21 Pages
English

The technological innovation of multinational corporations in the French regions - article ; n°1 ; vol.109, pg 9-28

-

Gain access to the library to view online
Learn more

Description

Revue d'économie industrielle - Année 2005 - Volume 109 - Numéro 1 - Pages 9-28
In a rapidly globalising economy, and particularly in the face of a process of economic integration such as that occurring in the EU, regions forge an increasing number of linkages with other locations within and across national boundaries through the local technological development efforts of multinational corporations (MNCs). By using patents granted to the largest industrial firms - arranged by the region (NUTS 1) host to the research facility responsible - the paper explores the location of innovative activities of MNCs in France, and the relationship between the profiles of technological specialisation of foreign-owned and indigenous companies in the French regions.
Dans le processus de globalisation de nos économies, et particulièrement face au processus d'intégration économique à l'œuvre en Europe, les régions mettent en place de plus en plus de liens avec d'autres lieux dans et en dehors des frontières nationales à travers les efforts de développement technologique des entreprises multinationales. En utilisant les brevets des grandes entreprises industrielles - appréhendées par la région (NUTS1) abritant les capacités de recherche - l'article explore la localisation des activités innovatrices des multinationales en France, et la relation entre les profils de spécialisation technologique des entreprises étrangères et des entreprises indigènes dans les régions françaises.
20 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Subjects

Informations

Published by
Published 01 January 2005
Reads 17
Language English
Document size 2 MB

John Cantwell
Simona Iammarino
The technological innovation of multinational corporations in the
French regions
In: Revue d'économie industrielle. Vol. 109. 1er trimestre 2005. pp. 9-28.
Abstract
In a rapidly globalising economy, and particularly in the face of a process of economic integration such as that occurring in the
EU, regions forge an increasing number of linkages with other locations within and across national boundaries through the local
technological development efforts of multinational corporations (MNCs). By using patents granted to the largest industrial firms -
arranged by the region (NUTS 1) host to the research facility responsible - the paper explores the location of innovative activities
of MNCs in France, and the relationship between the profiles of technological specialisation of foreign-owned and indigenous
companies in the French regions.
Résumé
Dans le processus de globalisation de nos économies, et particulièrement face au processus d'intégration économique à l'œuvre
en Europe, les régions mettent en place de plus en plus de liens avec d'autres lieux dans et en dehors des frontières nationales à
travers les efforts de développement technologique des entreprises multinationales. En utilisant les brevets des grandes
entreprises industrielles - appréhendées par la région (NUTS1) abritant les capacités de recherche - l'article explore la
localisation des activités innovatrices des multinationales en France, et la relation entre les profils de spécialisation technologique
des entreprises étrangères et des entreprises indigènes dans les régions françaises.
Citer ce document / Cite this document :
Cantwell John, Iammarino Simona. The technological innovation of multinational corporations in the French regions. In: Revue
d'économie industrielle. Vol. 109. 1er trimestre 2005. pp. 9-28.
doi : 10.3406/rei.2005.3062
http://www.persee.fr/web/revues/home/prescript/article/rei_0154-3229_2005_num_109_1_3062John CANTWELL
Rutgers Business School and University of Reading
and Simona IAMMARINO
SPRU, University of Sussex and University of Rome "La Sapienza"
THE TECHNOLOGICAL INNOVATION
OF MULTINATIONAL CORPORATIONS
IN THE FRENCH REGIONS
Mots-clés : entreprises multinationales, innovation technologie, systèmes régionaux.
Key words : Multinational Corporations, Technological Innovation. Regional Systems.
I. - INTRODUCTION
The nexus between global and local processes has been investigated quite
extensively by the literature of the most recent years. One particular and cru
cial aspect of such a relationship lies in the creation and diffusion of innovat
ion, which, more than other economic processes, show rather complex pat
terns of distribution across space.
Indeed, as emphasised by Dicken, '"global' and 'local' are not fixed scales;
rather, they represent the extreme points of a dialectical continuum of complex
mutual interactions" (Dicken, 1994, p. 103). As a consequence, neither the
orthodox approach - which traditionally considers both the (multinational)
firm and the local system as black boxes whose behaviours are determined by
exogenous factors; nor an entirely endogenous perspective - which tends to
explain structure and growth mechanisms as the result of purely internal
forces - seem appropriate to investigate the issue 'global versus local'. Rather,
structure and behaviour of the two "extreme points" need to be considered
within the context of their increasing interdependence, including both endoge
nous determinants and exogenous variables relevant to the analysis.
Following the line of our previous studies, this paper presents an analysis of
the location patterns of multinational firms' innovative activities in France.
REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, Ie- trimestre 2005 9 The paper is divided into five sections. The following section summarises the
conceptual framework for the empirical study carried out here. After a short
description of the data used for the empirical investigation, section three ana
lyses the geographical distribution of innovative activities of large firms, both
indigenous and foreign-owned, across the French regions in the period 1969-
95. Section four documents the characteristics of the hierarchy of regional
research centres in France, by testing the relationship between the profiles of
technological specialisation in foreign-owned and indigenous French firms. As
we have suggested elsewhere, differences in the regional capacity of attraction
of high value added activities serve as a prerequisite to unveiling the potential
technological communication (or technological spillovers) that may be in exis
tence within these regions and, indeed, across EU national boundaries. Finally,
section five highlights the main implications that can be drawn, and our futu
re research agenda.
II. - MULTI NATIONAL CORPORATIONS, TECHNOLOGICAL
INNOVATION AND EUROPEAN REGIONS
It is widely accepted that the organisation of innovative activities can no lon
ger be illustrated simply on the basis of concepts such as the dichotomy of
market-hierarchies and the transaction costs mechanism as fundamental expla
nations of internalisation/externalisation of capabilities, functions and assets.
Indeed, the trend has been increasingly observed for multinational corpora
tions (MNCs) to establish internal (intra-firm) and external (inter-firm) net
works for innovation, which are characterised by different levels of territorial
and social embeddedness with reference to the location which hosts them.
Regions, even belonging to the same nation-State, show different characterist
ics that determine the degree of attractiveness and the amount of spillovers
that a region is able to draw. The choice of a particular location for a MNC to
invest in research and development activities is thus driven by several factors,
which can be summarised as "social capability" and "technological congruen
ce" (Abramovitz, 1986; Fagerberg, Verspagen and von Tunzelman, 1994).
While the first refers to the overall ability of the region to engage in innovati
ve and organisation processes, the latter points to the distance of the region
from the technological frontier, or, in other words, its capacity to implement
the technical properties connected to the new knowledge.
Furthermore, it has been shown that MNC affiliates abroad have assumed a
predominant role in an increasing proportion of all the most advanced technol
ogies. The interpretation given is twofold: on the one hand, the ceaseless rel
evance of local innovation processes as reservoirs of different technical expert
ise in the globalisation era, and on the other, the outgrowth of an "organisa
tional capital" which allows the integration of several related technological
competencies across geographically dispersed units (Zander, 1997). Thus, the
development of cross-border corporate integration and intra-border inter-com
pany sectoral integration, as new forms of governance, makes it increasingly
1 0 REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 important to examine where and how innovative activity by MNCs is interna
tionally dispersed and regionally concentrated (1).
Whilst the establishment in a foreign location facilitates the monitoring of
developments in different technological fields, it also enables the extraction of
local knowledge for MNC global networks (Cantwell, 1992, 1995; Chesnais,
1992; Granstrand and Sjolander, 1992; Dunning & Wymbs. 1997). In the case
of the former, the firm is likely to be active abroad in technologies where comp
lementarity between the strength of the host economy and its own expertise
exists. In the case of the latter, a firm locates its research facility abroad to
exploit the technological advantage of the host region either to reinforce its
own competence or to alleviate its weakness at home. On this basis, it was the
refore suggested that, behind location choices related to innovation, there are
the two typical types of agglomeration forces, which operate differently across
regions. The first can be termed "localisation economies", which are sector-
specific and tend to intensify intra-border sectoral integration through local
external networks between foreign-owned affiliates, indigenous firms and
local non-market institutions. The second are referred to as general external
economies, or "urbanisation economies", which attract all kinds of economic
activities and tend to strengthen cross-border intra-firm integration, allowing
the feedback of knowledge, expertise and information within networks of affi
liates.
Arising from this differentiation of agglomeration economies, it became pos
sible to distinguish between higher order and intermediate regional
centres (2). Such centres arise "as a consequence of the interaction and the
intensity of general external economies and localisation economies, which in
turn depend upon the characteristics of the regional system considered"
(Cantwell and lammarino, 1998, p. 387). Whilst in the case of an intermediat
e location, knowledge in specific technological fields is accessed and injected
into the multinational network, affiliates located in higher order centres can
enjoy a broader range of spillovers from the local environment. However,
whilst there is evidence that much of the technology developed abroad by large
firms lies in their core areas of strength (Patel and Vega, 1999) (3), MNC
research in foreign locations is also increasingly associated with a higher pro-
(1) For an in-depth discussion on the linkages between the globalisation of innovation and
regions - which highlights the importance of location in the process and the
refore the supremacy of a regional approach when analysing this phenomenon - see
Cantwell and lammarino (2003).
(2) The other extreme is that of lower order regions, i.e. technologically weak and backward
regions that have an inadequate innovative base in order to compete with other locations
and to be attractive for external flows of knowledge and technology.
(3) This suggests that adaptation and technical support to foreign manufacturing plants conti
nue to be major explanatory factors.
REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 1 1 of entry into new and more distantly related fields of technology. Such bability
knowledge-seeking activity is undertaken to help define the future directions
in the evolution of the corporations' sources of competitiveness (Pearce,
1999).
The relationship 'global versus local' implies a mounting competitive bid
ding in order to attract external sources of knowledge and technological comp
etence, by which to enhance the regional base (4) and tap into the
full growth potential of globalisation. The sufficient condition to take part to
the competitive bidding, especially within the European economic arena, is
clearly the dynamics (upgrading/diversification) of regional technological
advantage and the ability to provide opportunities in the most promising tech
nological fields. Such an ability is strictly related to the regional capacity to
engage in "institutional learning", i.e. to adjust the local institutional structure
in order to support, sustain and enhance the development of new technologies
and to adapt to the prevailing technological paradigm. Thus, the "institutional
comparative advantage" of the regional economy is often the underlying rea
son for a more dynamic technological performance and of the gradual shift of
the regional specialisation towards the fastest growing areas of innovative acti
vity. Obviously, not all regional cores are able to adjust their profiles of spe
cialisation to the highest technological opportunities: the cumulative and loca
tion-specific nature of process might eventually imply the rise
and the decline of technological poles within Europe.
III. - MNC RESEARCH LOCATION IN THE FRENCH REGIONS
As in our previous studies, the geographical unit of analysis used to explore
the French case is based upon sub-national entities that derive from normative
criteria, as classified by Eurostat in the Nomenclature of Territorial Units for
Statistics (NUTS), providing a single uniform arrangement of territorial sys
tems. The comparable NUTS level chosen here for France is NUTS 1. The
empirical investigation uses patents granted in the US by the US Patent and
Trademark Office (USPTO) to the world's largest industrial firms for inven
tions achieved in their French-located operations over the 1969-1995 period,
classified by the host region in which the research facility responsible is loca
ted. Each patent was classified into one of 56 technological sectors derived by
mapping from the primary classification of the USPTO and organising patents
into common technological groups (see Appendix for the resulting 56 sectors).
For further discussion of the data and the régionalisation of the patent databa
se see Cantwell and Iammarino (1998, 2000, 2003).
(4) We use the word 'enhance' and not 'construct' because the existence of such a base is a
necessary, although not sufficient, prerequisite in order to have attraction capacity and
thus to be a regional centre of excellence, either as a higher order or as an intermediate
location.
1 2 REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 few selected indicators by region - relative to the end of the period to A
which our database on US patents refers - may provide a brief picture of the
relative size of the French regions. Ile de France appears to be the only region
well above the national average for most indicators: its economic relevance is
leading also when compared to the EU average - the index of GDP per capita
in the middle 1990s being above 175 with respect to the base (EU15=100).
More generally, the most economically advanced areas are certainly those
around the capital region, the centre and the centre-east of the country, which
show a GDP higher than the EU average.
Looking at innovation variables, according to the National Innovation
Survey the most innovative areas are identifiable within the southern and eas
tern belt of Ile de France (within the Bassin Parisien region) - where in fact
the greatest concentration of innovative firms is found by the Survey - whilst
the north-west and the south of the country appear to be relatively lagging
behind. Such a difference in the geographical spread of innovative activities is
mainly attributed to the regional industrial structure - i.e. the industrial sectors
most represented in the eastern part of France are those in which small and
medium enterprises (SMEs) show a relatively stronger propensity to
innovate - and to the effects of local policies for research and technological
development (Ministère de l'Industrie, 1994).
Even accounting for demographic and economic size, both in terms of R&D
expenditure (as a percentage of GDP) and of R&D personnel (as a percentage
of active population) île de France drives up the national figure: 3.3% against
a national average of 2.4% in the case of the former indicator, and 2.3%
against 1.5% with respect to the latter (5). It has to be remembered that, in
absolute terms, the highest R&D expenditure among all EU regions (at the
comparable geographical level) is indeed that of Ile de France: more than
11,400 million of euros in 1995 (the same figure, for example, for the South
East of the UK was approximately 6,000). Moreover, the French capital region
is at the top of the ranking among the EU regions which are both wealthy (per
capita GDP above the EU average) and innovative (R&D as a share of GDP
above the EU average) (Eurostat, 2000).
Turning to the indicator used in the present work, Table 1 reports the shares
of patents granted by the USPTO to large corporate inventors - both the lar
gest French firms and the largest French-located foreign-owned firms - attr
ibutable to research facilities based in the French regions (6). First of all, in
terms of absolute size of large firms' patenting activity, France, showing
(5) It is worth mentioning that, in 1995, the capital region accounted for nearly 50% of total
national R&D expenditure (Eurostat, 1999).
(6) The départements d'outre-mer are not considered here, as they do not register any US
patents by large firms in the period considered. Thus, there are 8 regions included in Table
1, instead of the 9 NUTS1 French regions.
REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, Ier trimestre 2005 1 3 28,106 patents in the period 1969-95 as a whole, represents less than one third
of the overall activity carried out in Germany (with 92,058 patents), lags well
behind the UK (with 35,219), but is far above Italy (with 7,040). The share of
foreign-owned firms in the overall total is 25.6%: this is consistent with other
studies on the economic role of foreign-owned affiliates, which in fact place
France in an intermediate position between the highly globalised character of
the research carried out in the UK and the endogenously-based strength of
German technological competence (OECD, 1999) (7).
It is interesting to report a few features which characterise corporate inno
vation in France, as they emerge again from the National Innovation
Survey (8). In general terms, foreign-owned affiliates show on average a stron
ger propensity towards product innovation than their nationally-owned count
erparts, which seem to be relatively stronger in process innovation. They dif
fer significantly also with regard to the sources of innovation: whilst foreign-
owned firms rely more on external sources, indigenous French firms build pr
imarily on in-house research (Dupont, 1994). To some extent, this might be
interpreted as a relatively more pronounced tendency by foreign-owned affi
liates to establish inter-firm networks for innovation as part of the overall stra
tegy of MNCs outside their country of origin. Moreover, firms
(7) France is ranked between the UK and Germany in terms of both the foreign affiliate share
of R&D expenditure and the share of production (turnover) in manufacturing (OECD,
1999).
(8) The Survey referred to the innovative activities undertaken in the period 1986-91. The
source of information is Dupont (1994).
Table 1 - Shares of US patents of both the largest French firms
and the France-located foreign-owned firms, attributable to research
in the French regions relative to France as a whole, 1969-95 (%)
REGIONS French firms Foreign firms Total
Ile de France 58.3 58.2 58.3
Bassin Parisien 8.4 14.0 9.8
Nord-Pas-de-Calais 1.3 1.0 1.2
Est 3.6 7.0 4.5
Ouest 2.3 2.2 2.3
Sud-Ouest 4.7 1.9 4.0
Centre-Est 17.4 6.9 14.7
Méditerranée 4.0 8.7 5.2
Total France (absolute nos.) 20902 7204 28106
14 REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 show on average a greater sensitivity to the association between technological
innovation and organisational innovation, which is a crucial aspect in the capac
ity of firms to source competencies in various foreign locations.
As emerges from Table 1, geographical agglomeration turns out to be as
outstanding in the French case as in other comparable economies: with refe
rence to the first regional core, more than 58% of the overall patenting activi
ty is concentrated in Ile de France, followed by the Centre-Est, with almost
15%, and Bassin Parisien, with slightly less than 10%. The latter region, howev
er, is the second most popular location for foreign research carried out in the
country (14%), whilst the Centre-Est ranks fifth in order of importance,
accounting only for 6.9% of foreign-owned France-based patenting, in spite of
having the second largest share of indigenous research (17.4%).
The high geographical polarisation of innovation generated in MNCs drove
our choice to restrict the analysis at a more detailed level to the three regions
mentioned above, since the absolute numbers of patents granted in the other
regional sites is too low for meaningful statistical analysis. It is necessary to
bear in mind, however, the differences in the degree of attractiveness of exter
nal resources that - even in comparison with the patterns of spatial distribution
of indigenous research - mark out these three regional systems, which per se
lend support, at least at first glance, to our hypothesis of the existence of a
regional hierarchy within national boundaries.
The sectoral distribution of patents demonstrates interregional differences to
an even greater extent. Table 2 reports the foreign shares by sector and region
(foreign-owned firms' percentage of total patents granted to large firms for
research located in the region) in the 27 years considered (9). The highest
contribution of foreign research to the regional total is recorded in Bassin
Passin (36.5%), which is much above that observed for île de France (25.6%,
identical to the national average); as expected, the Centre-Est is far less attrac
tive - with the foreign share accounting for just 12% of regionally -based
research.
It is interesting to note that, in all cases, the most relevant contributions of
foreign-owned research to the local totals are found in some fast-growing and
'general purpose' technological fields, which are likely to lie at the heart of
spillovers between indigenous and foreign-owned firms. This is particularly
the case of Information and Communication Technology (ICT) sectors - such
(9) Some of the technological sectors were dropped from the table on the grounds of the rela
tively small number of patents for corporate research in the country as a whole, the ou
tcome being that only 31 sectors are reported in Tables 2 and 3 and thus discussed in the
text. However, the total 56 refers to the total number of patents for all 56 technological
sectors. The key to the sectoral codes is given in the Appendix.
REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 1 5 2 - Foreign shares (foreign-owned firms' percentage Table
of total patents granted to large firms for local research in France),
by sector and region, 1969-95
REGIONS
SECTORS FRANCE
ILE DE FRANCE BASSEV PARISIEN CENTRE-EST
3 3.6 10.8 1.0 5.5
5 13.9 44.5 10.9 18.6
7 22.1 38.3 4.8 23.6
17.5 10.9 5.3 17.0 9
10 4.8 5.3 6.3 17.1
11.5 2.5 25.5 11 40.5
12 35.3 50.0 27.5 37.5
13 16.0 25.6 9.6 13.7
14 25.0 49.4 22.4 24.9
16 12.8 30.5 9.8 18.1
16.0 17 21.0 36.2 19.6
20 28.9 52.0 3.7 28.9
23 20.4 28.6 50.0 21.9
28 7.3 33.3 34.4 23.6
29 29.4 37.6 40.5 29.0
31 12.8 15.8 25.0 13.1
35.6 50.0 25.0 42.4 33
34 38.4 50.0 27.6 39.8
- 35 17.9 25.0 18.9
36 29.9 60.0 2.4 28.3
14.0 65.9 5.6 19.9 37
38 20.5 56.6 15.8 24.6
17.7 66.0 18.7 20.4 39
13.5 40 30.0 77.9 39.7
41 35.4 75.6 14.1 40.2
42 13.0 51.4 60.0 24.7
49 10.5 45.2 3.5 13.1
50 27.3 10.1 4.8 19.3
2.8 51 2.2 32.8 8.9
53 30.9 45.8 25.8 30.4
56 16.1 9.8 22.7 16.2
Total Tech56 25.6 36.5 12.0 25.6
as Telecommunications (33, 34), Semiconductors (40) and Office equipment
and data processing systems (41) - but also of Other general industrial equip
ment (29) and Other instruments and controls (53). A high foreign share at
both national and regional levels is observable in Pharmaceuticals and bio
technology (12).
As expected - and in line with what was observed for the South East in the
UK, Baden Württemberg in Germany and Lombardia in Italy - the sectoral
16 REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 of foreign-owned companies' shares of regional and national distribution
research activity shows a very high degree of correspondence between île de
France and the country as a whole. However, contrary to what was observed
in the UK and Italian studies, which, for the regional centres other than the
most prominent one, showed more focused and locally specific sectoral pat
terns in the distribution of foreign participation in regional research, the other
two French regions indicate foreign contributions that are relatively wides
pread at a sectoral level. This is especially the case of Bassin Parisien, which
shows remarkably high foreign shares in all sectors classified as various cate
gories of electrical equipment (and not only in those identifiable as ICTs) -
such as Illumination devices (37), Electrical devices and systems (38) and
Other general electrical equipment (39) - and also in some chemical processes
(5 and 7). The Centre-Est shows foreign shares that are well above average in
a different set of mechanical fields, such as for example Mining equipment
(23) and Other specialised machinery (28), which have instead relatively lower
values in the other regional cores and in the country as a whole.
Looking at the evolution over time of, respectively, the French-owned and
foreign-owned firms' total number of patents, it emerges that both French and
foreign patenting have followed a rather similar and discontinuous path:
increasing in the earliest 1970s, with a sharp downturn in the latter half of the
decade - which fell to a trough in 1979 - and then increasing again up to the
beginning of the 1990s, since which time the trend has been declining (see also
Cantwell and Kotecha, 1997). The evolution of foreign shares between 1969
and 1995 shows a great deal of differentiation across regions. île de France and
the country as a whole have rather stable foreign shares, whose development
over time is almost coincident, becoming divergent only in the last years
observed. Bassin Parisien has been characterised by a very high degree of fluc
tuation: although the foreign contribution to total regional patenting has remai
ned the highest throughout almost the whole period, in 1995 it was more or
less at the same level as at the end of the 1960s. The Centre-Est has instead
followed a steadily increasing trend which, in the early 1990s, has enabled this
region to overtake the capital core in terms of foreign share of research activit
y.
In general terms, the above picture confirms the traditional representation of
the French innovative and economic system, which is known to be one of the
most geographically hierarchical in Europe. The centralisation of actors and
activities - firms, capital, infrastructures, innovation, lobbies - within the Paris
region has been recognised as one of the key structural factor of the French
economic geography, and also the gradual process of decentralisation underta
ken during the 1970s and the 1980s was organised in hierarchical terms (some
times viewed rather as a 'metropolitanisation' process). Indeed, contrary espe
cially to the Italian examples of agglomeration - such as the famous industrial
districts - French local systems are much more 'urban systems', i.e. growth
poles represented by large industrial cities (a typical example is, Grenoble in
the Centre-Est), with weak connections with the rest of the territory, highly
REVUE D'ÉCONOMIE INDUSTRIELLE — n° 109, 1er trimestre 2005 1 J