National Culture and Research and Development Activities
Couto, João Pedro
The global context in which firms develop and implement business strategies has changed significantly. The knowledge-based economy has made multinational companies increase the transfer of innovation processes to their foreign subsidiaries in order to adapt their products and services to local needs and to make use of the knowledge locally available. The initiatives to optimise the innovation capabilities and obtain competitive advantages have led to a research and development effort at the subsidiary level. Indeed, the benefits of decentralized R&D have been mentioned in Pearce (1997), Birkinshaw (2000), Holm and Pederson (2000), Jones and Davis (2000) and Birkinshaw and Hood (1997, 1998a, 1998b).
This paper builds on the responses of subsidiaries’ managers and analyses R&D activities of the subsidiaries of multinational companies (MNC). The relationship of these activities with the culture of the host country and the management model, provided by the geographic origin of the multinational company, is explored. In order to analyse these relationships, we use a data set collected from a survey sent to MNC’s subsidiaries in five European countries: Germany, France, Sweden, United Kingdom and Portugal.
The results allow us to establish two findings. Firstly, cultural dimensions influence research and development activities at the subsidiary level. In particular, the higher the uncertainty avoidance and individualism and the lower the power distance and masculinity indexes the higher the level of research and development activities performed. Secondly, the type of management model influences the organization of research and development activities of the subsidiaries. These results contribute to this field of research by gathering empirical evidence that sustains and integrates national culture relations and R&D.
There exists a wide and growing body of references in the literature regarding the multinational firm motivations to internalise R&D, the decentralization process and its relation to location advantages (e.g., Casson and Singh, 1993; Pearce, 1989; Cantwell, 1989; Teece, 1976; Ronstadt, 1978).
The location of R&D can be attributed to various motives: (1) drivers for the internationalisation process; (2) firm’s geographic orientation in R&D activities; (3) strategic role of the foreign subsidiary (see Casson and Singh, 1993; De Meyer, 1992, 1993; Dunning and Narula, 1995; Cantwell, 1992). The R&D activities of subsidiaries can be very different. They can range from the role of receiver of technology from other locations of the multinational network, to an adaptive role to local needs, or an autonomous activity (Davis, 2000; Pearce, 1997; Taggart, 1997).
In the context of the subsidiary strategy, the role of the subsidiary’s manager in developing initiatives and the relation with the affirmation of R&D capability, has been focused on various studies (Birkinshaw and Hood, 1998a, 1998b, 1997; Holm and Pedersen, 2000). In this sense, the importance of R&D has been mentioned as a key factor to the determination of the subsidiary’s strategic role (Forsgren et al. 1992; Kogut and Zander, 1993; Dunning, 1992; Papanastassiou and Pearce 1996, 1997). The influence of national culture on the development of R&D activities is normally related to Hofstede’s (1987) cultural dimensions and has been associated with the processes of invention and innovation (Shane, 1992, 1993) and new product development (Nakata and Sivakumar, 1996; Morris et al. 1994).
The Hofstede’s (1987) cultural dimensions are five. The first dimension is “power distance” defined as the level of acceptance of an uneven distribution of power in the society. The second indicator is “individualism” which is defined as the importance of the individual as compared with collective goals and efforts. The third indicator is “masculinity” defined as the level of assertiveness that is promoted in the national culture by either gender. The fourth indicator is “uncertainty avoidance” which is related to the level of uncertainty with regards to future events that people from a specific national culture are willing to accept. Altogether, these indicators seem suitable to classify and distinguish national cultures. Hofstede also developed a fifth indicator in order to distinguish oriental cultures from the occidental counterparts, which is designated by Confucianism. This indicator relies on the importance of honour, virtue and the recognition of group values.
The influence of local culture on innovation and its importance to R&D competitive advantages has also been pointed out by Nakata and Sivakumar (1996), when they refer that implicit in the selection of any location is the selection of a national culture.
Several studies on the relationship between cultural dimensions and innovation activities have been conducted by Kedia et al. (1992), Morris et al. (1994), Nakata and Sivakumar (1996) and Shane (1992, 1993). These studies have suggested that low power distance and uncertainty avoidance and high masculinity and individualism can foster higher innovation, thus confirming Hofstede (1987) statement that societies with lower power distance and lower uncertainty avoidance exert a positive influence on innovation.
A global framework which integrates these factors is proposed by Jones and Davis (2000), who established a classification of the role of R&D for the subsidiaries based on contingency factors. The roles range from: locally-oriented support and adaptation; locally-oriented R&D; globally-oriented R&D; competitor market/technology scanning; to other non-technologically related roles.
They relate these roles to the contingency factors such as the type of motivation, geographic orientation, and R&D mission and establish a theoretical relationship with Hofstede’s cultural dimensions. According to these authors the differentiation from locally-oriented support and adaptation to a globally-oriented R&D, means a change from high power distance, uncertainty avoidance and masculinity to a lower level of these indicators. The individualism would change from low to high and the Confucian dynamism would remain positive in all technological roles.
Based on these studies we consider that the relation between national culture and research and development can be critical and that there can be advantages from national cultures to specific phases of the process which we evaluate on the following hypotheses.
In this paper we put forward two hypotheses: a first, regarding the effect of cultural dimensions on the R&D of the subsidiary; and second, concerned with the influence of the management model of the multinational firm on R&D activities of the subsidiaries.
In hypothesis one it is stated that: the national cultural dimensions influence the R&D activities of the subsidiaries.
This first hypothesis relies on vast literature related to the definition of the national culture where the works of Hofstede (1987), Ronen and Shenkar (1985), Trompenaars (1994), or Schneider and Barsoux (1997) are included. On the other hand, the features of the technology and innovation of the multinational company have been stressed by authors like Forsgren at al. (1992), Kogut and Zander, (1993), Dunning, (1992), Papanastassiou and Pearce (1996, 1997).
The technological capabilities have been considered as a key factor to the definition of strategic roles of the subsidiaries by Forsgren et al. (1992); Kogut and Zander (1993); Birkinshaw and Hood (1997); Dunning (1992); Dunning and Narula (1995); De Meyer (1993); Papanastassiou and Pearce (1996, 1997).
The relation of national culture to R&D can occur in two different aspects: first, the process of interdepartment cooperation and; second, the propensity for incremental versus radical innovation. These aspects have to do with the element of conflict and risk that are involved in the process of innovation and could be favoured by cultural adaptation. Roure (2001),
The relationship involved in the cooperation process needed in R&D can be found related to information availability (Brown and Eisenhardt, 1995); pluralism needed to better problem solving solutions (Albach, 1994); and competitive development (Hauptmann and Hirji, 1996). Marquis (1969) already stated the idea that there is a need for cooperation between the forces of “technology push” and the “market pull”. Empirical research shows that cooperation is needed between the various departments (Souder, 1987; Ayers et al. 1997). The process of cooperation inter-functions can be resumed by the aspects of open communication, a feeling of interdependence; the capacity to manage conflicts; the level of cohesion and a focus on the competitiveness with the outside environment. (Henke, 1993; Thamhain, 1990; Gladstein, 1984, Pinto and Pinto, 1990). The link between these aspects of interfunction cooperation and national culture are associated with the question of individualism and versus collectivism. This can be related with the importance of more group decisions (Hall and Hall, 1984, 1990) that are present German companies, and the more individualistic approach of personal contribution in French companies (Roure, 2001).
A second aspect of R&D that can be related to national culture is the manner innovation is produced. In fact, the aspect of uncertainty avoidance can be determinant on the fact that a subsidiary develops a more radical research or more development. If the new characteristics of a product are of the most important elements of R&D (Brown and Eisenhardt, 1995; Cooper, 1994), namely the development of superiority at the eyes of consumers (Cooper and Kleinschmidt, 1987; Song and Parry, 1997), which can foster an important competitive advantage (Foster, 1986; Vidal, 1995), still this process is riskier. The level of uncertainty avoidance in Japan and Germany tends to be higher that in Latin countries (Bollinger and Hofstede, 1987). This can be observed by the way German companies develop their business plans. Ali et al. (1993) refer that what can lead to less investment in some product’s innovation is the uncertainty of the commercial results of pioneers. The assumption of risk tends to be higher in American companies that in Japanese companies favouring a more radical innovation in America than in Japan.
Another aspect included in these relations between radical innovation and national culture is also the question of individualism. In fact the individualism leads to experimentation and more openness that collectivism (Shane, 1992, 1993). The spirit of competition present in individualistic societies can increase innovation (Herbig and Miller, 1992). The idea of making a difference and make personal contributions are very present in the process of radical innovation and not so much in the process of development.
In the same line of these arguments Hill et al. (1995) reinforce these ideas by recognising that American national culture of individualism and autonomy can undermine the process of reengineering and the innovation process. They state that these aspects, although positive in creation of new products and ideas, are less favorable when it comes to process innovation which is a fundamental element of implementing effective production response for these new products.
The inter-functional barriers and lack of trust common in the American competitive business environment reduce knowledge transfer and the development under intense competitive pressure leads to a culture of speed that is not always favourable to innovation implementation.
Another aspect is the question of national culture and consumer innovativeness developed by Steenkamp et al. (1999). These authors relate a more open perspective of innovations in consumers in countries where the levels of individualism and masculinity are higher. They relate this to more freedom of experimenting and more value on achievement and success. The level of uncertainty avoidance is associated with the presence of a more conservative ambience that tends to reduce experimentation. They also relate these aspects with the national tendency ethnocentrism, but refer that this is moderated by low collectivism.
The fact that experimentation in consumer behavior varies according to national culture elements, leads to the need to develop R&D in countries where the markets can foster higher levels of response and contribute to the process performance.
In light of these ideas we can argue that an individualist and more risk taking society will tend to produce more radical research, based on cooperation and inter-function communication while a more collectivist and uncertainty avoidance society will tend to do better, in developing a concept. This reinforces the element of subsidiary adequacy in R&D to national culture as an element determinant to the success of the and to the overall multinational R&D capacity. More specifically, in this hypothesis we consider that the R&D activities of the subsidiaries are going to be enhanced by their presence in a country that tends to have low power distance, high individualism, low masculinity and low uncertainty avoidance.
Hypothesis two states that: the type of management model of the parent company influences the R&D activities of the subsidiaries.
The second hypothesis introduces the type of management model of the parent company as a determinant of the subsidiary R&D activities. For this purpose, we consider the classification of the companies into three distinct models, according to the proximity of the parent: a European model, an American and a Japanese model.
The geographic proximity is usually viewed as a determinant of foreign direct investment and can be associated with cultural proximity. In this sense, the Japanese “Kaisha” and the Southeast Asia economies suggest that when compared with Western economies there is more than one way of organizing business activities and that particular institutional environments can promote local adaptations. In terms of market organization, Japanese firms develop a more extensive network of companies, associated with family and loyalty relationships, than the more professional management of the American firms, where activities are managed as “portfolio” with planning and control mechanisms (Whitley, 1992).
While in occidental countries competition is a common form of market organization in the oriental counterparts there are more cooperative efforts with companies relying on mutual dependence, sharing information and developing long-term relations (Orru et al., 1989). In terms of decentralization of functions and authority there is more diversity in American than in Japanese firms. A result of this model is that control is exercised in a more personalized manner and based on personal contact and common values that in formal processes (Kagono, et al. 1985). Calori and De Wool (1994) compared the American and Japanese managers with a distinctive European management model. Despite the diversity of countries in Europe they consider that the approximation to a social perspective of the market economy as well as a more active involvement of the workers in the decision process are distinguishing factors of a European management model. Furthermore, these authors state that American and Japanese companies operating in Europe do not follow this model.
Abramson et al. (1996) made a comparison of Canadian, American and Japanese managers regarding their cognitive process and concluded that Canadians are more theoretical orientated, and that Japanese are more long-run oriented than American managers. These authors state that these differences can be important to the type of actions taken, organization design and type of performance indicators selected.
Yip et al. (1997) examined the influence of management processes on the globalisation of industries and the definition of global strategies and concluded that there are significant differences due to national effects. Furthermore, they observe that European and Japanese companies use more global strategies than American companies do. In light of these ideas we consider that the element of national culture is present in R&D in another manner than the influence of local culture on subsidiary, but in terms on multinational organization.
This view relates national culture to organizational culture and its influence on R&D is determined by the manner the multinational company deploys its power in relation with the subsidiaries. This hypothesis considers that the more centralized HQ-Subsidiary relation in Japanese companies will determine that local subsidiaries perform less R&D and less contribution is derived from local host nations. The American type of management model will tend to be more open and therefore produce more R&D at subsidiary level. At the reverse of Japanese Multinationals we position European Multinationals. These will tend to be more focused in the gains of local subsidiary R&D, namely companies from more collectivist and feminine societies that combine these national culture characteristics with moderate uncertainty avoidance and low power distance.
METHODOLOGY OF THE STUDY
The first step in defining the sample was the identification of Hofstede’s indicators that were used to select the countries of this study. The countries selected were United Kingdom, Sweden, Germany, France and Portugal, representing different groups of European nations: Anglo-Saxonic, Nordic, Germanic, Central Europe and Southern Europe. To identify the countries used in the study, a sample of subsidiaries was taken according to the proportion of foreign subsidiaries existing in each country.
The main data source was a questionnaire sent to the foreign subsidiaries in the selected countries. This questionnaire was directed to the managers of the subsidiaries with the indication that they should themselves supply the information. Otherwise, the questionnaire should be filled by someone involved in the planning processes.
From this questionnaire we obtained information on the location of the parent company in order to determine the management model. With respect to the organisation of activities of research and development the following information was provided to us: 1= activity not performed; 2= activity performed only to the host country and 3= activity performed to several countries. The management model was classified as the European model, the American model and the Japanese model.
THE EMPIRICAL MODEL
The percentage of responses to the questionnaire was equal to 23.1% (see Table 1). We can also verify that the subsidiaries located in Portugal and the United Kingdom were the ones with higher response rates.
Analyzing the distribution of the responses by country we can observe that France was the country with the highest percentage of valid questionnaires (28.6%), followed by Germany (26.8%), United Kingdom (18.2%), Portugal (16.9%) and Sweden (9.5%). Given that that the distribution of responses by country do not diverge significantly from the initial setting of the questionnaires, we can consider that the distribution of responses does not undermine the significance of the sample.
The statistics for the average, standard deviation and correlations of the variables can be found in Table 2. As we can see, the more significant correlations are the ones observed between the different Hofstede indicators.
The estimation results are included in Table 3. The null hypothesis that the exogenous variables have no explanatory power is rejected at the 1% level, since value of the test (Chi-squared(6) = 79.6) is far above the critical value of 16.8.
The results indicate that the higher the individualism and the uncertainty avoidance indexes the higher the probability of a firm to choose the level 3 (activity performed to several countries) and the lower the probability of choosing the level 1 (activity not performed). The reverse is valid for masculinity and power distance indexes.
Regarding the geographic variables (i.e the management model), United States and Japan have a negative impact on the probability of choosing level 2, as compared with Western Europe (used as the reference category in Table 3). However, the difference between Europe and the United States is not statistically significant at 1% level.
The problem with the interpretation of the parameters of the ordered probit model is that the signs only indicate the sign of the impact of the corresponding variable on the probabilities of the lowest and the highest levels of the scale. In order to overcome this shortcoming and determine the effects on the intermediate levels (in this case, level 1) it is usual to calculate the marginal effects for a reference category.
Considering the average values of the variables and that the subsidiary is from Europe we can analyse the corresponding marginal effects. From Table 4 we can see that an increase in the levels of individualism and uncertainty avoidance reduce the probability of an average subsidiary to be included in level 1 (activity not performed) and increase the probability of being included in level 3 (activity performed to several countries), such as stated above. The levels of masculinity and power distance have the reverse effect. We find no evidence that these variables have an impact on the level 2, since any change on them switches the probability straightforwardly between the level 1 and level 3 (see Table 4).
The fact that the parent company is from Japan reduces the probability of being in level 3 and increases the probability of being in level 1. The same is true in the case of the United States, but in a much smaller degree. This reinforces the idea that Japanese multinationals centralise more their research and development activities than European multinationals. American multinationals align at an intermediate level that does not distinguish them from the European companies, in statistical terms.
DISCUSSION AND CONCLUSIONS
In this paper we have examined the effect of national culture and the type of management model on research and development activities performed by the subsidiaries of multinational firms. The results support: (1) the confirmation of the hypothesis that the cultural dimensions of the host country influence the type of research and development performed by the foreign subsidiaries; (2) the type of management model, associated with the origin of the multinational company, can also influence the nature of research activities performed by the subsidiaries. At this level, the influence of the Japanese model differs from that of the American and European counterparts. The latter two models are not statistically different.
These results are in accordance with the literature. Jones and Davis (2000) refer that a globally-oriented research and development would be associated with a lower power distance, uncertainty avoidance and masculinity and to a higher individualism. This type of association is also made by Shane (1993), Nakata and Sivakumar (1996). The results of the empirical analysis show a relation between national culture and R&D in two levels: first, the relation of national culture of the host country with the activities of R&D of the subsidiary and; second, the relation of national culture in the multinational, derived from the home country’s influence on corporate culture. In the first case we can verify that the high level of individualism leads to higher efforts in producing R&D contributions; the low power distance permits the inter-function cooperation and problem solving environment, the low masculinity leads to a type of more collaborative effort that seams to be important to the establishment of organized R&D at the subsidiary and finally, the high uncertainty avoidance is present showing that R&D at subsidiary level tends to be more present when subsidiaries develop and improve products to the local market rather than when it’s a more radical R&D contribution.
The fact that multinational companies tend to be influenced by the national culture of the country of origin shows that there is substantial connection in terms of R&D management between the multinational corporate culture and their home national culture. This is referred by Bartlett and Ghoshal (1989), who associate this type of organization to historical heritage, being characterized by centralization of capitals, resources and responsibilities and giving less freedom to create new products and define strategies by the local subsidiaries. This is an aspect that must be understood by multinational management in order to fully respond to R&D challenges. It is reasonable to argue that in some cases multinational management does not identify this type of problem as a limitation to company competitiveness.
As practical implications of this study we must highlight the following three levels: (1) management of the multinationals (2) management of the subsidiaries and (3) local governments. In the case of multinational management it is important to consider when choosing a location for the subsidiary that national culture has an impact on the results of research and development. The absence of this variable can reduce the understanding of why certain initiatives are more successful that others due to location. In the case of the managers of the subsidiaries, the knowledge of the influence of national culture on R&D can help them orient their efforts. In the case of local governments, this relationship between national culture and research and development activities can represent an additional argument to attract foreign investments to certain countries.
As questions requiring further research and in order to overcome some of the drawbacks of this study, we can point out the use of a broader sample of subsidiaries including other countries with a wider number of regions which could also allow for an evaluation of the effects of Confucianism.
Copyright St. Louis University, John Cook School of Business, Boeing Institute of International Business Spring 2004
Provided by ProQuest Information and Learning Company. All rights Reserved.