* The aim of this paper is to test whether the theoretical models
concerning international structures and strategic approaches explain the
behaviour of the internationalisation of firms from a newly liberalised
* Using a sample of 117 Spanish MNC, a high correlation between the
theoretical models and the development of the international strategy in
the sampled firms has been validated.
When a company establishes a foreign subsidiary new relationships
will appear, forcing that company to adapt its organisational
structures. In general, there is no one way to adapt to these new
conditions, but there are some patterns of behaviour that have been
observed in some cases. The diversification of products is related to
product divisions, the diversification of markets is related to area
divisions and both types of diversification are associated with matrix
or mixed structures. However, it will be insufficient just to change the
structure; a set of strategic variables has to be considered. These
variables shape the organisational processes that flow along the
structures (Bartlett 1986, Bartlett/Ghoshal 1989). Therefore, the
research of organisational processes which allow the co-ordination and
control of international activities has turned into one of the crucial
aspects to be considered in multinational companies (Ghoshal/Nohria
1993, Nohria/Ghoshal 1994). Both the strategic approach and the
organisational structure of the company have a decisive influence in
These initial models were established in the last two decades but
recently a new interest in these two fields of study has developed
(Egelhoff 1997, Hedlund/ Ridderstrale 1997, Hedlund 1999, Harzing 2000).
Following this stream of research, the purpose of this paper is to
analyse these "classical" organisational theories in the
context of a newly liberalised economy.
The paper attempts to answer the following question: Are existing
theories about international structures and international strategic
approaches valid for the strategies of foreign investors from a newly
integrated/liberalised country like Spain? The paper supports the
argument that the basic underlying international strategy-structure fit
seems to apply. Therefore, this research: 1) attempts to reconcile some
differences in existing theory, and 2) investigates how applicable
existing theory is to emerging Spanish MNCs. Moreover, the paper shows
the interaction between the formal structures and the Bartlett and
Ghoshal's research. This fact also contributes to overcome the
problem of the relative lack of empirical work verifying the extent of
such models (Harzing 2000).
The article is structured as follows. Section two reviews the main
characteristics of the international structures and strategic focuses.
It also establishes the hypotheses that test the relationships between
those structures, the growth strategy and the strategic focuses adopted
by companies. The next section discuses the methodology applied in the
research and the operationalisation of the variables whereas section
four presents the statistical analysis and a discussion of the results.
Finally, the last section reveals the main conclusions.
Literature Review and Hypotheses
Companies may adopt a variety of structural forms when they
implement an international strategy. These structures have been
identified and described in a great number of works (Stopford/Wells
1972, Robock/Simmonds 1973, Franko 1976, Galbraith/Nathanson 1978,
Egelhoff 1982, Dunning 1993). In the initial stage of the international
expansion, foreign sales are an extension of domestic sales. As long as
the firm increases its international presence by creating some sales
subsidiary, a mother-daughter structure can be adopted (Franko 1974). A
high level of independence from the corporate strategy and the transfer
of expatriates, that implies a control strategy based on socialisation,
characterise that structure (Edstrom/Galbraith 1977). The corporate
headquarters is in charge of each and every foreign unit that serves its
own market and has a fairly free hand in the operation of the day-to-day
business (Dunning 1993).
A second stage follows after the -already described--initial phase
in which companies tend to create an international division. In this
structure, all foreign subsidiaries report directly to that single
international division that is organisational-wise separated from the
domestic operations (Brooke/Remmers 1970). This division covers all
foreign activities and becomes the company's first source of
international expertise (Stopford/Wells 1972). This structure
facilitates information processing between the headquarters and the
foreign subsidiaries but hinders information processing at the parent
level between the domestic operations and the international division
As the international division grows and the head office recognises
the necessity of a global perspective in which there should not be any
difference between domestic and foreign activities, two new types of
international structures appear: the world-wide product division and the
A world-wide product division structure extends the
responsibilities of the domestic product divisions to cover their
product lines on a world-wide base. (Egelhoff 1982). Two structural
changes appear: authority shifts from subsidiaries and regional managers
to headquarters, and headquarters' responsibility shifts from a
single line of authority to multiple lines (Davidson/Haspeslagh 1982).
Area divisions divide the world into regions, each of which is
headed by a top executive at the corporate level. Each regional
headquarter is responsible for all of the company's products and
business within the corresponding geographical area (Egelhoff 1982).
Finally, global matrix or mixed structures appear as the most
developed and complex international structures. In a matrix structure,
there is a dual line of authority. For example, in an area division by
product division matrix structure, a subsidiary manager reports to both
the world-wide product division HQ and the geographical area HQ. Both HQ
managers will be at the same level and their responsibilities will
overlap (Dunning 1993). In a mixed structure, different substructures
work simultaneously within the organisation (Egelhoff 1988, Habib/Victor
The optimal, or perceived optimal, organisational structure of any
particular MNE may change over time as, for example, it widens or
reduces its product range or increases its degree of multinationality
(Dunning 1993). Indeed, some authors (Egelhoff 1988, Franko 1976,
Galbraith/Nathanson 1978, Habib/Victor 1991 Stopford/Wells 1972) have
shown a pattern of behaviour according to the growth strategy of the
firm. From this perspective, MNC structure follows strategy.
The first major study by Stopford and Wells (1972) found that, as
product diversity increases, companies moved from the simplest
structures to world-wide product division structure. On the other hand,
when foreign market involvement--measured by the percentage of foreign
sales--increased, firms tended to adopt worldwide area division
structure. Finally, when both--product diversity and foreign sales--were
high, an MNE would follow a matrix/mixed structure. In a parallel study,
Egelhoff (1988) re-examined the Stopford and Wells's model
introducing a new element of strategy: the relative size of foreign
manufacturing. This work found that both world-wide area divisions and
world-wide product divisions had relatively high foreign sales. The
crucial factor to discern both structures was the level of foreign
manufacturing. Egelhoff (1988) suggested that both, high levels of
foreign sales and foreign manufacturing, were essential for adopting
world-wide area divisions.
From the previous literature, the following hypotheses tested by
Stopford and Wells (1972), Egelhoff (1988) and Habib and Victor (1991)
are proposed in the Spanish context:
Hypothesis 1. Low levels of product diversification and foreign
market involvement will be associated with the simplest structures:
mother-daughter structure or international division.
Hypothesis 2. High levels of product diversification will be
associated with product divisions.
Hypothesis 3. High levels of foreign market involvement will be
associated with area divisions.
Hypothesis 4. High levels of both product diversification and
foreign market involvement will be associated with matrix/mixed
The formal structures described in the previous section are only a
partial representation of a world-wide organisation. A new paradigm
based on the physiology (core management processes) and psychology (the
"mind-set" of managers) of the organisations has emerged
(Bartlett 1986). Two variables appear to be decisive of this new model.
On the one hand, there is a "pressure for global integration",
including growing homogenisation of markets and homogenisation of
technology. On the other hand, there is a "pressure for national
differentiation", including state intervention, distinctive tastes
and preferences and distinctive marketing and distribution systems
(Ghoshal/Westney 1993). From this double dimension, researchers
(Bartlett/Ghoshal 1989, 1992, Hedlund 1986, 1993, Perlmutter 1969,
Porter 1986, Prahalad/Doz 1987, White/Poynter 1990) have categorised the
strategic focuses of the MNC into three basic models: global,
multinational and transnational.
A global strategy is based upon the centralisation of the main
assets, resources and responsibilities in the head office. These
companies think in terms of creating products for a world market and
manufacturing them on a global scale in a few highly efficient plants,
often at the corporate centre. This approach requires central
co-ordination and control (Perlmutter 1969, Bartlett/Ghoshal 1989,
Harzing 1998). On the other hand, a multinational strategy emphasises
the differences between national markets. The subsidiaries become more
independent and their managers become more host-country oriented.
Products or services are differentiated to meet different local demands,
and policies are differentiated to conform to different governmental and
market demands (Bartlett/Ghoshal 1989, Porter 1986). Finally, a
transnational strategy recognises that the necessity to be responsive to
local demands and the pressure to develop competitive efficiency on a
global scale are simultaneous. The resources and responsibilities are
dispersed but specialised. There is a reciprocal interdependence between
HQ and subsidiaries. New ideas may come up in many different countries
and later be exploited on a global scale. In some way, a transnational
strategy combines characteristics of both global and multinational
strategy (Bartlett/Ghoshal 1989, Hedlund 1986).
If we think about the relationship between the different
international structures and strategic focuses, some types of structure
are better suited to exploit a special category of strategic purpose
than others (Egelhoff 1982, Jones 1995, Pelmutter 1969, White/Poynter
International divisions or world-wide product divisions are able to
exploit globally base advantages. Both structures centralise world-wide
responsibility in the hands of a few product or business managers and
locate centres of innovation at the parent company (Davidson/Haspeslagh
On the other hand, the multinational strategic focus is typically
associated with mother-daughter structure or area divisions. In the
first case, the independence of managers from the HQ allows them to act
as entrepreneurial local managers. In the second case, all decisions are
centralised in a few managers, each of whom is responsible for a
particular geographical area. This specialised market knowledge
facilitates the adoption of a more flexible approach by modifying their
products and strategies country by country (White/Poynter 1990, Jones
Finally, the multidimensionality of organisation principles implies
that a matrix/mixed structure would probably be the most appropriate
formal structure to balance worldwide product and local market concerns
(Bartlett/Ghoshal 1992) (2). Therefore, we suggest the following
Hypothesis 5. Most of the companies with a multinational strategic
focus will adopt a mother-daughter or area division structure
Hypothesis 6. Most of the companies with a global strategic focus
will adopt a international division or product division structure.
Hypothesis 7. Most of the companies with a transnational strategic
focus will adopt a matrix/mixed structure.
Sample Description and Data Collection
The population to be studied consists of Spanish companies
currently active in international markets. Dun & Bradstreet provided
an international database in order to select Spanish companies with
subsidiaries outside Spain in 1996. This amounted to 498 multinational
companies. These firms represent the Spanish enterprises with the
highest degree of involvement in the international arena.
A preliminary questionnaire was discussed with five academics.
Based on their comments, some of the questionnaire items were modified.
The revised questionnaire was pre-tested with twelve randomly selected
firms from different industries. The purpose of this test was to confirm
that the items were understandable and unambiguous.
Questionnaires were mailed to CEOs of these Spanish companies
conducting FDI. After three months, we received 74 answers. A follow-up
questionnaire was distributed in order to improve the response. The
process was concluded with a total of 117 responses.
The sample was spread across the following industry groups which
represents every kind of commercial activity in accordance with the
structure of the Spanish Economy: construction and contracting (5), food
and kindred products (10), textile mill products, apparel and other
textile products (8), lumber, wood products, furniture and fixtures
(11), publishing (4), chemicals (10), rubber and plastic products (5),
leather (3), stone and glass products (8), metal products (6),
industrial machinery (15), electronic equipment (12), transportation
equipment (3), instruments (1), communication and public utilities (2),
retailing (6), finance, insurance and real estate (5) and other services
A description of the main variables in the sample is shown in the
Measurement of Variables
The main characteristics of the formal structures were introduced
in the questionnaire. Respondents classified their enterprises according
to these characteristics. It should be noted that "pure
structures", the way they have been classified here and in previous
research, are difficult to find. However, the pre-assigned
classification at the top management level was found to be useful
(Habib/Victor 1991). Table 2 shows the results.
Foreign Market Involvement
Three variables were used to measure the degree of foreign market
involvement: the percentage of foreign sales (Stopford/Wells 1972), the
percentage of foreign manufacturing (Egelhoff 1988), and the number of
countries in which the products of the company are sold. The last
variable was not included in previous models.
Foreign Product Diversity
The number of product lines was selected as a significant variable
defining the company's product strategy. Following Egelhoff (1988)
eight categories of foreign product diversity were introduced, where the
final category was "eight or more" product lines.
From Leong and Tan (1993), two sets of attributes associated with a
particular strategic dimension were introduced. Respondents had to
choose one of these attributes in each set. Table 3 reflects those
Two phases of analysis were conducted. In the first phase, the
contrast of hypotheses 1 to 4 was developed using the Mann-Whitney test
and a Kruskal-Wallis analysis of variance (3). Both analyses determine
the statistical significance of the differences between each group (as
defined by the international structures presented in Table 2), in both
product diversity and foreign market involvement. The chi-square value
resulting from the Kruskal-Wallis Anova reveals whether there is an
overall difference between the structures while the Mann-Whitney test
summarises which of the pairs are different.
In the second phase, cluster analysis was used to identify the
discrete categories or groups of strategic focuses. After the clusters
were established, a crosstabulation and the chi-square test were
conducted among the two sets of classificatory variables (structures and
strategic focuses). This analysis tests the relationships between
structures and strategic focuses expressed by the Hypotheses 5 to 7.
Results and Discussion
Tables 4 and 5 show the results of the analysis of variance and the
Mann-Whitney test. These results tend to support the concept of
strategy-structure fit proposed by Stopford and Wells (1972) and
Egelhoff (1988). Hypothesis 1 stated that MNCs with mother-daughter or
international division structures will tend to have the lowest levels of
product diversity and foreign involvement. This hypothesis is fully
supported for international division structures, whereas the
mother-daughter structure shows a high level of foreign manufacturing.
Hypothesis 2 stated that MNCs with product division structure would
tend to have high levels of product diversification. This hypothesis is
fully supported. Product division structure shows the highest level of
product diversity while international divisions and mother-daughter
structure show the lowest values in this variable.
Hypothesis 3 is fully supported when foreign involvement is
measured by percentage of foreign sales or number of countries. With an
increase in foreign involvement, Spanish MNCs choose area division
structure. If the percentage of foreign manufacturing is introduced, the
study also supports the conclusions of Egelhoff (1988). Area divisions
and matrix structures seem to be the structures associated with a high
level of foreign manufacturing. Strategies which have a high level of
foreign manufacturing cause high interdependencies between foreign
subsidiaries within a region. The area division and matrix/mixed
structures fit this kind of interdependencies providing a high level of
co-ordination and information exchange. This result helps to resolve
again the differences between the models of Stopford and Wells (1972)
and Egelhoff (1988).
Finally, Hypothesis 4 is also fully supported. When both product
diversity and foreign involvement, become relatively high, MNCs employ
the most complex structures such as matrix or mixed structures. This
result is also an important finding of this study. Until now theoretical
arguments have been forwarded for matrix/mixed structures, but empirical
support was inadequate (Habib/Victor 1991). This study empirically
contributes to the theoretical reasoning and thus positively adds to the
previous recommendations of Stopford and Wells (1972) and Egelhoff
However, two differences from those previous models have been
found. First, the relevance of mother-daughter structure. Like other
studies in the European context (Franko 1974, 1976), the subsidiaries of
some European firms in the initial stages of internationalisation have
almost complete autonomy and share with one another only ties of
ownership to a common parent firm. The importance of this type of
structure in the sample (25) reveals the classification of Spain as a
late investor (Dunning/Narula 1996).
Second, we have found in our analysis that the variable
"number of countries" is also a good predictor of foreign
market involvement. In fact, we think that this variable fits in samples
of European firms better than the variable "percentage of foreign
sales". European managers still tend not to view Europe as a single
market and they tend to respect and distinguish between the national
differences more than US firms. This fact could have some influence
concerning the organisational decisions and, for instance, it could be
more relevant to the geographical dispersion rather than the level of
total foreign sales.
The following figure shows the previous results in graphical form.
[FIGURE 1 OMITTED]
This phase of the study tested the relationships between strategic
focuses and structures. From data in Table 3, cluster analysis was used
to classify companies according to their strategic focuses. A
three-cluster solution was chosen to facilitate the comparison with the
theoretical model (Table 6). An examination was also made for two
cluster solutions. For this examination an aggregation of the
multinational and transnational groups was made.
The centre of clusters and means suggest that one group (Cluster 1)
confronts predominantly local responsiveness pressures. Subsidiaries are
more independent and innovations are developed in overseas units.
Another group (Cluster 2) confronts global integration pressures. In
this case, subsidiaries are totally dependent on the head office and
innovations are developed and retained in the parent company. Finally,
Cluster 3 confronts both pressures. Subsidiaries cooperate with the head
office and depend upon each other and R&D activities are conducted
jointly by parent company and subsidiaries.
The following crosstabulation shows the relationships between the
previous Clusters and the international structure groups established in
The chi-square test (p < 0.001) is statistically significant.
Therefore, the null hypothesis can be rejected and consequently, there
is some relationship of dependence between the two categorical
Hypothesis 5 suggested a relationship between mother-daughter and
area division structures with the multinational strategic approach. This
hypothesis is partially supported by the research. The highest
percentages in the multinational focus can be observed (44% and 58.3%
respectively) in mother-daughter and area division structures.
Nevertheless, these two structures are also used under other strategic
approaches. In the case of area division, the standardised residual is
greater than 1, which means that the probability of using a
multinational focus is higher than in the case of independence between
this particular structure and strategic approach.
Hypotheses 6 stated that firms with international division or
product division structures would tend to adopt a global approach. This
hypothesis is fully supported. There is a high association between
international division and global focus (stdresid 1.5). No case of
product division with a transnational focus is observed.
Finally, Hypothesis 7 stated that most of the firms with a
mixed/matrix structure will adopt a transnational strategic focus. This
hypothesis is fully supported. There is a strong association between
matrix structure and transnational strategic focus (38.9% and stdresid
In summary, the strongest associations found were area division
structure with multinational focus, international division structure
with a global focus and matrix structure with transnational focus.
According to these relations, we can show that the model of Stopford and
Wells is still valid for explaining the evolution of MNC in terms of the
typology of Bartlett and Ghoshal.
The purpose of this paper was to test whether the theoretical
models concerning the international structures and strategic approaches
explain the internationalisation behaviour of companies from a newly
liberalised economy such as Spain. The analysis of data suggests a high
correlation between the theoretical models and the internationalisation
of Spanish companies.
Basically, the models of fit proposed by Stopford and Wells (1972)
and Egelhoff (1988) between the growth strategy of the company and its
international structure have been validated. Nevertheless, some
differences with those models have been shown: a) the relevance of the
mother-daughter structure and b) the convenience of introducing the
variable "number of countries" in the model. Here, we have
argued that such differences are due to the specific characteristics of
the internationalisation of European firms. Also, empirical support for
matrix/mixed structures has been added.
On the other hand, the existence of three basic models of strategic
approaches has been supported again by the empirical analysis. In
addition, some new evidence of the interdependency between international
structures and strategic focuses has been found. The hypothesised proper
structure-strategic focus match was observed. These results contribute
to a) add some empirical work verifying the extent of such models in
large-scale research and b) insert the Stopford and Wells model inside
the typology of Bartlett and Ghoshal.
From our point of view, the increasingly international environments
are producing convergent models of MNE organisation. After these
results, we have shown that the organisational variables in the
internationalisation of firms from a late investor like Spain
(Dunning/Narula 1996) follow the same pattern as firms from other
nationalities more active in the international arena. Moreover, managers
could use these results and classification schemes as a starting point
for thinking about the core attributes of their organisational needs.
One last contribution of this study is to improve the understanding
of the internationalisation of Spanish companies. This is a new
phenomenon in the international field that has grown extremely rapidly
in recent years (specially in Latin America) and there is still no
empirical evidence on how Spanish companies, once they have developed
some physical presence abroad, organise their international activities.
Finally, the author recognises that some of the issues analysed are
too complex to be studied only through questionnaires. A second
limitation is the absence of some important environmental, contextual
and market variables (e.g. technology, market concentration, etc.) which
may have some impact on the strategy-structure fit. Moreover, our
results are based on the opinions of a single respondent in each firm. A
final limitation of the paper is that it does not relate
"theoretical fit" to performance. However, despite these
limitations, the study provides an extension of the research done in
other contexts. By revising and extending the previous models in the
context of a newly liberalised country, the present study has sought to
advance our knowledge of critical relationships between structure and
strategy in MNCs.
Table 3. Strategic Attributes
1. The most vital and strategic skills and resources of my
organisation tend to be located in the parent company.
2. The skills and resources of my organisation are located around
the world, but each overseas unit conducts its own operations.
3. My organisation locates specialised skills and resources around
the world, but our overseas units often cooperate with and depend upon
1. Research and development activities are conducted, and the
results retained, at parent company headquarters with little
dissemination to our overseas units.
2. The new knowledge developed in our overseas units tends not to
be transferred to other locations in which my organisation operates.
3. Research and development activities are typically conducted
jointly by parent company and overseas units with the knowledge gained
shared world-wide in my organisation.
Source: adapted from Leong and Tan (1993)
Manuscript received February 2000, revised July 2000, revised
(1) The author wishes to express thanks to Mark Casson, Juan J.
Renau, J. M. Campa, Esteban Garcia-Canal and two anonymous reviewers for
their helpful comments on an earlier draft of this paper. Financial
support from Generalitat Valenciana is gratefully acknowledged (project
(2) Further studies added some aspects such as lateral decision
processes, shared decision premises or horizontal networks to improve
the development of mixed/matrix structures (Hedlund, 1986, 1993; White
and Poynter, 1990).
(3) These analyses are non-parametric tests. They are used when the
distribution is not normal.
Bartlett, C. A., Building and Managing the Transnational: The New
Organizational Challenge, in Porter, M. (ed.) Competition in Global
Industries, Boston: Harvard Business School Press 1986.
Bartlett, C. A./Ghoshal, S., Managing Across Borders: The
Transnational Solution, Boston: Harvard Business School Press 1989.
Bartlett, C. A./Ghoshal, S., Transnational Management. Text, Cases
and Readings in Cross-Border Managemen, London: Irwin 1992.
Brooke, M. Z./Remmers, H., The Strategy of Multinational
Enterprise: Organisation and Finance, London: MacMillan 1970.
Davidson, W. H./Haspeslagh, P., Shaping a Global Product
Organization, Harvard Business Review, 60, 4, 1982, pp. 125-132.
Daniels, J. D./Pitts, R. A./Tretter, M. J., Organizing for Dual
Strategies of Product Diversity and International Expansion, Strategic
Management Journal, 6, 1985, pp. 223-237.
Dunning, J. H., Multinational Enterprises and The Global Economy,
Wokingham: Addison-Wesley 1993.
Dunning. J. H./Narula, R., Foreign Direct Investment and
Governments (eds.), London and New York: Routledge 1996.
Dun & Bradstreet, Duns 250.000 Principales Empresas Espanolas,
Edstrom, A./Galbraith, J. R., Transfers of Managers as a
Coordination and Control Strategy in Multinational Organizations,
Administrative Science Quaterly, 22, 1977, pp. 248-263.
Egelhoff, W. G., Strategy and Structure in Multinational
Corporations: An Information Processing Approach, Administrative Science
Quarterly, 27, 1982, pp. 435-458.
Egelhoff, W. G., Strategy and Structure in Multinational
Corporations: A Revision of the Stopford and Wells Model, Strategic
Management Journal, 9 1, 1988, pp. 1-14.
Egelhoff, W. G., The implications of Information Technology for
Coordination and Communication in Multinational Corporations, in Toyne,
B./Nigh, D. (eds) International Business: An emerging vision, Columbia:
University of South Carolina Press 1997.
Egelhoff, W. G., Introduction, in Egelhoff, W. G. (ed.),
Transforming International Organizations, Cheltemham: Edward Elgar 1998.
Franko, L. G., The move toward a Multidivisional Structure in
European Organizations, Administrative Science Quarterly, December, 1974
pp. 493- 506.
Franko, L. G., The European Multinationals, New York: Harper 1976.
Galbraith, J. R./Nathanson, D. A., Strategy Implementation: the
role of Structure and Process, New York: West Publising Company 1978.
Ghoshal, S., Of Cakes, Clothes, Emperors, and Obituaries, in Toyne,
B./Nigh, D. (eds) International Business. An Emerging Vision, Columbia:
University of South Carolina Press 1997.
Ghoshal, S./Nohria, N., Horses for Courses: Organisational Forms
for Multinational Corporations, Sloan Management Review, 34, 1993, pp
Ghoshal, S./Westney, E., Introduction and overview in Ghoshal,
S./Westney, E. (ed.), Organization Theory and the Multinational
Corporation, New York: St. Martin's Press 1993.
Habib, M./Victor, B., Strategy, Structure, and Performance of US
Manufacturing and Service MNCs: A Comparative Analysis, Strategic
Management Journal, 12, 1991, pp. 589-606.
Harzing, A. W., Configuration Analysis in International Management:
The way forward?, Conference Proceedings of the 25th Annual Conference
of the UK Chapter of the Academy of International Business, London,
Harzing, A. W., An Empirical Analysis and extension of the Bartlett
and Ghoshal Typology of Multinational Companies, Journal of
International Business Studies, 31, 2000, pp. 101-120.
Hedlund, G., The Hypermodern MNC- A heterarchy?, Human Resource
Management, 25 1, 1986, pp. 9-35.
Hedlund, G., Assumptions of Hierarchy and Heterarchy, with
Applications to the Management of the Multinational Corporation, in
Ghosal, S./Westney, E. (ed.), Organization Theory and the Multinational
Corporation, New York: St. Martin's Press 1993.
Hedlund, G., The Intensity and Extensity of Knowledge and The
Multinational Corporation as a Nearly Recomponsable System, Management
International Review, 39, Special Issue 1/1999, pp. 45-54.
Hedlund, G./Ridderstrale, J., Toward a Theory of the Selfrenewing
MNC, in Toyne, B/Nigh, D. (eds), International Business, An Emerging
Vision, Columbia: University of South Carolina Press 1997.
Hitt, M. A., The Evolution of Multinational Corporations:
Integration of International Diversification and Strategic Management,
in Toyne, B./Nigh, D. (eds), International Business: An Emerging Vision,
Columbia: University of South Carolina Press 1997.
Jones, G. R., Organisational Theory, Text and Cases, Reading:
Addison Wesley 1995.
Leong, S. M./Tang, C. T., Managing Across Borders: An Empirical
Test of the Bartlett and Ghoshal (1989) Organizational Typology, Journal
of International Business Studies, third quarter, 1993, pp. 449-464.
Martinez, J./Jarillo, J. C., The Evolution on Research on
Coordination Mechanism in Multinational Corporations, Journal of
International Business Studies, Fall 1989, pp. 489-514.
Nohria, N./Ghoshal, S., Differentiated Fit and Shared Values:
Alternatives for Managing Headquarters-subdsidiary Relations, Strategic
Management Review, 15, 1994, pp. 491-502.
Porter, M. E., Competition in Global Industries: A Conceptual
Framework, in Porter, M. E. (ed.), Competition in Global Industries,
Boston: Harward Business School Press 1986.
Perlmutter, H., The Tortuous Evolution of The Multinational
Company, Columbia Journal of World Business, 4, 1, 1969, pp. 9-18.
Prahalad, C. K./Doz, Y., The Multinational Mission, New York: The
Free Press 1987.
Robock, S. H./Simmonds, K., International Business and
Multinational Enterprises, Homewood. Illinois: Richard D. Irwin 1973.
Stopford, J. M./Wells, L., Managing the Multinational Enterprise.
Organization of the Firm and Ownership of the Subsidiaries, New York:
Basic Books 1972.
White, R./Poynter, T. A., Organizing for a World-Wide Advantage, in
Doz, Y./Bartlett, C. A./Hedlund, G. (eds.), Managing the Global
Corporation, London: Routledge 1990.
Jose Pla-Barber, Associate Professor of International Business,
Faculty of Economics, University of Valencia, Spain.
Table 1. Sample Description
Variables N % Mean Median Standard
Number of employees
1-50 25 21.4
50-250 40 34.2
250-500 12 10.3
> 500 37 31.6 2468 163 9830
Age (years since foundation)
1-10 8 6.8
11-20 25 21.4
21-30 22 18.8
31-40 17 14.5
>40 43 36.8 44 36 36
Number of products lines
1 13 13.2
2-3 35 35.7
4-5 25 25.5
6-7 11 11.2
>8 14 14.2 5.25 3.50 7.30
Foreign manufacturing (%)
0 55 56.7
1 1 1
2-9 10 10.3
10-30 19 19.5
>30 12 12.3 6.92 0 15.21
Foreign sales (%)
<25 27 23.1
26-50 30 25.6
51-75 35 29.9
>75 25 21.4 50.68 51.08 25.33
Number of countries
1-3 13 11.1
3-15 36 30.8
15-30 27 23.1
>30 38 32.5 28.97 20 28.9
Table 2. Structures
Structure N %
International divisions 52 44.4
Mother-daughter 25 21.3
Product divisions 10 8.5
Area divisions 12 10.2
Matrix/mixed 18 15.3
Table 4. Kruskal-Wallis Anova
Internation Structures Foreign Foreign Number of
sales (%) manufacturing product lines
(mean rank) (%) (mean rank) (mean rank)
International division 52.5 50.8 47.2
Mother-daughter 52.7 70.7 53.3
Product division 46.4 55.8 64.5
Area division 72.8 65.2 42.7
Matrix 55.5 63.8 54.21
Chi-square 4.24 9.82 ** 4.06
Internation Structures Number of
International division 55.0
Product division 61.5
Area division 72.1
Chi-square 5.25 *
* P < 0.1
** P < 0.05
Table 5. Mann-Whitney Test
% Number of
Internation foreign % foreign product Number of
Structures sales manufacturing lines countries
division (ID) ID < MD **
ID < DP *
ID < AD ** ID < AD ** ID < AD *
ID < M **
(MD) MD < PD *
MD < AD * MD < AD **
MD < M *
(PD) PD < MD *
PD < AD ** PD < AD *
(AD) AD < PD **
Matrix (M) M < AD **
* P < 0,1
** P <0,05
Table 6. Analysis of Clusters
Cluster 1 Cluster 2 Cluster 3
Multinational Global Transnational
42 60 12
Centre of clusters:
* Dependence (SET A) 2 1 3
* Location of innovations
(SET B) 3 1 3
Means of variables:
* Dependence 1.52 1.27 3
* Location of innovation 2.90 1.05 3
Source: own data base
Table 7. Relationship Structure-strategic Focus
Multinational Global Transnational Total
N 14 35 3 52
% 26.9 67.3 5.8 100
Stdresid. -1.1 1.5 -1.3
N 11 10 4 25
% 44 40 16 100
Stdresid. 0.7 -0.8 0.6
N 3 7 0 10
% 30 70 0 100
Stdresid. -0.3 0.8 -1.1
N 7 5 0 12
% 58.3 41.7 0 100
Stdresid. 1.3 -0.5 -1.2
N 7 4 7 18
% 38.9 22.2 38.9 100
Stdresid. 0.2 -1.8 3.3
p < 0.001