The organizational structures of companies vary synchronically as well as diachronically as a function of both their acquired skills and the transformations they must make. It therefore seems that each of them, to a certain degree, develops its organizational capabilities autonomously (Chandler, 1992), such that each company may develop its own particular trajectory, with its own particular bifurcations and blockages, in its attempts to secure competitive advantage. This evolutionary perspective suggests that multinationalization can be considered as a learning process (Casson, 1995) which leads companies through a particular succession of configurations. The transition to more complex organizational forms is then linked to the experiences particular to each company (path dependency) during a process which in reality has little to do with some process of selecting an optimal trajectory (Hedlund, 1993).
On the basis of this thesis, in the first section of our paper we propose a typology of forms of multinationalization, which will later be used to interpret the strategies of automobile producers. Key configurations are identified which are presented as possible stages of internationalization. The paper then subjects this framework to the test of the history of the multinationalization of the major automobile producers. Even if globalization represents a goal for all producers, few amongst them have actually reached this stage, and the issue of which trajectory should be pursued to reach it remains wide open. Due to their earlier multinationalization, the American companies have developed a clear trajectory towards globalization which can serve as a reference point, if not an example, for the examination of the other producers. However, the development of the American companies does not trace out some obligatory path for all companies to follow. Moreover, examination of the different paths which the American companies have themselves followed reveals an unsuspected variety in their geographical strategies. Similarly, Japanese producers embarking upon the internationalization of production in an environment of globalization are also following different paths. Their activities are both reinforcing and modifying the globalization which constrains all companies. Accordingly, the particular feature of these companies has been to embark on a strategy of globalization immediately, leap-frogging the intermediate phase which characterized the American companies. As far as European producers are concerned, they continue to define themselves principally in terms of a mono-regional space, reflecting the difficulties they have faced in integrating themselves into the process of globalization.
The framework for interpreting the multinationalization of companies thus depicts the wide variety of forms found within the automobile industry: from the earlier internationalization of the American producers with their own varied trajectories, to those of the Japanese producers obliged to adapt to local constraints, and finally the limited path pursued by the European producers
The spatial hierarchical organization of the multinational company - which encompasses both the principles of hierarchical organization and the level of hierarchical control.- provides an analytical path for the identification of key configurations.
The spatial hierarchical organization of the multinational company
As far as the principles of hierarchical organization are concerned, there is a qualitative difference between world-wide internationalization ('mondialisation') and globalization ('globalisation'), according to the distinction developed by Charles Albert Michalet (1990). The principle of world-wide internationalization refers to a process of expanding the company's sphere of operation beyond the country of origin without changing its initial structures. With world-wide internationalization, space is structured such that there is an asymmetric relationship between centre and periphery, accompanied by a flow of unilateral exchanges; the set of countries and regions outside the country of origin are considered to be endowed with generic resources (in the sense of Amendola and Gaffard, 1988) for the company. By contrast, globalization is accompanied by a criss-cross flow of exchanges within the context of a polycentric system, in which each centre is considered in terms of its own specific resources. The concept of specific resources takes into account the learning processes which contribute to the evolution of competencies or organizational capabilities developed locally. By combining the competencies of different spaces, the coordination of the learning processes, which is itself an organizational capacity of the company, becomes an important source of competitive advantage in a context where the environment requires a high level of ability to react to change.
Subsumed within each of these principles of hierarchical organization, there is also a variable degree of control exercised by the company's central management over its international activities (see Fujimoto, et alii, 1994). The level of control generally depends upon the degree to which the company is spatially homogeneous; the greater the degree of homogeneity, the greater the degree of hierarchical control. Conversely, heterogeneity makes centralised management inefficient, since when identical control procedures are used it is not possible to grasp all the pertinent information required to make appropriate decisions. Amongst the main factors behind heterogeneity are economic constraints such as the geographical variety of products, which is related to the characteristics of demand in the different national and regional markets, and the more institutional factors related to political constraints such as trade policies or national regulations on the activities of multinationals.
By combining these two modes of interpreting the spatial hierarchical organization of multinational companies, we can construct a framework of different forms of multinationalization (Bélis-Bergouignan, Bordenave, Lung, 1994) which suggests four major configurations: world-wide, multi-domestic, multi-regional and trans-regional (Figure 1).
|Principles of hierarchy||Degree of hierarchical control|
|world-wide internationalization||multi-domestic company||world-wide company|
|globalization||multi-regional company||trans-regional company|
The four configurations above are not equal to each other, since every process of multinationalization necessarily begins with the world-wide mode. Hence the key configurations are the possible steps in the internationalization of companies when they move from world-wide strategies towards global strategies, following different paths as they do so.
The dynamic of the world-wide company
The term world-wide company does not presuppose that the presence of the company is immediately spread across the whole world, but rather that the company's spatial organization is homogeneous and ethnocentric (in the sense of Perlmutter, 1965). The company reproduces its domestic mode of functioning at the international level: multinationalization is simply an expansion of the initial organizational structure, or a duplication, which is why the company is homogenous over space. Nevertheless, homogeneous does not mean identical, since the inevitable diversity of the countries and the regions in which the company invests is taken into account, and the company develops an international division of labour on the basis of it.
Given the diversity of possible forms of the world-wide company this configuration can be maintained through reorganizations which do not undermine the hierarchical principles structuring the company's space. Nevertheless, as a result of environmental change and the effects of competition, the multinational company may evolve to surpass this configuration and progress towards more complex forms of spatial organization. Yet this is only possible if it develops organizational learning processes capable of opening up new alternatives; the variety of organizational forms generated in this way attest to the competencies and modes of organizational flexibility which characterize each of them.
If the company evolves towards a weakening of hierarchical control while maintaining a spatial structure based on a centre-periphery relationship, the configuration becomes 'multi- domestic' (following Porter). If the company begins to recognize the interdependence among the different production regions which characterizes the globalization process, its spatial structure becomes polycentric. Within this general concept two major configurations can be identified, depending on the degree of hierarchical control in the company: the company.
The multi-domestic company
The multi-domestic company permits its different national subsidiaries a certain degree of autonomy in decision making, and so space is heterogeneous. The multi-domestic company thus adapts its products and its rules of functioning to local conditions, which leads to a strong geographical differentiation in the characteristics of goods offered in the different markets (specific ranges and products) and to a decentralization of company management.
And yet, although the space is heterogeneous, and despite its decentralized management, the company retains a single centre. Although they are granted a relative autonomy, the foreign investments remain marginal in that they are mainly viewed by the head office as sources of financial income. They are not viewed as possessing any specific competencies, notably in terms of technology and know-how, which could be transferred to the centre. Consequently, there remains a uni-lateral relationship of domination with no reciprocity between the spaces, which are managed in parallel with no interdependent production.
The multi-regional company
The multi-regional configuration is distinct from the previous configuration in that the company is organized into different regions that are inter-related. The disparities between countries and regions, together with the segmentation of markets, can impede homogeneous management and favour a decentralization of the company's principal economic functions to the level of the regional spaces. During its earlier learning necessary organizational capabilities have been developed, and each of the principal production regions can be managed autonomously (specific range, local design and production, management and alliances), even if various initiatives remain under the control of the headquarters.
However, global coordination is not limited to this passive form of control, which risks being inadequate in the face of the centrifugal forces of autonomization. The interdependence of the different regions is attained through the centralization of certain activities in a single region and through the exchanges of goods and information among the regions. It is nevertheless assumed that the forces tending towards local adaptation are now more powerful than those tending towards global integration.
The trans-regional company
Conversely, if the factors leading to the integration of all the company's activities at the world level (economies of scale, geographic convergence of markets, and so on) predominate, the configuration is that of the trans-regional company. In this case, hierarchical control is much greater, and the company's geographical organization tends towards homogeneity. Different regions are recognized as spaces of specific competencies which can be coordinated within the context of an immediately global approach to the company's activities and its network of alliances. A world-wide range of products is sold in the different markets. In this context, each region is involved in more than a limited sphere of activity, in a whole set of activities based on its own logic and dynamics, in relation to which it exercises its own competencies, such as product design and launch. This polycentrism does not exclude a preeminent headquarters for the strategic decision making which directs the future of the company; though this is a different type of command (a 'heterarchy', as suggested by Hedlund), and another form of hierarchical management.
These four key configurations help us place in perspective and interpret the history of the internationalization of the automobile industry.
Ford's trajectory serves as an example by reason of its seniority, the universalist claims it made from the start, and the company's apparently well planned process of internationalization. The particular position historically acquired by Ford is reflected in its proportion of foreign production, higher than that of its competitors. In 1994, for example, 37,1% of Ford's[5 ]production took place in Europe and 54,6% in North America (including Mexico) (see Appendix). This explains why analyses of the American 'model' grant a significant role to the development process of this company, and we shall attempt to describe the crucial sequences and bifurcations that characterize it. The Ford trajectory will nevertheless be contextualized by comparison with the trajectory of General Motors (GM), which represents an alter-ego that opted for a different strategy. Within the American model, GM has a pattern of internationalization as broad and long as Ford, but slightly less developed. The third major American company, Chrysler, represents a case of internationalization first aborted and then resumed. The experience of Chrysler will not be taken up in this paper, since its narrow scope does not provide sufficient elements to constitute an alternative American model.
The development of Ford's principles of spatial organization can be analyzed in terms of an historical trajectory describing a continuous cycle (Figure 2). Here, globalization represents both a deliberate strategy and a response to constraints linked to the evolution of competition in the market. Globalization provoked an early and innovative bifurcation in the multinationalization strategy. The first part of the trajectory (the sequence from 1 to 2), which spans the period between the 1920s and the 1960s, saw the company move from a world-wide export strategy to a multi-domestic structure based on an adaptation of spatial organization to the fragmentation of nations. The second trajectory (the sequence from 2 to 3, then from 3 to 4), linked to a perception of the world space as globalized, would be structured around attempts to elaborate an organization that could surpass national cleavages, on the basis of a strategy of regional integration.
The world-wide strategy and national fragmentation
International expansion formed an essential part of the Fordist trajectory from its inception, in the sense that Ford explicitly developed world-wide strategies from its beginnings. The patterns of geographical investments were tied to the intrinsic characteristics of the Fordist model. The manufacture of a single product, the application of the principle of standardization, homogeneity, and an indefinite expansion of demand at the world level generated a strategy based on exports followed by investments in overseas assembly plants so as to reduce transport costs. The main aim of the strategy was to seek external markets, and it was supported by the simultaneous possession during the 1920s of comparative advantages peculiar to the American industry compared to Europe, and competitive advantages peculiar to Ford compared to its American competitors.
During this period Ford was a pioneer (in the sense of Chandler's 'first mover'). The company had been manufacturing in Europe since 1911. Ford rigidly transferred its single product, the model 'T', together with its methods of organization and production as these developed in the parent company (Tolliday, 1996). Its European manufacturing plants, at Manchester in England and then in several other countries, at first simply assembled vehicles sent from the parent company in the form of component kits. However, following the first world war, protectionist trade barriers in most countries forced Ford to pursue locally integrated production. It is notable that a geographical form of organization based on the centre-periphery model represented the desired, if not entirely realized, transposition of the organizational form prevailing in the United States, where there was an identical pattern of assembly factories spread across the country, satellites of the central sites in Detroit. Hence, the dispersal of production, reflecting Ford's hegemonic claims, represents the form of managing space characteristic of an expansion of the standardization model.
The standardization model worked to Ford's advantage until the mid-1920s, largely due to its unidimensional coherence in contrast to the greater complexity of the GM structure. All the same, Henry Ford's intransigence over the single product and organizational forms soon brought dangerous consequences for the company in both overseas and domestic markets, where it was outpaced by competitors. GM was efficiently organized under the direction of Sloan (Chandler, 1962) and became firmly established as the leading automobile producer in the United States and the world. The opposition Ford encountered in the United States increased its vulnerability overseas.
The heterogeneous structure adopted by GM generated a less hierarchical approach, and its subsidiaries were more autonomous, based on a more flexible system of production (Hounshell, 1984). In contrast to the form of internationalization adopted by Ford, GM grew externally by absorbing competitors (Vauxhall and Opel in Europe, Holden in Australia). This is another example of the diffusion of a policy initiated in the United States to other countries.
During the 1930s, Ford was forced to abandon its original principle of a single product and incorporate into its organization certain methodslinked to the multiple range strategy initiated by Sloan at GM in the 1920s. The typically multi-domestic structure of GM led to progressive modifications in Ford's geographical organization (Dassbach, 1989). Hence, in 1932 and 1934, the Ford models 'Y' and 'C' appeared, the first vehicles to be especially designed for the European market. These models initiated an international divergence in model ranges between North America and Europe, consistent with the policy of local manufacture that was being followed simultaneously in response to protectionist pressures. This trend was further accentuated by the war, which split Ford and GM in two in Europe. With peace restored, Ford confirmed the trend by adopting a European structure with two separate domestic bases, Germany and the United Kingdom. Production processes and products were distinct, appearing quite similar to GM, which had become the point of reference. At the same time European and North American operations became almost independent of one another. Ultimately, the multinationalization of Ford, which started out under the auspices of a world-wide internationalization, ended in a form of multi-domestic organization which, while more hierarchical than GM, was also highly fragmented. The convergence of the principles of spatial organization of the two automobile giants during the 1950s doubtless reflected in part a mimetism related to their direct competition. Beyond this, it also represented an appropriate response to the conditions and opportunities of expansion in the automobile sector during this period.
The duality of global strategies
From the 1960s, the opening of borders, the diffusion of free- trade, and the creation of zones of regional integration defined a global context in which the pursuit of proximity to markets gave way to a desire to optimize the spatial distribution of activities as a whole.
The Fordist trans-regional strategy
With the creation of Ford of Europe in 1967, the latter half of the 1960s was marked by a bifurcation in Ford's trajectory. The implementation of a multi-regional strategy involved the decision to integrate European operations in terms of products and manufacturing, in parallel to developments in North America following the Auto Pact agreed between the United States and Canada. This reorganization occurred during the 1970s. It included both geographical expansion towards the South (industrial investments in France and especially in Spain) and a trans-national management of space based on the interdependence of production operations. It led to the relative weakening of the British pole in European assembly activities, with the company discovering that geographical restructuring could lessen the effects of the localized crisis of productivity there (Bordenave, Lung, 1988). A specialization of activities went hand in hand with a growth in intra-European trade within the company. A spatial hierarchy developed in which the production of prestige models was assigned to Germany, which became the dominant pole in Europe.
Over the course of the 1980s, with the competitive pressures of Japanese producers increasing, Ford focused more on the spatial reorganization of operations at the world level. Taking the world as a whole, the company redefined regional and intra-regional specializations while reinforcing coordination within regional spaces. This transition from multi-regional to trans-regional took place as an alliance was formed with the Japanese producer Mazda. The rapprochement of the two companies began at the start of the 1970s and was sealed when Ford took 25% of Mazda's capital in 1979. The two companies developed close cooperation based on shared product lines, designs and joint manufacturing in both the Asia-Pacific and North American regions. Mazda was included in the 'centres of responsibility' policy initiated by Ford in 1987, an initiative which reflected Ford's commitment to a trans-regional strategy. The policy involved dividing product development activities related to the principal models of a globalized range among the three main world regions: the smallest cars for the Japanese partner Mazda, the compact vehicles for Ford of Europe, and the large vehicles for Ford of America. This renewed confirmation of the world car strategy, after several false starts during the 1970s, was a response to the twin objectives of achieving economies of scale and utilising differentiated resources and competencies (Bordenave, 1996). It presupposed permanent trade between regions, and occurred within a context of continued hierarchical domination of the American pole, which remained the centre of gravity of the company's operations, over the other poles.
Ultimately, the centres of responsibility policy only created a single vehicle, launched in Europe in 1993 and the following year in North America.. It was replaced by a more ambitious plan for reorganization, announced in 1994 and baptized 'Ford 2000'. This plan radicalized the trans-regional orientation by totally integrating North American and European activities into a single structure covering all aspects of the production process. Five entities called 'vehicle centres', specialized according to product type, were created. One was based in Europe, and the other four in the United States. Once again the aim was to reduce overlaps by restricting the number of basic platforms, while increasing variety in the final product. A further aim was to better utilize resources by coordinating products, production and supplies at the world scale. Finally, Ford wanted to reinforce the importance of the product over that of the managerial functions at the highest level of the organizational structure. Both Asia-Pacific and Latin America remained outside the new organization, though both were expected to be integrated into it. This is now possible with Ford having strengthened its control over Mazda and become autonomous again in South America. In 1966, Ford raised its share of Mazda's capital to 33,4% and took control of the company by appointing one of its own managers to Mazda's helm as well as reinforcing the presence of American Ford staff in managerial positions. Meanwhile, Ford's Brazil and Argentina Autolatina joint-venture with VW, which had been launched in 1987, was dissolved in 1995.
Doubtless the evolution of Ford's organization is related to a long-standing concern with standardization, present from the company's inception, and from which there had been deviations historically. It may also be related to a desire to be more in line with the internationalization of Japanese companies. Although less advanced than their American counterparts in their internationalization of production, Japanese companies had the economic advantage of being able to offer world products when the American companies, by reason of their own historical trajectories, could only offer regional products. From this perspective, having been the first mover in the strategy of multinationalization may have become a handicap in the new context of market globalization and homogenization. However, this interpretation should be kept in context; the possession of a diffused industrial base continues to create advantages in terms of proximity to markets and access to diversified resources.
Ford's transition towards this new form of organization will most likely be long and not without difficulties. If the trans-regional orientation is one of the possible axes of future spatial structuring for automobile companies, it has not yet been taken up by GM, which remains a multi-regional company.
The bi-centrism of GM
The multi-dimensional structure of GM, combined with ongoing uncertainties about the recomposition of its domestic divisions, has strongly influenced GM's global strategy. Over the past two decades, the company has developed its multi-domestic strategy in the direction of a polarization into two major regional operations: North America, with 59,7% of world production, and Europe, with 30,1%. However, GM differs from Ford since the process of regional integration, notably European, occurred independently and in the absence of a guiding project at the world scale. It should be noted, moreover, that in contrast to Ford the association with Isuzu has contributed little to strengthening GM's presence in Asia Pacific.[14 ]And NUMMI, the joint venture formed with Toyota in California, gave GM a privileged view of Japanese methods of organization, but, since it was oriented towards the American market, did not help GM to enter Japan, as Mazda has, to a degree, done for Ford.
The autonomous policy followed in Europe has nevertheless enabled GM to catch up with its rival in this region through the rationalization and integration of its operations. GM has similar traits to Ford in Europe, such as investment programmes in southern Europe (Spain). The major difference is that European unification has involved relegating the British subsidiary, Vauxhall, to being a mere appendage of the German subsidiary, Opel. Ford's activities in the United Kingdom were far too important to begin with for the company to have adopted a similar solution, so that its research and development activities remain divided between the United Kingdom and Germany. From this standpoint GM's operations in Europe are more homogeneous than those of Ford. The success of GM's European integration enabled the company to achieve much better results than Ford during the early 1990s, in contrast to the situation that prevailed in the United States where GM's parent company experienced severe troubles. This has unquestionably contributed to GM's coordinating centre in Europe, created in Zurich in 1986, being given, at the start of the 1990s, the mission to coordinate all of GM's international activities outside North America - including the emergent countries in Brazil and in the Asia-Pacific region.
At present GM has a segmented poly-centric structure, which is being rationalized, without a truly globalized management of space: the American and European poles remain essentially autonomous. Within this overall bi-centric framework, there are projects to share products between geographic areas, but for now they do not possess the same systematic character and scope as Ford. Hence the small mass market European car, the 'Corsa', is to be manufactured in Latin America and exported to the Asia-Pacific region. In the long term (towards the end of the 1990s) it may also be manufactured in the United States. An American minivan is now being exported to Europe to be sold under the local brand names, Vauxhall and Opel. The platform of a prestige European model (Opel `Omega') is being used for an American model from the Cadillac division. Platform sharing, which is one of the axes of GM's rationalization in the United States, may progress further in future, within Europe between Opel and Saab (of which GM took a 50% share in 1989), and particularly across the Atlantic. The recent plan to share components across continents (engines and transmissions) is also a sign of movement towards a trans-regional configuration.
Hence GM started the phase of globalization later than Ford and by establishing its own forms. There is no guarantee that it will be drawn to the trans-regional model, even if this represents a possible orientation with which the company is partly experimenting. The geographical organization of GM might retain its basis in the multi-regional model, since the company does not seem to want to give a prominence to global cars; nor does GM possess the centralizing tradition of Ford. The paths marked out by American producers, who had drawn closer together in the post second world war period, have once again diverged; but for how long and on what basis will they meet again? At any rate, these trajectories may attract the attention of both Japanese and European companies.
JAPANESE STRATEGIES: ADAPTING TO LOCAL CONSTRAINTS
The massive export policy adopted by Japanese companies throughout the 1970s confirmed their emergence and firmly established them in the sector (Bélis-Bergouignan, Lung, 1994). This basic strategy most particularly characterized the initial stage of multinationalization, led by Toyota and, to a lesser degree, Nissan, with the examples set by these two firms later followed by other producers. Hence, the phrases Global 10 and Global 12,15 symbolizing Toyota's increasing international strength, summarized that company's expectations of multinationalization during this phase.
However, today the capture of an increasing share of overseas markets represents only one of many aspects of the Japanese spatial strategy, since these companies have followed the phase of market internationalization during the 1970s with a phase of internationalization of production, in the United States and then Europe (Sachwald, 1993). Thus in 1994 the Japanese produced 1,5 million passenger cars in North America and 342.000 in Europe, a volume which should rapidly increase, given the growth of the transplants' capacities in both regions. The production of the three Japanese producers in North America accounts for different proportions of their total output: 38,9% for Honda which has the highest level, 22,0% for Nissan and 12,7% for Toyota (see Appendix).
Towards a multi-regional configuration?
The Japanese producers are explicitly taking account of the regionalization of global space as they develop their strategies: now that exports have peaked they have to cope with the local constraints placed on their overseas investments. Only a satisfactory response to this challenge is likely to guarantee a durable internationalization of production. Whilst until that point Honda had been a follower, the company was the first to launch into the new phase. Honda's strategy deserves to be examined in more detail, for two reasons: first because it is far more advanced than the other producers in its internationalization of production and sales; and second because it has experimented with an innovative process for managing its activities over space. Honda's experience, certain aspects of which have been adopted and reproduced by its Japanese competitors, can illuminate our thinking about the potential of the Japanese companies to create their own particular model for the management of space.
Referring to the criteria set out in the analytical framework, Honda's trajectory seems to be evolving from a world strategy (Figure 3, Box 1) to a multi-regional strategy (Figure 3, Box 4). Honda's
traditional 'localization' (Suguira, 1990) is nowadays practised within a regional framework quite different from the multi-domestic configurations characteristic of the American companies when they launched their world-wide strategies. The spatial structure is developing through the coordination of operations organized in a multi-divisional form, which is similar to the multi-regional configuration ('World Headquarters Strategy'). Each regional division (United States, Europe, and Asia) is fully responsible for all activities from the research and design to the sale of the final product with the goal of creating regional structures autonomous with respect to the world headquarters. In this framework the latter retains decision-making over matters that affect the company as a whole, with the regional structures controlling and coordinating activities related to their own jurisdictions. As this trajectory deepens, some consider that Honda may develop, during the 1990s, a specific range of vehicles for each of the three world regions. Defined on the basis of in-depth, decentralized market studies, designed locally by research and development departments, and produced locally, these models would cover the high volume segments of the regional automobile markets. Products thus defined in response to local demand conditions could be exported, in modified versions, towards the other regions where they would be targeted at low volume niches. One recognizes here the picture of the Multiregional Motor Company, which the IMVP report established as the ultimate point of reference (Womack et al., 1990: chapter 8).
It is, however, worth noting that this vision has not yet been realised, even if it is at times promised by company managers, who willingly discuss the post-national company within a context of difficult negotiations over automobile trade between the United States and
Japan. In the current transition period this situation translates into levels of local integration and adaptation which vary by region, and by the maintenance of central control over the regional divisions. In the absence of a new balance with the regional headquarters, this confirmation of the hierarchical dimension, together with the appearance of inter-regional coordination, suggests a drift away from the publicly announced spatial strategy towards a strategy which is more trans-regional than multi-regional.
Or perhaps trans-regional?
If the Japanese producers have confirmed their shift away from a 'world car' strategy in terms of design, one ought not automatically to deduce from this an autonomization of resources and the creation of separate regional model ranges. On the contrary, the gains that have been made in terms of similarities between regions suggest that models can be designed at the world level (platform and/or architecture and/or components shared), with adaptation as a function of the specificities of local markets. Here too it is Honda that has opened up the trajectory; after having sold the same model in different regions, the company has moved pragmatically to develop variants for each region (United States, Europe, Japan, and even South East Asia) on the basis of the same vehicle architecture and a sharing of the principal mechanical components. Examples are the new Accord with its different American, European and Japanese versions, and Honda's version of the Asia Car, a small car derived from the Civic platform for production and sale in Thailand. After having experimented with competition between its design centres for the design of models, Honda is now increasing coordination among its studios dispersed across the world (Japan, California and Germany) in the development of future products.
Hence, the specializations adopted by Honda in terms of the geographical organization of its activities (Mair, 1994) may be expanded to design over the course of the 1990s. Given the organizational skills that have been developed over a number of years in the United States, it is not inconceivable that Honda will locate world-wide responsibility there for not only production (this is already the case for the North American version of
the Accord and Civic) but also for the design of certain products.
Honda's 'global local' model, to use Andrew Mair's term, is closer to the trans-regional than the multi- regional configuration.Other Japanese producers have been inspired by Honda's strategy. Although still preoccupied with the difficulties of world-wide internationalization, these companies are relaxing their world product strategies and attempting to make progress in terms of regional adaptation. Hence, Nissan has started to integrate regionally its operations in North America and in Europe. In this last region, Nissan has the most substantial production base of the Japanese producers. In South East Asia, the company was the first to start production and sale of a specific vehicle destined for these markets (Asia Car), on the basis of a platform shared with Japanese products. An organizational structure is emerging which is based on four world regions: Japan, United States, Europe and South-East Asia. Toyota has made significant regional advances as far as its Asia-Pacific operations are concerned, as it attempts to coordinate the activities of its various subsidiaries spread out among the ASEAN countries, creating a regional division of labour, and linking in the factories in Australia and even California into a network of criss-cross flows of complementary components. Faithful to its own tradition, Toyota continues to expand in markets it has already entered and influences through its exports, with the internationalization of its sales, based on the existence of a sales network, serving as a basis upon which to make investments in production. Moreover, the magnitude of the investments Toyota has announced in the United States may lead the company to overtake Honda and become the leading Japanese producer in North America. While Toyota remains characterized by a configuration which links a world-wide vision to a pronounced degree of spatial hierarchy in its activities (Figure 3, Box 1), its future trajectory may lead the company towards a new configuration on the basis of a pragmatic internationalization which evolves as a function of local problems to be resolved, local competencies which are gradually built up, and opportunities to be seized.
EUROPEAN COMPANIES: LATE IN GLOBALIZATION
European producers have barely started to move beyond the world- wide configuration. Their multinationalization continues to evolve along lines which are essentially unchanged (Figure 3, Box 1), even if possible developments are on the horizon.
The Euro-centred producers
The international structures of the European producers are characterized in the first place by a significant presence in only one of the triad regions: their region of origin. Not only does Europe represent their primary, if not exclusive, production base and their principal market (between 60% and 85% of world-wide sales) but their few manufacturing and sales activities beyond Europe are mostly situated in developing countries (particularly Latin America and the Mediterranean basin). As far as North American markets are concerned, and even more so Japan, the Europeans remain restricted to export strategies, which are more successful for the specialist producers focusing on low-volume market segments than for the generalist producers (see Appendix). Some of the European companies did attempt, as early as the late 1970s, to adapt their international production structure to the globalization of the automobile industry. Yet, arriving in North America at the same time as the Japanese, who benefited at that time from a significant competitive advantage, Renault's and Volkswagen's attempts to establish production bases in the United States ended in failure.
The last fifteen years have therefore been marked by a retreat from an embryonic polycentrism, and a renewed focus on Europe. The nuances which differentiate among the European producers (see below) cannot hide what appears to be their principal shared trait; in every case their international structure is rooted in a centre-periphery approach. It is based on a clear hierarchical division between Western Europe, where the companies concentrate all their key know-how in terms of research, design, and management, and the growth regions which have the status of supplementary markets or production sites with no autonomy. It is only within the European region that these companies have developed, and to variable degrees, a geographical division of production. This tendency towards a regional integration of activities has clearly weakened the strictly national character which long marked the European producers (Bordenave, Lung, 1996). They therefore appear to be archetypal ethnocentric companies, and remain very homogeneous internally, which keeps them coherent and simplifies their coordination. All the same, their uni-dimensional focus on a single region may prove damaging to their long-term prospects. Their lack of participation in globalization restricts their access to diversified and renewed resources.
Towards more innovative global strategies?
The restructuring now taking place within the world automobile industry may nevertheless lead to new trajectories for Europe's 'specialist' and 'generalist' producers. During the 1990s it has been the specialist producers which have taken up the role of investing in the United States: with their North American factories, BMW and Mercedes hope to counter penetration of the luxury market segment by the Japanese. By overturning price and quality norms, Japanese producers have effaced the competitive advantage in terms of product differentiation held by the German producers. The Europeans have been forced to relocate in order to improve their price competitiveness (high wage costs in Germany, the dollar/mark exchange rate), at the same time attempting to recreate a margin of product differentiation. In North America they will manufacture specific products designed locally, and aimed at American consumers; hence relocation corresponds to the penetration of new market segments. It is thus associated with the gradual disappearance of the 'specialist' category of producers targeting low volume luxury vehicles. It may still be too early to conclude that the world-wide configuration has been surpassed, but clearly the uni-directional export strategy has been abandoned and there is a move towards a global configuration.
The European generalists have focused their international investment at the margins of the triad, where the prospects for growth are strong and competitive positions are unstable. The penetration of new expanding markets ought to enable a further internationalization of production. The most hesitant producers in this regard are the French (PSA Peugeot-Citroën and Renault) which, while they are developing significant investment programmes in certain emerging markets (PSA in China and India, Renault in Brazil) are still operating according to a highly centralized logic in which the production of certain domestic models is shifted overseas. Their investments abroad remain below the critical mass liable to elicit specific competencies and trajectories.
Both the Volkswagen-Audi group (VAG) and Fiat have been more ambitious in their vision, consolidating their investments in the peripheral Southern or Central European countries (with VAG buying up Seat and Skoda, Audi investing in Hungary, and Fiat seeking to strengthen its presence in Poland) and, on the other hand, devising regional strategies oriented towards emerging countries. Thus Fiat has developed, in Italy, a vehicle specially designed for southern hemisphere markets (`Palio') as a response to the specific needs and conditions of production in these countries. Fiat hopes to produce this model on a scale of 750,000 units per year, in twelve different countries. Volkswagen is also looking, though only partially, at specific products (those of Skoda and Seat, the `Santana' and the `Gol' models developed by VW in Brazil) to penetrate first-time buyer markets in South America and Asia (India, China, etc.).
However, while Fiat remains a highly centralized organization, the trajectory adopted by VW appears more complex. One might imagine an innovative scenario based on locally developed organizational competencies: the production bases inherited from a long standing presence in South America or acquired through buying up Seat and Skoda, would be utilized to support a decentralized regional pole. In Brazil, the import substitution policies which prevailed until the 1980s forced the company to locally develop industrial know-how in terms of the manufacturing and design of specific models, often adapted from European versions. The international opening of this country offers several new possibilities, with the creation of regional markets (MERCOSUR) and the opportunity to export to similar markets (Africa, the Middle East). Similarly, VAG has discovered its own competencies in Spain and in the Czech Republic. If these resources were federated, a single pole would open up into a multi-polar geographical organization which would not be limited to three poles of the 'triad'; by integrating part of the southern hemisphere into poly-centrism, it would offer a new variant of the global configuration.
Yet, this scenario does not appear to be the most likely one, given the strategic orientations outlined over the past few months by Volkswagen, incorporating a world-wide integration of its activities, notably through the 'platform strategy', the primary objective of which is the rationalization of the product ranges of its brands, which had become inflated with the acquisition of Seat and Skoda (Jürgens, 1993). By reducing the number of platforms from 16 to 4, the VAG group will be joining in the effort being made by all producers to increase the sharing of components and architectures. This strategy extends beyond Europe the world-wide configuration, to which Volkswagen is now returning to integrate the emerging countries. Given the paucity of developed organizational competencies outside Europe, the tendency towards centralization remains strong even if certain opportunities may be seized, so that the Europeans remain confined to
The competitive context is tending to create a world-wide automobile industry. It would be wrong, however, to think that companies are necessarily constrained to adopt the same type of industrial model or the same type of multi-national configuration, whether this be the multi-regional, trans-regional or trans-national form (Bartlett, Ghoshal, 1993). On the one hand, the issue of to what extent the process of globalization is a reality remains to be established (Ruigrock, van Tulder, 1995; Hirst, Thompson, 1996), including for the automobile industry (Freyssenet, Lung, 1996), in the sense that this world-wide expansion may produce new sources of heterogeneity. On the other hand, the current work of the research network GERPISA has emphasized that companies may, depending on their trajectory, adopt different, but equally successful production configurations (Boyer, Freyssenet, 1995). This can now be verified in terms of internationalization strategies
Internationalization of the automobile producers
World production of passenger cars (only) in 1994
World sales of new passenger cars in 1994
(output in 1000s)
|Geographic distribution of production||Geographic distribution of sales|
|Producers||North America||Europe||Japan||Others||North America||Europe||Japan||Others|
|Ford (& Mazda)||4 928||46,6%||29,8%||16,7%||7,0%||49,3%||32,3%||5,5%||12,9%|
|General Motors||5 486||59,7%||30,1%||10,2%||61,4%||28,4%||0,4%||9,8%|
|GM (& Isuzu)||5 537||59,1%||29,9%||0,9%||10,1%||60,8%||28,1%||0,5%||10,6%|
|PSA Peugeot-Citroën||1 798||98,5%||1,5%||0,0%||84,9%||0,3%||14,7%|
Sources : Comité des Constructeurs Français d'Automobile, Automotive News
North America: United States, Canada, Mexico (NAFTA)
Europe: European Union
Others: All countries except North America, European Union and Japan
Ford: Ford, Jaguar, Aston Martin, Autolatina (Ford), AutoAlliance (Ford)
Ford & Mazda: Ford + Mazda (Ford held 25%), AutoAlliance Mazda)
GM: General Morors., Saab, Lotus, NUMMI (Geo)
GM & Isuzu: GM + Isuzu (controled by GM, 35%)
Toyota: Toyota, NUMMI Toyota)
VAG: Audi, Seat, Volkswagen, Skoda, Autolatina (V.W.)
Fiat: Alfa, Ferrari, Fiat, Innocenti, Lancia, Maserati, Sevel (Fiat), Tofas
Renault: Renault, F.A.S.A., Oyak, Revoz, CIADEA (ex Renault-Argentina)
PSA: Peugeot, Citroën, Sevel (Peugeot or Citroën)
BMW-Rover: BMW, Rover
Asanuma B., 1991, "Coordination between Production and Distribution in a Globalized Network of Firms: Assessing Flexibility Achieved in the Japanese Automobile Industry", Paper presented to Japan in a Global Economy. A European Perspective, Stocholm School of Economics, Stockholm, Sweden, 5-6 September.
Bartlett C. A., 1986, "Building and Managing the Transnational: The New Organizational Challenge", Competition in Global Industries, M.E. Porter (ed.), Harvard Business School Press, Boston, Mass., pp.367-401.
Bartlett C., Ghoshal S., 1989, Managing Across Borders. The Transnational Solution, Harvard Business School Press, Boston, Mass.
Bélis-Bergouignan M.C., Bordenave G., Lung Y., 1994, "Hiérarchie et multinationalisation. Une application à l'industrie automobile", Revue d'économie politique, Vol. 104, ndeg.5, pp.116-139.
Bélis-Bergouignan M.C., Lung Y., 1994, "Le mythe de la variété originelle. L'internationalisation dans la trajectoire du modèle productif japonais", Annales. Histoire, Sciences sociales, 49ème année, ndeg.3, pp.541-567.
Bordenave G., 1992, "Le modèle fordien et son espace", Actes du GERPISA-Réseau international, ndeg.5, Université d'Evry-Val d'Essonne, Juin.
Bordenave G., 1996, "Globalization at the Core of Changes in Organization Crisis and Recovery of Ford Motor Company", One best way ? Automobile Firms trajectories and the New Industrial Models, M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), Oxford University Press, Oxford (forthcoming).
Bordenave G., Lung Y., 1988, "Ford en Europe. Crises locales, crise globale du fordisme", Cahiers de recherche du G.I.P. "Mutations industrielles", ndeg.17, Paris.
Bordenave G., Lung Y., 1996, "New spatial configuration in the European automobile industry", European Urban and Regional Studies, (forthcoming).
Boyer R., Freyssenet M., 1994, "The Emergence of New Industrial Models", Actes du GERPISA, ndeg.15, Université d'Evry-Val d'Essone, pp.75-155.
Casson M., 1995, "Internationalisation as Learning Process: a Model of Corporate Growth and Geographical Diversification", Numéro exceptionnel de la Revue d'économie industrielle, "
Économie industrielle développements récents", Editions Techniques et Economiques, Paris, pp.109-134.
Chandler A. D., Jr, 1962, Strategy and Structure, M.I.T. Press, Cambridge, Mass
Chandler A. D., Jr, 1990, Scale and Scope. The Dynamics of Industrial Capitalism, The Belknop Press of Harvard University Press, Cambridge, Mass..
Chandler A. D., Jr., 1992, "Organizational Capabilities and the Economic History of the Industrial Enterprise", Journal of Economic Perspectives, Vol.6, ndeg.3, Summer, pp.79-100.
Chung M.K., 1996, "Internationalisation of Korean car makers", Communication à la Quatrième rencontre internationale du GERPISA, Paris, June 19-21.
Dassbach C. H. A., 1989, Global Enterprises and the World Economy: Ford, General Motors and IBM, the Emergence of the Transnational Enterprise, Garland Publishing, Inc., New York & London.
Freyssenet M., Lung Y., 1996, "Between Globalization and Regionalization: what issues for the automobile industry ?", Introduction to The Global Automotive Industry: Between Homogenization and Hierarchy, GERPISA's Fourth International Colloquium, Paris, 19-21 June.
Fujimoto T., Nishigushi T. et Sei S., 1994, " The Strategy and Structure of Japanese Automobile Manufacturers in Europe", Does Ownership Matter ? Japanese Multinationals in Europe, Mason M. & Encarnation D. (eds.), Oxford University Press, Oxford, pp.367-406.
Hedlund G., 1993, "Assumptions of Hierarchy and Heterarchy, with Applications to th e Management of the Multinational Corporation", Organization Theory and the Multinational Corporation, Ghoshal S., Westney D.E. (eds.), St. Martin's Press, New York & London, pp.211-236.
Hirst P., Thompson G., 1996, Globalization in Question, Polity Press, Cambridge.
Hounshell D. A., 1984, From the American System to Mass Production 1800-1932, John Hopkins University Press, Baltimore.
Jürgens U., 1992, "Internationalization Strategies of Japanese and German Automobile Companies with special emphasis on Volkswagen", Tokunaga S., Altmann N., Demes H. (eds), New Impact on Industrial Relations, Internationalization and Changing Production strategies, Ludicium, München.
Jürgens U., 1996, "VW at the Turning Point, Sucess and Crisis of a German Production Concept", , One best way ? Automobile Firms trajectories and the New Industrial Models, M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), Oxford University Press, Oxford (forthcoming).
Lung Y., 1992a, Global competition and Transregional strategy. Spatial reorganisation of the European car industry", Cities and Regions in the New Europe. The Global-Local Interplay and Spatial Development Strategies, M. Dunford and G. Kafkalas (eds.), Belhaven Press, London, 1992, pp.68-85.
Lung Y., 1992b, Industrial Change and Spatial Strategy of Multinational Enterprise. The Ford's "world economy", IERSO, Université Montesquieu-Bordeaux IV, unpublished manuscript.
Mair A., 1994, Honda's Global Local Corporation, Macmillan Press Ltd, Houndmills, Basingstoke.
Marx R., Zilbovicius M., Salerno M.S., 1996, "The "Modular Consortium" in a new VW Truck Plant in Brazil: New forms of assembler and suppliers relationship", Paper presented at The Global Automotive Industry: Between Homogenization and Hierarchy, GERPISA's Fourth International Colloquium, Paris, 19-21 June.
Michalet C.A., 1990, "Où en est la notion d'économie mondiale ?", dans Investissement international et dynamique de l'économie mondiale, M.Humbert (édit.), Economica, Paris, pp.13-34.
Perlmutter H.V., 1965, "L'entreprise internationale - trois conceptions", Revue économique et sociale, ndeg.23.
Porter M., 1986, "Competition in Global Industries: A Conceptual Framework", Competition in Global Industries, M.E. Porter (ed.), Harvard Business School Press, Boston, Mass., pp.15-60.
Ruigrok W. van Tulder R., 1993, "Regionalisation, globalisation or glocalisation : the case of the world car industry", The Impact of Globalisation on Europe's Firms and Industries, M.Humbert (édit.), Pinter Publishers, London & New York pp.22-33.
Sachwald F., 1993, "L'industrie automobile : la transplantation du système japonais à l'étranger", Les entreprises japonaises en Europe. Motivations et stratégies, F. Sachwald (éd.), I.F.R.I.-Masson, Paris, pp.155-192.
Sachwald F., 1994, "Mondialisation et circulation des connaissances dans l'industrie automobile", Les défis de la mondialisation. Innovation et concurrence, F. Sachwald (éd.), I.F.R.I.-Masson, Paris, pp.279-320.
Sugiura H., 1990, "How Honda Localizes its Global Strategy", Sloan Management Review, Fall 1990, pp.77-82.
Tolliday S., 1996, "Transferring Fordism : The First Phase of the Overseas Diffusion and Adaptation of Ford Methods, 1911-1939",Transfer and Hybridisation. The genesis and diffusion of new models of production in the automobile industry, R. Boyer, U. Jürgens, S. Tolliday and E. Charron (eds), Oxford University Press, Oxford (forthcoming).
Wilkins M., Hill Frank E., 1964, American business abroad, Ford on six continents, Wayne State University Press, Detroit.
Womack J. P., Jones D. T., Roos D., 1990, The Machine That Changed The World, Rawson Associates (Macmillan Publishing Co.), New-York.