Fombrun tichy and devanna 1984

O Academy of Management Review 1996, Vol. 21, No, 4, 959-385.

TOWARD AN INTEGRATIVE MODEL OF STRATEGIC INTERNATIONAL HUMAN

RESOURCE MANAGEMENT

SULLY TAYLOR Portland State University

SCHON BEECHLER Columbia University

NANCY NAPIER Boise State University

This article builds on previous work in intemational human resource management by drawing on concepts from the resource-based view of the firm and resource dependence to develop a theoretical model of the determinants of strategic intemational human resource manage- ment (SIHRM) systems in multinational corporations. The article then offers propositions concerning the relationships between a number of key determinants and the multinational corp>oration's overall SIHRM approach, the design of a particular affiliate's HRM system, and the HRM system for critical groups of employees within the affiliate.

As multinational corporations (MNCs) and their overseas affiliates have become increasingly important players in the global economy, inter- est in and research of the strategies and management practices of these firms have also grown (e.g., Hamel & Prahalad, 1985; Ohmae, 1990; Porter. 1986; Prahalad & Doz. 1987; Rosenzweig & Singh. 1991). Although research- ers have explored numerous issues internal and external to the MNC, there is a growing consensus that a key differentiator between the corpo- rate winners and losers in the 21st century will be the effectiveness of the human organization (Bartlett & Ghoshal, 1994, 1995; Pfeffer, 1994; Pucik, 1992; Tichy, Brimm. Charan, & Takeuchi, 1992; Ulrich & Lake, 1990).

In an effort to effectively leverage human resources to implement the intended strategies of organizations, researchers and practitioners alike have begun to explore the theoretical and empirical linkages between human resource management (HRM) and strategy both in the domes- tic (e.g., Butler, Ferris, & Napier, 1991; Fombrun, Tichy, & Devanna, 1984; Lengnick-Hall & Lengnick-Hall, 1988; Schuler & Jackson, 1987; Wright &

We are very grateful to those who commented on earlier versions of this article: Talya Bauer, Rot)ert Eder, Mauro Guillen, Stan Slater, and Pam Tiemey. We also acknowledge the insightful and constructive comments of the three anonymous reviewers.

959

960 Academy ot Management Review October

McMahan. 1992) and international contexts (e.g.. Adler & Ghadar. 1990; Kobrin. 1992; Milliman, von Glinow. & Nathan, 1991; Schuler, Dowling, & De Cieri, 1993). In this article, we draw on a number of recent models of strategic international human resource management (SIHRM). which have greatly furthered our understanding of the factors that influence the design of an MNCs SIHRM system. Work in the strategy and organizational theory fields also has generated a number of theoretical perspectives that provide valuable frameworks for understanding SIHRM. In this article, we draw on two such frameworks, the resource-based theory of the firm (e.g.. Barney, 1991; Conner. 1991; Penrose, 1959; Wernerfelt. 1984) and resource depen- dence (e.g.. Pfeifer & Salancik, 1978) and apply these frameworks to exam- ine SIHRM at three distinct but interrelated levels of analysis: the MNC as a whole, the affiliate, and individual groups of employees within the affiliate. Resource-based theory adds to prior models of SIHRM the funda- mental notion that in order to provide value to the business, the SIHRM system of global firms should be constructed around specific organiza- tional competencies that are critical for securing competitive advantage in a global environment (Pucik, 1992). The resource dependence framework helps identify those situations in which MNCs will exercise control over the SIHRM system of their affiliates.

This article begins with a definition of terms and a brief review of the contributions of prior models of SIHRM to the field. We then describe a model of SIHRM that elaborates a system that examines both the determi- nants and the evolution of SIHRM systems over time. After laying out the model, we suggest a number of propositions to guide future research. Finally, we conclude with suggestions for further development, both theo- retical and empirical, to expand our understanding in this important area.

DEFINITION OF TERMS

Before proceeding with an overview of current theories of SIHRM, it is important to distinguish between international human resource man- agement (IHRM) and strategic international human resource management (SIHRM). In keeping with Schuler and colleagues' (1993) and Lado and Wilson's (1994) work, we define the MNCs IHRM system as the set of distinct activities, functions, and processes that are directed at attracting, developing, and maintaining an MNCs human resources. It is thus the aggregate of the various HRM systems used to manage people in the MNC, both at home and overseas. By including headquarters (HQs) in this definition, we recognize that the parent company can become simply another one of the units of the MNC (cf. Hedlund, 1986).

Strategic Human Resource Management (SHRM), in contrast, is used to explicitly link HRM with the strategic management processes of the organization and to emphasize coordination or congruence among the various human resource management practices (Schuler & Jackson, 1987; Wright & McMahan, 1992). Thus, SIHRM is used to explicitly link IHRM

1996 Taylor, Beechler, and Napier 961

with the strategy of the MNC. Following Schuler and colleagues' work. (1993: 720). we define SIHRM as "human resource management issues, functions, and policies and practices that result from the strategic activi- ties of multinational enterprises and that impact the international con- cerns and goals of those enterprises." With these definitions in mind, we now provide a brief overview of current theories of SIHRM.

OVERVIEW OF CURRENT THEORIES OF SIHRM

In recent models ot SIHRM (Adler & Bartholomew. 1992; Adler 8E Ghadar. 1990; Evans & Lorange, 1989; Kobrin. 1992; Milliman. Von Glinow, & Nathan. 1991; Schuler et al.. 1993), authors adopted a strategic, macro perspective and focused on the SIHRM system as a way for MNCs to effectively manage and control their overseas operations. Although still in its infancy (Laurent, 1986), the recent work in SIHRM has been built on antecedents that are decades old (e.g., Edstrom & Galbraith. 1977; Franko, 1976; Heenan & Perlmutter. 1979; Perlmutter. 1969). Models found in the current literature have been used explicitly to tie the SIHRM system of the MNC to either its stage of internationalization (e.g., Adler & Ghadar, 1990; Milliman et al., 1991) or to its international strategy (e.g., Kobrin, 1992). In most of these models, researchers adopted a contingency perspec- tive and focused primarily on the fit of the SIHRM system with the goals of the firm.

In this literature (e.g.. Milliman et al., 1991), fit is defined as "the degree to which the needs, goals, objectives, and/or structure of one component are consistent with the needs, demands, goals, objectives, and/or structure of another component" (Nadler & Tushman. 1980: 40). As an example, Adler and Ghadar's (1990) model of SIHRM is based on the central idea that as an MNC moves through the product life cycle, it will face different pressures to respond to external and internal cultural diversity. As the authors stated, "The central issue for MNCs is not to identify the best international HRM policy per se, but rather to find the best fit between the firm's external environment, its overall strategy, and its HRM policy and implementation" (Adler & Ghadar. 1990: 245).

Some researchers in SIHRM recently addressed the need for flexibility as well as fit (e.g., Milliman et al., 1991; Schuler et al.. 1993). According to Milliman and colleagues, flexibility is defined as "the capacity of HRM to facilitate the organization's ability to adapt effectively and in a timely manner to changing or diverse demands from either its environment or from within the firm itself" (1991: 325). These authors argued that as the firm becomes increasingly internationalized, the need for flexibility will increase, and the most flexibility will be required in the most advanced stage of internationalization.

Schuler and colleagues' (1993) model is the most recent of the SIHRM models, and it captures all the important factors identified by the previous writers. Rather than tying SIHRM directly to either product/organizational

962 Academy of Management Beview October

life cycle or to the MNCs strategy, these authors build on the work of Prahalad and other strategy writers (Bartlett & Ghoshal, 1989; Prahalad, 1975; Prahalad & Doz, 1987; Prahalad & Hamel, 1990). Schuler and his colleagues postulated that the fundamental issue users of SIHRM must address is the tension between the need for interunit linkages (integration) and the challenges faced by each affiliate in order to operate effectively in its local environment (differentiation). The dual needs for integration and differentiation, orginally identified by Lawrence and Lorsch (1967), are parallel to the concepts of internal and external fit (Milliman et al.. 1991).

In addition to the strategic components. Schuler and colleagues (1993) specified that both exogenous and endogenous factors influence the de- sign of the SIHRM system of the firm. The exogenous factors these authors identified include industry characteristics and country/regional character- istics, whereas the endogenous factors they described include the struc- ture of international operations. HQs' international orientation, competi- tive strategy, and experience in managing international operations.

All of the models described previously have helped to advance re- searchers' theoretical understanding of the linkages between human re- sources and strategy in MNCs in a number of ways. First, they have shown how and why the stage of internationalization or international strategy of the MNC is tied to its approach to SIHRM. a linkage that has received some recent empirical support (Kobrin, 1992; Rosenzweig & Nohria. 1994). Second, they have helped to identify critical variables, in addition to strategy, that are important in determining the SIHRM of a firm, such as industry (Kobrin, 1992; Schuler et al., 1993), international experience of the MNC (Adler & Ghadar, 1990; Milliman et al., 1991; Schuler et al., 1993), organizational structure (Schuler et al., 1993), headquarters' (HQ) interna- tional orientation (Hedlund, 1986; Schuler et al., 1993), and the host country's cultural and legal environments (Adler & Ghadar, 1990; Milliman et al.. 1991; Schuler et al.. 1993). Third, these models have pointed to the need for the SIHRM system to address the tension between the dual imperatives of local responsiveness and global integration. Finally, as both Kobrin (1992) and Schuler and colleagues (1993) pointed out, the ability to design the SIHRM system such that it optimally balances these different forces will have performance implications for both the MNC and the individ- ual affiliate.

The model of SIHRM presented next is both an extension of and an addition to previous work. Although incorporating many aspects of SIHRM proposed in previous models, including a contingency approach, in the present model, we also draw on resource-based theory and resource de- pendence. These two frameworks help to identify four critical considera- tions that were not developed in previous SIHRM models: using the MNCs home HRM system as a resource for building its global SIHRM system; using SIHRM differentiation among types of affiliates; using SIHRM differ- entiation among types of employees; and being more explicit in how

1996 Taylor. Beechler. and Napier 963

SIHRM systems evolve over time. The next section describes our model of SIHRM, which is used to address these critical issues.

THE SIHRM MODEL

How a firm's systems, among its other attributes, enable it to achieve success relative to competitors is the central idea of the resource-based theory of the firm (Barney, 1991; Penrose, 1959; Wernerfelt, 1984). To generate a competitive advantage, the resource must be valuable, rare, imperfectly mobile, and inimitable (Barney, 1991; Conner. 1991). Lado and Wilson (1994: 699) described application of this theory to HRM in the following way: "The resource-based view suggests that human resource systems can contribute to sustained competitive advantage through facilitating the development of competencies that are firm specific, produce complex so- cial relationships, are embedded in a firm's history and culture, and gener- ate tacit organizational knowledge." The HRM function is unique because both the outputs (the employee behaviors) of the system and the system itself are potential sources of competence. Following this logic, we define HRM competence as the tangible (e.g.. HR planning systems, international sales training programs, selection tests) and intangible resources (e.g., shared mindset, ability to attract qualified employees to work for the firm, affiliate managers' experience in negotiating with the local government) that allow a firm to outperform its competitors.

We can distinguish between resources, including HRM competence, at three levels within the MNC First, there are the parent company's resources that originate from a particular configuration of economic, cul- tural, human, and other resources in a given country (Porter, 1990); these resources are not differentiated between firms domestically, but they can give MNCs an advantage when competing outside their home countries. For example, because of governmental tax incentives. German firms have highly developed employee training and development programs (Pfeffer, 1994) that result in skilled employees who can give the firm an edge over competitors from other countries.

The second source of potential competitive advantage for MNCs is at the parent company level and represents the unique bundle of assets and capabilities that the MNC has developed over its lifetime. Firms are idiosyncratic because throughout their unique histories they accumulate different tangible and intangible assets (Nelson & Winter, 1982). As an example. 3M's excellence in managing innovators represents a bundle of assets and capabilities at the parent company level. The combination of these two levels of resources (national and firm) is parallel to Bartlett and Ghoshal's (1989) notion of admin is (ra (ive heritage.

Finally, in addition to national- and firm-level resources, resources at the affiliate level may provide a source of competitive advantage for the MNC at the local, regional, or global levels. For example, a Japanese firm's Singapore affiliate may develop efficient HRM selection policies to

964 Academy of Management Review October

deal with the high labor market mobility there. If the policies are not useful for other affiliates (e.g.. those in the United States or ior the parent company), then this resource is useful only at the one affiliate. However, if the MNC can successfully transfer these policies to other environments that have high mobility in their labor markets, they may develop into parent company-level resources.

The origin of the resource—national, firm, or affiliate—is likely to affect its usefulness in other locations (see also Laurent. 1986; Schneider. 1988; Schuler et al., 1993). The usefulness of a particular element of an HRM competence, like any other resource, may be confined to its place of origin (it is context specific) or it may be effective across countries (it is context generalizable). For example, a U.S. MNC with a matrix structure that produces high-quality decisions may not be able to use this resource in its affiliate in France, because the French prefer a clear and fairly stringent hierarchical decision-making system (cf. Hofstede, 1980; Laurent. 1983). Thus, resources, including HRM competence, can be context specific or context generalizable. depending on their usefulness outside the loca- tion in which they were developed. This distinction is important in the choice of SIHRM orientation, as we will describe next.

The Parent Company's SIHRM System

Our model is used explicitly to examine the SIHRM system at three levels: the parent company, the affiliate, and specific employee groups within the affiliate (see Figure 1). The resource-based theory of the firm enables us to view the determinants of the SIHRM system at each of these levels and the evolutionary process of change over time from a new perspective. We begin at the organization level by asking: What determines the SIHRM system oi an MNC? In answering this question, we present the first part oi our model, which is focused on the parent company, and we discuss the determinants oi the SIHRM system at that level.

Prior work on SIHRM (e.g., Rosenzweig & Nohria. 1994; Schuler et al., 1993; Yuen & Kee, 1993) suggests that there are two key aspects of the SIHRM system at the parent company level: the parent company's SIHRM orientation and the parent company's SIHRM functional focus. Because the SIHRM functional focus is a minor component of the model we present, it will be discussed only briefly before we examine the SIHRM orientation.

An MNCs SIHRM functional focus is defined as those functional as- pects (e.g., selection techniques, compensation schemes) oi the parent company's home HRM system that the firm transfers to its affiliates or focuses attention on developing within them. SIHRM functional focus is most strongly influenced by the parent company's home HRM system. The parent firm will have developed its greatest resources in the HRM areas that top management believes are critical to the successful completion of organizational tasks, and hence these HRM functions will be the focus of their transfer to affiliates. We do not develop propositions concerning these relationships because, although the links can be surmised from

1996 BeechJer, and Napiez 965

d) d) >,

a

Ul

a.3

o Cr i

_d

3t S

im i

0

S) d;

e

M S

ys t

cd

"o

IC U

.

r.aa. d

'> •6 jn

• >

OT o

oy e

i E

m pl

0

a £

01

II

1 '

cfil

Pa r

i ' .

R

n

in

A fl

il

in III d

' M

<

n m(D

•3 3

O

u s n

D is

t

uB d U)

G

o

t^

ID

fi li

<

»

d

<

eg i

ra t

CO

d>

fi .c 01

X I d Ul

U

d)

CC

a Ot -

jn

"c di

E