Implementing international strategy at the business unit level: the role of managerial decision-making characteristics
Kendall Roth
Kogut, B. 1989. A note on global strategies. Strategic Management Journal, 10: 383-389.
Lawler, E.E., III. 1974. For a more effective organization-match the job to the man. Organizational Dynamics, 3(1): 19-29.
Lawrence, P.R., & Lorsch, J.W. 1967. Organization and environment. Boston, MA: Division of Research, Graduate School of Business Administration, Harvard University.
Lorsch, J.W., & Morse, J.J. 1974. Organizations and their The belief systems of an organization are exemplified in part through managerial decision making. These belief systems are not, however, inflexible or unalterable. Rather, part of the role of top management is reshaping and redirecting belief systems in a manner that supports strategic choices (Donaldson & Lorsch, 1983). From this perspective, research has attempted to identify belief systems that are consistent with different strategy types. Studies generally confirm that a higher level of organizational performance results when the decision-making characteristics within an organization and its strategy are aligned (e.g. Govindarajan, 1988, 1989; Gupta & Govindarajan, 1984; Miller, Kets de Vries & Toulouse, 1982; Miller & Toulouse, 1986).
The international strategy literature suggests that the management of decision making characteristics may be particularly important for businesses competing in global industries (Bartlett & Ghoshal, 1989; Hedlund & Rolander, 1990) because geographic and cultural distances between subsidiaries within the business unit inherently creates variability in the organization’s decision making. As a result, attention must be devoted to managing decision-making characteristics in order to avoid fragmentation and dissipation across locations. Furthermore, Doz and Prahalad (1988) assert that recent patterns of global competition cannot be fully explained without considering the influence of managerial belief systems. In an international context, capital, technological, and cost differentials between businesses are diminishing as businesses with dispersed operations have access to comparable resources and country-based factor endowments. Where asymmetries do exist, they may be offset quickly through strategic alliances, collaborative agreements, or other cooperative arrangements (Doz & Prahalad, 1988). Given comparable access to the fundamental sources of the advantage, a critical source of competitive advantage becomes managerial characteristics because such characteristics may create a differential or unique capability to implement a particular strategy, thereby becoming a firm-specific advantage.
Thus, a potential exists to further understand international strategy by considering the influence of the decision-making characteristics of top level management. As suggested by upper echelon theory, possible benefits of this approach would include not only a “greater power to predict organizational outcomes” but also a clearer direction for the design of management development programs and normative integration activities within the organization (Hambrick & Mason, 1984: 193). Furthermore, a better understanding of top management decision making may enhance understanding of organizations’ abilities to exploit strategic opportunities. Given these considerations, the purpose of this article is to determine if organizational outcomes can be predicted from patterns of businesses’ decision making characteristics and international strategy. In addition, the article examines whether congruence or contingency models of top management decision making and type of international strategy best predicts organizational outcomes.
In the following section, managerial decision-making characteristics are related theoretically to the strategic choice of a business. The specific strategic choice–global versus multidomestic international strategy–is introduced. Individual managerial decision-making characteristics are then related to the international strategy types. Finally, a systems argument is forwarded considering the collective effects of the decision-making characteristics.
Theoretical Background Hypotheses
Managerial Decision-Making Characteristics
Three decision-making characteristics of top level management are considered in this study: (a) willingness to take risks, (b) openness of decision making, and (c) consensus decision making. These three decision-making variables were selected based on the substantive theoretical base linking each to the alternate international strategy types and because each relates to the uncertainty, resource requirements, and interdependencies that are associated with the strategy types. This study takes the position that a consistency between these decision-making variables and the organization’s international strategy is desirable. This consistency should occur in both how the variables are linked or associated as well as in the different system states within which the variables take on characteristic values (Dubin, 1976; Fry & Smith, 1987). Thus, the integrity of the system is presumed to be maintainable as the system changes from one state to another (Dubin, 1976: 28).
The desirability of consistency within an international strategy and decision making characteristics system is supported by three theoretical arguments. First, a need apparently exists to match managers with specific characteristics to job requirements (Lawler, 1974), as congruence between job requirements and individual characteristics often leads to a higher level of performance (Lorsch & Morse, 1974; Griffin, 1980). Similarly, Govindarajan (1988; 1989) argues that different strategies imply different job requirements (Andrews, 1971; Galbraith & Nathanson, 1978). Thus, superior performance might be achieved through developing managerial competencies, including decision-making characteristics, that are congruent with the specific requirements of a particular strategy type (Gupta & Govindarajan, 1984; Hambrick & Mason, 1984).
Second, as suggested by upper echelon theory, complex decisions are partially a result of non-rational factors (Finkelstein & Hambrick, 1990). Bounded rationality results because complex decisions, such as strategic choices, are characterized by ambiguity, conflicting goals, and imperfect information (Janis & Mann, 1977; March & Simon, 1958). Reduced rationality allows increased levels of error to enter the decision process and, therefore, strategic choices will in part reflect the “idiosyncracies of decision makers” (Hambrick & Mason, 1984: 195). Thus, the cognitive base of top level management, partially reflected through their belief systems, becomes linked with strategic choices.
Third, managerial decision-making characteristics may also contribute directly to an organization’s competitive advantage. Barney (1986) argues that the organization’s culture–its values, beliefs, assumptions, and symbols–defines how an organization conducts its business. To support competitive advantage, the belief systems of an organization must not be common to other organizations and they must promote, within the organization, a capability to perform functional activities in a manner that increases the value-added of the firm. In addition, for incremental value creation, the belief systems must not be imitable because imitation would result in the source of advantage being dissipated (Barney, 1986). Thus, to the extent that an organization holds a core set of managerial values and they are valuable, unique, and non-imitable, they will have positive economic consequences. Managerial decision-making characteristics are, therefore, not only linked to strategy through creating a key capacity to implement a particular strategy type, but may be a direct source of competitive advantage as well (Deal & Kennedy, 1982; Ouchi, 1981). In summary, the preceding arguments suggest that the strategy of a business and its managerial decision-making characteristics are linked because these characteristics (a) are fundamental to effective implementation of the particular strategy choice, (b) are reflected directly in that choice, and (c) are a direct source of competitive advantage. Thus, strategy and managerial decision-making characteristics are posited as an interdependent and interactive system that affects organizational outcomes.
International Strategy Types
For the purpose of this study, the global-multidomestic conceptualization of international strategy was selected because this framework has received considerable attention in the literature (Fayerweather, 1969; Porter, 1986, Prahalad, 1975; Prahalad & Doz, 1987) and has been found to be internally consistent (Roth & Morrison, 1990). Within the framework, the two fundamental strategic choices in a global industry that are suggested are global and multidomestic strategies (Porter, 1986; Prahalad & Doz, 1987). A multidomestic strategy emphasizes the need to be responsive to each local environment. Consequently, a multidomestic strategy implies competing in industry segments that are most effected by local differences (Porter, 1986: 48)in order to maximize local competitive advantage (Yip, 1989). In contrast, a global strategy emphasizes the linking of competitive positions across national markets (Ghoshal, 1987). Competitive advantage is based predominantly on capturing global scale or scope economies through integration of the activities of the business and focusing on customer demands that are standardized across markets.
Willingness to Take Risk and International Strategy
Gupta and Govindarajan (1984) suggest that different strategies result in facing task environments that vary in their level of uncertainty and that implementing a particular strategy will require different levels of risk taking depending on the associated uncertainty. They found, for example, that “build” strategies, having higher external dependencies and environmental conflict, necessitated greater willingness to take risks as compared to “harvest” strategies. It is posited in this study that the choice of international strategy will also result in different levels of uncertainty. Two arguments suggest that a multidomestic strategy will result in a higher level of uncertainty, as compared to a global strategy. First, as discussed in the preceding section, global scale economies/efficiencies are a central source of competitive advantage for the global strategy (Ghoshal, 1987). At the product level, global scale economies are achieved through identifying homogenous market segments worldwide and offering a “standardized core product that requires minimal local adaptation” (Yip, 1989: 31). In contrast, businesses pursuing a multidomestic strategy attempt to compete within each country independently. Products are designed country-specific in order to tailor products to local needs. Govindarajan (1988: 832) notes that, whereas strategies based on efficiency must minimize product offerings over time to control costs, differentiation-based strategies must emphasize product innovation to sustain “uniqueness and exclusivity” (Dess & Davis, 1984, Govindarajan, 1986; Gupta, 1987a). Previous research has found that businesses emphasizing new product activities will face high uncertainty given that the market response to its innovation is unknown (Biggadike, 1979). Further, as discussed by Yip (1989), businesses pursuing a global strategy focus on a smaller number of products than if pursuing a multidomestic strategy. Researchers have noted that product breadth is also associated with high uncertainty, as it results in confronting a more complex task environment (Chandler, 1962; Govindarajan, 1988; Gupta, 1987b). Consequently, the product innovation and product breadth associated with a multidomestic strategy results in a higher level of uncertainty for the overall business unit, as compared to a global strategy.
Second, a global strategy has been associated with the integration of activities across locations. This integration is necessary because a business unit pursuing a global strategy attempts to exploit cross-national sources of advantage. Thus, within global industries, research has verified that a high level of intrafirm resource flows occur within the industry (Kobrin, 1991). In contrast, the multidomestic strategy is based on location-specific sources of advantage, primarily a high degree of responsiveness to the local environment. A high degree of local responsiveness is supported operationally by allowing the business unit to be largely autonomous and by developing capabilities within the business unit that are necessary to monitor and respond to the local market (Prahalad & Doz, 1987). Increased autonomy and capability to perform a major portion of the value-adding process implies that a business unit pursuing a multidomestic strategy will depend more on locally-sourced resources as opposed to inputs from affiliated business units. Consequently, the business that pursues a multidomestic strategy is likely to be more dependent on the local environment for its resources, as compared to a global strategy. However, the business will have to compete lot these resources in locations that vary in resource abundance, particularly with respect to the various activities constituting the value-adding process. As resource availability decreases, an increase in conflict and interdependence among organizations within the environment results, subsequently increasing the level of uncertainty (Pfeffer & Salancik, 1978). Thus, businesses pursuing a multidomestic strategy will face greater uncertainty than their global strategy counterparts.
The preceding arguments suggest that due to greater demands for product innovation, product breadth, and resource dependency, the multidomestic strategy faces greater uncertainty. As suggested by decision theory, managers confronting increased uncertainly must be willing to take greater risks to be effective (Gupta & Govindarajan, 1984). Therefore, it is hypothesized:
Hypothesis 1: Willingness to take risks will make a greater contribution to performance for business units pursuing a multidomestic strategy than in the case of business units pursuing a global strategy.
Openness of Decision Making and International Strategy
Two arguments are forwarded linking the openness of decision making within a business unit and its international strategy. The first argument is based on viewing the organization as an information processing system. The preceding discussion suggested that a multidomestic strategy will have a higher level of uncertainty than a global strategy. Galbraith suggests “the greater the task uncertainty, the greater the amount of information that must be processed among decision makers” (1973:4). Furthermore, product diversity and the level of product modifications across locations within the business unit also effect the information processing requirements directly. Egelhoff (1982) notes that, as product diversity increases, a higher level of environmental and technological complexity results, necessitating increased information-processing requirements within the organization (Lawrence & Lorsch, 1967; Galbraith, 1977).
The multidomestic strategy necessitates product diversity to support its locally-based competitive advantage. However, local responsiveness affects the complete value-adding process of the business unit because R&D, manufacturing, and distribution/marketing must be tightly integrated in order to rapidly and flexibly respond to the local context. The resulting product/process complexity and heterogeneity will require more decentralized communication patterns within the business unit (Perrow, 1967). In contrast, the global strategy will have extensive communication patterns across business units while having a simpler pattern within the business unit. Identification of homogenous market segments and the use of standardized products allows the production processes to be more standardized and repetitive in nature (Govindarajan, 1989). Repetitive tasks require less information to be processed within the business unit as compared to nonrepetitive processes (Tushman & Nadler, 1978). Thus, the information-processing requirements within the business unit pursuing a global strategy will be less than the information processing that must occur to support a multidomestic strategy. Research indicates that as information requirements increase, informal and lateral channels of communication must be created to support decision making, thereby supplementing or replacing impersonal and hierarchical channels (Bartlett & Ghoshal, 1989; Egelhoff, 1988; Lawrence & Lorsch, 1967; Tushman, 1978; Van de Ven, Delbecq, & Koenig, 1976). It is expected that increased information requirements of the multidomestic strategy will require informal and lateral communication patterns.
The second argument is based on a task analysis perspective. Abell (1978) found that as task uncertainty increases, decision-making responsibilities are dispersed more broadly throughout the organization. Presumably, the increase in uncertainty results in an increase in the discretionary content of decisions, given that the decisions cannot be relegated to a set of rules. This increase in the discretionary content leads to an increase in the value of local knowledge, which necessitates more participatory and open decision making. In essence, uncertainty creates a dependency within the organization on the individuals who are able to reduce problems that stem from the lack of predictability associated with the uncertainty (Heller, Drenth, Koopman & Rus, 1988: 17). Thus, uncertainty is associated with reduced power concentration and more participative decision making (Heller & Wilpert, 1981; Lorsch & Morse, 1974; Rus et al., 1977).
These two arguments suggest that a multidomestic strategy will require informal and lateral channels of communication and a more participative decisionmaking structure as compared to a global strategy. Additional communication and participation may be achieved through emphasizing openness in decision making. Openness in decision making occurs when multiple individuals or preferences are incorporated in the decision process (Heller, 1971). It is based on direct or indirect interaction of managers irrespective of organizational position or geographic location and thereby provides for additional nonhierarchical and lateral communication. Managers interact on the basis of shared problem-solving efforts rather than relying exclusively on formal reporting relationships (Gupta, 1987b). Furthermore. having multiple managers involved in the decision-making process provides an increase in the total amount of knowledge and information that can be incorporated into the process (Maier, 1967; Shaw, 1981). Thus, openness in decision making supports the implementation requirements of the multidomestic strategy.
Hypothesis 2: Openness in decision making will make a greater contribution to performance for business units pursuing a multidomestic strategy than in the case of business units pursuing a global strategy.
Consensus Decision Making and International Strategy
As discussed above, important sources of competitive advantage supporting a global strategy are global scale and scope economies. Essentially, the organization is attempting to capture synergy in its activities across locations. In addition, differences may exist in input and output markets among the organization’s locations (Ghoshal, 1987). Such differences are a source of competitive advantage in that they allow the organization to capture efficiencies based on factor endowment costs or through exploiting organizational learning across locations. Exploiting these sources of advantage, as well as capturing global scale/scope economies, necessitates resource flows among the subunits within the business unit. Increased resource flows within the business unit are, therefore, a central operational implication of a global strategy (Kogut, 1989; Porter, 1986) in contrast to the multidomestic strategy, where activities across locations remain essentially independent.
Resource flows lead to increased interdependencies within the organization and thereby necessitate greater coordination of the organizational components (Thompson, 1967; Van de Ven, Delbecq, & Koenig, 1976), a theme carried forward in the international literature as researchers stress the importance of coordination in pursuing a global strategy (Bartlett & Ghoshal, 1986; Ghoshal, 1987, 1989; Kogut, 1985a,b; Prahalad & Doz, 1987). McCann and Galbraith argue that coordination within the organization is beneficial then instituted in conjunction with developing and maintaining “shared appreciations” of the interdependencies because shared values “provide a basis for agreement about the performance and merits of alternative strategies for managing interdependencies” (1981: 67). The importance of this perspective to the management of an international organization can be traced back to Perlmutter’s notion of the geocentric film (Perlmutter, 1969). A worldwide approach and a strong source of unification, through shared managerial beliefs, are considered critical organizational attributes to be developed and managed if the organization intends to pursue a global strategy (Bartlett & Ghoshal, 1989: 66). Decision making cannot be parochial, nor can it be made independently at the business unit level without consideration of the worldwide perspective. From this perspective, implementing a global strategy is achieved optimally through managerial decision making that facilitates shared understandings. Thus, it is posited that consensus decision making contributes to the ability of an organization to implement a global strategy. Group consensus provides for business-wide managerial acceptance of pursuing a global strategy, building support for the unit as a whole as well as for the development of managers’ understanding of the necessary tradeoffs implicit in the resource flows between locations.
It may appear contradictory to suggest that openness and participatory decision making accompanies a multidomestic strategy, whereas consensus accompanies a global strategy given that participation and consensus are often considered to be positively associated (Maier, 1967; Shaw, 1981). For example, Bartlett and Ghoshal advocate that “institutionalizing managers’ participation in the deliberations leading to major choices” is a critical means of ensuring managerial acceptance of such operational implications of a global strategy (1989: 192). However, the appropriateness of participation depends on the decision-making context (Vroom & Yetton, 1973). In an international setting, increased participation has been found to increase the level of disagreement within organizations (Heller et al., 1988) because, as the decision process becomes more open, alternate views enter into the process. Although differing views may ultimately result in improving the quality of the decision outcomes, they do not necessarily promote consensus. Heller et al. conclude that “a moderate degree of conflict is consistently associated with more participative behavior” (1988: 8). Thus, it is hypothesized that:
Hypothesis 3: Consensus decision making will make a greater contribution to performance for business units pursuing a global strategy than in the case of business units pursuing multidomestic strategy.
Contingency Hypothesis
The preceding hypotheses focus on the relationship between separate managerial decision-making characteristics and international strategy and their effect on business unit performance. The hypotheses are “congruence” forms of coalignment and take a reductionistic perspective in that the effects of the characteristics are considered independently. However, as noted by Donaldson and Lorsch, the beliefs of an organization are a “tightly interrelated system” (1983: 81). Further, the decision-making characteristics and strategic choices “interact to determine organizational performance levels” (Hambrick & Mason, 1984: 197). Venkatraman and Prescott argue that if one is attempting to retain the holistic nature of coalignment within a system, then tests of the performance effects of coalignment must reflect the simultaneous and collective pattern of the proposed interlinkages (1990: 5). Consequently, it is a systems approach of “contingency hypothesis” that is the real test of the proposed relationships because such an approach examines the patterns of consistency among multiple dimensions (Drazin & Van de Ven, 1985). If each variable is linked theoretically to effective performance with respect to a particular strategy type, it follows that internal consistency among all the dimensions should be associated with a higher level of performance, as compared to the “nonmatched” business units.
The internal consistency of the international strategy and managerial decision-making system may be defined through specifying the interrelatedness of the system dimensions. It has been argued previously that a multidomestic strategy implies increased information-processing requirements, uncertainty, product diversity, and non-repetitive type processes. These characteristics suggest that the multidomestic strategy is a rather complex and multifaceted strategy. Complex strategies create additional points of conflict (Bartlett & Ghoshal, 1989, Bourgeois; 1980) and make consensus within the organization pursuing a multidomestic strategy unlikely. Although consensus may not be expected, the presence of strategic complexity implies that openness in decision making will be important because it provides managers access to flows of ideas and information within the business unit (Kerr & Jackofsky, 1989). Access to ideas and information is particularly critical in that, as previously argued, a multidomestic strategy implies considerable risk taking. In the context of risk, the quality of the decision-making process becomes important as decisions based on more complete information and analysis are expected to yield better results (Schmit & Roth, 1990; Tushman & Nadler, 1978). In addition, there is some evidence to suggest that the quality of decisions is higher for recommendations generated under conflict as compared to those generated under consensus (Schweiger, Sandberg & Ragan, 1986). For decision making under consensus, Dess & Origer (1987) argue that “normative pressures to conform” may lead to decisions made with an incomplete assessment of the available and relevant information (Janis, 1972; Whitney & Smith, 1983).
In summary, this argument suggests that an interrelated strategy/decision-making system presumably exists. A multidomestic strategy implies risk taking and a reduction of the associated uncertainty accompanying risk taking occurs through quality decision-making processes. Such processes are enhanced by allowing openness within the organization to facilitate information and idea exchange. This information is effectively evaluated in the absence of pressures for consensus. Thus, the managerial decision-making characteristics and international strategy collectively and interdependently influence performance outcomes. Therefore, it is hypothesized:
Hypothesis 4: A fit between international strategy, risk taking, openness, and consensus decision making will be positively associated with business unit performance.
Research Method
Data were collected from the senior manager of 82 business units competing in global industries. Global industries were the defined domain of the study in that only in such industries would both global strategies and multidomestic strategies exist. Global industries were identified through (a) examining intra-industry trade (Porter, 1986; Cvar, 1984), (b) conducting a review of the literature that previously identified global industries (Bartlett & Ghoshal, 1987; Cvar, 1984; Hout, Porter, & Rudden, 1982, Porter, 1980; 1986; Prahalad & Doz, 1987), and (c) verifying the presence of at least one global competitor (Hamel & Prahalad, 1985; Hour, Rudden & Porter, 1982; Porter, 1986). Although it is not asserted that this three-stage procedure resulted in the identification of an exhaustive set of global industries, the industries identified appeared to indeed be global in nature.(1) Furthermore, to avoid potential confounding influences of global industry variability, the selection criteria were designed to capture a single global industry type, the “truly global” based on Doz’s (1986) global industry typology.
A mail survey was the primary means of data collection. The questionnaire was developed through conducting field interviews, reviewing related research, and pretesting an initial version with both academicians and managers (see Roth, Schweiger & Morrison, 1991). The survey was conducted in two stages. First, using Dun and Bradstreet’s America’s Corporate Families, and The Directory of Corporate Affiliations, business units within the selected industries and the senior manager of each business unit were identified. The senior manager generally held the title of president (approximately 70% of the business units), although other titles included group vice president, senior vice president, general manager, and chief operating officer. This process resulted in a total of 322 potential respondents. An initial questionnaire, designed to collect information on the industry structure, goal structure, strategy, and performance of each business unit, was then sent to each manager. This mailout and two follow-up mailouts to nonrespondents resulted in 147 responses. In the next stage, a second questionnaire was sent to the 147 respondents. This questionnaire was designed to collect information on each business unit’s formal and informal structure and managerial decision-making characteristics. The initial mailout, with summary results of the previous survey, was again followed by two additional mailouts to nonrespondents, resulting in 82 responses to the second questionnaire. The business units were major international divisions with responsibility for subunits located internationally. The business units had an average total sales of approximately $730,000,000, an average number of employees of 3966, a mean foreign operating profit as a percentage of total profits of 26.5%, and they operated in an average of seven foreign countries.
Measurements
International strategy. An instrument was developed to measure international strategy based on the conceptual discussion of global and multidomestic strategies found in Porter (1986) and Prahalad and Doz (1987). Executives were provided five items designed to determine the strategic position of their business unit. Executives were asked to consider only the sector of the industry in which their business unit competes and then indicate how characteristic each of the following statements are in describing their industry position: (a) buyer/customer needs are standardized worldwide, (b) product awareness/information exists worldwide, (c) standardized product technology exists worldwide, (d) competitors exist that have a presence in all key markets, and (e) competitors market a standardized product worldwide. Each item was rated on a 5-point scale, ranging from “not at all characteristic” to “extremely characteristic.” An index was then created by averaging the responses for the five items. Thus, the index was designed to inlet business unit strategy from the strategic group to which the business unit belongs. This approach was selected in that executives, when directly asked to convey the strategy of their business unit, are more likely to report their intended strategy rather than the realized or actual strategy (Hambrick, 1979).
The construct validity and reliability of this measure was assessed three ways. First, the internal reliability (Cronbach’s coefficient alpha) was examined and found acceptable (|is proportional to~ = 0.73). Second, 15 business units were randomly selected. Using secondary information, two researchers independently classified the business units as pursuing either a global strategy or a multidomestic strategy. Agreement between the researchers occurred for 12 of the 13 businesses for which secondary information was available. Furthermore, the researchers’ classifications were consistent with the classifications based on the survey instrument. Third, two additional indicators of international strategy were assessed. In a separate section of the instrument, a self-typing measure was used in which respondents were asked to indicate the description that the best characterized the strategy of their business.(2) Values of the descriptive classification were correlated significantly (p |is less than~ .05) with values for the international strategy measure.
The second indicator of international strategy was based on Porter’s (1986) discussion of global strategy. Porter asserts that coordination is a necessary condition for the existence of a global strategy because a global strategy is based in part on integrating value activities across geographic locations. This integration would not occur in the absence of coordination. The specific form of coordination discussed by Porter is the extent to which similar functional activities are coordinated within the entire business unit across country locations (1985). Thus, a measure was developed to assess this aspect of coordination for nine activities composing Porter’s value chain. The international strategy measure was found to be positively correlated with the coordination measure (p |is less than~ .01).
Managerial decsion-making characteristics. The three managerial decision-making characteristics were measured with an instrument derived from research on top management decision making (Donaldson & Lorsch, 1983; Dutton & Duncan, 1987; Gordon & Cummins, 1979) and adapted from similar measures used by previous researchers (Hofstede, 1976; Weber, 1988). Each characteristic was measured with four items, resulting in a total of 12 items. Using a 7-point scale, each item was designed to elicit executives’ perceptions of the extent to which the item characterized the management of activities within the business unit. Specifically, respondents were instructed, “In this section questions are asked about the degree to which beliefs are commonly shared by managers within your business unit. Focusing on the managers of functional activities of your business unit, please indicate the extent to which these managers share these beliefs.”
Risk taking was measured with four items regarding the extent to which (a) innovative rather than conservative decision making, (b) taking chances on good ideas, (c) holding managers accountable for end results (rather than the means) through using (d) clear performance measures, characterize beliefs within their business unit. Openness in decision making was measured with four items regarding the extent to which managers (a) share information and communicate with other subunits, (b) maintain/develop relationships with managers in other departments, (c) make efforts to understand other manager’s problems, and (d) create/maintain effective communication and cooperation with peers. Finally, consensus decision making was measured with two reverse-scored items regarding the extent to which (a) autonomy in decision making is given to managers and (b) individual initiative and achievement is emphasized; and two items examined the importance of (c) consensus in decision making and (d) policies/practices that promote a sense of duty and loyalty to the business.
The internal reliability estimates (Cronbach’s alphas) for risk taking, openness of decision making, and consensus decision making were 0.67, 0.80, and 0.79, respectively. Factor analysis was also used to confirm that the conceptualized constructs were empirically verifiable. The factor analysis (available from the author) provided clear evidence that only three constructs underlie the data. Further, with the exception of a single variable, all variables loaded on the predicted construct. Thus, there is support for the validity of the three measures.
The senior manager was considered the most appropriate respondent because research suggests that subordinates often overestimate their role in the decision-making process (Heller, 1971). Furthermore, the manager of the business unit is expected to be knowledgeable, if not responsible, for the style of decision making within the unit. However, the hypotheses are based on decision-making characteristics that typify the top level management of the business unit and not a single individual. To ensure that the responses were not idiosyncratic to individual managers, the instrument was administered to 17 executives from five business units. The responses within each business unit were found to be highly consistent based on the coefficient of concordance (Kendall’s tau) of the responses as all correlations were statistically significant (p|is less than~.05). The limited number of respondents, however, suggests that this is a very tentative assessment of interjudge reliability.
Business unit performance. Two indicators of performance–return on assets and sales growth–were assessed with a subjective measurement scale adapted from similar scales used by Dess and Davis (1984) and Ghoshal and Nohria (1989). Executives were asked to indicate their business unit’s performance over the last three years compared to other businesses in the industry, where 1 = “lowest 20%”, 2 = “lower 20%”, 3 = “middle 20%”, 4 = “next 20%” and 5 = “top 20%”. Subjective scales were considered to be a reliable source of performance information, given the diversity of industries being studied as well as potential accounting TABULAR DATA OMITTED TABULAR DATA OMITTED differences arising from business unit variation and business unit nationality. Furthermore, the pretest indicated that executives would be hesitant to respond to self-reported objective measures of performance. The reliability of the performance measure was assessed by means of an examination of industry reports and industry estimates of performance for 25 businesses. Without exception, performance indicators from secondary data fell in the range provided through the self-reported measures.
Results
The summary statistics for the variables are provided in Table 1.
The first three hypotheses were examined by estimating a regression equation with an interaction term.(3) The regression equations, after controlling for business unit size, are presented in Table 2, Table 3, and Table 4.(4) As equations 1 and 2 in Table 2 indicate, Hypothesis 1 was supported for ROA. Support of the hypothesis is provided when the coefficient of the interaction term is significant, the coefficient of determination is significantly greater for the equation with the interaction term as compared to the initial equation, and the sign of the coefficient of the interaction term is as hypothesized. For H1 (ROA), the interaction was negative and significant; the introduction of the interaction term resulted in a significant increase in |R.sup.2~ Hypothesis 1 was not supported when performance was operationalized as sales growth, as indicated in equation 3 and equation 4. The results of TABULAR DATA OMITTED equations 5 and 6, as displayed in Table 3, did not support Hypothesis 2 as the interaction term was negative and significant, but the increase in explained variance was not statistically significant. Hypothesis 2 was also not supported when sales growth was the dependent variable (equations 7 and 8). Results of equations 9, 10, 11, and 12, displayed in Table 4, did not support the hypothesized impact of international strategy on the utility of consensus decision making. The interaction term was not positive, as expected, nor was the term significant.
An examination of the contingency hypothesis, Hypothesis 4, was conducted according to the methodology suggested by Van de Ven and Drazin (1985). First, all variables were standardized in order to establish a uniform scale. Second, ideal TABULAR DATA OMITTED TABULAR DATA OMITTED profiles of the decision-making characteristics and international strategy types were developed. Empirically-derived profiles were used because theoretically-based profiles fail to recognize that the characteristics may take values other than the endpoints of the scale. Furthermore, the overall lack of support for the preceding hypotheses suggests that the theoretical endpoints would likely be misspecifications. Venkatraman and Prescott’s (1990) approach to deriving the ideal profiles was used. A “calibration sample” of the five best-performing business units within each strategy type was developed with the mean of the international strategy index used to classify the businesses by strategy. The means for the decision-making characteristics of each group were then computed. The results, reported in Table 5, may be considered the ideal patterns of design based on the calibration sample. Third, a fit or “distance-effectiveness” index was calculated for each business unit, measured as the euclidean distance between the business unit’s score and the scores of its ideal profile. Thus, this fit index considers the consistency of the business unit strategy and the three decision-making characteristics to the ideal profile. Finally, the relationship between the fit index and business unit performance was assessed. Essentially, as the distance measure becomes greater, less of a fit is indicated and, therefore, lower performance would be expected. Conversely, as the business unit is closer to its ideal profile, the distance measure will be smaller and higher performance would be expected. Thus, a negative correlation between the distance measure is indicative of a consistent fit or design pattern. As suggested by Gresov (1989) and Venkatraman and Prescott (1990), in order to avoid the upward bias that would result from using the same observations in which the ideal profiles were derived to also test the hypothesis, the 10 high performing firms used to develop the ideal profiles were excluded in this step of the analysis.
The results of the systems analysis are reported in Table 6. When performance was defined as sales growth, Hypothesis 4 was supported because both the global strategy and the multidomestic strategy fit indices were negatively correlated to effectiveness. For the ROA performance measure, business units in the multidomestic strategy group showed a negative correlation between the fit index and performance. However, the distance-effectiveness correlation was not significant for the global strategy group. In order to further examine Hypothesis 4 and the results of the preceding analysis, the performance measures were regressed on the decision-making characteristics and the fit index for each international strategy type. The fit index for each equation was significant (p|is less than~.01) with the exception of the global strategy equation predicting ROA, in which the model was not significant. This analysis confirmed that the design deviation was a predictor of financial performance after the main effects of decision-making characteristics and strategy type were controlled.
Table 6
Correlations of Distance Measures and Performance
Performance
International Strategy Type ROA Sales Growth
Global 0.19 -0.42(**)
Multidomestic -0.45(***) -0.32(*)
* p |is less than~ .05.
** p |is less than~ .01.
*** p |is less than or equal to~ .001.
Discussion
The purpose of this study was to determine if an internally consistent design of managerial decision-making characteristics and international strategy is related to organizational performance. The congruence between international strategy and risk taking (H1), openness (H2), and group consensus (H3) was generally not found to influence organizational performance. However, considered as a system (H4), the fit among the managerial decision-making characteristics and international strategy does appear to effect performance. Support for the systems approach suggests that it is the simultaneous and holistic pattern of interlinkages between decision-making characteristics and international strategy that influences business unit performance. This result has two basic implications. First, in considering the design of the belief systems of their organization, managers should evaluate the logical consistency of the total system. It appears unlikely that any single design characteristic will effect performance directly. Second, the results also suggest that, for future research, contingency approaches will likely provide richer explanations of business unit outcomes regarding decision-making characteristics and international strategy, as compared to congruence approaches. At minimum, research examining fit would benefit from recognizing that alternate forms of fit exist and should be directly evaluated (Drazin & Van de Ven, 1985).
Two departures from the theoretical predictions should be noted. First, theoretical arguments suggested that willingness to take risks would be associated with a multidomestic strategy. However, deriving the ideal profiles from top performing business units indicated that though risk taking was indeed part of the decision-making orientation of business units pursuing multidomestic strategies, a substantially higher level of risk taking characterized business units pursuing global strategies. Two interpretations of this result may be suggested. First, a multidomestic strategy may be characterized by greater decision making at a local level. Therefore, less task uncertainty may exist from the focal unit’s perspective because more standardized approaches are feasible given that decisions are delegated to the local units. Second, it may be inappropriate to assume that product innovation and product breadth are more elemental to a multidomestic strategy than the global strategy. It can be argued that product innovation and market coverage, though not pursued for the purpose of differentiation, continue to be sources of advantage of the global strategy. The purpose of the innovation would be to meet or develop homogenous market segments across locations and to develop scale economies with respect to serving these segments. This interpretation implies that understanding the specific nature and purpose of innovation and product breadth are a necessary prerequisite for understanding the level of risk associated with alternate international strategy types. Alternately, it may be that a global strategy does indeed consist of a lower level of product innovation and breadth, but that the need for coordination and integration across locations separated geographically and culturally creates more task uncertainty than the amount reduced through standardized and narrow product offerings.
The second departure concerns the specific dimensions of performance associated with the contingency hypothesis. Support for the contingency hypothesis was found for both ROA and sales growth for the multidomestic strategy, whereas only sales growth was found to be significant for a global strategy. Furthermore, support for Hypothesis 1 existed for ROA, but no support was found for sales growth. Three possible explanations for these inconsistent results are offered. First, the pattern of international strategy and managerial decision-making characteristics may be further contingent on the performance objectives of the business unit. Although there was no clear theoretical rationale to suggest hypotheses specific to each performance dimension, it appears that different ideal designs may exist for different business unit performance objectives. Second, it is also possible that the managerial decision-making characteristics studied were not the critical characteristics to support financial performance for a global strategy. Other aspects of an organization’s belief systems, such as the tolerance for ambiguity, reward and motivational systems, or analytical sophistication, may need to be incorporated into future research designs to capture the relevant system to support a global strategy. Finally, the results may simply reflect the cross-sectional nature of the methodology used in this study. Global strategy is a rather recent phenomenon and many businesses may still be in the processes of developing or implementing their global strategy. Global strategy is recognized as being a long-term strategy requiring an extremely high level of organizational commitment and resources. Thus, the decision to pursue a global strategy may yield, in the short term, poor financial performance. The business is essentially trading-off short-term profits to gain global market share. Thus, the proper managerial decision-making characteristics may be in place but the relationship to financial performance will not be evidenced until a future date. Clearly, this is an area where further research could add significantly to our understanding of the nature of implementing a global strategy.
Considering potential managerial implications, the data of this study support the need to consider the management of the decision-making characteristics of the organization to support specific international strategy types. Managerial selection criteria, expatriate selection and training, and other normative integration practices may, therefore, need to include consideration of the international strategy of the business unit. Alternately, in the evaluation of the strategic choices of a particular business unit, the decision-making characteristics of management may need to be considered as the business unit may have a unique advantage in implementing a particular international strategy type. Furthermore, for the multinational corporation, considerable managerial diversity typically exists due to cultural influences on its international operations. Top management may, therefore, need to also consider the need for some degree of tailoring the strategy by individual business units rather than attempting to pursue totally standardized strategies across business units.
From the perspective of theory development, several limitations of the study suggest areas for further research. First, rather limited dimensions of international strategy were examined. Strategies of the multinational corporation are not necessarily captured fully through a multidomestic|right arrow~global strategy continuum because such an approach fails to consider the competitive positioning of the business from a longitudinal perspective. For example, the multinational corporation that is in the process of becoming a global corporation may require different decision-making characteristics vis-a-vis the corporation that is currently competing globally. Furthermore, the relatedness occurring among business units or between business units and corporate headquarters is not explicitly considered through the global-multidomestic strategy continuum and this relatedness may also affect the “appropriate” managerial characteristics.
Second, although the data suggest that the three decision-making characteristics that were examined are important, the belief system of an organization comprises other dimensions as well. Other characteristics, such as tolerance for ambiguity, interpersonal communication skills, cross cultural adaptiveness, leadership styles, openness of business unit to business unit or business unit to headquarters interdependencies, and attitude toward authority, may also influence the effectiveness by which the organization is able to pursue a particular international strategy. Finally, it is also important to recognize that the decision-making characteristics are embedded within the context of the formal organization. Additional inquiry is necessary to understand interactions between the formal structure/control systems and their influence on the decision making/strategy design.
Future studies could contribute to the understanding of the managerial decision-making characteristics-international strategy relationship through the strengthening of two methodological considerations. Although the reliability and validity assessment provided support for the measures, the use of self-report data limits the conclusions of the study. The use of multi-rater measurements and objective information from secondary sources would increase the confidence in the validity of the theoretical framework. Furthermore, the framework implicitly assumes causal linkages, although the cross-sectional nature of the data obviously precludes testing the directionality of the international strategy and decision-making characteristics linkages. Methodologies incorporating longitudinal data collection would provide for such an assessment.
1 The identified industries were balances, watches and parts, textile machinery, mining, machinery, oilfield machinery, certain consumer electronic products, semiconductors, sewing machinery, electro-medical and x-ray apparatus, synthetic insecticides and fungicides, civil aircraft and parts, and typesetting machinery.
2 Using descriptions of global and multidomestic strategies offered by Porter (1985) and Prahalad and Doz (1987), each respondent was asked to select the description that best characterized their business unit strategy: (a) a relatively slow rate of technological change, a high level of responsiveness on a country-by-country basis and products customized to meet local tastes and preferences, (b) a high rate of technological change, exploiting global scale economies, and responsiveness to international standardized product demand.
3 Readers are referred to Schoonhoven (1981) and Govindarajan and Gupta (1984) for a more complete discussion of the procedure. It is important to note that for equations estimated with the interaction term, the results vary with changes in the points of origin of the two main variables and therefore the coefficients of the independent variables are not interpretable (Southwood, 1978).
4 Previous research has indicated that business unit size is associated with performance differences (Cummins & King, 1973; Govindarajan, 1988). Thus it was considered important to test the hypotheses after controlling for the confounding influence of business unit size on performance.
References
Abell, P. 1978. The task and power determinants of decentralization between headquarters and United Kingdom subsidiaries of international corporations. In A. Negandhi & B. Wilpert (Eds.), Work organization research: American and European perspectives: 91-120. Kent, Ohio: Kent State University Press.
Andrews, K.R. 1971. The concept of corporate strategy. Homewood, IL: Dow Jones-Irwin.
Barney, J.R. 1986. Organization culture: Can it be a source of sustained competitive advantage? Academy of Management Review, 11: 656-665.
Bartlett, C.A., & Ghoshal, S. 1986. Tap your subsidiaries for global reach. Harvard Business Review, 64(6): 87-94.
Bartlett, C.A., & Ghoshal, S. 1989. Managing across borders: The transnational solution. Boston, MA: Harvard Business School Press.
Biggadike, R. 1979. The risky business of diversification. Harvard Business Review, 57(3): 103-111.
Bourgeois, L.J. 1980. Performance and consensus. Strategic Management Journal, 1: 227-248.
Chandler, A.D. 1962. Strategy and structure: Chapters in the history of the American industrial enterprise. Cambridge, MA: MIT Press.
Cvar, M. 1984. Competitive strategies in global industries. Unpublished doctoral dissertation, Harvard University.
Deal, T., & Kennedy, A.E. 1982. Corporate cultures. Reading, MA: Addison-Wesley.
Dess, G.G., & Davis, P.S. 1984. Porter’s (1980) generic strategies as determinants of strategic group membership and organizational performance. Academy of Management Journal, 27: 467-488.
Dess, G.G., & Origer, N. 1987. Environment, structure, and consensus in strategy formulation: A conceptual integration. Academy of Management Review, 12: 313-330.
Donaldson. G., & Lorsch, J.W. 1983. Decision making at the top. New York: Basic Books.
Doz, Y.L. 1986. Government policies and global industries. In M. Porter (Ed.) Competition in global industries: 225-266. Boston: HBS Press.
Doz, Y.L. & Prahalad, C.K. 1988. Quality of management: An emerging source of global competitive advantage. In N. Hood & J. Vahlne (Eds.), Strategies in global competition: 345-369. London: Croom Helm.
Dubin, R. 1978. Theory building, New York: Macmillan.
Dutton, J.E., & Duncan, R.B. 1987. The creation of momentum for change through the process of strategic issue diagnosis. Strategic Management Journal, 8: 279-295.
Egelhoff, W.G. 1982. Strategy and structure in multinational corporations: An information processing approach. Administrative Science Quarterly, 27: 435-458.
Fayerweather, J. 1969. International business management: A conceptual framework. London: George Allen & Unwin.
Finkelstein, S., & Hambrick, D.C. 1990. Top-management-team tenure and organizational outcomes: The moderating role of managerial discretion. Administrative Science Quarterly, 35: 484-503.
Fry, L. & Smith, D.A. 1987. Congruence, contingency, and theory building. Academy of Management Review, 12: 117-132.
Galbraith, J.A. 1973. Designing complex organizations. Reading, MA: Addison-Wesley.
Galbraith, J.A. 1977. Organization design. Reading, MA: Addison-Wesley.
Galbraith, J.A., & Nathanson, D.A. 1978. Strategy implementation: The role of structure and process. St. Paul, MN: West Publishing.
Ghoshal, S. 1987. Global strategy: An organizing framework. Strategic Management Journal, 8: 425-440.
Ghoshal, S., & Nohria, N. 1989. International differentiation within multinational corporations. Strategic Management Journal, 10: 323-337.
Gordon, G.C., & Cummins, W.M. 1979. Managing management climate. Lexington, MA: Lexington Books.
Govindarajan, V. 1986. Decentralization, strategy, and effectiveness of strategic business units in multi-business organizations. Academy of Management Review, 11: 844-856.
Govindarajan, V. 1988. A contingency approach to strategy implementation at the business unit level: Integrating administrative mechanisms with strategy. Academy of Management Journal, 31: 828-853.
Govindarajan, V. 1989. Implementing competitive strategies at the business unit level: Implications of matching managers with strategies. Strategic Management Journal, 10:251-269.
Griffin, R.W. 1980. Relationship among individual, task design, and leader behavior variables. Academy of Management Journal, 23: 665-683.
Gresov, C. 1989. Exploring fit and misfit with multiple contingencies. Administrative Science Quarterly, 34: 431-453.
Gupta, A.K. 1987a. Matching managers to strategies: Point and counterpoint. Academy of Management Journal, 25: 215-234.
Gupta, A.K. 1987b. SBU Strategies, Corporate SBU relations, and SBU effectiveness in strategy implementation. Academy of Management Journal, 30: 477-500.
Gupta, A.K., & Govindarajan, V. 1984. Business unit strategy, managerial characteristics, and business unit effectiveness at strategy implementation. Academy of Management Journal, 27: 25-41.
Hambrick, D.C. 1979. Operationalizing the concept of business-level strategy in research. Academy of Management Review, 5: 567-575.
Hambrick, D.C., & Mason, P.A. 1984. Upper echelons: the organization as a reflection of its top managers. Academy of Management Review, 9: 193-206.
Hamel, G., & Prahalad, C.K. 1985. Do you really have a global strategy? Harvard Business Review, 63(4): 139-147.
Hedlund, G., & Rolander, D. 1990. Action in heterachies: New approaches to managing the MNC. In C. Bartlett, Y. Doz, & G. Hedlund (Eds.), Managing the global firm: 15-46. London: Routledge.
Heller, F.A. 1971. Managerial decision-making: A study of leadership styles and power-sharing among senior managers. London: Tavistock.
Heller, F., Drenth, P., Koopman, P., & Rus, V. 1988. Decisions in organizations: A three-country comparative study. London: Sage.
Hofstede, G. 1976. Nationality and espoused values of managers. Journal of Applied Psychology, 61(2): 148-155.
Hout, T., Porter, M., & Rudden, E. 1982. How global companies win out. Harvard Business Review, 60(5):98-108.
Janis, I. 1972. Victims of groupthink: A psychological study of foreign-policy decisions and fiascos. New York: Houghton Mifflin.
Kerr, J.L., & Jackofsky, E.F. 1989. Aligning managers with strategies: Management development versus selection. Strategic Management Journal, 10: 157-170.
Kobrin, S. 1991. An empirical analysis of the determinants of global integration. Strategic Management Journal, Summer: 17-31.
Kogut, B. 1985a. Designing global strategies: Comparative and competitive value-added chains. Sloan Management Review, Summer: 15-28.
Kogut, B. 1985b. Designing global strategies: Profiting from operational flexibility. Sloan Management Review, Fall: 27-38.
COPYRIGHT 1992 JAI Press, Inc.
COPYRIGHT 2004 Gale Group
