 9-12.
Bae, J., & L awler, J. J. 2000. Organizational and HRM strategies in Korea: Impact on
firm performance in an emerging economy, Academy of Management Journal, 43:
502-517.
Baron, R. M., & K enny, D . A. 1986. The mod erator-mediator variable distinction in
social psychological research: Co nceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology. 51 : 1173-1182.
Bart, R. 2002. M anaging customer services: Hum an resource practices, quit rates, and
sales growth. Academy of Management
Journal,
45 : 587-597.
Batt, R., & Valcour, P. M. 2003. Hum an resources practices as predictors of work-family
outcomes and employee turnover. Industrial Relations, 42: 189-220.
Blau, P.M. 1964. Exchange and Power in Social Life. W iley, New York.
Bowen, D. E., & Ostroff, C.I. 2004. Understanding HRM-FIRM performance linkage:
the role of the strength of the HR M system. Academy of Management Review,
29:203-221.
Cappelli, P. 2000. A m arket-driven ap proach to retaining talent. Harvard Business
Review, 8: 103-113.
Cascio, W. F. 1995. Man aging Hum an Resources: Productivity, quality of work life,
profits. New York: McGraw-Hill.
Cohen, J., & Cohen, P.
1983.
Applied m ultiple regression/correlation analysis for the
behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.
Collins, C. J., & S mith, K. G. 2006 . Know ledge exchange and com bination: The role of
35
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 42/147
human resource practices in the performance of high-technology firms. Academy
of Management Journal, 49 : 544-560.
Colquitt, J. A., C onlon, D . E., Wesson, M . J., Porter, C. O. L. H., & Ng, K. Y. 200 1.
Justice at the Millennium : A m eta-analytic review of 25 years of organizational
justice research. Journal of Applied Psychology, 86: 425-445.
Dalton, D. R., Krackhardt, D . M., & Porter, L. W. 19 81. F unctional turnover: an
empirical Assessment. Journal of Applied Psychology, 66: 716-721.
DeFrank, R . S., & Ivancevich , J. M. 1998. Stress on the job : An executive update.
Academy of Management Executive, 12: 55-66.
Delery, J. E., Doty, D . H. 1996. Mo des of theorizing in strategic human resource
ma nagem ent: tests of universalistic, contingency, and configurational
performance predictions. Academy of Management
Journal,
39 : 802-835.
Delery, J. E., & Sh aw, J. D. 20 01 . The strategic managem ent of people in work
organiza tions: review , synthesis, and extension. In: G. R. Ferris (Eds), Research
in personnel and human resources Management (Vol. 7, pp. 165-197). Greenwich,
CT: JAI Press.
Eaton, S. C. 2003. If you can use them: Flexibility policies, organizational comm itment,
and perceived performance. Industrial Relations, 42: 145-167.
Eisenberger, R., Fasolo, P., & Davis-LaM astro, V. 1990. Perceived organizational
support and employee diligence, commitment, and Innovation. Journal of Applied
Psychology, 75: 51-59.
Eisenberger, R., Hun tington, R., Hutchison, S., & Sowa, D. 1986. Perceived
organizational support.
Journal of Applied Psychology,
71 : 500-507.
36
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 43/147
Folger, R., & Green berg, J. 1985. Procedural justice: An interpretive analysis of
personnel systems. In: K. M. Rowland & G. R. Ferris (Eds), Research in
personnel and hum an resources management (V ol. 3, pp. 141-183). Greenwich,
CT:
JAI Press.
Frank, R. H. 1999. Luxury Fever: Why mo ney fails to satisfy in an era of excess. New
York: The Free Press.
Gaertner, K. N . & Nollen , S. D. 1989. Career experiences, perceptions of emp loyment
practices, and psychological commitment to the organization. Human Relations,
42:975-991.
Gardner, T. M. 200 5. Interfirm comp etition for human re sources: Evidence from the
software industry. Academy of Management
Journal,
48 : 237-256.
Godard, J. 200 1. High perform ance and the transformation of work? The implications of
alternative work practices for the experience and outcomes of work.
Industrial
and Labor Relations Review, 54: 776-805.
Gouldner, A. W. 1960. The norm of reciprocity: A preliminary statement. Am erican
Sociological Review, 25: 161-178.
Gould-Williams, J. 2004. The effects of 'high commitment' HRM practices on employee
attitude: The views of public sector workers. Public Administration, 82: 63-81.
Greenberg, J. 2001.T he seven loose can(n)ons of organizational justice. In: J. Greenberg
& R. Cropanzano (Eds.), Advances in organ izational justice (pp. 245-271). Palo
Alto, CA: Stanford University Press.
Griffeth, R. W ., Horn, P. W. 1995. The employee turnover process. In G. R. Ferris (Eds),
37
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 44/147
Research
in
personnel and human resources management, 13: 245-294.
Greenwich, C T: JAI.
Griffeth, R.W ., Horn, P.W., G aertner, S. 2000. A m eta-analysis of antecedents and
correlates of employee turnover: Update, moderator tests, and research
implications for the next millennium. Journal of Management, 26: 463-488.
Guest, D. E. 1997. Human resource management and performance: a review and research
agenda. International Journal of Human Resource Management, 8: 263-276.
Guest, D. E. 2001. Human resource management: when research confronts theory.
International Journal of Human Resource Management, 12: 1092-1106.
Guest, D. E. 2002. Human resource management, corporate performance and employee
wellbeing: Building the worker into HRM . T he Journal of Industrial Relations,
44: 335-358.
Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity:
Evidence from New Zealand. Academy of Management Journal, 44: 180-190.
Guzzo, R. A., & Noonan, K. A. 1994. Human resource practices as communications and
the psychological contract. Hum an Resource Management, 33: 447-462.
Hollenbeck, J. R., & William s, C. R. 1986. Turnover functionality v ersus turnover
frequency: A note on w ork attitudes and organizational effectiveness. Journal of
Applied Psychology, 71: 606-611.
Homans, G. C. 1974. Social B ehavior: Its elementary forms (revised edition). New York:
Harcourt Brace Jovanovich.
Huselid, M. A. 1995. The impact of human resource management practices on turnover,
productivity, and corporate financial performance. Academy of Management
38
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 45/147
Journal,
3 8: 635-672.
lies, P., Mab ey, C , & R obertson, I. 1990. HRM practices and employee com mitment:
Possibilities, pitfalls and paradox es. British Journal of Managem ent, 1: 147-157.
Jackofsky, E. F. 1984. Turnover and job performance: An integrated process model.
Academy of Management Review, 9: 74-83.
Keller, R. T. 1997. Job involvement and organizational commitment as longitudinal
predictors o f jo b perform ance: A study of scientists and engineers. Journal of
Applied Psychology, 82: 539-545.
Lawler, E. E., Porter, L. W. 1967. The effect of performance on job satisfaction.
IndustrialRelations, 7: 20-28.
Lee, T. W., & Mow day, R. T. 1987. Voluntarily leaving an organization: An em pirical
investigation of Steers and Mowday's model of turnover. Academy of
Management Journal, 3 0: 721-743.
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:
31-48.
Litt, M. D. 1 988. Cognitive m ediators of stressful experience: Self-efficacy and perceived
control. Cognitive Therapy and Research, 12: 241-260.
Lucas, J. W. 1 999. Behavioral and em otional outcomes of leadership in task groups.
Social Forces, 7 8: 747-776.
Mattel, L. 2003. Finding and keeping high perform ers: Best practices from 25 best
companies. Employment Relations Today, 30: 27- 43.
Mathieu, J. E., & Zajac, D . M. 1990. A review and meta-analysis of the antecedents,
39
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 46/147
correlates, and consequences of organizational commitment. Psychological
Bulletin, 108: 171-194.
Meyer, J. P., & Allen, N . J. 1997.
Comm itment in the workplace.
Thousand Oaks, CA:
Sage Publications.
Meyer, J. P., & Smith, C. A. 2000. HRM practices and organizational commitment: Test
of a mediation model. Canadian Journal of Administrative Science, 17: 319-331.
Michaels, E. Handfield-Jones, H., & A xelrod, B. 20 01. The war for talent. Boston:
Harvard Business School Press.
Mobley, W. H. 1982. Some Unanswered Questions in Turnover and Withdrawal
Research. Academy of Management Review, 7: 111-116.
Ogilvie, J. R. 1986. The role of human resource m anagem ent practices in predicting
organizational com mitment. Group Organ ization Studies,
11 :
335-359.
O'R eilly, C. A., Pfeffer, J. 2000. Hidden value. Boston: Harvard Business School Press.
Pfeffer, J. 1994. Competitive advantage through people. Boston: Harvard Business
School Press.
Pfeffer, J. 1998. The human equation, Boston: Harvard Business School Press.
Pfeffer, J. 200 1. Fighting the war for talent is hazardous to your o rganizatio n's health.
Organizational Dynamics, 29: 248-259.
Pfeffer, J. 2006. W orking alone: Whatever happ ened to the idea of organizations as
communities? In E. E. Lawler & J. O'Toole (Eds.), America at
work:
C hoices and
challenges: 3-21. New York: Palgrave M acmillan.
Porter, L. W., Steers, R. M., Mo wday, R . T., & Boulian, P. W. 1974. O rganizational
40
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 47/147
commitment, job satisfaction, and turnover among psychiatric technicians.
Journal of Applied Psychology, 59: 603-609.
Randall, D. M. 1987. Commitment and the organization: The organization man revisited.
Academy of Management Review, 12 : 460-471.
Rao, H., & Drazin, R. 2002. Overcoming resource constraints on product innovation by
recruiting talent from rivals: A study of the mu tual fund industry, 1986-94.
Academy of Management Journal, 45 : 491-507.
Rhoades, L. Eisenberger, R., & Armeli, S. 2001. Affective commitment on the
organization: The contribution of perceived organizational support. Journal of
Applied Psychology, 86: 825-836.
Rynes, S. L., Gerhart, B., & Minette, K.A. 2004. The importance of pay in employee
motivation: Discrepancies between what people say and what they do. Human
Resource Management, 43 : 381-394.
Schaubroeck, J., Merritt, D. E. 199 7. Divergent effects of job control on coping with
work stressors: The key role of self-efficacy. Academy of Management Journal,
40: 738-754.
Shalley, C. E., & Perry-Sm ith, J.E. 20 01 . Effects of social-psychological factors on
creative performan ce: The role of informatioinal and controlling expected
evaluation and modeling experience. Organizational Behavior and Human
Decision Processes, 84: 1-22.
Shaw, J. D., Gupta, N., & D elery, J. E. 200 5. Alternative conceptualizations of the
relationship between voluntary turnover and organizational performance.
Academy of Management
Journal,
48 : 50-68.
41
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 48/147
Smits, S. J., McLean , E. R., & Tanner, J. R. 1993. Man aging H igh-achieving information
systems professionals, Journal of Management Information Systems, 9: 103-120.
Snell, S. A., Dean, J. W . 1992. Integrating m anufacturing and human resource
managem ent: A hum an capital perspective. Academy of Management Journal, 35:
467-504.
Somers, M. J., & Birnbaum, D. 1998. Work-related commitment and job performance:
it's also the nature of the performance that counts. Journal of Organizational
Behavior, 19: 621-634.
Staw, B. M. 1980. The consequence of turnover. Journal of Occupational Behavior, 1:
253-273.
Steers, R. M. 1977. Antecedents and outcomes of organizational commitment.
Administrative Science Quarterly, 22: 46-56.
Steers, R. M., & Mowday, R. T. 1981. Employee turnover and post-decision
accomm odation processes. In L. L. Cummings & B. M. Staw (Eds.), Research in
organizational behavior, vol. 3: 235-281. Greenwich, Conn.: JAI Press.
Steers, R. M., & Porter, L. W . 1983. Motivation and work behavior. New Y ork:
McGraw-Hill.
Sturman, M. C , Trevor, C. O., Boudreau, J. W., & Gerhart, B. 2003. Is it worth it to win
the talent war? Ev aluating the utility of performance-based pay. Personnel
Psychology. 56: 997-1035 .
Tett, R. P., & M eyer, J. P. 1983. Job satisfaction, organ izational co mm itment, turnover
intention, and turnov er: Path analyses based on meta-analytic findings.
Personnel
Psychology, 46: 259-292.
42
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 49/147
Trank, C. Q ., Rynes, S. L., & Bretz, R. D , Jr. 2002 . Attracting applicants in the war for
talent: Differences in work preferences am ong high achievers. Journal of
Business and Psychology, 16: 331-34 5.
Trevor, C. O., Gerhart, B., & Boudreau, J.W. 1997. Voluntary turnover and job
performan ce: Curvilinearity and the moderating influences of salary growth and
promotions. Journal of Applied
Psychology,
82:
44-61.
Truss, C. 2001. Complexities and Controversies in linking HRM with organizational
outcomes.
Journal of Management Studies,
38: 1121-1149.
Tsui, A. S., Pearce, J. L., Porter, L.W., & T ripoli, A.M. 1997. Alternative approaches to
the employee-organization relationship: Does investment in employee pay off?
Academy of Management
Journal,
40: 1089-1121.
W allace, J. E. 1997. Beck er's side-bet theory of com mitm ent revisited: Is it time for a
moratorium or a resurrection? H uman Relations, 50: 727-749.
Wh itener, E. M. 20 01 . Do high comm itment hum an resource practices affect employee
commitment? Across-level analysis using hierarchical linear modeling. Journal of
Management,
27: 515-535.
Wood, R., & Bandura, A. 1989. Social cognitive theory of organizational management.
Academy of Management Review, 14: 361-384.
Wright, P. M., McM ahan, G. C , & McW illiams, A. 1994. Hum an resources and
sustained competitive advantage: a resource-based perspective. International
Journal of Human Resource M anagement, 5: 301-326.
Wright, P. M., Gardner, T. M , & Moynihan, L. M. 2003. The impact of HR practices on
43
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 50/147
the performance of business units. Human Resource Management Journal. 13:
21-36.
Wright, P. M., McCormick, B., Sherman, W. S., & McMahan, G. C. 1999. The role of
human resource practices in petro-chemical refinery performance. International
Journal of Human Resource Management,
10: 551-571.
44
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 51/147
TABLES
TABLE 2.1: Demographic Characteristics of Samples
Total
Top Performers
Other levels of Performers
Number of respondents
Tenure (Year)
Mean
S.D.
Age
(1)30 to 34
(2) 35 to 39
(3) Over 40
Education
(1) High school
(2) Bachelor (2 year)
(3) Bachelor (4 year)
(4) Masters
(5) Doctorate
Profession
(0) R&D (Scientists)
(1) Managers
589
11.45
5.72
52
207
330
21
13
314
148
93
214
375
363
10.50
6.04
31
131
201
7
5
168
108
75
141
222
226
12.97
4.80
21
76
129
14
8
146
40
18
73
153
45
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 52/147
T
A
2
2
M
e
S
a
d
D
a
o
a
C
e
a
o
O
1
A
e
v
O
2
P
o
m
3
T
e
4
1
A
D
m
1
4
2
A
D
m
2
4
3
A
D
m
3
5
E
o
6
P
o
e
o
7
S
v
s
o
s
u
8
H
9
T
n
n
e
o
M
e
3
5
0
6
1
4
0
0
0
3
0
5
1
2
0
6
3
7
2
1
1
8
S
D
0
5
0
4
5
7
0
2
0
4
0
4
2
3
0
4
0
6
3
3
1
0
1
.
0
.
3
-
2
_
1
.
2
-
2
.
2
5
.
6
-
4
2
-
2
-
0
.
0
-
0
.
2
-
0
1
.
1
-
0
3
.
3
-
2
4
-
4
.
1
.
1
.
2
-
0
4
1
.
2
.
3
1
-
2
-
0
-
0
1
4
2
-
8
.
0
-
0
-
0
1
.
0
4
3
-
1
.
1
.
0
.
1
-
0
5
-
4
.
1
-
1
.
0
6
1 1
.
1
7
8
.
6
-
2
-
2
*
0
*
0
1
T
p
o
m
1
o
h
e
o
p
o
m
0
A
1
3
o
3
y
2
3
o
3
y
3
m
h
3
y
D
m
1
=
1
A
=
1
o
h
w
=
0
D
m
2
=
1
A
=
2
o
h
w
=
0
D
m
3
=
1
A
=
3
o
h
w
=
0
3
M
a
1
S
e
0
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 53/147
TABLE 2.3: Hierarchical Regressions Results in the relationship between HCHRPs and
Affective OC
All respondents (N:5S9)
Tenure
Age : 35-39
Age : 40 and over
Education
Profession
Supervisor Support
HCHRPs
Top Performers
HCHRPs * Top
performers
Model 1
0.12**
0.09
0.27***
-0 .05
0.06
0.46***
Model 2
0.05
0.11*
0.26***
-0 .04
0.05
017***
047***
Model 3
0.03
0.12*
0.28***
-0 .03
0.05
0.17***
0 49***
- 0.07*
Model 4
0.03
0.12*
0.28***
-0.03
0.04
0.17***
0.41***
- 0.06*
0.11*
R
2
AR
2
FforAR
2
Overall F
0 400
23.8
0.518
0.118
139.6***
36.0***
0.521
0.004
4.5**
34.5***
0.526
0.004
5.0**
33.2***
***p <
0.001;
**p < 0 .01; *p<0.05. 10 company dumm ies are included in the regression m odels.
Note. Standardized p coefficients are reported.
47
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 54/147
TABLE 2.4: Mediation by Affective OC in the relationship between HCHRPs and
Turnover Intention
All Respondents (N=580')
Top Performers (N=357)
Model 5
(HCHRPs
=>T1)
Model 6
(OC => Tl)
Model 7
(HCHRPs &
PC => TI)
Model 8
(HCHRPs
=>TI)
Model 9
(OC => TI)
Model 10
(HCHRPs &
OC => TI)
Tenure
0.02
0.04
0.04
•0.08
•0.07
•0.06
Age : 35-39
Ag e : 40 and
over
Education
Profession
Supervisor
Support
HCHRPs
Affective OC
-0 .12
-0 .1 4 +
- 0 .09
T
- 0 . 1 0
T
-0 .15**
-0 .16**
-0 .07
-0 .04
-0 .10*
-0 .08
- 0 .08
T
-0 .38***
-0 .07
-0 .04
-0 .11*
-0 .08
- 0.09
T
0.02
- 0 . 38* '
•0.08
•0.00
- 0 . 1 2
T
-0.08
-0.13
1
-0 .21**
•0.04
0.07
•0 .12
1
•0.08
- 0 . 1 3
•0 .30
t
•0.04
0.06
-0.12
-0.07
-0.11
-0.06
- 0.27*
0.15
0.22
0.22
0.17
0.20 0.20
Overall F
5
97***
9.53*
9.00***
3 9g*** 494* **
4
69***
***p < 0.001; **p < 0.01; *p<0.05;
T
O.1 0 .
Note. 1) 9 subjects did not report their turnover intention. Standardized p coefficients are reported. 10 company
dummies are included in the regression m odels.
48
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 55/147
FIGURES
FIGURE 2.1: Research Framework on HCHRPs, Affective OC, and Turnover Intention
between Top Performers and Others
Top Performers versus Other
Levels of Performers
High Commitment
Human Resource
Practices
H2
v
H I
Affective
Organizational
Commitment
H3
Turnover
Intention
49
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 56/147
FIGURE 2.2: The Relationship between HCHRPs and Affective OC of Top versus Other
Levels of Performers
High
Affective OC
Low
—
Top Performers
- Others
-1SD
+1SD
HCHRPs
50
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 57/147
CHAPTER 3
HIGH PERFORMANCE WORK SYSTEMS, TURNOVER, AND FIRM
PERFORMANC E: WHOSE TURNOVER MATTERS?
Research in strategic human resource management (SHRM) suggests that high
performance, commitment, involvement, and human resource systems enable
organizations to enhance employees' skills, organizational commitment (OC), and
motivation, which results in higher cooperation among organization members, lower
turnover, and ultimately , higher organizational performance (e.g., Arthur, 1994; Batt,
2002;
Collins & Smith, 2006; Com bs, Liu, Hall, & Ketchen, 2006; Evans & Davis, 2005;
Huselid, 1995; Pfeffer, 1994, 1998; Sun, Aryee, & Law, 200 7; Takeuchi, Lepak, Wang,
& T akeuchi, 200 7). This research is built upon macro and micro theories from diverse
fields of the social sciences, and has dedicated significant effort to investigating whether
there exists a direct relationship between high performance work systems (HPWS) and
firm performance. Conseq uently, a few studies found prom ising results in this
relationship (e.g., Bae & Lawler, 200 0; Delery & Doty, 1996; Ichiniowski, Shaw, &
Prennush i, 199 7; Youndt, Sn ell, Dean Jr, & Lepak, 1996). Batt (2002) and Huselid
(1995) further buttressed this relationship by showing a mediating role of total workforce
turnover, which partly addresses a 'black box' mechanism in the relationship between
hum an resource (HR ) practices and firm performance (Becker & Gerhardt, 1996). In this
stream of research, the impact of HPWS on turnover has been, implicitly or explicitly,
regarded as one of the key mechanisms for explaining why HPWS influences firm
performance.
51
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 58/147
Ho weve r, this stream of research contains several limitations (e.g., Delery &
Shaw, 20 01 ; W right, Dunford, & Sn ell, 2001), one of which is that few studies have
systematically explored the q uality of employee turnover that H PW S influences. Cascio
(1995) stated, The crucial issue in analyzing turnover . . . is not how many em ployees
leave but rather the performance and replaceability of those who leave versus those who
stay (p. 582). Hu selid (1995) in his seminal work recognized the importance of
addressing whose turnover H PW S influence, stating that the use of High Performance
Work Practices can... enhance retention of quality employees while encouraging
nonperformers to leave the firm (p. 635). Also, Shaw, Dineen, Fang, and Vellella (in
press) called for a study that ex plicates the relationships between HR p ractices, the
quality (performance levels) of employee turnov er, and firm performance. However, to
date, few studies have paid attention to investigating the main effect of HP W S on high
performer turn over and its differing impact on high performer turnover comp ared to
relatively lower performer turnover.
This q uestion n eeds to be addressed in order to buttress and elaborate the findings
on the HRM-firm performance relationship. First, previous SHRM studies have described
the impact of HPW S on emp loyee turnover or the mediation effect of employee turnover
in the relationship between HPWS and firm performance as evidence that HRM
positively influences firm performance. T he rational behind this argument is largely
based on the assum ption that low employee turnover is universally beneficial for firm
performance. However, the low level of total employee turnover may not necessarily
result in superior firm performance and, in fact, high turnover of low er performers may
enhance firm performance (Cascio, 1995; Dalton & Toder, 1979; Mobley, 1982; Schwab,
52
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 59/147
1991; Staw, 1980; Steel, Griffeth, & Horn, 2002; Sturm an, Trevor, Boudreau, & Gerhart,
2003).
Second, a few recent studies suggest that the inter-organizational competition for
highly talented em ployee s has intensified, as organizations have increasingly recognized
the importance of high performing emp loyees in organizational value creation activities
(Gardner, 2005; Cappelli, 2000; Rao & Drazin, 2002). For example, Gardner (2005)
observed that more than 20 percent of organizations experienced 'purposive talent
raiding' from their competitors, which implies that high performers may leave
organizations m ore frequently than other levels of performers. In fact, existing theoretical
and emp irical work in dicates that high performers are more likely to leave their
organizations than relatively lower performers (Jackofsky, 1984; Trevor, Gerhart, &
Boud reau, 1997), although there is on-going debate regarding the relationship between
individual performance and turnover decision (Salamin & Horn, 2005).
Therefore, the aforem entioned studies imply that without empirical evidence in
the relationship betw een H PW S and performance levels of leavers, prior empirical
studies may rem ain suscep tible to an alternative interpretation that the found negative
effect of HP W S on em ployee turnover could be the result of high turnover among high
performers and low turnover among low performers. Consequently, SHRM research may
continue to suffer from a lack of coherent theoretical and em pirical explanations based on
the negative impact of HP W S on employee turnover in the HRM-firm performance
relationship.
This paper aims to resolve this issue in the SH RM research by simultaneously
investigating: 1) if HPW S influences total workforce turnover, 2) if HPW S reduces high
53
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 60/147
performer turnover, 3 ) if HP W S reduces turnover of high performers more than that of
the rest of work force, 4) how total and high performer turnover m ediate the relationship
between HPWS and firm performance. This study contributes to enriching theoretical and
empirical explanations for the HRM-firm performance linkage by taking into account the
relationships between HPWS, employee turnover, both of high and relatively lower
performers, and firm performan ce. It also provides practical implications for firms that
have dedicated m uch effort to d eveloping effective retention strategy for high performers,
who significantly contribute to creating value for organizations but at the same time are
exposed to a high turnover threat.
PREVIOUS LITERATURE AND HYPOTHESES
HPWS has been conceptualized in various ways and researchers have labeled a
different com bination of human resource practices as high performance, innovative,
involvement, and commitment human resource practices. Despite varying labels of
HPWS,
the commonly shared perspective in HRM literature is that HPWS enables
organizations to develop skillful, m otivated, and comm itted em ployees, who can create
superior value for the organ izations. In general, HPWS constitutes extensive selection,
intensive training and development, pay for performance compensation, rigorous
performance appraisal, flexible job design, employee participation in organizational
activities, and open co mm unication such as information sharing (Appelbaum , Bailey,
Berg, & Kalleberg, 2000; Arthur, 1994; Delery & Shaw, 2001; Guest, 1997; Huselid,
1995; Pfeffer, 1994; Whitner, 20 01 ; W right, Gardner, & Moynihan, 2 003). In particular,
the impacts of HPW S on OC and turnover have played pivotal roles in empirical studies
54
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 61/147
in the SHRM literature (e.g., Arthur, 1994; Batt, 2002; G uthrie, 200 1; Huselid, 1995). I
briefly review a few salient streams of research in the previous SHRM research,
emphasizing the centrality of employee commitment and turnover.
The m ain inquiry of research on HRM and its relationship to firm performance
has evolved from emp irically investigating the relationships betw een HR practices and
firm performance to providing theoretical explanations and exploring intermediate
linkage ( black box ) on this relationship. Part of this mo vem ent stems from researchers'
motivation to ove rcom e criticism that HRM literature has weak theoretical explanations
for why HR prac tices influence firm performance (e.g., Becker & Gerhart, 1996). Figure
1 presents several salient studies in the HRM literature which have con tributed to
understanding the complex relationships between HR practices and firm performance.
Direct investigation between HPWS and Firm Performance
Most studies in the first category in Figure
1
explored either the direct
relationship between HPWS and firm performance or contingent relationship between
HPW S and firm strategy based on a resource-based view of the firm (e.g., Barney, 1991;
W right & McM ahan, 1992). A fundamental p remise of the studies was that human
resources or hum an resource p ractices could be a source of sustained com petitive
advantage because of their distinctive characteristics: valuable, rare, nonsubstitutable, and
inimitable (e.g., Becker & Gerhart, 1996; Wright, McMahan, & McWilliams, 1994).
Although this viewpoint was shared by many previous studies on SHRM, the theoretical
explanations for the relationship between HPWS and firm performance were given by
three different perspec tives: universal, contingent, and configurational (see Delery, 1998;
Delery & D oty, 1996, for m ore details). They found a positive, though not unam biguous,
55
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 62/147
relationship between HPW S and firm performance in different industries and countries
(e.g., Bae & Law ler, 2000 ; Delery & Do ty, 1996). These studies, however, received
severe criticism d ue to the lack of theoretical explanations for why H PW S influenced
firm performance and through what mechanisms (e.g., Becker & Gerhart, 1996; Wright et
al.,
2001).
The mediation of turnover in the relationship between HPWS and Firm
Performance
Studies in the second category investigated the intermediate linkage between
HPWS and firm performance, focusing on HPWS' impact on total workforce turnover
through which H R practices influence firm performance (Batt, 200 2; Huselid, 1995). The
common premise in these studies was that HPWS would increase employee skill as well
as commitment, which results in lower turnover and higher productivity, ultimately
leading to higher organizational performance. Based on that idea, Huselid (1995) found
that HPWS leads to a lower level of employee turnover which mediates the relationship
between HPWS and firm performance in
U.S.
firms across industries. Also, a recent
study by Batt (2002) observed that high-involvement HR practices including training,
employee participation, and various forms of incentive compensation have a positive
impact on reducing employee turnover, which in part mediates the relationship between
the HR practices and sales growth.
The interaction effect between HPWS and Turnover on Firm Performance
In the third category of studies, researchers explored the relationship between HR
practices and turnover, and interaction effects between H PW S and turnover on firm
performance (Arthur, 1994; Guthrie, 2001). The theoretical argument behind these
studies stems from the assumption that high commitment (involvement) HR practices
56
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 63/147
enhance em ployee skill and O C, which result in lower workforce turnover. In addition,
they argued that the impact o f the lower turnover on firm performance may depend on the
types of HR practices - high commitment (involvement) versus control-oriented HR
practices - because of different costs involved in the two types of H R practices. They
observed the expected results: (1) high commitment (involvement) HR practices are more
likely to be associated w ith a lower level of turnover than control-oriented HR practices;
(2) the relationship betw een turnove r and firm performance is moderated depending on
the types of HR practices (Arthur, 1994; Guthrie; 2001).
The Impact of HR practices on Employees Organizational Commitment
Unlike the thre e categories of studies mentioned a bove, studies in the fourth
category have d edicated significant effort to investigating whether HP W S influences
employee OC, which has been a key variable in the relationship between HPWS,
turnover, and firm performance in macro-level HRM studies. They argued that although
macro-level HRM studies recognize the importance of employees' OC, which results in
high discretionary effort and reduced turnover, the studies did not pay much attention to
how employees respond to HR practices. They claimed that the unitary assumption in the
macro-level studies should be confirmed empirically in order to buttress the relationship
between HR practices and firm performance (e.g., Guest, 2002, Wright & Boswell, 2002).
Consequently, they found that high performance, involvement, commitment, and HR
practices indeed increase employee OC (Agarwalar,
2003;
Appelbaum et al., 2000;
Gould-W illiams, 2004 ; Eaton, 2001; Meyer & Smith, 2000; Whitener, 20 01; Wright et al,
2003).
In explaining this relationship, most studies relied on social exchange and
perceived organizational support theories (e.g., Whitner, 2001).
57
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 64/147
Summary of the previous studies
Surveying previou s studies in the HRM literature, one can see that the literature
on the relationship between HPWS and firm performance has generated promising
theoretical perspectives and empirical evidence, where OC and turnover played central
roles. In addition, on e could recognize that the diverse streams of research at the different
levels of analysis were closely interrelated and one stream of research would often m ake
up for weak nesses in the other streams of research. Based o n theoretical perspectives and
empirical results of the p revious studies, I replicate the following hypotheses.
Hypothesis la: High performance work systems will be positively related to firm
performance.
Hypothesis lb: High performance work systems w ill be negatively related to total
employee turnover.
Hypothesis lc: Total employee turnover will mediate the relationship between high
performance work system and firm performance.
Despite the significant contribution of previous studies to understanding the
relationship between HRM and firm performance, one could also recognize that previous
studies in HRM appear to be mon otonou s without sufficiently reflecting diverse
perspectives in the H RM -firm performance relationship. First of all, the literature on the
HRM-firm performance linkage has regarded the impact of HPWS on turnover as strong
evidence for that linkage. However, several studies in HRM and turnover literature
suggest that the low level of turnover m ay not result in higher firm p erformance (e.g.,
Cascio, 1 995; Dalton, K rachhardt, & Porter, 1981 ; Dalton & Toder, 1979; Delery &
Shaw, 2001; Lepak & Snell, 1999; Staw, 1980). Staw (1980) proposed that turnover may
58
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 65/147
have both positive and negative impacts on firm performance. He argued that although
turnover may negatively affect firm performance due to selection and training costs,
operational disruption, and demoralization of organizational membership, turnover may
also have a positive impact on firm performance. In particular, he emphasized that costs
accrued by turnover may not be universal, but depend on wh o leaves and who joins, and
pointed out that successful replacement of low performers with high performers could
have a positive impact on firm performance. Second, although some studies, which show
the positive impact of HPWS and employee OC, help to predict the relationship between
HPWS and employee turnover, they also consider a high level of OC to be universally
beneficial. However, this viewpoint may not be necessarily applicable across contexts.
For example, Steers and Porter (1983) and Randall (1987) argued that high OC of low
performers m ay have a negative im pact on firm performance due to reducing workforce
flexibility. Therefore, the se contingent aspects of O C and turnover m ay lead scholars and
practitioners to question monotonous theoretical explanations in the HRM literature for
why HPWS influences firm performance.
HPWS and High Performer Turnover
In the previou s section, I pointed out that in order to enrich theoretical
explanations in the HR M literature, it is necessary to explore how H PWS influences
turnover of high perform ers, g oing beyond just investigating the relationship between
HPW S and total workforce turnover. Although there are few studies that directly explain
the relationship between HPWS and high performer turnover, theoretical and empirical
evidence in OC and turnover helps explain this relationship.
59
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 66/147
Research in OC suggests that OC is the strongest predictor of employee turnover
(e.g., Mathieu & Zajac, 1990), and points out that when em ployees perceive that
organizational rew ards and treatment are favorable and fair to them, they develop OC as
reciprocation toward o rganizations (e.g., Mey er & Allen, 1997; Steers, 1977). The
overarching theoretical rationale for the development of OC draws upon social exchange
theory and Gouldner's (1960) 'norm of reciprocity'' that people should help those who
have helped them (p. 171). Building upon social exchange theory, Eisenberger and his
colleagues (Eisenberger, Fasolo, & Davis-LaMastro, 1990; Eisenberger, Huntington,
Hutchison, & Sowa, 1986) argued that emp loyees regard organizations like a person and
form a global belief abou t how the o rganization recognizes their contributions and cares
for their well-bein g. They theo rized that if employees perceive their organizations as
supportive, they tend to reciprocate w ith increased OC , and suggested that organizational
practices such as pay, job enrichm ent, and participation, affect em ployee OC because
these practices signal the degree to wh ich organizations value and support their
employees. Based on these perspectives, a few studies in HRM found that HPWS
positively influences OC (Agarwalar, 2003; Appelbaum et al., 2000; Gould-Williams,
2004;
Meyer & Smith, 2000; Whitener, 2001) or turnover intention (Batt, 2003).
Social exchange theory also has implications for the relationship between HPWS
and high perfo rm ers' OC and turnover. According to social exchange theory, the pattern
and degree of reciprocation in exchange relationships may depend on different
individ uals' n eeds, preferences, and situations because these differences influence the
extent to which indiv iduals perceive the value given by the other party (Gouldner, 1960;
Homans, 1974). Meyer and Allen (1997) also suggested that the relationship between
60
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 67/147
work practices and OC can be moderated by individual differences (e.g., personality,
values, needs, expectations).
There is some evidence that high performers and other levels of performers may
differ in terms of work preference (Smits, McLean, & Tanner, 1993; Trank, Rynes, &
Bretz, 2002) and the levels of self efficacy (Ackerman, Kanfer, and Goff, 1995; Wood &
Bandura, 1989). First, Trank, Rynes,
and
Bretz (2002) observed that high achievers,
compared to low achievers, are more likely to value challenging work, pay for
performance compensation, and the opportunity for training relative to low achievers.
Second,
a
study
by
Wood
and
Bandura (1989) suggests that high performers, relative
to
other levels of performers, may have a higher level of self-efficacy, which is defined as
belief in one's capabilities to mobilize the motivation, cognitive resource, and course of
action needed to meet given situational demands (p. 408) because of previous
performance success. Ackerman, Kanfer,
and
Goff (1995 ) em pirically found that
previous performance success is the strongest predictor of the degree of individuals' self-
efficacy.
When the individual differences between h igh performers and other levels of
performers and contingent perspective of the development of OC are taken into account
together, one may predict that HPW S - employee participation, job discretion, pay for
performance compensation, rigorous performance appraisal, intensive training
and
development, extensive staffing,
and
open com munication
-
could increase high
performers' OC and, consequently, reduce their turnover. First, a job design practice
emphasizing employee job discretion and an employee participation practice, common
components of
HPWS,
may enhance high performers' OC and, thus, ultimately reduce
61
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 68/147
their turnover becau se those w ork practices meets their work preferences and can be
interpreted m ore favorably by high performers than other levels of performers due to
different le vels of self-efficacy. Som e studies lend support for this predictio n. Godard
(2001) and Truss (2001) argued that employees may not necessarily perceive job
autonomy and employee participation as rewarding practices. In this regard, Defrank &
Ivancevich (1998) pointed out that whether employees interpret their work environments
positively or negatively may depend on their levels of self-efficacy. Schaubroeck and
Merritt (1997) emp irically observed that wh ereas people with high self-efficacy prefer a
high level of jo b co ntrol in a highly job demand ing context, people w ith low self-efficacy
prefer a low level of job control in the same job context. These studies suggest that, when
organizations provide employees with job discretion and participation in decision making,
high performers with a high level of self-efficacy are more likely to perceive that
organizations are supportive than o ther levels of performers, which results in higher OC
and lower turnover of high performers than relatively lower performers.
Second, pay for performance comp ensation, rigorous performance appraisal, and
intensive training and development practices, which are common components of
HPWS,
also increase em ployee OC (e.g., Appelbaum et al., 2000; O gilvie, 1986; Whitener, 200 1;
W right et al., 200 3); and in ad dition are more likely to affect the OC and turnover of high
performers relative to those of other performers. R esearch in organizational justice argues
that distributive and procedural justice perceptions influence employee OC (e.g., Colquitt,
Conlon, W esson, Porter, & Ng , 200 1; Greenberg, 20 01). In particular, organizational
structure and hum an resource prac tices such as comp ensation and performance appraisal
have been observed to influence individuals' (injustice perceptions (Folger & Greenberg,
62
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 69/147
1985; Schiminke, Cropanzano, & Rupp, 2002). Considering the theoretical perspective of
organizational justice, one may think of a hypothetical scenario in which em ployees are
paid on the b asis of their seniority regardless of their performance, or they are paid
according to their performance measured by accurate performance appraisal. In the
former case, high perform ers are less likely to perceive organizations as fair because the
organizations do n ot recognize their contribution. Th is injustice perception m ay lead high
performers to p erceive that their organizations are not supportive and, consequently,
reciprocate perceived lower organizational supportiveness with reduced OC. On the
contrary, in the latter case, it would b e oppo site.
Aside from the implications of OC for the relationship between HPWS and high
performer turnover, som e evidence in turnover literature also suggests that HPW S that
recognize an individual's performance may reduce high performer turnover. Trevor,
Gerhart, and Boudreau (1997) observed that although turnover is higher for high
performers than average performers, this relationship is moderated by relative wage
increases. Also, most of the studies that support the negative relationship b etween
performance and turnover found a significant mo derating impact of pay for performance
on this relationship (e.g., Griffeth, H orn, & Gaertner, 2000; W illiams, 1999; Williams &
Livingstone, 1994). M aertz & Griffeth (2004) argued that good performance feedback is
a signal to the em ployee that he or she is valued by the organization, particularly under
performance-contingent pay (p. 677).
Hypothesis 2a: High performance work systems w ill be negatively related to high
performer turnover.
63
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 70/147
Hypothesis 2b: The relationship between high performance work systems and high
performer turnover will be stronger than the relationship between high performance
work systems and rest of workforce turnover.
Mediation of High Performer Turnover in the HPWS and Firm Performance
Relationship
The relationship betw een turnover and firm perform ance has been explored from
diverse perspectives. Consequently, there are a few competing theoretical models and
empirical results, which pred ict that the relationship between turnover and firm
performance is (1) linearly negative (e.g., Kacm ar, Andrew s, Van R oony, Steilberg, &
Cerrone, 200 6; Osterm an, 1987), (2) attenuated negative (e.g., Shaw, Gupta, & Delery,
2005),
(3) curvilinear (e.g., Abelson & Baysinger, 1984; Glebeek & Bax, 2004), or (4)
contingent on con texts of turnover (e.g., Cascio, 1995; Dalton & Toder, 1979; Mobley,
1982;
Staw, 1980). Despite varying predictions about the relationship between turnover
and firm p erforma nce, these m odels share a com mo n idea that high performer turnover is
generally detrim ental for firm performance not only because high performers contribute
to firm perform ance mo re than other level of performers do but also because high
performers' turnover imposes high organizational replacement costs (e.g., Hollenback &
Williamson, 1986; Dalton, et al.,
1981;
Price, 1977). Price (1977) argued the importance
of high performer turnover on firm performance as follows:
If the leavers are disproportionately found among individuals who are high in performance,
turnover is likely to have an adverse result on effectiveness. However, if the leavers come
relatively more often from among individuals who are low in performance, turnover is
likely to increase effectiveness. The performance level of the leavers and stayers appears to
be critical for effectiveness (p. 113).
In addition, Shaw , Duffy, Johnson, and Lochhart (2005) and Dess and Shaw
(2001) argued th at social capital theory has im plications for the relationship between
64
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 71/147
turnover and firm performa nce. T hey pointed out that turnover of individuals who have
rich social capital is mo re detrimental for organizational performance because their
turnover leads to the loss of organization s' intangible assets embedded in relationships
among organizational members. From this perspective, high performer turnover may
have a significant n egative effect on firm performance because high performers tend to
have central network positions and ties (Cross & Cummings, 2004). Furthermore, given
knowledge-based organizational competition, high performer turnover becomes
increasingly harmful for organizational performance due to the threat of valuable
knowledge transfer by high performers to competing organizations (Cappelli, 2000).
Besides the above the oretical perspectives, a few simu lation studies report that high
performer turnover is detrimental for firm performance (Boudreau & Berger, 1985;
Schmidt & Hunter, 1983).
Hypothesis 3: The relationship between high performance work systems and firm
performance will be mediated by high performer turnover with a control for res of
workforce turnover.
METHODS
The data for this study w as collected by a prestigious research institute in South
Korea. Sou th Ko rea provide s an attractive research co ntext for testing the hypotheses in
this study. First, since the Asian financial crisis, firms op erating in South Korea have
increasingly adopted HPWS and previous research found a positive relationship between
HPW S and firm performance (e.g., Bae & Lawler, 2000 ). Second, firms in South Korea
have paid increasing attention to attracting and retaining high performers (Sung et al,
65
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 72/147
2004), which has helped firms respond to high performer turnover as well as total
workforce turnover.
Sample and Procedures
In July 20 06, the research institute distributed survey questionnaires to a senior
human resource manager of 994 firms via e-m ail. The reasons that the institute used e-
mail survey w ere that first, human resource m anagers, who responded to a pilot study,
preferred e-mail survey s to a paper-based mailing. Second, the institute explained that in
order to increase the re sponse rate for the survey, delivering the questionnaire to the right
person was particularly impo rtant because the questionnaire items such as total and high
performer turno ver rates tended to be considered confidential. The
address of
a
senior human resou rce m anager of the 994 firms w as identified by using the contact
information of the members affiliated with Korean Senior Human Resource Managers'
Association and H uman Resource Management Comm unity of an economic research
institute. The cover letter of the e-mail survey addressed that one of the purposes of the
survey was to examine the relationship between human resource management and
organizational performance. Each organization's response to the survey was guaranteed
to be treated as confidential and used only for research purp oses. A pilot study was
conducted in February 2006 for nine organizations in order to ensure the clarity of the
survey questionnaire. A total of 190 organizations participated in the survey, which
resulted in a
19.1%
response rate.
The total samp le size included in for hypothesis testing, however, w as reduced to
153 organizations after ex cluding 37 organizations for the following two reasons. First,
when organizational performance information was not available for participating
66
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 73/147
organizations, participating organizations failed to respond to q uestions such as high
performer tu rnover rate, or participating organ izations were governm ent subsidiaries, the
organizations w ere excluded from the analysis. Second, four organizations reported that
they changed parts of hum an resource practices in 2006. Including these four
organizations in the analyses could result in biased estimates of variables of interest in
this study because of
a
measu reme nt timing issue: measuring organ izational performance
and voluntary turnover in 2005 but HPW S in 2006 (e.g., Guthrie, 20 01 ; Wright, Gardner,
Mo ynihan, & A llen, 200 5). Therefore, I excluded these four organ izations in our analyses
to eliminate this problem . The 153 participating organizations operated in diverse
industries: manufacturing (46% ), construction (10% ), finance such as banking, insurance,
and security (9% ), and other non-manufacturing industries such as telecommunication
and distribution (35%). Their mean employment size was 2,449: 31 to less than 100
employees (8.7% ), 100 to less than 500 em ployees (34.8% ), 501 to less than 1,000
employees (19.9%), more than 1,000 employees (36.6%).
Measures
High performance work systems were m easured by adapting existing established
scales from previous studies (e.g., Wright et al, 2003; Bae & Lawler, 2000; Delery &
Doty, 1996; Snell & Dean, 1992). All survey questions were constructed with a 5 point
Likert-scale, whe re 1 meant strongly disagree and 5 meant strongly agree. Seven HR
practices were me asured w ith multiple items. In order to measure organization-wide
usage of each HR practice, a senior HR manager of each participating organization was
asked to evaluate the degree to w hich each practice was used for managerial and non-
managerial work ers sepa rately. Then, I averaged the scores of each practice for
67
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 74/147
managerial and non-managerial workers and used the averaged scores to test item
reliability. The seven organizational practices included in HPWS were extensive
selection, intensive training and development, pay for performance compen sation,
rigorous performance appraisal, flexible job design, employee participation in
organizational activities, and open communication.
Extensive selection was measured with three items (a = 0.81). An exemplary item
was This organization selects people according to highly refined selection criteria and
proced ures. Intensive training and developm ent was measured with three items (a =
0.93). An exem plary item was This organization spends a lot of mo ney on employee
training and developm ent. Pay for performance comp ensation was assessed with four
items (a = 0.76), where an exem plary item was This organization bases pay raise
decisions on emp loyee perform ance. Rigorous performance appraisal was measured
with four items (a = 0.88). An exemp lary item was This organization has an effective
formal performance appraisal system to evaluate employees' performance and
comp etencies. Flexible job design (a = 0.68) was measured with two items. An
exem plary item was This organization gives employees a lot of job discretion.
Employee participation in diverse organizational activities was measured with two items
(a = 0.73). An exem plary item was This organization utilizes formal practices so that
emp loyees can participate in organ izational activities (suggestion system, w ork-council,
quality-circle, etc) . Open comm unication was measured with two items (a = 0.76). An
exemplary item was This organization shares various information with employees (ex.
business strategy and financial status). All the items used to measure HPW S are shown in
Appendix A.
68
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 75/147
After m easuring the extent to which the organizations use the seven different
dimen sions of H PW S, I checked if all these practices can be loaded distinctively as
factors using a confirmatory factor analysis (CFA ). The results of CFA indicated that the
seven factor m odel, in wh ich each factor represents each of the seven HR practices
composing HPWS, fits decently with the data (
2
/df= 2.09; CFI = 0.91, IFI = 0.91,
RM SEA = 0.08). Then, I conducted a second-order confirmatory factor analysis in order
to test that the seven factors loaded onto their second-order construct, HPW S. The results
of the second-order factor analysis showed a decent fit {y?ldf= 2.21; CFI = 0.90, IFI =
0.90, RMSEA = 0.09). In addition, the Cronbach's alpha of the seven HR practices was
0.88. Based on these statistics, I adopted an additive approach in measuring H PWS by
aggregating the scores of each HR practice to form a HPWS index (e.g., Bae & Lawler,
2000 ; Becker & H uselid, 1998; Macky & Box all, 2007; Sun et al., 2007 ; Takeuchi et al.,
2007). The HPWS index in this study indicates the extent to which organizations used
HP W S. I interpreted this ind ex in a way such that the higher the aggregated scores of
HPW S in organizations, the more the organizations used HPW S.
Dependent variables. T otal workforce turnover rate was the average voluntary
turnover rate of permanent employees in 2005.
High performer turnover
and the
rest of
workforce turnover rates w ere the annual voluntary turnover rates am ong the highest
performing 20 % of employ ees and the other 80% employ ees, respectively. In order to
measure the th ree turno ver rates - overall turnover, high performer turnover, and the rest
of workforce turn over - two separate questions were asked to a senior human resource
manager of each firm. Total turnover rate was measured by asking a senior human
resource manager W hat is an annual workforce turnover rate of perman ent employees in
69
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 76/147
200 5? An additional explanation was specified in order to assess only voluntary
turnover as follows: Turno ver occurring due to such reasons as age limitation, disease,
and etc. is not included in co mpu ting an annual turnover rate . Then, in order to measure
a high performer turnover (top 20 percent performers) rate and a rest of workforce
turnover (the other 80 percen t performers) rate, a second question was asked to him/her
In the annual w orkforce turnover rate, what percentage is from the turnover of top 20
percent perform ers? To help respondents better understand this question, the following
formula was provided: (number of leavers, who were high performers) / (total number of
leavers) = ( ) %. Based on the responses to these two questions, I computed both high
performer turn over and rest of workforce turnover rates. For exam ple, if total w orkforce
turnover is 40 percent and the percentage of high (top 20 percent) performer turnover
within the total turnover is 10 percent, a high performer turnover rate is 20 percent by
compu ting [(0.4 X 0.1) / 0.2] and a rest of workforce turnover is 45 percent by computing
{[0.4 X (1 - 0.1)] / 0.8}. I used return on assets (ROA ) as a firm performance measure,
which w as a pop ular m easure in past studies (e.g., Huselid, 1995). ROA data was taken
from a database built by the Korea Information Service (KIS) which collaborates with
Moody's to provide information on organizations operating in South Korea for an
international audience (Chang, 2003).
Control Variables.
Va riables used in the mod els as controls were organization
age, organization size, union presence, capital intensity of each organization, and industry.
Organization a ge, which was m easured as the number of years in operation, was included
in order to control for any advantages related to length of business operation (e.g.,
Guthrie, 20 01 ; Hu selid, 1995). Firm size, which w as measured as the logarithm of the
70
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 77/147
number of employees, was controlled because larger organizations may have advantage
such as economy of scale relative to small organizations (e.g., Pfeffer & Salancik, 1978).
Union presence w as controlled because it may influence turnover and firm performance
(e.g., Freeman & Medoff, 1984). Capital intensity (logarithem of fixed assets / the
numbers of employees) of each organization (Datta, Guthrie, & Wright, 2005), four
industries (manufacturing, construction, finance, and other non-manufacturing), and
country of origin (K orean firm v ersus foreign firms) were also controlled in the
regression models (Bae & Lawler, 2000).
RESULTS
Table 1 shows me ans, standard deviations, and variable intercorrelations. The
results of simple intercorrelations present that HPW S and high performer turnover rate
(r=-0.19.
p < 0.05) are negatively related, and total turnover and high performer turnover
(r=0.60, p<0.01) are po sitively correlated, which suggests that in an organizations where
total turnover is high, hig h performers tend to leave their organizations m ore frequently.
In addition, high performer turnover w as also significantly p ositively correlated w ith rest
of workforce turnover (r=0.30, p < 0.01).
Tables 2 , 3, and 4 present the results of a series of multivariate hierarchical linear
regression analyses. First, I tested Hypotheses la , lb , and l c that replicate the findings of
the previous studies. Hyp othesis la predicted that HPW S will be positively related to
firm perform ance. The result of the regression, Model 2 in Table 2, shows that HPW S
explains a significant amo unt of additional variance of ROA (A R
z
0.024; F for A R\ p <
0.05) and its coefficient is significant for ROA (p < 0.05). Therefore, Hypothesis la is
71
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 78/147
supported. Hypothesis lb predicted a negative relationship between HPWS and total
turnover. The regression results of Model 1 in Table 2 show that HPW S explains
additional variance of total turnover (A R 0.024; F for A R , p < 0.05) and the
coefficient of HPWS is significant (p < 0.05). Therefore, the results support Hypothesis
lb . In order to test Hy pothesis lc tha t predicted the m ediation of total turnover in the
relationship between HPWS and firm performance, I conducted series of
OLS
regressions,
following sugg estions by Baron and Kenny (1986 ). Mod el 1 (HPW S
—
total turnover)
and Model 2 (HPWS —> firm perform ance) of Table 2 meet the first two conditions to
support Hypothesis lc. Lastly, as shown in Model 4 of Table 2,1 regressed firm
performance, ROA, on both HPWS and total turnover. The results of Model 4 show that
whereas the significance of the coefficient of HPW S decreases from a = 0.05 to a = 0.01,
the coefficient of total turn ove r rema in significant at the 0.01 leve l. A Sobel test was also
significant (p < 0.05). Therefore, the results of the series of regressions show partial
mediation effect of total turnover in the relationship between HPW S and firm
performance, which supports Hypothesis lc.
The nex t set of m odels in Table 3 and 4 relate to testing the relationships between
HPWS and high performer turnover and between HPWS and high performer turnover
versus the rest of workforce turnover (Hypothesis 2a, 2b), and the mediation effects of
high performer turnover in the relationship between HPWS and firm performance
(Hypo thesis 3). The regression results of M odel 5 in Table 3 indicate that HPW S explains
a significant am ount of add itional variance of high performer turnover (A
R
2
0.022;
F
for
A R
2
, p < 0.05) and that its coefficient is significant for high performer turnover (p <
0.05), which supports Hypothesis 2a.
72
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 79/147
In order to test Hypothesis 2b which predicted a stronger negative effect of
HPWS on high performer turnover than on the rest of workforce turnover, I used a
seemingly unrelated reg ression m ethod (Zellner, 1962). The regression results of Model 6
in Table 3 show that HP W S (p < 0.05) is significantly n egatively associated with high
performer turno ver. A lso, M odel 7 indicates that HPW S (p < 0.1) is negatively related to
rest of workforce turnover. Then, I tested the difference of coefficients of HPWS in
Mo del 6 and 7. The result indicated that although the direction of the difference was
negative, as I expected, the difference was not statistically significant. Therefore,
Hypothesis 2b is not supported.
Hypothesis 3 was tested by conducting the series of OLS regressions in Table 4.
In order to test the mediation effect, I followed the procedures suggested by B aron and
Kenny (1 986). First, Model 8 in Table 4 shows the significant relationship between
HPW S and RO A (p < 0.05). Second, Mod el 5 in Table 3 indicates the significant
association between HPWS and high performer turnover (p < 0.05). Third, Model 9 in
Table 4 shows the significant relationship between high performer turnover and ROA (p
< 0.01). The results of the three reg ressions m eet the three initial cond itions to test the
med iation effect of high performer turnover in the relationship between H PWS and RO A.
In the last step, I performed Mo del 10 in Table 4 where H PW S and high performer
turnover are loaded tog ether. Th e results show that, while the significance of the
coefficient of HPW S w as substantially reduced (p < 0.05 —> p < 0.10), that of high
performer turnov er rem ained at the same level (p < 0.01). Therefore, these results suggest
that high performer turnover partially mediates the relationship between HPWS and firm
performance with a control for rest of workforce turnover, which suppo rts Hypothesis 3.
73
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 80/147
DISCUSSION AND CONCLUSION
This study is an attempt to respond to the call for research that explicates the
quality of turnover through which HPWS influences firm performance (e.g., Huselid,
1995;
Shaw et al., in press), with consideration of the argum ent that the relationship
between turnover and firm performance may be contingent on the performance levels of
leavers. Investigating this issue in the HPWS-firm performance literature is important and
necessary because prior work on the HPWS-firm performance linkage has focused on
investigating only the relationships am ong HP W S, total turnover, and firm performance
without consideration of the performance level of leavers. This study fills this research
gap. The results of this paper indicate that, first, HP W S is negatively related not only to
total workforce turno ver, as previous studies observed, but also to high performer
turnover. Second , they show that the relationship between H PW S and firm performance
is mediated by high performer turnover with a control for rest-of-workforce turnover.
How ever, I could not find statistical support for the hyp othesis, which predicted a
stronger association of HPW S with high performer turnover than rest-of-workforce
turnover. Th is result seems to imply that there could be a meaningfully different impact
of HPW S on high (top 20 percen t) performer turnover relative to low (bottom 20 percent)
performer turn over rather than o n high performer turnover relative to rest-of-workforce
(the other 80 percen t) turnover. F uture research, which d istinguishes the turnover rates of
high performers, average perform ers, and low performers, could help better understand
the different impacts of HPW S. How ever, I would like to note the practical significance
of the impact of HPWS on turnovers between high performer turnover and rest-of-
workforce turnov er. Add itional analyses indicated that if
all
other variables are held at
74
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 81/147
the mean values, one standard deviation increase of HPWS above the mean reduces 48
percent of high performer turnover and 13 percent of rest of workforce turnover.
This study contributes to accumulating knowledge in the HPWS-firm
performance literature from several perspectives. First, from the con tingent perspective of
turnover on firm performance, the empirical evidence of previous SHRM studies on the
basis of the impact of H PW S on turnover and firm performance (Batt, 2002 ; Huselid,
1995) seemed to be questionable because they ignored the impact of HPWS on the
quality of turnover and its subsequent impact on firm perform ance. In fact, the evidence
from previous studies tends to be susceptible to an alternative interpretation such that the
low level of total turnover in previous studies may result from high turnov er among high
performers and low turno ver am ong low er performers. The results of this study, however,
indicate that HPWS reduces high performer turnover and, also, HPWS reduces high
performer turnover no less than relatively lower levels of performers. Th us, this study
suggests that the alternative interpretation is not valid, w hich buttresses the argument of
the past studies that HPW S influences firm performance through its impact on employee
turnover.
Second, this paper theoretically explains why HPWS reduces high performer
turnover. I reasoned that building up on the OC and turnover literatures, HPW S tends to
build fair and favorable exchange relationships between organizations and high
performers because HPWS could meet their work preferences and provide appropriate
work en vironm ents for high perform ers, who tend to have a high level of self-efficacy.
Although th is explanation is plausible, further research is required to investigate ho w
75
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 82/147
HPW S influences the O C of high performers and its impact on the OC of high performers
relative to other levels of performers at the organizational level.
Third, this paper helps understand an intermediate mechanism in the relationship
between HPWS and firm performance. Research has shown that high performers
significantly co ntribute to firm performance (e.g., Axelrod, Handfield-Jones, & W elsh,
2001;
Cap pelli, 2000 ; Sturman, Trevor, & Bo udreau, 2003), but the inter-organizational
comp etition for highly talented employees leads organizations to face the difficulty of
retaining their high perform ers (Cappelli, 200 0; Gardner, 2005). However, few studies
successfully have explained how to retain their high performers. T he results of this study
show that HPW S could reduce high performer turnover and shed light on the fact that one
of the 'black box' mechanisms through which HPWS influences firm performance is
through its impact o n the retention of high performers, who play critical roles in
enhancing firm performance.
Fourth, in terms of the im pacts of HR practices on the quality of em ployee
turnover, Shaw and his colleagues (in press) argued that a firm' HR practices, which can
be bundled to two HRM practices - inducement and investment practices and
expectation-enhan cing p ractices - reflect different exchange relationships between
employees and orga nizations and tend to influence the patterns of turnover between high-
performers and low -performe rs. They predicted the main effects of inducemen t and
investment prac tices (pay level, benefits level, training, job security, and p rocedural
justice) and expectation-enhancing practices (pay-for-performance, performance
appraisal, and m onitoring) on high performer turnover and the interaction effects of the
two HR practices on high performer turnover. However, they could not find support for
76
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 83/147
any of the three hy potheses in a study 2, where high performer turnov er rate meant the
turnover rate of people whose job performance was in the highest 20 % for full-time
employees), which is comparable with high performer turnover in this study.
Interestingly, unlike their study, this study found that HPWS had a significant association
with high performer turnover. T he finding of this study appears to imply that the human
resource system of a firm as a whole, like HPW S, may have a significant impact on high
performer turn over rather than tw o different HR systems within a firm. I hope that future
research investigates the inconsistent findings between this study and Shaw et al.' study
in order to better understand the relationship betw een H R practices and high performer
turnover.
Limitations
Althou gh this paper contributes to enriching theoretical perspectives w ith
prom ising results in the HR M and firm performan ce literature, the audience should be
cautious in interpreting th e findings of this paper due to the limitations involved in this
study. First, this study measu res a firm's HPW S from one senior HR manager at each
firm, so this study may be susceptible to measurement error (Gerhart, Wright, McMahan,
& Snell, 2000). Nevertheless, measuring HR practices from a senior HR manager could
be an effective way of assessing them because he or she tends to be in a better position to
be able to assess the HR practices (Huselid & Becker, 200 0). Second, this study uses the
cross-sectional data, which lacks the ability to demonstrate a causal relationship between
independent and depen dent va riables. Third, this study used a sample of firms operating
in South K orea and , thus, the results of this study may have a generalizability issue in
different coun tries. For exam ple, the results of this study may not hold wh en tested in
77
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 84/147
countries such as United States, due to the different national culture. How ever, the
positive effects of HPWS on individual commitment and firm performance were
observed in U .S. context (e.g., Appelbaum et al., 2000) so that our findings tested in an
East Asian context m ay no t be different in a U.S context. Furthermore, the universal and
conditional aspects of social exchange mechanisms engendered by HPWS between
employers and high perform ers and relatively lower performers m ay be stronger in
countries such as U.S. which value individualism than countries which value collectivism.
Thus,
I expect that wh en this study is replicated in U.S. context, the results would be
stronger. Fourth, this study used a sample collected across industries, meaning that
industry specific effects may influence this relationship. However, the fact that our
sample did represent mu ltiple industries does suggest the generalizability of our findings.
Practical Implications
In a current knowledge driven economy, firms have increasingly recognized the
importance of attracting and retaining high performers. This leads firms to be involved in
hyper-com petition for high p erformers in the labor market, which often leads firms to be
faced with difficulties in the attempt to retain their high performers. In fact, it has been
observed that firms often fail to retain high performers (Gardner, 2005 ). This study
provides firms w ith insights regarding a way of managing high performer retention and
turnover. This study suggests that firms can reduce high performer turnover by utilizing
HP W S, because it mee ts high performers' needs and work preferences and signals to high
performers that their o rganizations recog nize their performance and care for their well-
being. In addition, I could ob serve that one of practitioners' key con cerns about retaining
high performers was that, although firms could increase the retention of high performers
78
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 85/147
by providing the m with special treatment, this might cause a significant organizational
problem be cause this may create undesirable conflicts and com petition between high
performers and relatively low er performers and high turnover from the rest of workforce.
This study suggest that applying H PW S for all employees could be an effective
workforce reten tion strategy, which enables firms to retain high performers but also not
to lose many of the other levels of performers.
In conclusion , this paper took into account the implications of high performer
turnover in the HRM-firm performance literature. Barney (2001) argued that sustained
competitive advantage can be achieved not only through developing competitive
advantage but also through nurturing and maintaining the current source of competitive
advantage. Th is paper shed s light on the fact that HPW S may create and sustain
competitive advantage through people because HPWS enables organizations to retain not
only entire workforce that organizations invested to dev elop, but also high performers,
who play critical roles in organizational value creation a ctivities.
REFERENCES
Abelson, M.A., & Baysinger, B.D. 1984. Optimal and dysfunctional turnover: Toward an
organizational level model. Academy of Management Review, 9: 331-341.
Ackerman, P. L., Kanfer, R., & Goff, M. 1995. Cognitive and noncognitive determinants
and consequences of complex skill acquisition. Journal of Experimental
Psychology: Applied.
1: 270-304.
Agarwala, T. 2003. Innovative human resource practices and organizational commitment:
79
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 86/147
An empirical investigation.
International Journal of Human Resource
Management, 14: 175-197.
Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. 2000. Manufacturing Advantage:
Why high-performance work systems pay off. Ithaca, NY : Cornell U niversity
Press.
Arthur, J.B. 1994. Effects of hum an resource systems on manufacturing performance and
turnover. Academy of Management
Journal,
37 : 670-687.
Axelrod, E. L., Handfield-Jones, H., & W elsh, T. A. 20 01. War for talent, part two.
McKinsey Quarterly, 2: 9-12.
Barney, J. B. 19 91. Firm resou rces and sustained competitive advantage. Journal of
Management, 17: 99-120.
Barney, J. B. 200 1. Is the resource based view a useful perspective for strategic
management research? Yes. Academy of Management Review, 26 : 41-56.
Baron, R. M., & K enny, D . A. 1986. The moderator-m ediator variable distinction in
social psychological research: C onceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51: 1173-1182.
Bae, J., & L awler, J. J. 2000. Organizational and HRM strategies in Korea: Impact on
firm performance in an emerging economy. Academy of Management
Journal,
43:
502-517.
Batt, R. 2002. Managing customer services: Human resource practices, quit rates, and
sales growth. Academy of Management
Journal.
45 : 587-597.
Becker, B. E., & Gerhart, B . 1996. Hum an Resources and organizational performance:
progress and prospects. Academy of Management Journal, 39:
779-801.
80
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 87/147
Boudreau, J. W., & Berger, C . J. 1985. Decision-theoretic utility analysis applied to
employee separations and acquisitions. Journal of Applied Psychology. 70: 581-
612.
Cappelli, P. 2000. A market-driven approach to retaining talent. Harvard Business
Review, 78: 103-113.
Cascio, W.F. 1995. Man aging Hum an R esources: Productivity, quality of work life,
profits. New York: McGraw-Hill.
Chang, S. 2003. Financial crisis and transformation of Korean business groups.
Cam bridge, UK: Cambridge University Press.
Collins, C. J., & Smith, K. G. 2006. Knowledge exchange and combination: The role of
human resource practices in the performance of high-technology firms. Academy
of Management
Journal,
4 9: 544-560.
Colquitt, J. A., Conlon, D . E., Wesson, M. J., Porter, C. O. L. H., & Ng , K. Y. 2 001 .
Justice at the Millennium: A meta-analytic review of
25
years of organizational
justice research. Journal of Applied P sychology, 86: 425-445.
Com bs, J., Liu, Y., Hall, A., & Ketchen, D . 2006. How m uch do high-performance work
practices m atter? A m eta-analysis of their effects on organizational performance.
Personnel Psychology, 59: 501-528.
Cross, R., & Cummings, J. N. 2004. Tie and network correlates of individual
performance in knowledge-intensive work. Academy of Management
Journal,
47e: 928-937.
Dalton, D. R., Krackhardt, D. M., & Porter, L. W. 1981. Functional turnover: an
empirical assessment. Journal of Applied Psychology, 66: 716-721.
81
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 88/147
Dalton, D. R., & Todor, W. D. 1979. Turnover turned over: An expanded and positive
perspective. Academy of Management Review, 4: 225-235.
Datta, D. K., Guthrie, J. P., & Wright, P. M. 2005. Human resource management and
labor productivity: Does industry matter? Academy of Management Journal, 48:
135-145.
DeFrank, R . S., & Ivancevich, J. M . 1998. Stress on the job : An executive update.
Academy of Management Executive, 12: 55-66.
Delery, J.E., & Shaw, J.D. 2001. The strategic management of people in work
organization s: review, sy nthesis, and extension. In: G. R. Ferris (Eds), Research
in personnel and human resources Management (Vol. 7, pp . 165-197). Greenwich,
CT:
JAI Press.
Delery, J.E., Doty, D.H. 1996. Mo des of theorizing in strategic human resource
man agem ent: tests of universalistic, contingency, and configurational
performance predictions. Academy of Management
Journal,
39: 802-835.
Dess, G. G. & Shaw, J. D. 2001. Voluntary turnover, social capital, and organizational
performance, 26: 446-456.
Eaton, S.C. 2003. If you can use them : Flexibility policies, organizational comm itment,
and perceived performance. Industrial Relations, 42: 145-167.
Eisenberger, R., Fasolo, P., & Davis-LaM astro, V. 1990. Perceived organizational
support and employee diligence, commitment, and Innovation. Journal of Applied
Psychology, 75:51-59.
Eisenberger, R, Hun tington, R., Hutchison, S., & Sow a, D. 1986. Perceived
organizational support, Journal of Applied Psychology, 71 : 500-507.
82
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 89/147
Evans, W, R., & Dav is, W. D. 2005. High-performance work system s and organizational
performan ce: T he med iating role of internal social structure. Journal of
Management, 31: 758-775.
Folger, R., & Green berg, J. 1985. Procedural justice: An interpretive analysis of
personnel systems. In: K. M. Rowland & G. R. Ferris (Eds),
Research in
personnel and human resources management (V ol. 3, pp. 141-183). Greenwich,
CT: JAI Press.
Freeman, R. B., & Medoff, J. L. 1984. W hat do unions do? New York: Basic Books.
Gardner, T. M. 2005. Interfirm competition for human resources: Evidence from the
software industry. Academy of Management Journal, 48: 237-256.
Glebeek, A. C , & Ba x, E. H. 2004. Is high employee turnover really harmful? An
empirical test using company records. Academy of Management Journal, 4 7: 277-
286.
Godard, J. 200 1. High perform ance and the transformation of work? The implications of
alternative work practices for the experience and outcomes of work. Industrial
and Labor Relations Review. 54: 776-805.
Gouldner, A. W . 1960. The norm of reciprocity: A preliminary statement. Am erican
Sociological Review, 25: 161-178.
Gould-Williams, J. 2004. The effects of 'high commitment' HRM practices on employee
attitude: The views of public sector workers. Public Adm inistration. 82:
63-81.
Greenberg, J. 2001 .The seven loose can(n)ons of organizational justice. In: J. Greenberg
& R. Cropanzano (Eds.), Advances in organizational justice (pp. 245-271). Palo
Alto, CA: Stanford University Press.
83
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 90/147
Griffeth, R.W ., Horn, P.W ., Gaertner, S. 2000. A meta-analysis of antecedents and
correlates of employee turnover: Update, moderator tests, and research
implications for the next millennium. Journal of Management, 26: 463-488.
Guest, D. E. 2002. Human resource management, corporate performance and employee
wellbeing: Building the worker into HRM . T he Journal of Industrial Relations,
44:
335-358.
Guest, D. E. 1997. Human resource management and performance: a review and research
agenda. International Journal of Human Resource Management, 8: 263-276.
Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity:
Evidence from New Zealand. Academy of Management Journal, 44: 180-190.
Hollenbeck, J. R., & W illiams, C. R. 1986. Turnover functionality versus turnover
frequency: A note on work attitudes and organizational effectiveness. Journal of
Applied Psychology, 7 1 : 606-611.
Hom ans, G. C. 1974. Social Behavior: Its elementary forms (revised edition). New Y ork:
Harcourt Brace Jovanovich.
Huselid, M.A. 1995. The impact of human resource management practices on turnover,
productivity, and corporate financial performance. Academy of Management
Journal,
3 8: 635-672.
Ichniow ski, C , Shaw, K., & Prennushi, G. 1997. The effects of hum an resource
man agem ent practices on productivity: a study of steel finishing lines. Am erican
Economic Review, 87: 291-313 .
Jackofsky, E. F. 1984. Turnov er and job perform ance: A n integrated process model.
Academy of Management Review, 9: 74-83.
84
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 91/147
Kacmar, K. M , Andrew s, M. C , Van Roony, D. L., Steilberg, R. C , Cerrone, S. 2006.
Sure employee can be replaced . . . but at what cost? Turnover as a predictor of
unit-level performance. Academy of Management Journal, 49: 133-144.
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:
31-48.
McDuffie, J. P. 199 5. Hum an resource bundles and manufacturing performance -
organizational logic and flexible production system s in the w orld auto industry.
Academy of Management Journal, 48:197-221.
Ma ertz, C. P., Griffeth, R. W . 2004. Eight motivational forces and voluntary turnover: A
theoretical synthesis with implications for research. Journal of Management, 30:
667-683.
Mathieu, J. E., & Zajac, D. M . 1990. A review and meta-analysis of the antecedents,
correlates, and consequences of organizational commitment. Psychological
Bulletin, 108: 171-194.
Meyer, J. P., & Allen, N . J. 1997. Com mitment in the workplace. Thousand Oaks, CA:
Sage Publications.
Meyer, J. P., & Sm ith, C. A. 2000. HRM practices and organizational com mitment: Test
of a mediation m odel. Canad ian Journal of Administrative Science. 17: 319-331.
Michaels, E. Handfield-Jones, H., & A xelrod, B. 2001. The war for talent. Boston:
Harvard Business School Press.
Mobley, W. H. 1982. Some unanswered questions in turnover and withdrawal behavior.
Academy of Management Review, 7: 111-116.
85
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 92/147
Ogilvie, J.R. 1986. The role of hum an resource managem ent practices in predicting
organizational commitment. Group Organ ization Studies, 11: 335-359.
Osterman, P . 19 87. Turnov er, employm ent security, and the performance of the firm. In
M. Kleiner (Ed.), Human resources and
the
performance of the firm (pp. 275-317).
Madison, WI: Industrial Relations Research Association.
Pfeffer, J. 1994. Competitive advantage through people. Boston: Harvard Business
School Press.
Pfeffer, J. 1998. The human equation, Boston: Harvard Business School Press.
Pfeffer, J. 20 01 . Fighting the w ar for talent is hazardous to your o rganiza tion's health.
Organizational Dynamics, 29:248-259.
Pfeffer, J., & Salancik, G. R. 1978. The external control o f organizations: A resource
dependence perspective. New York: Harper & Row.
Price, J. L. 1977. The study of turnover. Ames: Iowa State University Press.
Rand all, D.M . 1987. Comm itment and the organization: The organization man revisited.
Academy of Management Review, 12: 460-471.
Rao, H., & Drazin, R. 2002. Overcoming resource constraints on product innovation by
recruiting talent from rivals: A study of the mutual fund industry, 1986-94.
Academy of Management Journal, 45: 491-507.
Rhoades, L. Eisenberger, R., & Armeli, S. 2001. Affective commitment on the
organization: The contribution of perceived organizational support. Journal of
Applied Psychology, 86: 825-836.
Schaubroeck, J., Merritt, D. E. 1997. Divergent effects of job control on coping with
86
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 93/147
work stressors: The ke y role of self-efficacy. Academy of Management Journal,
40: 738-754.
Schm inke, M., Cropa nzano , R. & Rupp, D.E. 2002 . Organization structure and fairness
percep tions: The mod erating effects of organizational level. Organizational
Behavior and H uman Decision Processes, 89: 881-905.
Schmidt, F. L., & H unter, J. E. 1983. Individual differences in productivity: An em pirical
test of estimates derived from studies of selection procedure utility. Journal of
Applied Psychology. 68: 407-414.
Schwab, D. P. 1991. Contextual variables in employee performance-turnover
relationships. Academy of Management Journal, 34: 966-975.
Shaw, J. D., Dineen, B . R., Fang, R., & V ellella, R. F. in press. Em ployee-organization
exchang e relationship s, HRM practices, and quit rates of good and poor performer.
Academy of Management Journal.
Shaw, J. D., Gupta, N ., & Delery, J. E. 20 05. Alternative co nceptualizations of the
relationship between voluntary turnover and organizational performance.
Academy of Management Journal, 48: 50-68.
Smits, S. J., McLean , E. R., & Tanner, J. R. 1993. Managing H igh-achieving information
systems professionals,
Journal of Management Information Systems,
9: 103-120.
Snell, S. A., Dean, J. W. 1992. Integrating manufacturing and human resource
managem ent: A hu man capital perspective. Academy of Management Journal, 35:
467-504.
Staw, B. M. 1980. The consequence of turnover. Journal of Occupational Behavior, 1:
253-273.
87
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 94/147
Steel, R. P., Griffeth, R. W ., & Horn, P. W . 2002 . Practical retention policy for the
practical m anager. Academ y ofMana gement Executive, 16: 148-164.
Steers, R. M. 1977. Antecedents and outcomes of organizational commitment.
Administrative Science Quarterly, 2 2: 46-56.
Steers, R. M., & Porter, L. W . 1983. Motivation and work behavior. New Y ork:
McGraw-Hill.
Sturman, M. C , Trevor, C . O., Boudreau, J. W., & Gerhart, B. 2003. Is it worth it to win
the talent war? Evaluating the utility of performance-based pay.
Personnel
Psychology. 56: 997-1035.
Sun, L., Ary ee, S., & Law , K. S. 2007 . High-performance hum an resource practices,
citizenship behavior, and organizational performance: A relational perspective.
Academy of Management Journal, 50: 558-577.
Sung, S., Chae, S., & Ba e, S. 2004 . Expanding workforce diversity and org anizations'
coping strategy. CEO Information, Samsung Economic Research Institute, Seoul.
(In Korean)
Takeuchi, R., Lepak , D. P., Wang, H., & Takeuchi, K. 2007. An emp irical examination
of the mechanisms mediating between high-performance work systems and the
performance of Japanese organizations. Journal of Applied Psychology, 92: 1069-
1083.
Trank, C. Q ., Rynes, S. L., & Bretz, R. D, Jr. 2002. Attracting ap plicants in the war for
talent: Differences in work preferences amon g high achievers. Journal of
Business and Psychology, 16: 331-3 45.
Trevor, C O ., Gerhart, B., & Boudreau, J.W. 1997. Voluntary turnover and job
88
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 95/147
performan ce: Curvilinearity and the moderating influences of salary growth and
promotions.
Journal of Applied
Psychology, 82: 44-61.
Truss, C. 2001. Complexities and Controversies in linking HRM with organizational
outcomes. Journal of Management Studies. 38: 1121-1149.
Williams, C. R. 1999. Reward contingency, unemployment, and functional turnover.
Human Resource Management Review, 9: 549-576.
Williams, C.R., & Livingstone, L.P. 1994. Another look at the relationship between
performance and voluntary turnover, Academy of Management
Journal,
37: 269-
298.
Wh itener, E. M. 20 01 . Do high comm itment hum an resource practices affect employee
commitment? Across-level analysis using hierarchical linear modeling. Journal of
Management, 27: 515-535.
Wood, R., & Bandura, A. 1989. Social cognitive theory of organizational management.
Academy of Management Review, 14: 361-384.
Wright, P. M., M cM ahan, G. C. 1992. Alternative theoretical perspectives for strategic
human resource management. Journal of Management, 18: 295-320.
Wright, P. M., McMahan, G . C , & McW illiams, A. 1994. Hum an resources and
sustained competitive advantage: a resource-based perspective.
International
Journal of Human Resource Management, 5: 301-326.
W right, P. M ., Dunford, B. B., & Snell, S. A. 20 01 . Hum an resources and the resource
based view of the firm. Journal of Management, 27: 701-721.
Wright, P. M., & Boswell, W. R. 2002. Desegregating HRM: A review and synthesis of
89
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 96/147
micro and macro human resource management research. Journal of Management,
28 : 247-276.
Wright, P. M., Gardner, T. M ., & Moynihan, L. M. 2003. The im pact of HR practices on
the performance of business units.
Human Resource M anagement Journal,
13:
21-36.
W right, P. M., Gardner, T. M , Mo ynihan, L. M., & Allen, M. R. 2005. The relationship
between HR practices and firm performance: Examining causal order. Personnel
Psychology,
58: 409-446.
Yound t, M. A., Sn ell, S. A., Dean Jr, J. W., & Lepak, D. P. 1996. Hum an resource
management, manufacturing strategy, and firm performance. Academy of
Management Journal, 39: 836-866.
Zeller, A. 1962. An efficient m ethod of estimating seemingly unrelated regression and
tests for ag gregation bias. Journal of the Am erican S tatistical Associatioin, 57:
348-368.
90
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 97/147
T
A
3
1
M
e
S
a
d
D
a
o
a
C
e
a
o
^
v
1
O
g
z
o
S
z
2
O
g
z
o
A
3
C
a
n
e
y
4
U
o
P
5
M
a
u
n
6
C
u
o
7
F
n
8
N
m
u
n
9
C
y
o
O
g
n
1
T
a
T
n
1
H
g
P
o
m
T
n
1
R
o
w
k
o
T
n
1
H
W
S
1
R
M
e
6
6
2
3
1
0
0
4
0
4
0
1
0
0
0
3
0
0
0
0
0
0
0
0
2
3
7
9
S
D
1
4
1
7
1
2
0
5
0
5
0
3
0
2
0
4
0
2
0
0
0
1
0
0
4
4
1
7
1
.
3
.
3
.
3
-
0
-
0
.
2
-
0
-
1
-
3
-
2
-
3
.
2
-
1
2
.
3
.
2
.
1
.
0
.
1
-
2
-
1
-
1
-
1
-
0
-
0
-
1
3
.
2
.
1
-
0
.
0
-
1
-
2
-
3
-
2
-
3
.
0
-
0
4
.
1
-
0
.
1
-
2
.
0
-
3
-
1
-
3
-
2
-
0
5
-
3
-
2
-
6
-
0
-
0
-
0
-
1
-
1
-
0
6
-
1
-
2
-
0
.
1
.
1
.
0
-
2
.
0
7
-
2
-
0
-
1
-
0
-
1
.
0
-
1
8
.
1
.
1
-
0
.
1
.
2
.
1
9
.
0
.
0
.
0
.
1
-
0
1
.
6
.
9
-
1
-
1
1
.
3
-
1
-
2
1
-
0
-
1
1
.
1
w
r «
*
<
0
0
w
a
e
e
*
0
0
w
a
e
e
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 98/147
TABLE 3.2: Results of OLS Regression for HPWS, Total Turnover, and ROA
Variables
Organization Size
Organization Age
Union Presence
Capital Intensity
Country of Origin
Industry
Construction
Finance
Non-manufacturing
HPWS
Total Turnover
R
2
AR
2
FforAR
2
Overall F
N
Total Turnover
Model 1
-0.1451
0.101
-0.327**
-0.195**
0.025
0.071
-0.036
0.057
-0.181*
0.245
0.024
4.599*
5.835**
153
Model 2
-0.114
-0.152*
0.127f
0.007
-0.138t
0.058
-0.126f
0.116t
0.182*
0.076
0.024
3.775*
1.758t
153
ROA
Model 3
-0.112
-0.129t
-0.020
-0.044
-0.105
0.058
-0.128t
0.144t
-0.297**
0.142
0.067
11.095**
2.638**
153
Model 4
-0.154t
-0.125t
0.037
-0.047
-0.131t
0.078
-0.136t
0.132f
0.132 |
-0.274**
0.155
0.079
6.611**
2.597**
153
Note. Standardized regression coefficients are shown. The omitted industry variable was manufacturing,
t p < 0.10, one-tailed test
*p < 0.05, one-tailed test
**p< 0.01, one-tailed test
92
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 99/147
TABLE 3.3: Results of OLS Regression for HPW S and High Performer and Rest of
workforce Turnover
Variables
Organization Size
Organization Age
Union Presence
Capital Intensity
Country of O rigin
Industry
Construction
Finance
Non-manufacturing
High Performer
Turnover
Model 5
a
-0 .126t
-0.020
-0.168*
-0.108
0.018
0.134t
0.004
-0.039
High Performer
Turnover
Model 6
b
-0.01 I t
-0.000
-0.044*
-0.011
0.009
0.058*
0.001
-0.010
Rest of Workforce
Turnover
Model 7
b
-0.006t
0.00 I t
-0.051**
-0.012*
0.007
0.007
-0.012
0.013
HPWS
-0.174 -0.005
-0.003t
AR
2
FforAR
2
Overall F
x
n
0.153
0.022
3.695*
2.871**
153
0.153 0.231
27.65**
153
46.05**
153
Note.
a
Standard ized regression coefficients are shown. Unstan dardized regression coefficients are shown.
The omitted industry variable was manufacturing,
tp < 0.10, one-tailed test
*p < 0.05, one-tailed test
**p <
0.01,
one-tailed test
93
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 100/147
TABLE 3.4: Results of OLS Regression for HPWS, High Performer and Rest of
workforce Turnover, and ROA
ROA
Variables
Model 8 Model 9 Model 10
Organization Size
Organization Age
Union Presence
Capital Intensity
Country of Origin
Industry
Construction
Finance
Non-manufacturing
Rest of workforce Turnover
HPWS
High Performer Turnover
-0.137f
-0.129f
0.068
-0.028
-0.133t
0.064
-0.135t
0.132t
-0.185*
0.155*
-0.125t
-0.145t
0.010
-0.043
-0.107t
0.079
-0.124f
0.127f
-0.146t
-0.268**
-0.163*
-0.140f
0.041
-0.046
-0.130f
0.096
-0 .131 |
0.118f
-0.133t
0.118t
-0.253**
R
2
AR
2
FforAR
2
Overall F
0.126
0.017
2.763*
2.044*
153
0.168
0.060
10.177**
2.876**
153
0.178
0.069
5.947**
2.779**
153
Note. Standardized regression coefficients are shown. The omitted industry variable was manufacturing,
tp < 0.10, one-tailed test
*p < 0.05, one-tailed test
**p < 0.0 1, one-tailed test
94
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 101/147
FIGURE
FIGURE
3.1:
Studies in the Relationship between HPWS and Firm Performance
Categories
Linkage between HR practices and Firm Performance
HR practices
Organizational
Commitment
Turnover Firm Performance
(1)
HR practices /
HR practices *
strategy
—»Firm
Performance
Bae & L awler (200 0), Ichiniowski, et al. (1997), Delery & Doty (1996), McDuffie
(1995) etc.
(2)
HR practices
—»Turnover
—»Firm
Performance
Batt (2002), Huselid (1995)
(3)
a. HR practices
—»Turnover
b. HR
Practices
x Turnover
—>Firm
Performance
a.
b -
Guthrie (2001), Arthur (1994)
(4)
HR practices
(perception to
HR practices)
Organizational
Commitment
Williams (2004), Agarwala (2003), Whitener (2001), Eaton (2001), Meyer & Smith
(2000), Appelbaum et al (2000), Oligva (1986), etc.
means that studies assumed the theoretical relationship.
95
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 102/147
CHAPTER 4
A RELATIONSHIP BETWEEN HIGH PERFORMER TURNOVER AND FIRM
PERFORMANCE: A FIRM-LEVEL FIELD STUDY
Traditionally, em ployee turnover has been viewed as problematic for firm
performance because employees' participation in organizational activities was regarded
as a necessary condition for effective organizational functioning (Barnard, 1938; March
& Simon, 1958). How ever, a stream of research has pointed out that whereas employee
turnover negatively influences firm performan ce, this relationship m ay depend on the
performance levels of the employees who leave (Cascio, 1995; Dalton & Todor, 1979;
Hollenback & W illiams, 198 6; Jackofsky, 1984; Jeswald, 1974; Mob ley, 1982;
Mu chinsky & Turtle, 1979; Price, 1977; Schwab, 1991 ; Staw, 1980; Steel, Griffeth, &
Horn, 2002; Sterman, Gerhardt, Boudreau, & Trevor, 2003). For example, Schwab (1991)
articulated that only if high performers leave are the consequences of turnover
necessarily bad for an organiza tion (p. 966). Gardner (2005) observed that, in practice,
more than 2 0% of organizations have experienced and suffered from high performer
turnover due to 'purposive talent raiding' from their competitors as a consequence of the
organizations' dynamic competition for attracting and retaining high performing valuable
emp loyees, not only to increase their organizational effectiveness but also to reduce their
competitors' competitive advantage.
W hereas both sch olars and practitioners have viewed high performer turnover as a
critical organizational problem , a few im portant theoretical and empirical questions
remain un answe red. First, am ong o thers, there are surprisingly few studies to empirically
96
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 103/147
investigate the relationship between high performer turnover and firm performance.
Although scattered theoretical perspectives suggest a negative relationship between high
performer turn over and firm performance, concrete empirical evidence is required to
enhance the confidence of said theoretical perspectives in this relationship (M obley,
1982; Price, 1977 ). Second, although high performer turnover may neg atively influence
firm performance, studies up to this point hav e paid little attention to the roles of varying
organizational co ntext in this relationship. As a result, little is know n of
if
and
which
organizational con textual variables influence this relationship. Investigating these
questions is critical not only to accumu late knowledge on research on turnover but also to
provide valid insights and practical guidelines for organizations, which have dedicated
much effort to d eveloping strategic workforce strategy in order to succeed in the dynamic
market competition for valuable human capital.
To this end, this pap er first theoretically explains the negative relationship
between high performer turnover and firm performance and empirically tests this
relationship. Second , this paper investigates how and in what directions a firm's
investment in employee training and development (ETD) and a firm's reputation, both of
which are critical organizational contextual variables, moderate the relationship between
high performer turnover and firm performance. This paper contributes to better
understanding not only of the overall relationship between high performer turnover and
firm performance but also of the moderating roles of a firm's investment in ETD and firm
reputation in this relationship. In addition, by considering heterogeneous organizational
context, it provides practical im plications for organizations that have striven to develop
high performer attraction and retention strategies.
97
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 104/147
RESEARCH FRAMEWORK
Accumulating theoretical and empirical work shows that overall employee
turnover is generally detrimen tal for organizational performance (Alexander, Bloom, &
Nucho ls, 1994; Brown &
Medoff,
1978; McElory, Morrow, & R ude, 200 1; Osterman,
1987; Price, 1977; Shaw, G upta, & Delery, 20 05; Yanadori & K ato, 2007). In turn, it also
points out that the performance levels of leavers and the organizational context in which
turnover o ccurs could hav e significant implications for the relationship b etween
employee turnover and firm performance (Abelson & Baysinger, 1984; Arthur, 1994;
Dalton, Todor, & Krachhardt, 1982; Mobley, 1982; Price, 1977; Schwab, 1991; Shawet
al.,
2005; Staw, 1980). However, most previous studies assumed that high performer
turnover is negatively associated with firm performance without empirical evidence. In
addition, they paid little attention to the possibility that this relationship can be moderated
by varying organizational contexts.
This paper fills this research gap in turnover research, focusing on the relationship
between high performer turnover and firm performance. Figure 1 depicts the research
framework of this study. First, building upo n hum an capital theory, social capital theory,
and a cost-benefit approach to the co nsequences of turnover, this study predicts and
empirically tests the negative relationship between high performer turnover and firm
performance. Second, this paper argues that this negative relationship can be m oderated
by the degree of a firm's investment in ETD and the degree of firm repu tation, because
these two org anizational contextual variables may affect the turnover costs of high
performers, including thro ugh ensuring a high level of workforce quality to mitigate their
turnover costs, and throug h ensuring o rganizational ability to find replacem ent workers.
98
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 105/147
Finally, this paper explains and tests two competing theoretical predictions about the
directions of mod erating effects of these two contextual variables on this relationship.
THEORETICAL BACKGROUND AND H YPOTHESES
High Performer Turnover and Firm Performance
A m ajority of organizational level studies examining the relationship between
turnover and firm performan ce argue that high performer turnov er negatively influences
firm perform ance (e.g., Dalton & T odor, 1979; Schwab, 1991). In line with this argument,
individual level studies on turnover have investigated how individual performance and
turnover is related and w hat factors could influence high performers' turnover (e.g., Allen
& G riffeth, 20 01 ; Jackofsky, 1984; Shaw & Gupta, 2007 ; Trevor, Gerhart, & Boudreau,
1997). Ho weve r, there are few studies that have explicitly exp lained why high performer
turnover is harmful for firm performance. I explain this theoretical question building
upon hum an capital theory, social capital theory, and cost-benefit approach to turnover.
Human capital theory argues that a firm's investment in employee training makes
workers m ore produ ctive by dev eloping app ropriate skills and abilities (Becker, 1964).
The accum ulation of firm-specific hum an capital embodied in a workforce determines
firm perform ance (Strober, 1990), and thus performance should decline as turnover
increases because an orga nizatio n's firm-specific hum an capital accumulation is depleted
by turnover. A decent am ount of studies based imp licitly or explicitly on huma n capital
theory suppo rt the negative linear relationship between turnover and firm performance
(Alexander et al., 1994; Bart, 2002; Huselid, 1995; Osterman, 1987). This prediction
based on hum an cap ital theory can be directly extended to the relationship between high
99
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 106/147
performer turnover an d firm performance. Shaw et al (2005) argued that high
performance signifies more hum an capital (p. 65). Therefore, turnover of high
performers re lative to other levels of performers will deplete the even larger amounts of
hum an capital of a firm, w hich at the very least results in negative firm performance.
The negative relationship betw een high performer turnover and firm performance
can be also explained from a social capital theory perspective, which explains the value
in social relationships in netw orks (Shaw, Duffy, Johnson, & Lochhart, 2005, p. 595).
Focusing on the collective social capital, defined as a bonding form of social capital
comp rising internal ties within collectivities (Adler & Kwo n, 2002, p. 19), Shaw et al
(2005b) argued that, whereas social capital created by continuous interactions among
organization members could be a valuable resource for organizations, turnover may
deplete the valuable social capital by damaging the internal social structure and fabric of
a firm. In particular, they em phasized that turnover among individuals who occupy key
structural and relational network positions could lead to significant loss of social capital
in organizations resulting in lower firm perform ance. E xisting theoretical perspective and
empirical evidence that individual performance and the network centrality and amount of
ties are highly associated (Burt, 1992; Cross & Cum mings, 2004) suggests that high
performer turnover will negatively influence firm performance because their turnover
results in significant amount of social capital loss.
Lastly, from a slightly different perspective, cost-benefit approach in the
consequence of turnover offers a rationale behind the negative relationship between high
performer turnover and firm performance. This approach argues that turnover may not
necessarily be harmful for firm performance but rather depends on the performance level
100
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 107/147
of leavers and varying organ izational context (e.g., Dalton, Todor, & Krachhardt, 1982;
Dalton & Tod er, 1982; Mobley, 1982; Price, 1977; Staw, 1980). This approach especially
emph asizes the implications of the performance level of leavers on the consequences of
turnover. It argues that, although turnover imposes various organizational costs, turnover
by lower performers can benefit firms because it allows firms to upgrade their human
capital pool by replacing lower performers with highly qualified people from outside. For
example, Dalton, Todor, and Krachhardt (1982) pointed out that performance and
replacebility tend to be orthogonal so that high perform ers' turnover is dysfunctional for
firm perform ance but low perform ers' turnover in fact can be functional for firm
performance. Although this approach advocates optimality (Inverted-U-shape) in the
overall relationship betw een turnov er and firm, in general, they suggest that, all things
being equal, high performer turnover results in lower firm performance. Building upon
the aforementioned theoretical perspectives, I propose the following hypothesis.
Hypothesis 1 : The relationship between high performer turnover and firm performance
will be nega tive.
Moderating Relationships between High Performer Turnover and Firm
Performance
A plethora of studies suggest that the relationship between turnover and firm
performance may be contingent on the performance levels of leavers and varying
organizational conditions (Cascio, 1995; Hollenback & Williamson, 1986; Jackofsky,
1984; Mo bly, 1982; Price, 19 77; Staw, 1980; Shaw et al., 200 5; Schwab, 1991). However,
the majority of theoretical and empirical studies tend to take the view either that high
performer turn over is universally harmful for firm performance or that varying
organizational conditions could moderate the relationship between overall employee
101
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 108/147
turnover and firm perform ance, w ithout much con sideration of the interaction effects
between high performer turnover and varying organizational conditions on firm
performance. For exam ple, a few em pirical studies either mention the possibility of a
mod erating role of performance level of leavers on overall employee turnover and the
firm's performance (Shaw et al., 2005b) or report the mo derating role of human resource
system as an organizational contextual variable in the relationship between employee
turnover and firm performance (Arthur, 1994; Guthrie, 2002).
How ever, it may n ot necessarily be true that the magnitude of the impact of high
performer turnover on firm performance is homogenous across firms because they
manage their workforce in varying ways. For example, Dalton, Todor, and Krachhardt
(1982) stated as follows:
For example, it is quite possible for a truly stellar (very high quality) employee to leave
the work unit; yet, because of the nature of the work or the work force, many equally or
even better qualified people are available to replace him/her. Such turnover is not a
threat to the organization even though it involves high quality performers
(p. 121).
These authors suggest that, although high performer turnover, in general, negatively
influences firm performance, the costs of high performer turnover varies depending on
diverse organizational factors such as quality of workforce and replacebility of high-
performing leavers. In the proceeding sections, I address how organizational context,
defined as the situational oppo rtunities or constraints that affect functional relationships
between variab les (Johns, 2006: 386), could influence the functional relationship
between h igh performer turnover and firm performance due to its impacts on the costs of
high performer turnover, quality of workforce, and replacebility of high-performing
leavers. Particularly, I investigate how this negative relationship can be m oderated by (1)
a firm's investment in ETD and (2) firm reputation, which are, among others, two critical
102
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 109/147
organizational contextual variables that may influence this relationship. In terms of the
direction o f the mo deration effects of these two contextual variables on this relationship,
how ever, there are a numb er of plausible but competing theoretical predictions, each of
which I will explore in the proceeding sections.
A Firm s Investment in Employee Training and Development
On the one hand, research suggests a negative association between employee
turnover and firm performance because employee turnover imposes diverse
organizational c osts related to training, socialization, replacement, e tc. On the other hand,
it also points out that these costs tend to be different among organizations so this
relationship can vary among them as well.
For instance, hum an capital and co st-based perspectives on the consequences of turnover
predict that employee turnover, particularly among high performers, depletes the
opportunity for orga nizations to receive return from their investment in ETD and, also,
imposes costs for retraining replacem ent w orkers (e.g., Dalton et al., 1982; Shaw et al,
2005 b; Price, 1 977). At the sam e time, they also imply that a difference in a firm 's
investment in ETD may moderate the relationship between overall employee turnover
and firm perform ance. Like the moderating effect of a firm's investment in ETD on the
overall employ ee turnov er and firm perform ance, I propose that the negative relationship
between high performer turnover and firm performance can be also moderated by a
firm's investment in ETD . How ever, theoretical predictions on the directions of its
mod erating effects on this relationship are not unequivocal due to their exclusive focus
on its impact on either the turnover costs of high performer turnover or on the quality of
workforce. Thus, I investigate both predictions below.
103
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 110/147
Stronger Negative Effect. High performer turnover m ay be more detrimental for
firms that invest more in ETD, because firms with high investment in ETD may lose
more o pportunities to receive return from highly trained employees and, in addition, they
need to invest more in retraining new ly acquired em ployees. In this vein, strategic human
resource man agem ent literature sugg ests that whereas firms that invest more in ETD can
enhance firm perform ance, employee turnov er in such organizations, relative to those
with less investment in ETD , may result in stronger negative effects on firm performance
due to the loss of employees w ho are highly trained and developed in the organizations
(Arthur, 1994; Guthrie, 200 1). Arthur (1994) specifically showed that firms that heavily
invested in developing employee skills by using a commitment based human resource
system, suffered mo re from employ ee turnover than firms that invest less.
Hypothesis 2a: There will be a stronger negative relationship between high performer
turnover and firm performance in firms that invest more in employee training and
development than in firms that invest less in employee training and development.
Weaker Negative E ffect. In co ntrast, the negative association of high performer
turnover w ith firm performance could be weaker in firms that invest more in employee
training and development. Unlike the former theoretical perspective that takes into
account only the different costs incurred due to high performer turnover, this prediction
can be explained by considering that high investm ent firms may po ssess a high quality
pool of hum an capital as a who le, which may enable the firms to effectively mitigate the
negative con sequenc es of high performer turnover (D alton et al., 1982). For instance,
Wright, McMahan, and McWilliams (1994) argued that firm performance is influenced
not only by top ma nagem ent or key managers b ut also by the quality of the total
104
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 111/147
organizational poo l of hum an capital. Also, Pfeffer (2001) argued that in order for
organizations to be successful, they should invest in enhancing the total quality of the
human capital pool rather than exclusively focusing on retaining high performers. The
argum ents of these autho rs suggest that a high quality pool of human capital, developed
through high investment in employee training and development, may help firms to reduce
the negative impa ct of high performer turnover on firm performance and to sustain
superior firm perform ance even though a portion of high performers leave.
Hypothesis 2b: There will be a weaker negative relationship between high performer
turnover and firm performance in firms that invest more in employee training and
development than in firms that invest less in employee training and developm ent.
Firm Reputation
Research points out that the replaceability of leavers is one of the key contingent
factors in the relationship between employee turnover and firm performance (e.g.,
Abelson & B aysinger, 1984; Mobely, 1982; Dalton et al., 1982; Price; 1977; Staw, 1980).
The replaceability of leavers can be largely affected by (1) the quantity and quality of
potential work ers wh o desire to join an organization and (2) the extent to which newly
hired employ ees successfully perform the tasks that leavers conducted. In this regard,
firm reputation, defined as a perceptual representation of a firm's overall appeal
compared to other leading rivals (Fombrun , 1996, p. 72), may influence the negative
relationship betwee n high performer turnover and firm performance due to its impact on
these two facets of the replacebility of
leavers.
However, like a firm's investment in ETD,
the direction of the modera tion effect of firm rep utation involves am biguity, since firm
repu tation's effects on bo th facets of the replaceability of leavers generate conflicting
105
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 112/147
predictions. Below , I explain both explanations for the moderating effect of firm
reputation on the negative relationship between high performer turnover and firm
performance.
Weaker Negative E ffect. High performer turnover can be negatively associated
with firm perform ance. How ever, firms that easily attract a high num ber of qualified
employees may reduce various costs of high performer turnover such as the recruitment,
retraining, and socialization. In a certain case, those firms can bring in people with even
higher quality skills and innovative ideas. Abelson and Baysinger (1984) argued that a
labor supply condition of organizations could influence the consequences of turnover
because it affects both turnove r and retention c osts. A key point implied in their argument
is that the quality and q uantity of replacem ent w orkers is a critical contingent factor in
the relationship between employee turnover and firm performance. In order words, even
though high performer turnover may have negative consequences for firm performance
due to the costs incurred due to the losses of human and social capital, the negative
consequences may also be conditioned by a firm 's ability to attract a high quantity of
qualified replac em ent w orkers. For instance, a meta-analytic study by Farrell and
Hakstian (2001) shows that firms with more job applicants can hire more productive
workers than firms with fewer job applicants.
Considering that the high performer turnover-firm performance relationship can
be moderated by the q uantity and quality of potential employees that firms can attract
(Abelson & B aysinger, 1984; Dalton et al., 1982), firm reputation m ay influence this
relationship, becau se a firm with high repu tation is more likely to attract a high q uantity
and quality of potential work ers than a firm w ith low reputation (C ollins & Han, 2004;
106
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 113/147
Gatewood, G owan , & Lautenschlager, 1993; Turban & Cable, 2003). Social identity
theory and signaling theo ry provide theoretical rationale regarding why firm reputation
affects the quantity of po tential employees (Turban & C able, 2003). Social identity
theory points out that peo ple are m otivated to enhance their self-esteem and classify
themselves into social categories such as group mem bership and organizational affiliation
(Ashforth & Me al, 1989 ; Tajfal, 1972). It suggests that the attributes and social status of
the social category, w here peo ple belong to, influence their self-esteem (Dutton,
Duk erich, & Harquail, 1994). Applying the implications of social identity theory to
recruitment literature, Turban and Cable (2003 ) argued that firms w ith positive reputation
attract more job applicants, w ho are mo tivated to enhance their self-esteem. In addition,
they claimed that signaling theory, which suggests that job applicants do not have
complete information before making
a
job application decision and they infer a firm's
working condition from the available information, supports this prediction because job
applicants may interpret firm reputation as a signal for the firm 's w orking condition. The
empirical results of their study strongly supported the hypothesis that firm reputation
affects the quantity o f jo b app licants of
a
firm.
Firm rep utation can also influence the quality of potential em ployees of a firm.
Building upon expectancy theory (Vroom, 1964), Rynes (1991) suggested that in the
processes of job search and choice behavior, applicants evaluate both the attractiveness
among job alternatives and expectation of attainability of the job in order to effectively
invest their limited tim e and reso urces. Based on this perspective, Turban and Cable
(2003) argued that firm reputation influences the degree of expectation of ap plicants'
obtaining a job based on their quality so that applicants with high qualification will be
107
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 114/147
more likely to be attracted to firm with high reputation than applicants with low
qualification. E mpirical results of their study supported their hypothesis.
In sum, the aforementioned theoretical perspectives and empirical works suggest the
following hypothesis:
Hypothesis 3a: Th e relationship betw een high performer turnover and firm performance
will be moderated by firm reputation, in such a way that the negative relationship
between high performer turnover and firm performance will be weaker in firms with
higher firm reputation than in firms with lower firm reputation.
Stronger Negative Effect. Hypo thesis 3 a predicts a weaker negative relationship
between high perform er turnover and firm performance in firms with relatively high
reputation, focusing on the firm reputation's effect, ceteris paribus, on the quantity and
quality of potential w orkers w ho desire to join the firms. How ever, this prediction does
not pay mu ch attention to the possibility that firm reputation m ay influence the quality of
existing employ ees in the firm s. It is plausible that the quality of employees m ay be
higher in firms w ith high repu tation than in firms with low repu tation, because in the past
firms with high reputation cou ld select employees from larger numbers of qualified
potential work ers, as addressed above. Thus, finding qualified replacement w orkers is
more difficult and costly in firms w ith high reputation than firms w ith low reputation.
Consequently, high performer turnover results in a stronger negative impact on firms
with relatively high reputation. T his prediction is in line with resource-based view of firm,
which argues that, whereas organizational resources such as high quality of workforce
can be a source of comp etitive advantage, nurturing and m aintaining them is necessary
108
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 115/147
for firms to take advan tage of these resources (e.g., Barney, 1991 , 20 01 ; W right et al,
1994).
In addition, Fom brun (1996) pointed out that some of the significant benefits that
firms receive from high reputation include high levels of cooperation among
organizational m em bers, teamw ork, and a sense of shared destiny. These benefits can be
viewed as a social capital advantage created by firm repu tation. Turnover research
suggests that employ ee turnov er m ay result in the severe loss of social capital, which
exists as relational form of trust or sense of bonding am ong organizational m embers
(Dess & Shaw,
2001;
C oleman , 1988). Furthermo re, in order to recover the loss of social
capital, the organizations need to invest a significant amount of resources to rebuild
social capital, which develops through continuous interaction among members over a
long period of time (C olem an, 1988; Nahap iet & Gh oshal, 1998). From this perspective,
it is plausible to predict that turnover of high performers m ay be m ore detrimental for
firms with high reputation compared to low reputation, because, in firms with higher
reputation, the loss of social capital present in the relationships among organizational
mem bers and the costs to repair dam age to that social capital due to turnover are larger,
and it may take much longer time for new replacement workers to become assimilated
into the high quality social relationships, trust, and cooperative way of thinking and
behaving present among organizational members. That is, replacing high performer
turnover is mo re costly in firms with high reputation than in firms with low reputation.
These theoretical perspectives suggest the following h ypothesis:
Hypothesis 3b: T he negative relationship betw een high performer turnover and firm
performance will be moder ated by firm reputation, in such a way that the negative
109
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 116/147
relationship betw een high performer turnover and firm performan ce will be stronger in
firms with higher firm repu tation than in firms with lower firm reputation.
METHODS
Sample and Procedures
The sam ple of this study is firms operating in South K orea. In 2006, a prestigious
research institute in South Ko rea distributed survey questionnaires to a senior human
resource manager of 994 firms via e-mail. E-mail survey was em ployed because of the
following two reaso ns: first, senior human resource m anagers, who had responded to a
pilot study, preferred e-mail survey to a paper-based m ailing; second, delivering the
questionnaire to the right person was particularly im portant because the questionnaire
items such as overall and high performer turnover rates tended to be considered
confidential. T he co ver letter of the e-mail survey addressed that one of the purposes of
the survey was to examine the relationship between human resource management and
firm performance. Each f irm 's response to the survey was guaranteed to be treated as
confidential and used o nly for research purposes. A pilot study was conducted in
February 2 006 for n ine firms in order to ensure the clarity of the survey questionnaire. A
total of 190 firms participated in the survey, which resulted in a
19.1%
of response rate.
The final samp le size included for hypothesis testing, however, was reduced to
158 firms after excluding 32 organizations for the following two reasons. First, when
firm performance information of participating organizations was not available, when
participating firms failed to respond to questions such as high performer turnover rate, or
when participating firms were government subsidiaries, the firms were excluded from the
110
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 117/147
analyses. Second, four organizations reported that they implemented new human resource
practices such as em ployee training and participation in 2006. Including these four
organizations in our analyses could result in biased estimates of the moderating effect of
firms' investment on employee training and development in the relationship between
high performer turnover and firm performance because of a measurement timing issue:
measuring firm performance and high performer turnover in 2005 but firms' investment
in employee training and development in 2006 (e.g., Guthrie,
2001;
Wright, Gardner,
Moynihan, & Allen, 2005). Therefore, we excluded these four organizations in our
analyses to avoid this prob lem. T he 158 participating firms operated in diverse industries:
manufacturing (46%), construction (10%), finance such as banking, insurance, and
security (9%), and other non-manufacturing industries such as telecommunication and
distribution (35%). Their mean employment size was 2,449: 31 to less than 100
employees (8.7% ), 100 to less than 500 em ployees (34 .8%), 501 to less than 1,000
employees (19.9%), more than 1,000 employees (36.6%).
Measures: independent variables
High performer turnover rate was a turnover rate among top 20 % performers in
2005 (Shaw , Dine en, Fang, & Vellella, in press) and non-high performer turnover rate,
which was controlled in the regression analyses, was a turnover rate among employees
who did not belong to the top 20% performing group. Both turnover rates measured
turnover rates of permanent employees. In order to measure high performer turnover and
non-high performer turnover rates, two separate questions were asked to a senior human
resource man ager of each firm. First, overall turnover rate was measured by asking a
senior hum an resource manager W hat is an annual workforce turnover rate of permanent
111
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 118/147
employees in 200 5? An additional explanation was specified in order to assess only
voluntary turnover as follows: Turnove r occurred by such reasons as age limitation,
disease, and etc. is not included in com puting an annual turnover rate . Then, in order to
measure a high performer turnover (top 20 percent performer) rate and a non-high
performer turn over (the other 80 % percent performers) rate, a second question was asked
to him/her In the annual workforce turnover rate, what percentage is from the turnover
of top 20 percent perfo rm ers? To clarify this question, the following formula was
provided: (numbers of leavers, who were high performer) / (total numbers of leavers) =
( ) %. Based on the responses to the two questions, I computed both high performer
turnover and non -high perform er turnov er rates. For example, if overall turnover is 40
percent and the percentag e o f high (top 20 percent) performer turnover in the overall
turnover is 10 percent, a high performer turnover rate is 20 percent by computing [(0.4 X
0.1) / 0.2] and a non-h igh perform er turnov er rate is 45 percent by com puting {[0.4 X (1
- 0 . 1 ) ] / 0 . 8 } .
Firms' Investment in Employee Training and Development was measured by a
senior HR manager's responses to three items, adapted from the measures used by Bae
and Lawler (2000) and Snell & Dean (1992). All three items were constructed with a 5
point Likert-scale, where 1 meant strongly disagree and 5 mean t strongly agree . The
senior HR m anager of each firm w as asked to assess these items separately for
manag erial and non-man agerial w orkers and I used average scores of each item. The
inter-item reliability of th is measure of this variable was 0. 93. The three items were:
This organization spends a lot of mon ey on employee training , This organizations
112
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 119/147
provide employ ees with a variety of training and development program s , and This
organizations provide employees with structured training and development programs .
Firm reputation was directly measured by surveying 398 undergraduate students
in South Korea in April 2007, following a method used in a study of Rindova,
Williamson, Petkova, and Sever (2005). The survey asked the undergraduate students to
assess firm reputation of the 190 participating firms in the organizational-level survey
and nom inate ten firms in term s of their firm reputation. The logarithm of the numbers of
nom ination that a firm received from u ndergraduate students was used as a measure of
firm reputation. This firm reputation measure is different from the measure of firm
reputation indirectly assessed by using sources from such as Business Week and Fortune
(e.g.,, Turban & Cable, 2003; Flanagan & O'Shaughnessy, 2005 ; Gatewood, Gowan, &
Lautenschlager, 1993). In the research context of this study, directly measuring firm
reputation by surveying undergradu ate students appears more appropriate than indirectly
measuring it by using second-han d sources for the following reason s. First, a firm
reputation m easure from the second-hand sources is available only for large firms so that
a firm reputation m easure for the small- or med ium-size participating firms in the survey
of this study w as not ava ilable. Also, it may not be necessarily true for small- and
medium-size firms have low firm reputation. Second, undergraduate students are in a
good position to m easure firm reputation, which reflects the overall appeal of firms to
multiple stakeho lders, because they are consumers of firm s' d iverse products and services
as well as potential employees.
113
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 120/147
Measures: dependent variables
Firm performan ce wa s assessed by using two indicators, return on asset (ROA)
and return on equity (RO E). Both RO A and RO E were taken from a database built by the
Korea Information Service (KIS ), which is equivalent to Compu stat in U S. KIS
collaborates with Moody's to provide information on organizations operating in South
Korea for an international audience (Chang, 2003).
Measures: control variables
Variables used in the mod els as controls were organization age, organization size,
union presenc e, capital intensity of each organization, and industry. Organization age,
which w as measured as the number of years in operation, was included in order to control
for any ad vantages related to length of business o peration (e.g., Guthrie, 2001 ; Huselid,
1995). Firm size,
which was measured as the logarithm of the number of employees, was
controlled because larger organizations may have advantage such as economy of scale
relative to small organizations (e.g., Pfeffer & Salancik, 1978). Union presence was
controlled because it may influence firm performance (e.g., Freeman & Medoff, 1984).
Capital intensity, w hich is measured by the logarithm of fixed assets/the num bers of
employ ees, was controlled beca use a recent study show s that this may influence firm
performance (Datta, Guthrie, & Wright, 2005). Industry - manufacturing, construction,
finance, and other non-ma nufacturing - was controlled in the regression models because
firm perform ance may be influenced by the industry effect, and country of origin (Korean
firm v ersus foreign firms) wa s also controlled in the regression m odels (Bae & Law ler,
2000).
114
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 121/147
Analysis
W hen testing hy potheses in this study, I employed a hierarchical regression
method that is designed to a ssess whether a single variable or set of variables explains
additional variances o ver the variance explained by previous sets of variables (Cohen,
Cohen, W est, & Aike n, 20 03). In order to test hypothesis 1, which investigates an overall
relationship betw een h igh performer turnover and firm performan ce, I first loaded control
variables in step 1, including firm size and age, unionization, capital intensity, industry,
country of origin, and non-high performer turnover rate. In step 2,1 added high performer
turnover rate to the step 1 regression model.
In order to test moderating effects of a firm's investment in ETD and firm
reputation on the relationship between high performer turnover and firm performance, I
added interaction terms - a firm's investment in ETD x high performer turnover, firm
reputation x high performer turnover, and both interaction terms - to the based model
with all control variables, and checked the significance of coefficients of each variable as
well as R chang e. In order to reduce any multicollinearity problem, I standardized high
performer turnover rate, a firm's investment in ETD, and firm reputation before
computing two interaction terms (Cohen et al. 2003).
RESULTS
Table 1 contains the m eans, standard deviations, and a correlation m atrix of
variables in this study. As expected, the correlations between high performer turnover
and RO A and be tween h igh performer turnov er and RO E were -0 . 23 (p < 0.01), -0.43 (p
< 0.01), respectively, wh ich indicated significant n egative relationships between those
115
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 122/147
variables. Table 2 reports the results of two sets of hierarchical regressions, which tested
Hypo thesis 1, predicting the ov erall negative relationship between high performer
turnover and firm perform ance - ROA and RO E. Models 1 and 3 in Table 2 show that a
set of control and organizational difference variables explained a considerable amount of
variance in ROA (R
2
= 0.105, p < 0.10), but they did not explain much variance in ROA.
Mo dels 2 and 4 in Table 2 , the second step in the hierarchical regressions, show that
when hig h performer turnover rate was added to the regression w ith a previous set of
variables, high performer turnover rate explained a significant amount of additional
variance in both ROA (A R
2
= 0.060, p <
0.001)
and RO E (A R
2
=
0.201,
p < 0.001). The
coefficients of high performer turno ver rate were significant for both ROA (p < 0.001)
and RO E (p =0.001 ). Therefore, the results of a series of hierarchical regressions support
Hypothesis 1 predicting the negative relationship between high performer turnover and
firm performance.
Table 3 reports the results of the hierarchical regressions, which tested
Hypo theses 2a, 2b, 3a, and 3b, which predict the moderating effects of a firm's
investment in ETD and firm reputation on the relationship between high performer
turnover and firm perform ance. In order to test whether these two variables moderate a
high performer turnover-firm performance relationship, I ran five separate regressions for
each dependent v ariable in Table 3. For RO A, mod el 9 shows that two interaction terms,
firm's investment in ETD x high performer turnover and firm reputation x high
performer turnover, explain a significant amount of additional variance (AiT = 0.035,p<
0.05) beyond the varian ce explained b y variables in model 6. In addition, models 7 and 8,
where each of two interaction term s was added separately, also indicate that, although
116
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 123/147
these variables are margin ally significant, they m oderate the negative relationship
between high performer turnover and firm performance. For ROE, model 14 shows that
the two interaction term s explain a significant proportion of additional variance (A
R =
0.181, p < 0.001) beyond the variance explained by variables in model 11. Models 12 and
13 also point out that a firm's investment in ETD and firm rep utation moderate the
negative relationship between high performer turnover and firm performance. In terms of
the direction of mod erating effects of both variables, mo dels 9 and 14 indicate a weaker
negative relationship be tween h igh performer turnover and firm performance in firms that
invest more in ET D than in firms that invest less in ETD (Hypothesis 2b ); a stronger
negative relationship betw een high performer turnover and firm performance in firms
with high reputation th an in firm w ith lower reputation. (Hypothesis 3b ). Figure 2 and 3
graph the mode rating effects of a firm's investment in ETD and firm reputation on the
relationship between high performer turnover and firm performance (ROA, ROE), which
lend support for Hypotheses 2b and 3b.
DISCUSSION AND CONCLUSION
Research in employee turnover has thought that high performer turnover
negatively influences firm performance. In practice, firms under the current turbulent
economy have competed for valuable human capital not only to increase their own
performance but also deplete the com petitive adv antage of their com petitors. This
dynamic market competition for valuable human capital among firms stems from the
belief that attracting and retaining high performers is a critical factor for firms' success
and survival. However, to my knowledge, there is no single empirical study that
117
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 124/147
explicitly investigates h ow high performer turnover is associated w ith firm performance.
Furtherm ore, sparse scholarly attention has been paid to the possibility that this
relationship can be m oderated by varying o rganizational contextual variables. This paper
is the first emp irical study that investigates the overall relationship betw een high
performer turn over and firm performance. In addition, this study explored the fact that
two organ izational contextual variables, a firm 's investm ent in ETD and firm reputation,
could moderate this relationship.
The results of this study showed that high performer turnover and firm
performance wer e negatively related. This finding confirms the assum ption of research
on turnover reg arding this relationship (e.g., Hollenback & W illiams, 1986; Schwab,
1991;
Steel, Griffeth, & Ho rn, 2002; Sterman, Gerhardt, Boudreau, & Trevor, 2003).
Further analyses sho wed that one standard deviation increase in high performer turnover
above the mean was associated with 39% and 106% decrease in ROA, ROE, respectively.
In addition, this study found the moderating effects of a firm's investment on ETD and
firm rep utation on the relationship between high performer turnover and firm
performance. As show n in Figure 2 and 3, this study found as follows: the negative
relationship between h igh performer turno ver and firm performance was weaker in a firm
that invests more in ETD than a firm that invests less; this negative relationship was
stronger in firms with high er firm rep utation than in firms w ith lower firm reputation
This study prov ides several theoretical implications for the research on employee
turnover. First, this study suggests that in order to enhance the understanding the
relationship be tween emp loyee turnover and firm performance, it is necessary for
scholars to explicitly take into account the roles of organizational context. To date, most
118
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 125/147
studies in the turnov er and firm performance literature have recognized the m oderating
roles of organizational context, because it influences the am ounts of employee turnover
costs and, in certain case s, may provide firm with opportunities to receive benefits from
employee turnover. However, only a few studies have considered the impact of
organizational context on the relationship between em ployee turnover and firm
performance and even these studies investigated a single organizational context, firms'
human resou rce system s (Arthur, 1994; Guthrie, 200 1; Shaw et al., 2005b). Furthermore,
regarding the relationship betw een high performer turnover and firm performance, no
studies have taken into account the m oderating roles of organizational context. This study
provides evidence that the relationship between high performer turnover and firm
performance is moderated by two organizational contextual variables, a firm's investment
in ETD and firm reputation . How ever, different organizational context is not limited to
only those variables. For exam ple, other organizational contextual variables such as
production technology and work independence (e.g., Mobley, 1982; Staw, 1980) could
moderate not only the relationship between overall employee turnover and firm
performance but also the relationship between high performer turnover and firm
performance.
Second, this study provid es a theoretical insight regarding the debate on how
employee turno ver is related to firm performance. Research claims that either this
relationship is negative, negative bu t attenuated, inverted U -shape, or contingent on
organ izations' investm ent in emp loyee. Empirical evidence on this relationship is not
unequivocal. For example, research shows that turnover generally hurts organizational
performance (e.g., Batt, 200 2; Huselid, 1995; Kacmar, An drews, Van Roo y, Steilberg, &
119
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 126/147
Cerrone, 200 6), that the negative effects are attenuated as turnover increases (e.g., Shaw,
Gupta, & D elery, 20 05), or that at least some level of turnover is necessary and benefits
organizational performanc e (e.g., Glebbeek & B ax, 2004), and that organizational
investment in hum an resources m oderates this relationship (Arthur, 1994; Guthrie, 2001).
The am biguity regardin g the shape of this relationship may stem from the fact that overall
turnover m easured in the aforementioned studies juxtapose s two different types of
turnover, high performer turn over and non-high performer turno ver, which have differing
impacts on firm performan ce. For exam ple, although this study did not develop a
different mo del for the impact of turnover between high performer and non-high
performers, the reg ression results in Table 2 indicate that high performer turnover m atters
for firm performance more than non-high performer turnover. Future research
distinguishing employee turnover from high performers versus non-high performers may
help better understand the relationship between employee turnover and firm performance.
It also could be p ossible that, just as turnovers from full-time versus part-time employees
have differently shaped relationships w ith firm performance (Siebert & Zubanov, in
press),
high performer turnover and non-high performer turnover could be associated
with firm performance in a different shape.
Third, individual-level studies on employee turnover have argued that the
relationship betwee n individual performance and turnover is an inverted-U shape, which
means that high performer and low performers are more likely to leave organization than
average level performers (Jackofsky, 1984; Trevor et al., 1997; Salamin & Horn, 2005).
How ever, the results of this study show that an average high performer turnover rate and
non-high perform er turnover rate is approximately 5 percent, 9 percent, respectively,
120
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 127/147
which suggests that high performers are less likely to leave firms than n on-high
performers. T he difference between this study and previous studies in terms of the
performance effect o n an em ploy ee's turnover decision could be related to the different
research contexts, namely employees in Western countries versus Korea. However, this
difference may im ply that the relationship between individual performance and turnover
may no t be either linear or curvilinear. Rather, this relationship can be d etermined by
how fairly firms re cognize and reward individual performance. That is, if firms treat high
performers in a fair and favorable way, the high performers may stay in their firms.
Otherw ise, they m ay be m ore likely to leave their firms than other levels of performers,
because they tend to have m ore job alternatives. The results of this study imply this
possibility. For exam ple, I observed that there is a huge variation in high performer
turnover rate amo ng firm s, and the correlation between high performer turnover rate and
non-high performer turnover rate is 0.3. These results point out that high performer
turnover rate may not be necessarily either higher or lower than average or low
performers bu t rather depends on the firms' fair treatment of high performers. Future
research that takes into account the interaction effect between h igh performance and
firms'
recognition of high performance would contribute to accumulating knowledge on
the relationship between an employee's performance and turnover.
Study Limitations
This paper p rovides p romising results and future research d irections related to the
relationship between high performer turnover and firm performance. However, the
audience should interpret current findings of this study cautiously. First, this study may
be susceptible to measurem ent error, because it measures a firm's investment in ETD
121
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 128/147
from the perspective of one senior HR manager at each firm (Gerhart, Wright, McMahan,
& Snell, 2000). Nevertheless, as Huselid and Becker (2000) argued, measuring HR
practices from the perspective of a senior HR manager could be an effective way of
assessing them b ecause he or she is in a better position to assess the HR practices. Second,
the cross-sectional data lacks the ability to address causal relationships between
independent and dep enden t variables. Third, this study used a sample of firms operating
in South K orea and, thu s, the results of this study may have a generalizability issue in
different countries. I think that m ore studies are required to broaden scientific knowledge
on this topic.
Practical Implications
The results of this study have a few im plications for practitioners, who strive to
develop strategic plans to manage high performer turnover. This study shows empirical
evidence that high performer turnover at the firm level is negatively associated with two
firm performance me asures, ROA and ROE. This finding points out that firms need to
develop effective high performer attraction and retention strategies. Otherwise, their
com petitors may hire aw ay valuable high performers, wh ich could result in significant
damage on firm performance.
Aside from the general im plications for firm performance of high performer
turnover, this study also provides p ractitioners w ith specific con ditions that they should
consider in developing effective strategic plans for managing h igh performer turnover.
This study indicates that the degree of the negative effect of high performer turnover on
firm performance may depend o n diverse organizational c ontexts. This study suggests
that for firms that invest m ore in ET D, high performer turnov er results in a less negative
122
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 129/147
impact on firm perform ance. Practitioners in these firms m ay decide to reduce investment
in ETD in order to m inimize the costs of high performer turnover. This strategy seems
plausible from a theoretical perspective focusing on the costs occurred by human capital
loss. How ever, this may not necessarily be an appropriate strategic choice because
reducing investment in ETD may result in degrading a firm's human capital pool of
employees w ho significantly influence firm performance. This study recomm ends that
the practitioners k eep in m ind that adjusting investment in ED T in order to mitigate the
negative impact of high performer turnover on firm performance could have both
negative and positive consequ ences. This study also suggests that practitioners need to
consider their firm repu tation in m anaging high performer turnover issues. In particular,
practitioners in firms w ith high reputations should pay m ore attention to retaining high
performers, w ho play significant roles in firms' value creation activities, because firms
with high reputation m ay suffer m ore from high performer turnover than firms with low
reputation.
In conclusion, I tested the relationship between high performer turnover and firm
performance, wh ich has been recognized as a critical issue in turnover research but has
not been emp irically tested. Furthermo re, I took into account the moderating roles of the
salient organ izational contextual v ariables in this relationship. This study unv eiled
complexities involved in the relationship between high performer turnover and firm
performance and prov ides practical implications for m anagers in developing strategic
plans for m anagin g high performer turnover. Future turnover research that takes into
account diverse organ izational con texts and the implication of the performance levels of
leavers will not only contribute to the accumulating bo dy of knowledge about the
123
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 130/147
complex relationship between employee turnover and firm performance but also provide
practitioners with useful guidelines regarding how em ployee turnover could be better
managed.
REFERENCES
Abelson, M. A., & Baysinger, B. D. 1984. Optimal and dysfunctional turnover: Toward
an organizational level model. Academy of Management R eview, 9: 331-341.
Adler, P. S. & Kwon, S-W. 2002. Social Capital: Prospects for a new concept. Academy
of Management Review, 27: 17-40.
Alexander, J. A., Bloo m, J. R., & Nu chols, B. A. 1994. Nursing turnover and hospital
efficiency: An orga nizational-level analysis. Industrial Relations, 3 3: 505-520.
Allen, D. G., Griffeth, R. W . 200 1. Test of mediated performance-turnover relationship
highlighting the moderating roles of visibility and reward contingency. Journal of
Applied Psychology, 86: 1014-1021.
Arther, J. B. 1994. Effects of human resource systems on manufacturing performance and
turnover. Academy of Management Journal, 37: 670-687.
Ashforth, B . E., M eal, F. 1989. Social identify theory and the organization. Academy of
Management Review, 14: 20-39.
Bae, J., & Law ler, J. J. 200 0. Organizational and H RM strategies in Korea: Impact on
firm performance in an emerging economy, Academy of Management
Journal,
43:
502-517.
Barnard, C. I. 1938. The functions of the executive. Cambridge, MA: Harvard University
Press.
124
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 131/147
Barney, J. B. 199 1. Firm resources and sustained com petitive advantage. Journal of
Management, 17: 99-12 0.
Barney, J. B. 20 01 . Is the resource based view a useful perspective for strategic
management research? Yes. Academy of Management Review, 26: 41-56.
Batt, R. 2002. Managing customer services: Human resource practices, quit rates, and
sales growth. Academy of Management Journal, 4 5: 587-597.
Becker. G. 1964. Human capital. New York: Columbia University Press.
Brown, C , &
Medoff,
J. 1978 . Trade unions in the production pro cess. Journal of
Political Economy, 86: 355-378.
Burt, R. S.1992. Structural holes. Cambridge, MA: Harvard University Press.
Cascio, W. F. 1995. Man aging Hum an R esources: Productivity, quality of work life,
profits. New York: McGraw-Hill.
Chang, S. 2003. Financial crisis and transformation o f Korean business groups.
Cam bridge, UK: C ambridge University Press.
Cohen, J., Cohen, P ., West, S. G., & Aiken, L. S. 2003. Applied m ultiple
regression/correlation analysis for the behavioral science
(3
rd
ed.). Mahw ah, NJ:
Erlbaum.
Coleman , J. S.1988. Social Cap ital in the creation of human capital. American Journal of
Sociology, 94: 95-120.
Collins, C. J., & Han, J. 20 04. Exploring applicant pool quantity and q uality: The effects
of early recruitmen t practice strategies, corporate advertising, and firm reputation.
Personnel Psychology,
57: 685-717.
Cross,
R., & Cummings, J. N. 2004. Tie and network correlates of individual
125
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 132/147
performance in knowledge-intensive work. Academy of Management Journal, 47:
928-937.
Dalton, D. R., & Todor, W . D. 1979 . Turnover turned over: An expanded and positive
perspective. Academy of Management Review, 4: 225-235.
Dalton, D. R., Todor, W. D. 1982. Turnover: A lucrative hard dollar phenomenon.
Academy of Management Review, 7: 212-218.
Dalton, D. R., Todor, W . D., & Krachhardt, D. M. 1982. Turnover overstated. The
functional taxonomy,
Academy of Management Review,
7: 117-123.
Datta, D. K., Guthrie, J. P., & Wright, P. M. 2005. Human resource management and
labor productivity: Does industry matter? Academy of Management
Journal,
48:
135-145.
Dess, G. G. & Shaw, J. D. 200 1. Voluntary turnover, social capital, and organizational
performance, 26: 446-456.
Dutton, J. E., Dukerich, J. M., & H arquail, C. V. 1994. Organizational Images and
Member Identification. Administrative Science Quarterly, 39: 239-263.
Farrell, S., & Hakstain, A. R. 2001. Improving salesforce performance: A meta-analytic
investigation of effectiveness and utility of personnel selection procedures and
training interventions. Psychology Marketing, 18: 281-3 16.
Flanagan, D . J., & O 'Shaughnessy, K. C. 2005 . The effect of layoffs on firm reputation.
Journal of Management, 31: 445-463.
Fombrum, C. 1996. Reputation: Realizing value from the corporate image. Boston, MA:
Harvard Business School Press.
Gardner, T. M. 20 05. Interfirm com petition for hum an resources: Evidence from the
126
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 133/147
software industry. Academy of Management Journal, 48: 237-256.
Gatewood, R. D., Gowan, M. A., & Lautenschlager, G. J. 1993. Corporate image,
recruitment image and initial job choice decisions. Academy of Management
Journal, 36: 414-427.
Gerhart, B., Wright, P. M., McMahan, G. C., & Snell, S. 2000. Measurement error in
research on human resources and firm performance: How much error is there and
how does it influence effect size estimates? Personnel Psychology, 53: 803-834.
Glebbeek, A. C , & Bax , E. H. 2004. Is high employee turnover really harmful? An
empirical test using company records. Academy of M anagement Journal, 38: 291-
313.
Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity:
Evidence from N ew Zealand. Academy of Management Journal, 44: 180-190.
Hollenbeck, J. R., & William s, C. R. 1986. Turnover functionality versus turnover
frequency: A note on work attitudes and organizational effectiveness. Journal of
Applied Psychology, 71:
606-611.
Huselid, M. A. 1995. The impact of human resource management practices on turnover,
productivity, and corporate financial performance. Academy of Management
Journal,
38: 635-672.
Huselid, M. A., Becker, B. E. 2000 . Com men t on Measu rement error in research on
human resources and firm performance: How much error is there and how does it
influence effect size estima tes? by Gerhart, Wright, MaM ahan, and Snell.
Personnel Psychology, 53: 835-854.
Jackofsky, E. F. 1984. Turnover and job performance: An integrated process model.
127
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 134/147
Academy of Management Review, 9: 74-83.
Johns, G. 2006. The essential impact of context on organizational behavior. Academy of
Management Review, 31: 386-408.
Leana, C. R., & Van B uren, H. J. 1999. Organizational capital and employm ent practices.
Academy of Management Review, 2 4: 538-555.
Lepak, D. P., & Snell, S. A. 1999. The human resou rce architecture: toward a theory of
human capital allocation and development.
Academy of Management
Review, 24:
31-48.
Napap iet, J., & Gh oshal, S. 1998. Social capital, intellectual capital, and the
organizational advantage. Academy of Management R eview, 23: 242-266.
March, J. G., & Simon, H. 1958. Organizations, New York, Wiley.
McE lory, J. C , & Morro w, P. C., & Rude , S. N. 200 1. Turnover and organizational
performance: A comparative analysis of voluntary, involuntary, and reduction-in-
workforce turnover. Journal of Applied
Psychology,
86: 1294-1299.
Mobley, W. H. 1982. Some unanswered questions in turnover and withdrawal research.
Academy of Management Review, 7: 111-116.
Muchinsky, P. M., & Turtle, M. L. 1979. Employee turnover: An empirical and
methodological assessment, Journal of Vocational Behavior, 14: 43-77.
Osterman, P . 1987. Turnover, em ploym ent security, and the performance of the firm. In
M. Kleiner (Ed.), Human resources and the performance of the firm: 275-317.
Madison, WI: Industrial Relations Research Association.
Pfeffer, J. 20 01 . Fighting the w ar for talent is hazardous to your organ ization's health.
Organizational Dynamics, 29: 248-259.
128
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 135/147
Pfeffer, J., & Salancik, G. R. 1978. The external control of organizations: A resource
dependence perspective.
New York: Harper & Row.
Price, J. L. 1 977. The study of turnover. Ames: Iowa State University Press.
Rindova, V . P., Williamson , I. O., Petkova, A. P., & Sever, J. M. 20 05. Being good or
being known: An empirical examination of the dimensions, antecedents, and
consequences of organizational reputation. Academy of Management Journal, 48:
1033-1049.
Rynes, S. L. 1991 . Recruitm ent, job choice, and post-hire consequ ences: a call for new
research directions. In M. D. Dunnette, & L. M. Hough (Eds.), Handbook of
industrial and organizational psychology, 2
n
edn (vol, 2. pp. 399-444). Palo Alto,
CA: Consulting Psychologists Press.
Salamin, A., & Horn, P. W. 200 5. In search of the elusive U-shaped Performance-
Turnov er Re lationship: Are high performing Swiss bankers more liable to quit?
Journal of Applied Psychology, 90:1204-1216.
Schwab, D. P. 1991. Contextual variables in employee performance-turnover
relationships. Academy of Management Journal, 34 : 966-975.
Shaw, J. D., Dineen, B . R., Fang, R., & Vellella, R. F. In press. Em ployee-Organization
exchange relationships, HRM practices, and quit Rate of good and poor
performers. Academy of Management
Journal.
Shaw, J. D., Duffy, M , K., Johnson , J. L., Lochhart, D. E. 2005a. T urnover, social capital
losses, and performance. Academy of Management Journal, 4 8: 594-606.
Shaw, J. D., & G upta, N . 2007 . Pay system characteristics and quit patterns of good,
average, and poor performers. Personnel Psychology, 60: 903-928.
129
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 136/147
Shaw, J. D., Gupta, N ., & De lery, J. E. 2005b. A lternative con ceptualizations of the
relationship between voluntary turnover and organizational performance.
Academy of Management Journal, 4 8: 50-68.
Siebert, W . S., & Zuban ov, N . In press. Searching for the optimal level of employee
turnover: A study of a large UK retail organization. Academy of Management
Journal.
Staw, B. M. 1980. The consequence of turnover. Journal of Occupational Behavior, 1:
253-273.
Steel, R. P., Griffeth, R. W ., & Horn, P. W . 2002 . Practical retention policy for the
practical manager. Academy of Management Executive, 16: 148-164.
Sturman, M. C , T revor, C. O., Boudreau, J. W., & Gerhart, B. 2003. Is it worth it to win
the talent war? Ev aluating the utility of performance-based pay. Personnel
Psychology. 56: 997-1035.
Strober, M. H. 1990. Human capital theory: Implications for HR managers. Industrial
Relations,
29: 214-239.
Tajfel, H . 1972. Social C ategorization, English m anuscript of La categorization sociale,
in S. Moscov ici (ed.) Introduction a la Psychologie Sociale, vol. 1. Paris:
Larousse. (see A bram s and hogg , 1990. Social Identity theory, p. 29)
Trevor, C O ., Gerhart, B., & Boudreau, J.W. 1997. Voluntary turnover and job
perform ance: Curvilinearity and the moderating influences of salary growth and
promotions.
Journal of Applied
Psychology, 82: 44-61.
Turban, D. B., & Cable, D. M. 2003. Firm reputation and applicant pool characteristics.
Journal of Organizational Behavior, 24: 733-751.
130
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 137/147
Vroom, V. 1964. Work and motivation. N ew York: W iley.
Wright, P. M., Gardn er, T. M ., Moynihan , L. M., & Allen, M. R. 2005 . The relationship
between HR practices and firm performance: Examining causal order. Personnel
Psychology, 58: 409-446.
Wright, P. M., McM ahan, G. C , & McW illiams, A. 1994. Hum an resources and
sustained competitive advantage: a resource-based perspective. International
Journal of Human Resource Managem ent, 5: 301-326.
Yanadori, Y., & Kato, T. 2007. Average employee tenure, voluntary turnover ratio, and
labour productivity: Ev idence from Japanese firms. International Journal of
Hum an Resource Managem ent, 18: 1841-1857.
131
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 138/147
T
A
4
1
C
e
a
o
a
D
p
v
1
O
g
z
o
S
z
2
O
g
z
o
A
3
C
a
n
e
y
4
U
o
P
e
5
C
y
o
O
g
n
6
H
g
P
o
m
T
n
7
N
H
g
P
o
m
T
n
8
n
m
n
E
m
o
T
a
n
n
&
D
o
m
9
F
m
R
a
o
1
R
u
n
o
A
R
1
R
u
n
o
E
y
R
M
e
6
5
2
2
1
0
0
4
0
0
0
0
0
0
3
6
2
1
7
8
1
8
:
S
a
s
c
S
D
1
4
1
7
1
1
0
4
0
2
0
1
0
0
0
8
1
3
1
6
3
9
1
.
3
.
3
.
3
-
1
-
2
-
3
.
3
.
5
-
0
.
0
2
.
3
.
2
-
1
-
1
-
0
.
0
.
1
-
1
-
1
3
.
2
-
2
-
2
-
2
.
1
.
2
-
0
-
0
4
.
0
-
1
-
3
-
1
.
1
-
0
.
0
5
.
0
.
0
.
0
.
1
-
0
.
0
6
.
3
-
1
-
1
-
2
-
4
7
-
1
-
1
-
1
-
1
8
.
2
.
0
.
0
9
-
0
.
0
1
.
7
BLES
n
=
1
e
o
R
n
*
p
<
0
0
w
a
e
e
*
p
<
0
0
w
a
e
e
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 139/147
TABLE 4.2: Results of Hierarchical Regression for High Performer Turnover and Firm
Performance
Variables
ROA ROE
Model 1 Model 2 Model 3 Model 4
Organization Size
Organization Age
Union Presence
Capital Intensity
Country of Origin
Industry
Construction
Finance
Non-manufacturing
Non-high Performer Turnover
-0.068 -0.109
-0.140t -0.153*
-0.017 -0.028
-0.019
-0.091
-0.038
-0.095
0 034
-0.041
• O . l l l t
0.054
- 0 . 0 8 5
0.025
- 0 . 1 3 3 f
0.034
- 0 . 1 2 1 f
0.018
0.042
• 0 . 1 2 5 t
0.134t
- 0 . 2 1 2
0.083
- 0 . 1 2 3 f
0.117|
- 0 . 1 5 0
0.080
- 0 . 0 2 6
0.189
- 0 . 1 4 5 f
0.155
0.030
0.159
- 0 . 0 3 5
High Performer Turnover
-0.269***
-0.493***
R
2
AR
2
FforAR
2
Overall F
1.105
. 9 3 5 t
158
0.165
0.060
1 0 . 4 8 8
2 . 9 0 2
158
0.074
1.281
155
0.275
0.201
3 9 . 8 8 5
5 . 4 5 0
155
Note. Standardized regression coefficients are shown. Standard errors are reported in the parentheses .The omitted
industry variable is manufacturing.
tp < 0.10, one-tailed test
*p < 0.05, one-tailed test
**p < 0.01, one-tailed test
***p < 0.001, one-tailed test
133
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 140/147
T
A
4
3
H
e
a
c
c
R
e
o
R
u
s
f
o
h
M
o
a
e
R
a
o
h
p
b
w
H
g
P
o
m
T
n
a
F
m
P
o
m
R
R
V
a
e
—
—
—
—
—
—
—
—
—
—
—
O
g
z
o
S
z
O
g
z
o
A
U
o
P
C
a
n
e
y
C
y
o
O
g
n
C
u
o
F
n
N
m
u
n
N
h
g
P
o
m
T
n
H
g
P
o
m
T
n
(
H
T
n
n
&
D
o
m
(
T
&
F
m
R
a
o
F
H
P
&
D
H
R
A
F
o
A
O
v
F
M
o
5
-
0
1
-
0
1
-
0
0
-
0
0
-
0
0
0
0
-
0
1
0
1
-
0
1
-
0
2
0
1
2
9
M
o
6
-
0
1
-
0
1
-
0
0
-
0
0
-
0
1
0
0
-
0
1
0
0
-
0
1
-
0
2
0
1
0
0
0
1
0
0
1
1
2
6
M
o
7
-
0
1
-
0
1
0
0
-
0
0
-
0
1
0
0
-
0
1
0
0
-
0
1
-
0
1
0
1
0
0
0
1
0
1
0
0
1
9
2
5
M
o
8
-
0
2
-
0
1
0
0
-
0
0
-
0
1
0
0
-
0
1
0
0
-
0
0
-
0
5
0
1
0
0
-
0
3
0
1
0
0
2
4
2
6
M
o
9
-
0
2
-
0
1
0
0
-
0
0
-
0
1
0
0
-
0
1
0
0
-
0
1
-
0
5
0
1
0
0
0
2
-
0
4
0
2
0
0
3
1
2
7
M
o
1
-
0
0
-
0
1
0
0
-
0
1
0
0
0
1
0
0
0
1
-
0
0
-
0
4
0
2
5
4
M
o
-
0
0
-
0
1
0
0
-
0
1
0
0
0
1
0
0
0
1
-
0
0
-
0
4
0
0
-
0
0
0
2
0
0
0
5
4
5
M
o
1
-
0
0
-
0
1
0
0
-
0
1
-
0
0
0
1
0
0
0
1
-
0
1
-
0
1
0
0
-
0
0
0
4
0
3
0
0
1
8
6
1
M
o
1
-
0
0
-
0
1
0
0
-
0
1
0
0
0
1
0
0
0
1
0
0
-
1
1
0
0
-
0
1
-
0
6
0
3
0
0
9
6
5
2
M
o
1
-
0
0
-
0
1
0
0
-
0
1
0
0
0
1
0
0
0
1
0
0
-
0
9
0
0
-
0
1
0
5
-
0
9
0
4
0
1
2
5
8
5
N
o
e
n
=
1
o
R
1
o
R
S
a
d
z
e
e
o
c
c
e
s
a
e
s
h
w
t
p
<
0
1
o
a
e
e
*
<
0
0
o
a
e
e
*
<
0
0
o
a
e
e
*
<
0
0
o
a
e
e
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 141/147
FIGURES
FIGURE 4.1: Research Framework
Training & D evelopment
• Hum an Capital Loss
• Higher Quality of Total
Human Capital Pool
High Performer
Turnover Rate
• Human Capital Loss
• Social Capital Loss
• Diverse Turnove r Costs
H2a H2b
Firm Reputation
• Quantity and Quality of
potential workers
• Replac ebil ity of leavers
H3a
T
1
H3b
'
H I
Firm Performance
• ROA
• ROE
135
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 142/147
FIGURE 4.2: Moderating Effect of in Firms' Investment on Employee Training and
Development on the Relationship between High Performer Turnover and Firm Performance
(a)
High
ROA
Low
Low 0 percent +1 s.d. High
Hiph Performer Turnover
•• Firm high on investment on employee training & development 1 s.d. above sample mean)
- Firm low on investment on employee training & development
(1 .
s.d. below sample mean)
(b)
High
ROE
Low
Low 0 percent +1 s.d. High
Hiph Performer Turnover
Firm high on investment on employee training & development 1 s.d. above sample mean)
Firm low on investment on employee training & development
(1 .
s.d. below sample mean)
136
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 143/147
FIGURE 4.3: M oderating Effect of Firm Rep utation on the Relationship between High
Performer Turnover and Firm Performance
(a)
High
ROA
Low
Low 0 percent +1 s.d.
Hiph Performer Turnover
High
Firm high on reputation (1 s.d. above sample mean)
Firm low on reputation (1 . s.d. below sample mean)
(b)
High
ROE
Low
Low 0 percent +1 s.d. High
Hiph Performer Turnover
Firm high on reputation (1 s.d. above sample mean )
Firm low on reputation (1 . s.d. below sample mean)
137
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 144/147
CHAPTER 5
CONCLUSION
Organ izations have striven to reduce the turnover of high performers, who play a
significant role in org anizational value creation activities. Scholars have also paid attention
to understanding th e diverse im plications of high performer turnover and retention on firm
performance. How ever, there hav e been few studies that systematically address this topic.
This lack of studies has no t only im pinged on the validity of mo st prior studies that did not
take into account the qu ality of employee turnov er and retention, but also widened the gap
between science and prac tices. This dissertation aimed to address this issue from diverse
theoretical perspectives at the both individual and organizational levels.
In order to und erstand th is issue, in chapter 2,1 investigated at the individual level
the impact of HCHRPs on organizational commitment of high performers and relatively
lower performers with co nsideration of both universal and conditional aspects of social
exchange theory. I found that HCHRPs increases employees' organizational commitment
and also is more likely to increase high performers' organizational comm itment than that of
lower levels of performers. In chapter 3,1 m oved up the level of analysis to the organization
on the basis of the findings and theoretical perspectives of chapter 2 .1 observed that HPWS
reduces high performer turnover, and the relationship between HPWS and firm performance
is med iated by high performer turnover. In chapter 4,1 questioned, how ever, why high
performer turnover negatively influence firm performance, which is a fundamental
assumption in both chapter 2 and chapter 3.1 theoretically addressed the negative
relationship between high performer turnover and firm performance, and found empirical
138
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 145/147
evidence that high performer turnover is indeed negatively associated w ith firm performance.
In addition, I observed that even the negative relationship betwee n high performer turnover
and firm performance can be moderated by the degree of a firm's investment on employee
training and development and the degree of firm reputation.
These three interrelated studies in my dissertation co ntribute to better u nderstanding
the complex relationships among human resource systems, high performer turnover, and
firm p erforman ce. H owev er, the reader should recognize the limitations in the three studies,
chapters 2, 3, and 4 , which w ere described in detail in each chapter. In conclusion, my
dissertation took into account the im plications of high performer retention and turnover in
the micro and m acro H RM literature, and provided theoretical and practical implications for
diverse audience . I hope m y dissertation sparks efforts to further understand the antecedents
and consequences of the retention and turnover of high performers, who create superior
values for organizations.
139
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 146/147
APPENDIX
(for chapter 3)
Appendix A: Items Measuring High Performance Work Systems
Extensive Selection (a=0.81)
1.
This organization selects peo ple according to highly refined selection criteria and procedures.
2. This organization hires peop le by utilizing different kinds of selection tools (ex. interviews, aptitude
test, written exam, etc.).
3. This organization spends much money in order to select right people.
Intensive Training and Development (a=0.93)
1.
This organization pro vides employees w ith a variety of training and de velopment opportunities.
2. This organization spends a lot of money on employee training and development.
3. This organization provides employee with structured formal training and development programs.
Rigorous Performance Appraisal (a=0.88)
1. This organization has an effective formal performance appraisal system to evaluate employees'
performance and competencies.
2. This organization appra ises employ ees' performance with objective and qua ntitative criteria
(ex. MBO, BSC , KPI, etc.).
3.
This organization utilizes the results of performance appraisal in deciding pay raises or promotions of the
employees.
4. This organization appraises employees' performance based on their objective achievement.
Pay for Performance Compensation (a=0.76)
1. This organization bases pay raise decisions on employee performance.
2.
This organization has wide range in pay within a same job grade.
3. This organization extensive ly utilizes a compan y-wide profit-sharing and/or a gain-sharing program.
4. This organization utilizes seniority-based rewards practices ®.
Flexible Job Design (a=0.68)
1. This organization gives em ployee a lot of job discretion.
2. This organization pro vides employees w ith opportunities to work flexibly (ex. flexible work sch edule).
3. This org anization flexibly assigns the scope and responsibilities of jobs , based on employees' skills and
needs.
Employee Participation (a=0.73)
1. This organization utilizes formal programs through which em ployees can participate in organizational
activities (ex. work-co uncil, em ployee suggestion, qua lity-circle, etc.)
2. This organization provides employees with opportunities to participate in decision-making and
problem-solving related to job and work environment.
Open Communication (a=0.76)
1.
This organization shares v arious information with employees (ex. business strategy and financial
status).
2.
This organization listens to em ployees' opinions through different kinds of formal or informal
programs (ex. attitude survey, grievance system)
140
7/25/2019 out formatting
http://slidepdf.com/reader/full/out-formatting 147/147
AUTHOR S BIOGRAPHY
Ki-Wook Kwon earned his Bachelor of Science degree in Business Administration
from Hanyang Univeristy, Seoul, Korea and his Master of Arts in Human Resource and
Industrial Relations from University of M innesota, Twin C ities. His work has been accepted
for publication in
Management International Review.
He currently h as a manuscript under
2
n
review in Human Resource Managem ent. He has wide ranging work experience
including co-founding a business, researching and consulting at Samsung Econo mic
Research Institute, and serving as a staff mem ber in the human resource department of a
manufacturing firm. After the co mpletion of his Ph.D., he will teach at the University of
Illinois at Urbana-C ham paign as a Visiting Assistant Professor in the School of Business
and the School of Labor and Employment Relations.