Conditions influencing headquarters and foreign subsidiary ... - SSRN

Journal of International Business Studies (2003) 34, 567–585

& 2003 Palgrave Macmillan Ltd. All rights reserved 0047-2506 $25.00 www.jibs.net

Conditions influencing headquarters and foreign subsidiary roles in marketing activities and their effects on performance K Hewett1, MS Roth2 and K Roth2 1

Management and Marketing Department, College of Business Administration, Winthrop University, Rock Hill, USA; 2Moore School of Business, University of South Carolina, Columbia, USA Correspondence: Dr K Hewett, Management and Marketing Department, College of Business Administration, Winthrop University, 324 Thurmond Hall, Rock Hill, SC 29733, USA. Tel: þ 1 803 323 4817; Fax: þ 1 803 323 3960; E-mail: [email protected]

Received: 16 April 2002 Revised: 12 August 2003 Accepted: 26 August 2003 Online publication date: 30 October 2003

Abstract In this study, we examine the extent to which foreign subsidiaries are responsible for the development and implementation of marketing activities, as compared with these activities being controlled by headquarters. We propose and test a model of conditions that affect such headquarters and subsidiary roles, and explore the extent to which the alignment of these roles with certain conditions is associated with product performance. Our findings suggest that the more closely headquarters and subsidiary roles in marketing activities are aligned with relational, industry, and market conditions, the greater market share tends to be. Journal of International Business Studies (2003) 34, 567–585. doi:10.1057/palgrave. jibs.8400054 Keywords: standardization; marketing; strategy

Introduction An area of increased attention in the international business literature is the role of subsidiaries in multinational firms (Birkinshaw and Morrison, 1995; Birkinshaw and Hood, 1998). Subsidiaries can take on various roles to develop and leverage capabilities that are a source of value creation in the subsidiary’s market and possibly in other markets served by the firm (Bartlett and Ghoshal, 1991). A fundamental challenge for multinationals is to determine the conditions under which foreign subsidiaries may assume these value-creating roles. Put another way, under what conditions should specific activities and capabilities be the responsibility of subsidiaries rather than headquarters? We argue in this study that both external and internal conditions are determinants of foreign subsidiary roles. External conditions pertain to the local environment in which the subsidiary operates. Internal conditions include organizational attributes that characterize the relationship between the subsidiary and headquarters. An important consideration in examining subsidiary roles is defining the unit of analysis. As Frost et al. (2002) discuss, the subsidiary level can be too aggregate a unit of analysis, because of the increased complexity and sophistication of multinationals’ value chains and the various ways in which a subsidiary can contribute to them. Rather, particular functional specializations within subsidiaries should be studied (Kim et al., 2003). In so doing, specific subsidiary activities in areas such as marketing, operations and R&D should be examined in terms of the capabilities they offer

Conditions influencing headquarters and subsidiary roles

K Hewett et al

568

the multinational firm, and the conditions under which such capabilities can best be leveraged for competitive advantage. Thus our interest is in examining how performance might benefit from aligning subsidiaries’ roles with the best conditions for those roles. In this paper we focus on subsidiaries’ vs headquarters’ roles in marketing activities. Marketing activities are the processes of developing promotional campaigns, making pricing decisions, engaging in distribution decisions, and so forth (Zou and Cavusgil, 2002). In particular, we examine the extent to which foreign subsidiaries are responsible for the development and implementation of marketing activities, vs these activities being controlled by headquarters – that is, the use of locally customized vs centrally standardized processes for managing marketing activities. The coordination of marketing activities within the multinational is an important consideration in designing a global marketing strategy (Zou and Cavusgil, 2002), but little is known about the conditions related to headquarters’ vs subsidiaries’ roles in managing these activities. Furthermore, the extent to which conditions, internal and external to the subsidiary, affect the relationship between these roles in marketing activities and product performance has not been explored. Our goal is to fill these gaps in understanding headquarters’ and subsidiaries’ roles in marketing activities as well as the potential performance implications of successfully aligning roles with the internal and external conditions. The remainder of this paper is organized as follows. First, we develop a model of conditions associated with headquarters’ and subsidiaries’ roles in marketing activities. Next, we report the results of a study on relational, industry, and market conditions, and the extent to which they are related to headquarters and subsidiary roles in managing marketing activities. In addition, we propose and test ideal profile models of success when the subsidiary is responsible for marketing activities, and when headquarters controls activities. Fourth, and finally, we discuss major findings and their theoretical and managerial implications.

organization (IO) and institutional theories to specify conditions that may be determinants of these roles. IO and institutional theories are complementary in the sense that they both expect, in response to environmental pressures, organizations to adopt similar practices over time. From an IO perspective, the adoption process is motivated by technical efficiency: firms adopt ‘best’ practices because they are thought to promote the economic goals of the firm. Technically efficient practices are identified as those that allow the firm to become aligned with or fit its external market and industry conditions, thereby resulting in better performance (Venkatraman and Prescott, 1990). Thus the external market and industry context is central to the determination of desired firm behaviors. In contrast, adoption processes in institutional theory emphasize the need to conform to pressures within the firm’s environment and thereby gain legitimacy in relation to that environment (Tolbert and Zucker, 1983; Scott, 2001). For example, regulatory institutions mandate behaviors and manipulate sanctions so as to influence the adoption of specific practices. Similarly, social values, norms, belief systems, cultural frames, etc. establish conventional or preferred ways of doing things. Social actors may internalize practices or they may adopt practices because other social actors with high status embrace them. In either case, firms that conform to mandated and socially preferred ways of doing things will be viewed as legitimate. Legitimacy provides the firm with access to societal resources and thereby ensures its long-term viability (Zucker, 1987). Although IO and institutional theories are consistent in focusing on pressures in the environment as a basis for understanding firm behaviors, Gooderham et al. (1999) assert that introducing the crosscultural context results in divergent predictions from the two perspectives. They argue that

A model of conditions influencing headquarters’ and foreign subsidiaries’ roles in marketing activities

Their study found more support for the institutional explanation, compared with the rational model, as practices were influenced largely by national environments. However, these findings were based on comparing local firms from different countries. Considering the multinational corporation (MNC), as we do in this study, introduces

Theoretical foundations In developing our model of headquarters vs subsidiary marketing roles, we draw on industrial

Journal of International Business Studies

while the rational model assumes that organizational practices are universal across national borders, institutionalism is sensitive to the possibility of cross-national institutional differences, which in turn generate significant cross-national differences in managerial systems (p. 507).


K Hewett et al

569

additional complexity as these firms span multiple countries. Within the MNC, foreign subsidiaries operate in a host country environment and, at the same time, they are part of an ‘MNC environment.’ A foreign subsidiary must maintain legitimacy with its local environment and headquarters as well as adopt practices viewed within the firm as being technically efficient. As described by Westney (1993): The MNC organization is the source of strong isomorphic pulls towards similarity across the organizational structures and processes of subsidiariesy and yet each subsidiary is also operating within a local organizational field that exerts a range of isomorphic pulls on its organization (p. 67).

Thus, beyond the market and industry environment, we posit that it is important to consider the relational context between headquarters and foreign subsidiaries, as this context will influence the way in which pressures from headquarters are interpreted and perceived by a foreign subsidiary (Kostova and Roth, 2002). The successful development of activities and capabilities in subsidiaries is thereby influenced through the nature of their relationships with headquarters, i.e., an internal relational condition. In summary, in this study, we bring together IO and institutional theories in an effort to understand the pressures, external and internal, that influence the adoption of specific subsidiary marketing roles. We posit that the role of the subsidiary in marketing activities (e.g., the degree to which marketing strategies are developed and implemented using standardized vs customized processes) will be influenced by: (1) internal relational conditions; (2) external industry conditions; and (3) external market conditions. To develop specific attributes relevant to the conditions within a marketing role, we build on the marketing strategy and global marketing literatures. Consistent with Quelch and Hoff (1986) and Zou and Cavusgil (2002), we examine the subsidiaries’ vs headquarters’ roles in key marketing activities, including the development and implementation of branding, positioning, advertising, promotions, pricing, channels of distribution, and customer service strategies. Specifically we examine the extent to which these activities are driven by the subsidiary (locally customized processes) or mandated by headquarters (centrally standardized processes). We then study the extent to which the subsidiary or headquarters playing the central role

in these marketing activities under certain relational, industry, and market conditions enhances product performance in the subsidiary’s market.

Relational conditions Subsidiaries will experience pressure to adopt practices handed down from headquarters. We argue, based on institutional theory, that the relational context between a subsidiary and headquarters will influence the way in which such pressures are perceived by the subsidiary (Kostova and Roth, 2002). In their framework for the adoption of organizational practices by subsidiaries, Kostova and Roth view the key issues of importance as the subsidiary’s trust in the headquarters, such that it results in conformity to the headquarters’ practices, the perceived level of dependence of the subsidiary on the headquarters, and the extent to which the subsidiary feels as if it is a part of the parent organization. In this study, we similarly view this relational context as key in determining the subsidiary’s role in marketing activities vs that of the headquarters. Guided by Kostova and Roth (2002) and international marketing literature, we consider three aspects of the headquarters–subsidiary relationship: (1) the level of cooperation between subsidiary marketing managers and the marketing operation at headquarters, (2) the extent to which the subsidiary perceives itself as being dependent on the headquarters, and (3) the degree to which the subsidiary is involved in setting the product-level marketing objectives for its market. Cooperation Cooperation is defined as the subsidiary marketing operation’s perceptions of the communications, interactions, and relationship with its headquarters marketing operation (see Anderson and Narus, 1990, 45). Given the subsidiary’s need to balance the pulls of both the local market and the MNC organization, it may at times act autonomously and at times be motivated to follow directives from headquarters. From an institutional theory perspective, cooperation between the subsidiary and parent organization will reduce the uncertainty regarding the decisions being handed down, and will make adoption of practices from headquarters more likely (Kostova and Roth, 2002). Similarly, Jain (1989) proposes that conflict, or a poor relationship, between marketing functions at an MNC’s headquarters and its subsidiaries may discourage the transfer of global marketing programs to foreign markets. Under conditions in which the



K Hewett et al

570

subsidiary and headquarters have a positive relationship, open communications, and regular interactions, we expect that the objectives of the planning system will be better understood, and adopted with less resistance by the subsidiary. Relatedly, Rau and Preble (1987) propose that the level of communication between the headquarters and subsidiary is an important determinant of the degree of standardization of marketing practices, which represents greater headquarters as opposed to subsidiary roles. Cooperation, although still important, is less critical when the subsidiary plays a greater role in managing marketing activities. Thus we propose that cooperation between the two marketing operations will be greater when marketing activities are controlled by headquarters and less when foreign subsidiaries take on this role. H1: The level of subsidiary–headquarters cooperation will be more positively related to the headquarters’ role than the foreign subsidiary’s role in marketing activities.

Vertical dependence Vertical dependence is defined as the extent to which the subsidiary relies on headquarters to fulfill important needs (Rusbult and Van Lange, 1996; Kostova and Roth, 2002). Drawing on institutional theory, dependence can be seen as influencing compliance with institutional pressures (i.e., pressures from the parent organization; Rosenzweig and Singh, 1991). The level of perceived dependence of one party on another is an important feature of dyadic relationships (Anderson and Narus, 1990; Gundlach and Cadotte, 1994; Smith et al., 1997), and is related to the degree of control that one party has over another (Molm, 1994). According to DiMaggio and Powell (1991), increased isomorphism results from power or influence over other actors in organizational relationships. Thus pressures to adopt practices handed down by the headquarters to subsidiaries will be greater when the subsidiary perceives itself as dependent on the headquarters. Kostova and Roth (2002) argue that, when dependence is high, subsidiaries will tend to comply with mandates from the parent organization. Relatedly, support for the importance of dependence on the standardization–adaptation issue can be found in a study by Picard et al. (1998), who found that customization is positively associated with low levels of headquarters control over subsidiaries. Similarly, Kirpalani et al. (1988) found a significant relationship between headquarters control over the subsidiary and the


likelihood of standardization in advertising, a result confirmed in a later study by Laroche et al. (2001). When the subsidiary perceives itself to be dependent on the headquarters, pressures from the headquarters will also be perceived to be greater, and thus we expect the headquarters to be more successful in handing down standardized marketing processes, and so the subsidiary’s role in marketing activities will be less. H2: Vertical dependence will be more positively related to the headquarters’ role than the foreign subsidiary’s role in marketing activities.

Participation in goal setting The headquarters–subsidiary relationship is also reflected by the extent to which subsidiaries participate in setting goals for a particular product. Participation in goal setting is defined as the extent to which the subsidiary is involved in setting goals for the product in its particular market. From an institutional theory perspective, when a subsidiary is involved in goal setting for a product, it is more likely to view the marketing decisions concerning that product as reflecting its own beliefs as opposed to those of the parent organization. Thus the subsidiary is likely to perceive itself as having a greater role when it enjoys such participation. Further, the subsidiary is likely to feel pressure to impose beliefs consistent with the local institutional context (Westney, 1993), and participation in goal setting for the product is one way in which the subsidiary can impose such beliefs. Additionally, when decision-making occurs primarily at the headquarters, this implies a more hierarchical structure, which suggests less of a role for subsidiaries. Relatedly, Pappavassilou and Stathakopolous (1997) argue that standardization of advertising strategy is more likely when decision-making occurs primarily at headquarters. Picard et al. (1998) similarly found that a higher level of product policy standardization was reported by firms in which headquarters was primarily responsible for decisions regarding strategy implementation. Thus, when subsidiary involvement in setting goals for the products is high, the subsidiary is also more likely to be directly involved in developing and implementing the marketing processes for those products, or to take on a greater role in those activities. H3: Participation in goal setting for the product will be more positively related to the foreign subsidiary’s role than the headquarters’ role in marketing activities.


K Hewett et al

571

Industry conditions Consistent with both IO and institutional theories, the external context of an organization will be important in defining subsidiaries’ roles in marketing activities. Three characteristics reflecting industry conditions within which the subsidiary operates are considered. Two (market turbulence and market concentration) represent characteristics of the industry in the subsidiary’s particular market, and the third (technological turbulence) reflects the industry as a whole. Market turbulence Market turbulence is defined as the rate of change in the composition of customers and their preferences (Jaworski and Kohli, 1993; Slater and Narver, 1994; Moorman and Miner, 1997). This variable captures the changing needs and product preferences of customers in individual markets, as well as the changing types of customers that the subsidiary serves. From an institutional theory perspective, a turbulent local market can be seen as a source of pressure on the subsidiary to become ‘isomorphic with the local institutional context’ in its efforts to better serve a rapidly changing customer base (Scott, 1991; Kostova and Roth, 2002). In highly turbulent markets, organizations must continually modify their product and service offerings in order to satisfy the changing preferences of customers (Jaworski and Kohli, 1993). Thus, in turbulent markets, subsidiaries may find it beneficial to rely on their own systems, crafted through experience and familiarity with their local customer base, rather than rely on practices or policies mandated by a headquarters located in a different market. Hence, under conditions of turbulence, subsidiaries are likely to have a greater role in marketing activities. Indeed, Frost et al. (2002) argue that the dynamic nature of a subsidiary’s market may even provide ‘latent’ opportunities for multinational firms to learn from the experiences of the subsidiary, suggesting a greater role for the subsidiary under these conditions. Additionally, Frost et al. suggest that customer needs in the local markets can influence the development of subsidiary roles. Garnier (1982) and Quester and Conduit (1996) also describe such variability in an individual market as a form of risk, and suggest that such risk will make it more important for decisions to be made at the subsidiary level. Thus we propose that turbulent markets will be associated with greater roles for subsidiaries.

H4: Market turbulence will be more positively related to the foreign subsidiary’s role than the headquarters’ role in marketing activities.

Market concentration Market concentration reflects the share of sales accounted for by a market’s top competitors (Scherer, 1980). In a highly concentrated market, few competitors have a large share of the market. Thus, according to Porter (1980), a high concentration ratio indicates that a high concentration of market share is held by the largest firms and, with only a few firms holding a large market share, the competitive landscape is less competitive (i.e., closer to an oligopoly). Under these conditions, one or a few dominant firms lead the market and often play a coordinating role in the industry (Porter, 1980). Competitors must often closely follow the moves of these dominant firms. A low concentration ratio indicates that many rivals, none of whom has a significant market share, characterize a more competitive market. Under these conditions, competitors are more prone to fighting and retaliation, creating greater market instability (Porter, 1980), which in turn leads to a more intensively competitive environment with greater market uncertainty and unpredictability (Gupta et al., 1986). From an institutional theory perspective, a highly competitive local market places pressures on the subsidiary to react and respond to the local industry context. For example, under intensified competition, competitor information gathering is essential to combat greater market uncertainty and unpredictability (Gupta et al., 1986). Therefore, when market concentration is low, input from local managers would be necessary to reduce uncertainty and provide the flexibility needed to effectively compete, suggesting greater roles for subsidiaries. In contrast, with high levels of concentration, monitoring competitors is less of a necessity (Kohli and Jaworski, 1990) and the subsidiary roles may be less important. In addition, under these conditions, Jain (1989) suggests that the use of standardized marketing strategies handed down from headquarters can be effective. Thus we expect that, in highly concentrated markets, subsidiaries will take on less of a role in marketing activities. H5: Market concentration will be more positively related to the headquarters’ role than the foreign subsidiary’s role in marketing activities.



K Hewett et al

572

Technological turbulence Technological turbulence is defined as the rate of change in technology associated with the development of products (Menon et al., 1997). A perception of a rapid pace of technological change is said to create uncertainty (Weiss and Heide, 1993). Technological turbulence is reflective of the overall industry, and not specifically the state of that industry in the subsidiary’s local market. Thus the uncertainty associated with technological turbulence is born by the MNC as a whole, and subsidiaries could expect to reduce their own risk by relying on the industry and category experiences amassed at headquarters. From an IO theory perspective, because the subsidiary is motivated to draw on the experiences of headquarters in order to reduce risk, adopting practices developed at headquarters may be viewed as more efficient than taking on a greater role in marketing activities. As technological advances make their way into product and marketing mix element designs, the headquarters would also be motivated to achieve economies of scale and learning by transmitting them across the firm’s multiple locations. Given the perceived risks associated with operating in a technologically turbulent market, pressures from the overall MNC for the subsidiaries to comply in implementing marketing activities developed at headquarters would likely be greater than pressures from the subsidiary’s market to be more locally responsive. Thus, from an institutional theory perspective, the MNC environment would exert more pressure than the local environment. Therefore, under these conditions, subsidiaries are expected to have less of a role in marketing activities than under conditions of low levels of technological turbulence. Further support can be found in Cavusgil et al. (1993), who argue that the greater the extent to which the industry has a technology orientation, the more the marketing strategies of a firm’s products are suited for a global approach. This argument was based on the belief that preferences for products in technology-intensive industries will tend to vary less than those for consumer goods (Douglas and Urban, 1977). Thus we expect that the degree of technological turbulence will be more positively (negatively) associated with the headquarters’ (subsidiary’s) role in marketing activities. H6: Technological turbulence will be more positively related to the headquarters’ role than the foreign subsidiary’s role in marketing activities.


Market conditions We expect that, as foreign markets differ from the home market, subsidiaries will take on greater roles. Consistent with IO and institutional theories, when differences between the subsidiary’s market and the parent location are greater, we expect there to be greater pressure from the local environment for the subsidiary to respond to the market’s needs than from the headquarters. According to Scott (1991), subsidiaries will attempt to relate to their local environment, and in doing so will make choices that may differ from those of the headquarters. We suggest that two specific market conditions reflect foreign market and home market differences: (1) cultural distance and (2) economic distance. Cultural distance Because of culture’s influence on customers’ underlying values, attitudes, and behavior (Hofstede, 1980; Schneider and De Meyer, 1991; Shane, 1994; Tayeb, 1994), the greater the cultural differences, or distance between the two markets, the more difficult it will be for the headquarters to make decisions regarding the subsidiary’s market without subsidiary input. In addition, institutional theory suggests that the institutional forces of the parent company in the home country are indirect, whereas the forces from the local market in which the subsidiary operates are more direct (Kostova and Roth, 2002). This suggests a greater role for the subsidiary under high-distance conditions as the subsidiary attempts to maintain its legitimacy in the local market. Jain (1989) similarly proposed that the degree of standardization will be higher across markets in which the behavior and lifestyles of customers are alike. Some preliminary evidence of the effects of cultural differences on the role of subsidiaries in marketing activities can be found in the marketing literature. For example, cultural differences were found by Schiffman et al. (1981) to influence consumer acculturation, affecting the likelihood that consumers will accept a product with features that were originally developed in a different market. Additionally, Mueller (1991) found that a standardized approach to advertising strategy, handed down by headquarters, is more likely when there are low levels of cultural distance between countries. We expect that greater cultural distance between the subsidiary and headquarters markets will increase the importance of the subsidiary’s role.


K Hewett et al

573

H7: Cultural distance will be more positively related to the foreign subsidiary’s role than to the headquarters’ role in marketing activities.

Economic distance Economic distance is defined as the difference between the economic conditions of the individual foreign markets and the market in which the headquarters is located. Similar to the arguments for differences in culture, we expect that the greater the economic differences between the two markets, the greater the role the subsidiary should play vis-a`-vis headquarters in marketing activities. In addition, from an institutional theory perspective, the more direct forces from the local market for the subsidiary to maintain its legitimacy would be greater than the more indirect institutional forces of the parent company in the home country. Evidence in support of the importance of the difference economic conditions between the headquarters and subsidiary markets can be found in the marketing literature. In an export context, Armine and Cavusgil (1986), for example, found that economic differences between markets positively influence the degree of customization of marketing program elements. Roth (1995b) also found that socioeconomic conditions in foreign markets influence the degree to which a common brand image is pursued. In addition, economic similarity of markets was suggested by Jain (1989) to reflect common incomes, lifestyles and attitudes that in turn influence elements of the marketing mix. Similarity in income levels, for example, can influence strategies for setting prices as well as marketing communication strategies (Baalbaki and Malhotra, 1992). We expect that, the greater the economic distance between headquarters and subsidiary markets, the greater the importance of the subsidiary in marketing decisions. H8: Economic distance will be more positively related to the foreign subsidiary’s role than the headquarters’ role in marketing activities.

Ideal profiles and product performance In the emerging tradition of international marketing strategy (e.g., Roth, 1995a, b) and MNC headquarters–subsidiary role (e.g., Frost et al., 2002) research, the investigation of effects on performance is of critical importance. In addition, the ability of multinational firms to effectively manage

dispersed capabilities has been suggested to be a source of competitive advantage, such that performance benefits would be expected (Bartlett and Ghoshal, 1991; Nohria and Ghoshal, 1997; Ghoshal and Nohria, 1989; Frost et al., 2002). We suggest that the performance benefits of having subsidiaries vs headquarters take on more significant roles in marketing activities are contingent on the prevailing relational, industry, and market conditions. In reality, the conditions we hypothesize are not fully independent, nor are organizations always able to address each individually or in a step-wise fashion. This suggests a collective pattern as well as independent linear effects of conditions on the ultimate role of the subsidiary or the headquarters. In this study, we focus on product performance in the subsidiaries’ individual markets, given our concentration on the marketing activities for a particular product. A systems-based approach for examining contingency effects can be found in Drazin and Van de Ven (1985) and Doty et al. (1993). Following Drazin and Van de Ven (1985) we suggest that deviation from an ideal profile for either a greater headquarters or a greater subsidiary role in marketing activities will result in lower product performance in individual markets. That is, products that achieve high performance do so because global firms have successfully aligned the roles taken on by headquarters and subsidiaries to the system in which they operate. Systems theory suggests that an ideal set of conditions exists for the roles of subsidiaries and headquarters in marketing activities, and that these conditions are best identified by decomposing the conditions associated with high-performance products in markets where the subsidiary or headquarters takes on a greater role. The conditions associated with high-performance products are referred to as an ideal profile. Products with alignment of conditions and subsidiary vs headquarters role (referred to as fit) should experience better performance than products with weak condition–role alignments. Put another way, as the distance between the relational, industry, and market conditions, and the role of the subsidiary vs headquarters in marketing activities decreases, higher product performance is expected. H9: The fit between the role of headquarters vs the foreign subsidiary and relational, industry and market conditions (i.e., the role–condition alignment) will be positively related to product performance in individual foreign markets.



K Hewett et al

574

Research methods Survey procedures and sample The survey instrument was initially based on existing measures. The instrument was then pretested among 16 academic experts and marketing managers. Modifications were made based on participant feedback. The sample comprised subsidiaries of US-based MNCs, and it was verified that English was used as the business language. Key informants were subsidiary-based marketing managers responsible for the planning and/or implementation of marketing activities for one or more products sold in their respective markets. Informants were instructed to focus on a specific product, and on one geographic market, and to indicate on the questionnaires the name of the product, its product category, and country market for which they responded. In developing the sample, the International Directory of Corporate Affiliations (1997), which contains hierarchies of US-based firms, was used to identify multinational firms and subsidiaries. Next, telephone calls were placed to identify marketing managers with the appropriate responsibilities in foreign subsidiaries. We identified marketing managers for 395 subsidiaries. Subsequent correspondence indicated that, in some cases, the subsidiary was not appropriate for this study (e.g., no substantial marketing activities, or the manager was new in the position). Elimination of such locations resulted in a final target sample of 366 managers at subsidiaries in 48 countries. A questionnaire package was sent to each of the informants. Reminder postcards were sent after 3 weeks, and follow-up questionnaires were sent after 6 and 10 weeks (Dillman, 1978). Respondents were also offered a report of findings. Across the three waves of mailings, we received usable questionnaires from 128 managers for a 36% response rate. These 128 responses represent 35 different country markets. The respondents represented more than 30 industries, including durables (e.g., automobiles, computers, household appliances, and tractors), nondurables (e.g., cosmetics, food and beverage products, and tobacco products) and services (e.g., telecommunications services and software development). The median number of subsidiary employees was 150, and the average sales for the subsidiary were US$533 million. The respondents averaged 3.33 years of experience in their particular positions and 7.90 years of experience with the subsidiary.


Nonresponse bias was first examined using the procedures recommended by Armstrong and Overton (1977). Responses from the first mailing were compared with the responses from the third mailing by testing for mean differences on all variables, including subsidiary characteristics. The results showed no significant differences across the waves of mailing (P-values ranged from 0.20 to 0.87). We also gathered secondary data on subsidiary characteristics for both responding and nonresponding firms using America’s Corporate Families and International Affiliates (1998). Comparisons across the numbers of employees and subsidiary sales also yielded no significant differences (Po0.77 for employees and o0.43 for sales).

Measures and measure validation A number of variables were assessed using multipleitem scales (final scale items and Cronbach’s a values are provided in Appendix A). The vertical dependence scale included items intended to assess the extent to which managers perceive that their subsidiary depends, in general, on the effective functioning of the headquarters in order to perform its own tasks. On the participation in goal setting scale, respondents indicated the level of participation of their subsidiaries in establishing five different performance objectives for the product. The cooperation between the subsidiary and headquarters scale measured perceptions of the communications, interactions, and relationships between subsidiary and the headquarters marketing operation. For the market turbulence scale, respondents rated their agreement with five statements regarding changes in customer preferences and expectations, and in the composition of their customer base. For technological turbulence, respondents rated their agreement with five statements regarding technological change in the development of their products, as well as technological developments in their industry in general. The remaining constructs were assessed using measures other than multiple-item scales. Hofstede’s (1980) indices were used to calculate a measure of cultural distance between the subsidiary and US headquarters (Kogut and Singh, 1988, 422). Economic distance from the US was calculated from gross domestic product (GDP) per capita data. Market concentration was assessed as the sum of the focal product’s market share and the shares of its top three competitors: the higher the value, the higher the level of concentration. An index of 14 marketing elements (e.g., packaging, advertising


K Hewett et al

575

message; Quelch and Hoff, 1986) was used to assess the subsidiary’s role in marketing activities compared with that of the headquarters. Positive responses on a five-point scale indicate greater subsidiary roles in the marketing activities. Product performance in the markets was operationalized as the market share of the product over the most recent annual period (Samiee and Roth, 1992; Roth, 1995a, b). The scaled multiple-item measures (i.e., all measures except the GDP distance, cultural distance, market concentration, and headquarters–subsidiary role in marketing activities) were subjected to purification procedures designed to evaluate dimensionality, reliability, and discriminant validity (Fornell and Larcker, 1981; Anderson, 1987; Gerbing and Anderson, 1988). Across all the scales, three items with low factor loadings (lo0.50) were identified and subsequently dropped from further analyses. Using the PROC CALIS procedure in SAS, we first used confirmatory factor analysis to assess the psychometric properties of the multiple-item scales. Because of our limited sample size and the overall number of indicators, we constructed two separate measurement models using subsets of our constructs. One set included constructs representing market conditions (e.g., market turbulence and technological turbulence), and one set included constructs representing relational characteristics (e.g., cooperation, vertical dependence, and participation) (see Doney and Cannon, 1997 for an example of this procedure). For the second model, the participation construct was estimated as a higher-order factor with two sets of indicators representing participation in setting goals for sales volume (sales and market share) and for profitability (ROI, ROA, and profit margin). Although w2 statistics for the two models were both significant (w2¼18.797, d.f.¼19 for the first model, and w2¼129.08, d.f.¼60 for the second), we considered other fit indices, given the sensitivity of these estimates to sample size (Gerbing and Anderson, 1993; Sharma, 1996). Other fit indices do suggest that the models provide a reasonable fit to the data (GFI¼0.96 for the first model and 0.85 for the second; RNI¼0.99 for the first model and 0.94 for the second; TLI¼0.99 for the first model and 0.92 for the second; CFI¼0.94 for the first model and 0.99 for the second; and DELTA2¼1.00 for the first model and 0.94 for the second). According to Sharma, indices such as the Tucker Lewis Index (TLI) and the Relative Noncentrality Index (RNI), which are both relative fit indices, based on a

comparison of the fit of the hypothesized model relative to a null model, are among the indices that are least sensitive to sample size. In addition, the TLI is not sensitive to the number of parameters in the model (Bollen and Long, 1993), and has been found to be comparable across data sets (Gerbing and Anderson, 1993). According to Gerbing and Anderson (1993), the DELTA2 index is also particularly insensitive to sample size. Finally, we examined the factor loadings of each indicator, and found them all to be significant (t-values all 44.5). Cronbach’s a, composite reliabilities, and variance extracted were computed for each scale to assess internal consistency. Cronbach’s a for all final scales exceeded 0.70, providing evidence of internal consistency (Peter, 1979). Additionally, composite reliability scores based on the item loadings from confirmatory factor models ranged from 0.73 to 0.93 (for these same variables). Discriminant validity was first assessed using the procedure recommended by Gerbing and Anderson (1988). Confirmatory factor models with two factors involving each possible pair of constructs were run. In the first model, the f coefficient was constrained to 1.0, and then was estimated freely in the second model. In all cases, the model with the free f coefficient was found to be superior to the model with the fixed f coefficient. Secondly, following the procedure recommended by Fornell and Larcker (1981), the variance extracted from each construct in a given pair was compared with a f2 estimate. Variance extracted estimates greater than the f2 values provide further evidence of discriminant validity. In all cases, the variance extracted estimates exceeded the f2 values (see Appendix B for a summary of these analyses). Lastly, we constructed confidence intervals around the f coefficient estimates using two times the standard error of the f coefficient for each pair of constructs. In none of the cases did the confidence interval contain 1.0, providing additional evidence of discriminant validity.

Control variables Based on previous research, a number of control variables were entered in our model. The control variables included: (1) key informant experience, operationalized as the number of years for which the respondent had been in his/her position (Cavusgil et al., 1993);



K Hewett et al

576

(2) subsidiary size (Quester and Conduit, 1996), operationalized as total sales for the most recent annual period;1 and (3) the product’s time in the market, operationalized as the number of months since the introduction of the product.

for the most successful products. We standardized the data and derived an ideal profile for each group by calculating the mean level of the eight conditions for the top 10 performing products in each group (based on market share).2 To verify that the profiles were distinct by the level of subsidiary vs headquarters role, we performed a multivariate analysis of variance, using the group (greater vs lesser role for the subsidiary) as the independent factor, and the conditions as dependent variables. The overall multivariate F-statistic was significant (F(8,81)¼2.83; Po0.008), providing evidence of different profiles based on the eight different conditions. Third, we calculated a fit measure for the remaining firms in each group by using the squared Euclidean distance of each observation from the group’s ideal profile over the eight conditions. Finally, a Pearson correlation coefficient was calculated between the fit and the market share. A significant negative relationship between the fit measure and performance would support H9, as greater distance represents lower fit, indicating that deviations from the ideal profile are associated with lower product performance. The results of the hypothesis-testing procedures are discussed below.

Hypothesis testing procedures The summary statistics and correlations for all variables are reported in Table 1. Regression analysis was used to assess H1–H8. The largest of the resulting variance inflation factors was 2.70, well below 10 – the level suggested by Mason and Perreault (1991) to signal detrimental multicollinearity. Additionally, residual plots indicated no outliers, and a normal probability plot suggested no violation of the normality assumption (Neter et al., 1990). Lastly, tests of the linearity of Q–Q plots for each variable confirmed that the normality assumption was not violated (Sharma, 1996). In testing H9, we employed the fit analysis procedure recommended by Drazin and Van de Ven (1985) for testing contingency hypotheses in interactive systems. First, we divided our sample based on a one-third split for the level of subsidiary vs headquarters role. The top group (greater subsidiary roles, indicated by high levels of customization) and bottom group (greater headquarters role, indicated by high levels of standardization) were retained for the remaining portion of the analysis. Second, for each group, we profiled the level of market, industry, and relational conditions Table 1

Results Findings regarding relational, industry, and market conditions The results of the regression analysis used to test H1–H8 are provided in Table 2. The coefficient for cooperation between the subsidiary and

Variable means, standard deviations, and intercorrelations

Variable 1 Cooperation 2 Vertical dependence 3 Participation 4 Market turbulence 5 Market concentration 6 Technological turbulence 7 Cultural distance 8 Economic distance 9 Informant experience 10 Subsidiary sales 11 Time in market 12 HQ–sub. role Mean Standard deviation: Cronbach’s a:

1

0.44** 0.16 0.04 0.02 0.02 0.17 0.02 0.05 0.20* 0.03 0.12 4.71 1.24 0.86

*Po0.05. **Po0.01.


2

3

4

0.03 0.09 0.16 0.06 0.02 0.12 0.10 0.10 0.48** 0.10 0.11 0.08 0.06 0.09 0.11 0.01 0.01 0.14 0.05 0.13 0.06 0.03 0.22* 0.20* 0.28** 0.20* 0.10 4.48 5.28 4.35 1.31 1.54 1.23 0.77 0.93 0.72

5

6

0.12 0.13 0.09 0.19* 0.03 0.04 0.17 69.34 32.60 —

0.03 0.11 0.12 0.08 0.18* 0.18* 4.68 1.57 0.90

7

8

9

10

11

12

0.28** 0.21* 0.17 0.02 0.07 0.06 0.06 0.06 0.32** 0.01 0.06 0.06 0.03 0.04 0.15 49.56 10914.4 0.37 89.40 314.70 3.21 33.08 7815.6 0.36 1497.20 709.10 0.73 — — — — — —


K Hewett et al

577

Table 2

Results of regression analysis

Predictiona

Standardized parameter estimatesb

t-Statistic

Probability

Relational conditions Cooperation Vertical dependence Participation in goal setting

(H1 ) (H2 ) (H3 +)

0.042 0.219 0.171

0.42 2.29 1.91

0.658 0.027 0.069

Industry conditions Market turbulence Market concentration Technological turbulence

(H4 +) (H5 ) (H6 )

0.184 0.207 0.331

1.84 2.39 3.39

0.051 0.034 0.001

Market conditions Cultural distance GDP distance

(H7 +) (H8 +)

0.020 0.099

0.22 1.10

0.985 0.229

0.062 0.059 0.160 0.173 3.259

0.67 0.67 1.73

0.534 0.519 0.105

Independent variables

Control variables Key informant experience Subsidiary sales Product time in market Adjusted R2 F-statistic a

Predictions are indicated as + for effects on the subsidiary’s role and – for effects on the headquarters’ role in marketing activities. Dependent variable indexes on five-point scales anchored by 1¼100% standardized by headquarters marketing operation (greater headquarters role) and 5¼100% customized by the subsidiary marketing operation (greater subsidiary role). b

headquarters marketing operations was not significant, and therefore H1 was not supported. The coefficient for vertical dependence was significant (Po0.05) and in the hypothesized direction, providing support for H2. The coefficient for participation in goal setting was significant (Po0.10) and in the hypothesized direction. Hence H3 is supported. The coefficients for all three of the industry conditions are significant (at Po0.05) and in the expected direction. Thus H4–H6 are supported, indicating the importance of industry conditions. Last, as shown in Table 2, neither the cultural distance condition nor the economic distance condition is significantly associated with the headquarters’ nor subsidiary’s role, and therefore H7 and H8 are not supported.

Ideal profile contingency hypothesis The results of the fit analysis procedure are reported in Table 3. This shows the correlation of fit (based on the distance between the ideal profiles and the remaining cases for each type of role) and market share. Our results reveal a significant negative correlation between market share and the fit calculation for products where the headquarters takes on a greater role in marketing activities (Pearson

Table 3 sharea,b

Profile analysis results: correlations of fit with market

Market share Fit for greater subsidiary role Fit for greater headquarters role

0.153 0.452*

a Fit is the summation of the squared Euclidean distance of each observation from the group’s ideal profile across all eight conditions. b Two-tailed significance values indicated by *Po0.01.

correlation¼0.452; Po0.01). Alternatively, the correlation between market share and fit for subsidiaries with greater roles is negative but not significant (Pearson correlation¼0.153).

Product and promotion aspects of marketing activities Although we focus in this study on the role of the subsidiary or headquarters in implementing all marketing activities for a particular product, we recognize that there may be differences in roles across the various marketing activities, even under similar conditions. In addition, past studies have explored specific aspects of the marketing mix (cf. Cavusgil et al., 1993). Therefore we conducted



K Hewett et al

578

additional analyses to determine whether our findings generalize to the specific product and promotional activities for the product. We identified the items pertaining to product and promotion marketing components, and tested our hypotheses for both the product and promotion subgroups. The results of these tests are provided in Appendix C, and reveal a few different patterns in the relationships between the conditions and role of the subsidiary or headquarters from when the analysis was on the overall marketing process. For the overall marketing process, market turbulence and market concentration are important conditions related to the subsidiary vs headquarters role, yet neither is related to the product component. Another interesting difference is that participation in goal setting is significant for the promotion sub-component, and to a lesser extent the overall marketing program, but not for the product sub-component. Implications of these and our other results are discussed below.

Discussion and implications The findings reported here suggest that relational and industry conditions are more important in influencing the role of the subsidiary vs headquarters than market conditions, supporting our institutional and IO theory-based model of marketing roles in MNCs. We also found that role–condition fit is increasingly important as MNCs attempt to hand down marketing processes developed at headquarters, as is frequently done when marketing strategies and/or implementation plans are standardized pan-regionally or globally. We explore these findings and their implications in more detail below. Effects of relational, industry, and market conditions Significant effects of two relational conditions were found. Vertical dependence and participation in goal setting are significantly associated with the roles of subsidiaries vs headquarters in marketing activities. We conceptualized and measured vertical dependence as reflecting the level of headquarters control over the subsidiary. Our results imply, then, that global marketing managers taking on greater roles in attempting to hand down marketing activities to foreign subsidiaries may be more successful the greater the degree of perceived vertical dependence among foreign subsidiaries. Additionally, we consider our finding that participation in goal setting is significant for the


promotion activities, but not for the product subcomponent, to be particularly interesting. This result may be due to greater headquarters involvement in developing products compared with the promotion activities and the overall marketing program in general. We speculate that, even when a subsidiary is actively involved in goal setting, some of the activities for developing and implementing product mix elements such as product design and branding may still be managed by the headquarters. As a result, whether or not participation in goal setting takes place may not affect certain product-related decisions. One surprising result from our study is that headquarters–subsidiary cooperation was not associated with greater subsidiary roles. In an effort to understand this unexpected result, we examined findings from previous studies of headquarters– subsidiary relationships. Roth and Nigh (1992), for example, explored the role of coordination in establishing effective headquarters–subsidiary relationships. In their study, coordination was described as allowing integration of subsidiary and headquarters activities through intra-organizational associations rather than the use of authority, which is consistent with our definition of cooperation. Roth and Nigh (1992) conclude that subsidiary managers do not see coordination as a constraint on their autonomy. In the context of our findings, this suggests that subsidiary marketing managers might not necessarily feel a lack of autonomy, and therefore cooperation may be equally likely when subsidiaries or headquarters take on greater roles in marketing activities. It is also likely that additional characteristics of the headquarters–subsidiary relationship could influence these roles, and our examination of only the three studied here represents a limitation of our study. We also found that industry conditions are significantly associated with subsidiary vs headquarters roles, indicating the importance of market, competitive, and technological issues. Our results are consistent with previous research from the market orientation literature (Jaworski and Kohli, 1993) that demonstrates industry effects on marketing strategy in single-country settings. In addition, the significance of market turbulence, market concentration, and technological turbulence found in our study deepens our understanding of industry drivers on international marketing program management and performance (Szymanksi et al. 1993). One caveat, however, is that the effects of industry


K Hewett et al

579

conditions seem to be sensitive to specific types of marketing activities (e.g., product or promotion). Our finding that market turbulence and market concentration are important for the overall marketing process, but that neither is related to the product component, may be due to the fact that product issues are inherently more resource intensive and related to production strategies and constraints that are often outside the subsidiary marketing manager’s control. As a result, local industry conditions are not integrated in productrelated processes as much as they may be in promotion, distribution, service, and other aspects of marketing. In contrast, technological turbulence, which we argue is a global as opposed to a local industry condition, is integrated into product as well as other aspects of the marketing process. Our findings also suggest that market conditions are less important in influencing subsidiary roles than industry or relational conditions. This is an important finding given that previous studies have often found cultural and economic conditions to affect local marketing strategies in multinational firms (e.g., Cavusgil et al., 1993; Roth, 1995b). A unique aspect of our research is that we empirically test influences on strategy across three types of conditions. The significant effects of market conditions in previous research may be an artifact of applying uni- or bi-condition measures. Alternatively, the level of specificity at which we measured culture and economics compared with previous research may explain differences in effects. Prior research has explored the influence of specific cultural dimensions (e.g., individualism–collectivism) rather than simply the overall differences in the cultures of two or more geographic markets (cf. Roth, 1995b). Specific cultural dimensions may be more closely related to subsidiary roles in certain situations than an overall cultural distance measure. Regarding economic conditions, findings from other studies seemed to differ based on the manner in which the conditions in individual markets were considered. Hite and Fraser (1988), for example, found that the rate of economic growth has decreased in importance for transferability of advertising messages, whereas Roth (1995b) found moderating effects of socioeconomic characteristics on the performance of customized vs standardized brand image strategies. Thus simply examining aggregate differences between the headquarters’ market and that of foreign subsidiaries may not reveal the true influence of these market conditions on subsidiary vs headquarters roles.

Our goal in this study was to better understand subsidiary vs headquarters roles in marketing activities in MNCs. We used institutional and IO theories to develop and test our model. From a theoretical perspective, our results confirm that both theories offer complementary perspectives that collectively enrich our understanding of this phenomenon. Both the efficiency mechanisms of IO theory embedded in the external industry conditions, and the legitimacy motivations found in the internal relational conditions, help to explain why MNCs can and do benefit from organizing marketing activities at subsidiary vs headquarter locations. Therefore institutional and IO theories provide unique and complementary perspectives to guide further studies on management roles and practices in MNCs.

Condition: strategic role fit and performance The contingency framework that we developed and tested supports previous findings (Cavusgil et al., 1993; Roth, 1995a, b) that international marketing performance is often contingent on external and internal factors. Unlike previous studies, however, we developed and tested a multidimensional (multi-condition) contingency model as opposed to the more traditional approach in which the interaction of two variables is examined independent of other variable interactions (Drazin and Van de Ven, 1985; Doty et al., 1993). The systems perspective rather than an interaction perspective is critical given the different types of conditions (relational, industry, market) that we investigate. To examine alignment differences in further detail we examined the condition means of the 10 low-performing and 10 high-performing products for each group. Table 4 shows the highand low-performance profiles for products where the subsidiary vs the headquarters takes on a greater role in marketing activities. The high-performing products for subsidiaries where the headquarters takes on a greater role were associated with high levels of cooperation, vertical dependence, and participation in goal setting as well as high market concentration and technological turbulence. Note also that these products tended to be marketed in culturally distant locations. In contrast, conditions associated with high-performance products in subsidiaries with greater roles are more industry specific. Greater subsidiary roles appear to be most successful when the market is turbulent, but with low levels of technological turbulence. In



K Hewett et al

580

Table 4

Means for greater subsidiary vs headquarters roles

Overall meansa

Variables Greater subsidiary roles Low performers (n¼10) Relational conditions Cooperation Vertical dependence Participation in goal setting Industry conditions Market turbulence Market concentration Technological turbulence Market conditions Cultural distance GDP distance

Greater headquarters roles

High performers (n¼10)

Low performers (n¼10)

High performers (n¼10)

4.44 4.43 6.06

4.62 4.50 5.02

4.88 4.20 4.58

5.34 4.87 5.60

4.18 52.40 4.20

4.25 96.30 4.00

3.88 47.90 4.40

4.00 101.70 4.85

56.27 8872.00

53.95 10202.00

39.18 9986.00

65.51 10582.00

a Values are the mean scores of the lowest and highest performing profiles for products where the subsidiary vs the headquarters takes on a greater role in marketing activities.

general these results are consistent with our hypotheses and regression results. It is interesting to note the pattern of mean scores for participation in goal setting shown in Table 4. As hypothesized, participation in goal setting is greater for firms in which subsidiaries have greater roles compared with those where the headquarters has a greater role. When broken down by low and high performers within each group, however, we observe that low performers at subsidiaries with greater roles have higher participation than high performers, and that high-performing products where the headquarters has a greater role have higher participation than low performers. So although the aggregate means by group are consistent with our hypothesis and regression results, the within-group analysis reveals that participation in goal setting may have an inverted-U effect, as opposed to a linear effect, on performance. Participation in goal setting benefits subsidiaries with greater roles, but too much participation may become detrimental. Likewise, although less participation in goal setting is needed when headquarters is attempting to hand down marketing strategies such that subsidiaries are taking on a less significant role in marketing activities, too little participation may negatively affect performance. The more conditions deviate from those faced by subsidiary marketers of successful products who take on a particular role, the lower product performance tends to be. The ability of global


managers to determine the extent to which they should foster greater roles for particular subsidiaries should also be enhanced in that these managers, unable to control market, industry, and/or relational conditions, can better forecast success by determining whether greater or lesser subsidiary roles are more aligned with these conditions. Performance assessment will also be improved by recognizing the obstacles faced in situations where alignment may not be ideal, such as when the headquarters is attempting to hand down processes under adverse industry conditions.

Conclusion Our findings provide new insights into the management of marketing programs in international markets. We have provided a greater understanding of the importance of headquarters vs subsidiary roles in a marketing context, and enhanced understanding of the interrelationships of relational, industry, and market conditions when implementing marketing strategy worldwide. In particular, our results suggest that global managers should focus on aligning industry and relational conditions with the decision whether headquarters or foreign subsidiaries should take on the role of managing marketing activities. Thus our findings represent a contribution to the literature on roles for headquarters and sub-units within the MNC context as well as the global marketing literature. In particular, our study verifies the complementary nature of


K Hewett et al

581

institutional and IO theories, in particular the efficiency and legitimacy mechanisms, for understanding role adoption and success in MNCs. As with any empirical study, however, we must acknowledge the possibility of study design limitations on our findings. Although steps were taken to ensure that the appropriate key informants were identified for this study, the potential for singlerespondent bias still exists. An additional concern related to our sample is the fact that only US-based MNCs were included in this study. Although our results are based on the strategies of subsidiaries located in 35 different countries, we specifically sampled only US-based MNCs to control for effects of parent-firm nationality and culture. Crossnational marketing studies have explored managerial decision-making (Tse et al., 1988) and corporate values (Norburn et al., 1990), but research has yet to explore differences in the headquarters–subsidiary relations as a function of parent-firm location. Evidence of parent-firm nationality effects on management styles and internal relationships could limit the generalizability of our findings. Future studies are needed in key strategic marketing areas such as order of entry (e.g., first mover vs follower), pricing (e.g., penetration vs skimming), distribution (direct vs indirect), and so on. Studies that focus on specific marketing strategies such as these will help to test the generalizability of extant theories and models as well as providing more strategic insights to international marketing managers.

Acknowledgements This research was funded by a grant from the Center for International Business Education and Research at the Moore School of Business, University of South Carolina. The authors would like to thank the Departmental Editor, G Tomas M Hult as well as three anonymous JIBS reviewers for their helpful comments on earlier versions of the manuscript. In addition, the study benefited from valuable input from William O Bearden and Thomas J Madden of the University of South Carolina. Notes 1 In an effort to control for firm size, we gathered information on subsidiary sales for the previous year as well as the number of employees. Since the two measures were significantly correlated (Pearson correlation¼0.40; Po0.001), we ran the regression analyses using each measure separately, and received consistent results. Therefore we report only the results using the subsidiary sales measure. 2 Note that this procedure was replicated using three alternative methods including: (1) a median split of the sample, and an ideal profile built using the top 10 performing firms; (2) a median split, with an ideal profile built using the top five performing firms; and (3) a 1/3 split with an ideal profile built using the top five performing firms. The results using these three additional analyses were consistent with those reported in our Results section. 3 Items were reverse-coded.

References America’s Corporate Families and International Affiliates: Corporate Affiliations Plus (1998) America’s Corporate Families and International Affiliates: Corporate Affiliations Plus, Reed Reference Electronic Publications: New Providence, NJ. Anderson, J.C. (1987) ‘An approach for confirmatory measurement and structural equation modeling of organizational properties’, Management Science 33: 525–541. Anderson, J.C. and Narus, J.A. (1990) ‘A model of distribution firm and manufacturing firm working partnerships’, Journal of Marketing 54: 42–58. Armine, L.S. and Cavusgil, S.T. (1986) ‘Consumer Market Environment in the Middle East’, in E. Kaynak (ed.) International Business in the Middle East, Walter de Gruyter: Berlin, pp: 163–176. Armstrong, J.S. and Overton, T.S. (1977) ‘Estimating nonresponse bias in mail surveys’, Journal of Marketing Research 14: 396–402. Astley, W.G. and Zajac, E.J. (1990) ‘Beyond dyadic exchange: functional interdependence and sub-unit power’, Organization Studies 11(4): 481–501. Baalbaki, I.B. and Malhotra, N.K. (1992) ‘Marketing management bases for international market segmentation: an alternate look at the standardization/customization debate’, International Marketing Review 10(1): 19–44.

Bartlett, C.A. and Ghoshal, S. (1991) Managing Across Borders, Harvard Business School Press: Boston, MA. Birkinshaw, J. and Hood, N. (1998) ‘Multinational subsidiary evolution: capability and charter change in foreign-owned subsidiary companies’, Academy of Management Review 23(4): 773–796. Birkinshaw, J. and Morrison, A. (1995) ‘Configurations of strategy and structure in subsidiaries of multinational corporations’, Journal of International Business Studies 26: 729–754. Bollen, K. and Long, J.S. (1993) Testing Structural Equation Models, Sage Publications: Newbury Park, CA. Cavusgil, S.T., Zou, S. and Naidu, G.M. (1993) ‘Product and promotion adaptation in export ventures: an empirical investigation’, Journal of International Business Studies 24(3): 479–506. Dillman, D.A. (1978) Mail and Telephone Surveys: The Total Design Method, John Wiley & Sons: New York. DiMaggio, P.J. and Powell, W.W. (1991) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, in W.W. Powell and P.J. DiMaggio (eds.) The New Institutionalism in Organizational Markets, University of Chicago Press: Chicago: 63–82. Doney, P.M. and Cannon, J.P. (1997) ‘An examination of the nature of trust in buyer–seller relationships’, Journal of Marketing 61: 35–51.



K Hewett et al

582

Doty, D.H., Glick, W.H. and Huber, G.P. (1993) ‘Fit, equifinality, and organizational effectiveness: a test of two configurational theories’, Academy of Management Journal 36(3): 1196–1250. Douglas, S.P. and Urban, C.D. (1977) ‘Life-style analysis to profile women in international markets’, Journal of Marketing 41(3): 46–54. Drazin, R. and Van de Ven, A.H. (1985) ‘An examination of alternative forms of fit in contingency theory’, Administrative Science Quarterly 30: 514–539. Fornell, C. and Larcker, D.F. (1981) ‘Evaluating structural equation models with unobservable variables and measurement error’, Journal of Marketing Research 18: 39–50. Frost, T.S., Birkinshaw, J.M. and Ensign, P.C. (2002) ‘Centers of excellence in multinational corporations’, Strategic Management Journal 23: 997–1018. Garnier, G. (1982) ‘Context and decision making autonomy in the foreign affiliates of US multinational corporations’, Academy of Management Journal 25: 893–908. Gerbing, D.W. and Anderson, J.C. (1988) ‘An updated paradigm for scale development incorporating unidimensionality and its assessment’, Journal of Marketing Research 25: 186–192. Gerbing, D.W. and Anderson, J.C. (1993) ‘Monte Carlo Estimations of Goodness-of-Fit Indices’, in K.A. Bollen and J.S. Long (eds.) Testing Structural Equations, Sage Publications: Newbury Park, CA, pp: 40–63. Ghoshal, S. and Nohria, N. (1989) ‘Internal differentiation within multinational corporations’, Strategic Management Journal 10(2): 323–337. Gooderham, P.N., Nordhaug, O. and Ringdal, K. (1999) ‘Institutional and rational determinants of organizational practices: human resource management in European firms’, Administrative Science Quarterly 44: 507–531. Gundlach, G.T. and Cadotte, E.R. (1994) ‘Exchange interdependence and interfirm interaction: research in a simulated channel setting’, Journal of Marketing Research 31: 516–532. Gupta, A.K., Raj, S.P. and Wilemon, W. (1986) ‘A model for studying R&D–marketing interface in the product innovation process’, Journal of Marketing 50(2): 7–17. Hite, R.E. and Fraser, C. (1988) ‘International advertising strategies of multinational corporations’, Journal of Advertising Research 28(4): 9–17. Hofstede, G. (1980) Culture’s Consequences, Sage Publications: Beverly Hills, CA. International Directory of Corporate Affiliations: Corporate Affiliations Plus (1997) International Directory of Corporate Affiliations: Corporate Affiliations Plus, Reed Reference Electronic Publications: New Providence, NJ. Jain, S.C. (1989) ‘Standardization of international marketing strategy: some research hypotheses’, Journal of Marketing 53(1): 70–79. Jaworski, B.J. and Kohli, A. (1993) ‘Market orientation: antecedents and consequences’, Journal of Marketing 57: 53–70. Kim, K., Park, J.-H. and Prescott, J.E. (2003) ‘The global integration of business functions: a study of multinational business in integrated global industries’, Journal of International Business Studies 34: 327–344. Kirpalani, V.H., Laroche, M. and Darmon, R.Y. (1988) ‘Role of headquarters control by multinationals in international advertising decisions’, International Journal of Advertising 7(4): 323–333. Kogut, B. and Singh, H. (1988) ‘The effect of national culture on the choice of entry mode’, Journal of International Business Studies 19(3): 411–432. Kohli, A. and Jaworski, B.J. (1990) ‘Market orientation: the construct, research propositions, and managerial implications’, Journal of Marketing 54(2): 1–18. Kostova, T. and Roth, K. (2002) ‘Adoption of an organizational practice by subsidiaries of multinational corporations: institutional and relational effects’, Academy of Management Journal 45(1): 215–233. Laroche, M., Kirpalani, V.H., Pons, F. and Zhou, L. (2001) ‘A model of advertising standardization in multinational corporations’, Journal of International Business Studies 32(2): 249–266.


Mason, C. and Perreault, W.D. (1991) ‘Collinearity, power, and interpretation of multiple regression analysis’, Journal of Marketing Research 28: 268–280. Menon, A., Jaworski, B.J. and Kohli, A.J. (1997) ‘Product quality: impact of interdepartmental interactions’, Journal of the Academy of Marketing Science 25(3): 187–200. Molm, L.D. (1994) ‘Dependence and risk: transforming the structure of social exchange’, Social Psychology Quarterly 57(3): 163–176. Moorman, C. and Miner, A.S. (1997) ‘The impact of organizational memory on new product performance and creativity’, Journal of Marketing Research 34: 91–106. Mueller, B. (1991) ‘Multinational advertising: factors influencing the standardized vs specialized approach’, International Marketing Review 8(1): 7–18. Neter, J., Wasserman, W. and Kutner, M.H. (1990) Applied Linear Statistical Models: Regression, Analysis of Variance, and Experimental Designs, 3rd edn. Irwin: Burr Ridge, IL. Nohria, N. and Ghoshal, S. (1997) The Differentiated Network, Jossey Bass: New York. Norburn, D., Birley, S., Payne, A. and Dunn, M. (1990) ‘A four nation study of the relationship between marketing effectiveness, corporate culture, corporate values, and market orientation’, Journal of International Business Studies 21(3): 451–468. Pappavassilou, N. and Stathakopolous, V. (1997) ‘Standardization versus adaptation of international advertising strategies: toward a framework’, European Journal of Marketing 31(7): 504–527. Peter, J.P. (1979) ‘Reliability: a review of psychometric basics and recent marketing practices’, Journal of Marketing Research 16: 6–17. Picard, J., Boddewyn, JJ. and Grosse, R. (1998) ‘Centralization and autonomy in international marketing decision making: a longitudinal study (1973–1993) of US MNEs in the European Union’, Journal of Global Marketing 12(2): 5–24. Porter, M.E. (1980) Competitive Strategy: Techniques for Analyzing Industries and Competitors, The Free Press: New York. Quelch, J.A. and Hoff, E.J. (1986) ‘Customizing global marketing’, Harvard Business Review 64: 59–68. Quester, P.G. and Conduit, J. (1996) ‘Standardisation, centralisation and marketing in multinational companies’, International Business Review 5(4): 395–421. Rau, P.A. and Preble, J.F. (1987) ‘Standardisation of marketing strategy by multinationals’, International Marketing Review 4: 18–28. Rosenzweig, P. and Singh, H. (1991) ‘Organizational environments and the multinational enterprise’, Academy of Management Review 16: 340–361. Roth, K. and Nigh, D. (1992) ‘The effectiveness of headquarters– subsidiary relationships: the role of coordination, control, and conflict’, Journal of Business Research 25(2): 277–301. Roth, M.S. (1995a) ‘The effects of culture and socioeconomics on the performance of global brand image strategies’, Journal of Marketing Research 32: 163–175. Roth, M.S. (1995b) ‘Effects of global market conditions on brand image customization and brand performance’, Journal of Advertising 24(4): 55–75. Rusbult, C.E. and Van Lange, P.A.M. (1996) ‘Interdependence Processes’, in E.T. Higgins and A.W. Kruglanski (eds.) Social Psychology: Handbook of Basic Principles, The Guilford Press: New York, pp: 564–596. Samiee, S. and Roth, K. (1992) ‘The influence of global marketing standardization on performance’, Journal of Marketing 56(2): 1–17. Scherer, F.M. (1980) Industrial Market Structure and Economic Performance, 2nd edn. Houghton Mifflin: Boston. Schiffman, L.G., Dillon, W.R. and Ngumah, F.E. (1981) ‘The influence of subcultural and personality factors on consumer acculturation’, Journal of International Business Studies 12: 137–143.


K Hewett et al

583

Schneider, S. and De Meyer, A. (1991) ‘Interpreting and responding to strategic issues: the impact of national culture’, Strategic Management Journal 12(4): 307–320. Scott, W.R. (1991) ‘Unpacking Institutional Arguments’, in W.W. Powell and P.J. DiMaggio (eds.) The New Institutionalism in Organizational Markets, University of Chicago Press: Chicago. Scott, W.R. (2001) Institutions and Organizations, 2nd edn. Sage: Thousand Oaks, CA, pp: 143–163. Shane, S. (1994) ‘The effect of national culture on the choice between licensing and direct foreign investment’, Strategic Management Journal 15(4): 627–642. Sharma, S. (1996) Applied Multivariate Techniques, John Wiley & Sons: New York. Slater, S.F. and Narver, J.C. (1994) ‘Does competitive environment moderate the market orientation-performance relationship?’, Journal of Marketing 58(1): 46–55. Smith, P.M., Ross, E.S. and Smith, T. (1997) ‘A case study of distributor–supplier business relationships’, Journal of Business Research 39(1): 39–44. Song, X.M., Montoya-Weiss, M.M. and Schmidt, J.B. (1997) ‘Antecedents and consequences of cross-functional cooperation: a comparison of R&D, manufacturing, and marketing perspectives’, Journal of Product Innovation Management 14(1): 35–47. Szymanski, D.M., Bharadwaj, S.G. and Varadarajan, P.R. (1993) ‘An analysis of the market share-profitability relationship’, Journal of Marketing 57(3): 1–18. Tayeb, M. (1994) ‘Organizations and national culture: methodology considered’, Organization Studies 15(3): 429–446. Tolbert, P.S. and Zucker, L.G. (1983) ‘Institutional sources of change in the formal structure of organizations: the diffusion of Civil Service reform, 1880–1935’, Administrative Science Quarterly 28: 22–39. Tse, D.K., Lee, K.-H., Vertinsky, I. and Wehrung, D.A. (1988) ‘Does culture matter? A cross-cultural study of executives, choice, decisiveness, and risk adjustment in international marketing’, Journal of Marketing 52: 81–95. Venkatraman, N. and Prescott, J.E. (1990) ‘The market share– profitability relationship: testing temporal stability across business cycles’, Journal of Management 16(4): 783–806. Weiss, A.M. and Heide, J.B. (1993) ‘The nature of organizational search in high technology markets’, Journal of Marketing Research 30: 220–233. Westney, E (1993) ‘Institutionalization Theory and the Multinational Corporation’, in S. Ghoshal and E. Westney (eds.) Organization Theory and the Multinational Corporation, St Martin’s Press: New York, pp: 53–76. Zou, S. and Cavusgil, S.T. (2002) ‘The GMS: a broad conceptualization of global marketing strategy and its effect on firm performance’, Journal of Marketing 66: 40–56. Zucker, L. (1987) ‘Institutional theories or organization’, Annual Review of Sociology 13: 443–464.

Appendix A Final scale items Cooperation Adapted from Song et al. (1997) (Cronbach’s a¼0.86)

(b) There is open communication between the marketing operations at headquarters and your subsidiary. (c) The marketing operations at headquarters and your subsidiary have similar goals. (d) Overall, your subsidiary’s marketing operation is satisfied with its interaction with the marketing operation at headquarters. (e) There is a give-and-take relationship between the marketing operations at headquarters and your subsidiary.

Vertical dependence Adapted from Astley and Zajac (1990) (Cronbach’s a¼0.77) Please indicate the extent to which each of the following statements describes your subsidiary’s marketing operation using a 7-point scale where 1¼to a very little extent and 7¼to a very great extent.

(a) To perform its own tasks effectively, the marketing operation at your subsidiary relies on the effective functioning of the marketing operation at headquarters. (b) Knowledge gained in the marketing operation at headquarters is transferred to the marketing operation at your subsidiary. (c) Work in the marketing operation at your subsidiary is connected to the work of the marketing operation at headquarters.

Participation in goal setting Adapted from performance measure used by Moorman and Miner (1997) (Cronbach’s a¼0.93) Please rate the level of participation of your subsidiary in establishing the following performance objectives for the brand/product you indicated on page 1 using a 7-point scale where 1¼no participation at all and 7¼full participation.

(a) (b) (c) (d) (e)

Market share Sales Return on assets Profit margin Return on investment

With regard to your subsidiary’s brand or product management activities, please rate the extent to which you agree with the following statements using a 7-point scale where 1¼strongly disagree and 7¼strongly agree.

Market turbulence Adapted from Jaworski and Kohli (1993) (Cronbach’s a¼0.72)

(a) People from the marketing operations at both headquarters and your subsidiary regularly interact.

Please rate your level of agreement with each of the following statements regarding the local market you specified on page 1 using a 7-point scale where 1¼strongly disagree and 7¼strongly agree.



K Hewett et al

584

(a) In our business, customers’ preferences change substantially over time. (b) Our customers continually expect product enhancements each year. (c) We are witnessing demand for our products or services from customers who never bought them before. (d) New customers have needs that are different from those of our existing customers.

Technological turbulence Adapted from Menon et al. (1997) (Cronbach’s a¼0.90) Please rate your level of agreement with each of the following statements regarding the local market you specified on page 1 using a 7-point scale where 1¼strongly disagree and 7¼strongly agree.

(a) The technology behind the development of our products changes rapidly. (b) Technological changes provide big opportunities in our industry. (c) A large number of new product ideas have been made possible through technological breakthroughs in our industry. (d) Technological developments in our industry are rather minor.3

Headquarters vs subsidiary role in marketing activities Think about the brand/product you indicated on page 1. For each of the following marketing elements, please approximate the extent to which headquarters has developed standardized processes that it requires you to use in your market vs allowing your subsidiary’s marketing operation to develop and implement market- or country-specific marketing processes. (Check ‘N/A’ if an item does not apply.) Scale: 1¼100% standardized by HQ marketing operation, 2¼75% standardized/25% customized, 3¼50% standardized/ 50% customized, 4¼25% standardized/75% customized, 5¼100% customized by subsidiary marketing operation.

(a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l)

Product design Product positioning Brand name used Packaging Price Basic advertising message Creative expression Sales promotion Media allocation Salesforce responsibilities Management of salesforce Use of middlemen


(m) Type of retail outlet (n) Customer service.

Appendix B Summary of analysis for multiple-item scales (see Table A1). Appendix C Testing of hypotheses for product and promotion sub-components In testing our hypotheses separately for factors related to product and promotion elements, we first used principal components analysis with varimax rotation to identify items related to the two components. Three factors emerged using the eigenvalue-greater-than-one and scree plot rules (Sharma, 1996). One factor, labeled product, contained five items (product design, product positioning, brand name, packaging, and price), and a promotion component contained four items (basic advertising message, creative expression, sales promotion, and media allocation). The remaining items loaded on a factor labeled sales and customer service. We focus on the product and promotion components, based on prior research (Cavusgil et al., 1993). We ran regression models to assess effects of the three sets of conditions on the subsidiary vs headquarters role for the product and promotion elements. Overall F-statistics for the product and promotion components were 3.015 (Po0.01) and 2.969 (Po0.01), respectively. For product, significant coefficients were found for cooperation (t¼1.71; Po0.10), vertical dependence (t¼3.66; Po0.01), and technological turbulence (t¼3.11; Po0.01). For promotion, significant coefficients were found for vertical dependence (t¼2.25; Po0.05), participation in goal setting (t¼1.80; Po0.10), market concentration (t¼1.98; Po0.05), technological turbulence (t¼2.46; Po0.05), cultural distance (t¼1.70; Po0.10), and GDP distance (t¼1.81; Po0.10). Next, the profile analysis was repeated for the two components. Results revealed significant negative correlations between the fit measure and performance when the subsidiary has a greater role in the product planning process (Pearson correlation¼0.413; Po0.01), and when the headquarters has a greater role in the product planning process (Pearson correlation¼0.292; Po0.05). Neither of the correlations for the promotion component was significant (Pearson


K Hewett et al

585

Table A1

Discriminant validity analysis for multiple-item scales.

Cooperation

Cooperation Vertical dependence

Participation in goal setting

Market turbulence

Technological turbulence

0.55 0.55 63.30 66.22 0.13 294.06 159.83 0.00 103.81 66.26 0.02 323.31 51.95

Vertical dependence

Participation in goal setting

Market turbulence

Technological turbulence

0.57

0.00 112.72 112.60 0.06 104.32 16.81 0.06 113.73 20.80

0.72

0.15 102.47 126.54 0.12 320.89 134.22

0.42

0.60 60.31 18.80

0.71

Entries below the diagonal show: (1) f; (2) difference in w2 from fixed (f¼1.00) model, and free (f estimated) model; and (3) w2 for free model. Shared variance values are provided on the diagonal.

correlations¼0.064 and 0.156 for greater subsidiary and headquarters roles, respectively).

About the authors K Hewett earned her Ph.D. in Marketing from the University of South Carolina. She is currently an Assistant Professor of Marketing at Winthrop University. Her research has appeared in the Journal of Marketing, the Journal of the Academy of Marketing Science, and the Journal of International Marketing among others. MS Roth earned his Ph.D. at the University of Pittsburgh and is Associate Professor of International

Business, Moore School of Business, University of South Carolina. His research on global marketing strategy has been published in journals including the Journal of International Business Studies, Journal of Marketing Research, and Journal of Advertising. K Roth earned his doctorate degree in international business from the University of South Carolina in 1986 and has served on the faculty at USC since that time. He is professor of international business, chair of the international business department and the J. Willis Cantey Chair of International Business and Economics.

Accepted by G. Tomas Hult, Departmental Editor, 26 August 2003.
