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



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



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



To my parents,

to my w ife,

an d to my son and daughter

i i i



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



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


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


References 124

Tables 132

Figures 135

CHAPTER 5: CONCLU SION 138

APPENDIX 140

AUTH OR'S BIOGRAPHY 141

v



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



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



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



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



human capital allocation and development. Academy of Management Review, 24:

31-48.

Mobley, W. H. 1982. Some unanswered questions in turnover and withdrawal behavior.


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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



= 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



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



< 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



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



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



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



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



'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



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



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



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



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




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,


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




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


Eisenberger, R., Hun tington, R., Hutchison, S., & Sowa, D. 1986. Perceived

organizational support.

Journal of Applied Psychology,

71 : 500-507.

36



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



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


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



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.



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


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


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



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.


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



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.


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


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



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.


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


42



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?


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.


Wright, P. M., McM ahan, G. C , & McW illiams, A. 1994. Hum an resources and


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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



2004), which has helped firms respond to high performer turnover as well as total

workforce turnover.


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

e-mail

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



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



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



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



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



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



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



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




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



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



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



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



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.


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



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



An empirical investigation.

International Journal of Human Resource


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.


Barney, J. B. 19 91. Firm resou rces and sustained competitive advantage. Journal of


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



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,


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


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



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



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


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



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.


Evidence from New Zealand. Academy of Management Journal, 44: 180-190.

Hollenbeck, J. R., & W illiams, C. R. 1986. Turnover functionality versus turnover


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,


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.


84



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


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:


Mobley, W. H. 1982. Some unanswered questions in turnover and withdrawal behavior.


85



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.


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.



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


Schaubroeck, J., Merritt, D. E. 1997. Divergent effects of job control on coping with

86



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



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.


253-273.

87



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.


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



performan ce: Curvilinearity and the moderating influences of salary growth and

promotions.

Journal of Applied


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


Wood, R., & Bandura, A. 1989. Social cognitive theory of organizational management.


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



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



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



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



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 <

0.01,

one-tailed test

93



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,



**p < 0.0 1, one-tailed test

94



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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



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


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



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



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



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



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



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



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



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.


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



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



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



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



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



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.


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



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



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


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



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,


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



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.


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:


Gardner, T. M. 20 05. Interfirm com petition for hum an resources: Evidence from the

126



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.


Evidence from N ew Zealand. Academy of Management Journal, 44: 180-190.

Hollenbeck, J. R., & William s, C. R. 1986. Turnover functionality versus turnover


Applied Psychology, 71:

606-611.

Huselid, M. A. 1995. The impact of human resource management practices on turnover,


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.


127




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.


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.


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



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



Shaw, J. D., Gupta, N ., & De lery, J. E. 2005b. A lternative con ceptualizations of the


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.


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


Turban, D. B., & Cable, D. M. 2003. Firm reputation and applicant pool characteristics.

Journal of Organizational Behavior, 24: 733-751.

130



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


Wright, P. M., McM ahan, G. C , & McW illiams, A. 1994. Hum an resources and


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



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



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.



**p < 0.01, one-tailed test

***p < 0.001, one-tailed test

133



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



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



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



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



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.


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



Firm high on reputation (1 s.d. above sample mean )

Firm low on reputation (1 . s.d. below sample mean)

137



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



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



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



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.

out formatting

Documents

Transcript of out formatting