Technological Forecasting & Social Change 76 (2009) 410–421

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

The role of networking in the competitiveness of firms

Isabel Álvarez ⁎,a, Raquel Marin b, Antonio Fonfría a
a
Departamento de Economía Aplicada II, and Instituto Complutense de Estudios Internacionales, Universidad Complutense de Madrid, Madrid, Spain
b
Universidad Europea de Madrid and Instituto Complutense de Estudios Internacionales, Universidad Complutense de Madrid, Madrid, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Two main forces characterise production systems today: on the one hand, the increasingly
Received 24 March 2008 global nature of markets and economies has resulted in increasing competition and a new,
Received in revised form 6 October 2008 more global division of labour. On the other, the greater complexity of technology makes
Accepted 9 October 2008
innovation a key aspect in the competitiveness of manufacturing firms. The establishment of
cooperative networks seems to be important in both processes. This paper aims to explore
JEL:
these aspects by analysing the competitiveness of firms in four different sectors of the
D24
D85 manufacturing industry: food, chemicals, electronics and vehicles. Data have been obtained
O32 from a survey conducted specifically for this purpose at company level in Spain. Findings from
the empirical analysis, based on the application of the Polytomous Logistic Universal Model
Keywords: (PLUM), confirm the positive effects that the ability to network has on company performance.
Networks In particular, among all the potential organisations that work as partners of the firms, the paper
Innovation shows the importance of intra-firm cooperation, the user–producer relationships and the
Competitiveness cooperation between competitors.
Inter-firm cooperation
© 2008 Elsevier Inc. All rights reserved.

1. Introduction

The increasingly global nature of markets and economic activities has resulted in increasing competition and a new and more
global division of labour. On the other hand, the greater complexity of technology makes innovation a fundamental and
differentiated element for the competitiveness of firms. The establishment of networks seems to be important in both processes.
The aim of this paper is to analyse the impact of relationships between manufacturing firms and other independent organisations
on the improvement of competitiveness, by studying firms' preferences in choosing their partners, and analysing their networking
abilities — essential elements to try and provide a solid framework for understanding whether or not firms are willing to engage in
sharing knowledge with potential competitors.
The concept of competitiveness has been much discussed by academics.1 Competitiveness normally applies as a relative term,
not reflecting absolute performance, but generally having a double meaning related to economic well-being and trade
performance of economies [2]. Considering the impact of knowledge-based economies, competitiveness is defined as a nation's
ability to command significant world market share in high technology products while maintaining the living standards of its
citizens [3]. This leadership connects industries and competitiveness because it derives from the interactions between firms,
regions and countries, and the factors influencing competitiveness are not only resources and market conditions but also
institutions and inter-company networks [4]. Networks are understood as a hybrid form of organisation defined by interactions
among agents, institutions and environmental conditions [5–7]. They include strong and weak ties among agents, manifested in

⁎ Corresponding author. Departamento de Economía Aplicada II, Facultad de Económicas, Universidad Complutense de Madrid, Campus de Somosaguas, 28223-
Pozuelo de Alarcón, Spain. Fax: +34 913942458.
E-mail address: isabel.alvarez@ccee.ucm.es (I. Álvarez).
1
For a critical view see [1].

0040-1625/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.techfore.2008.10.002
 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421 411

formal contractual relationships such as subcontracting, strategic alliances or industrial consortia as well as informal ties, differing
in duration and stability and the specificity of the task [8,9].
As far as current evidence allows, one objective of this paper is to observe whether network indicators differ between industries
in fostering company competitiveness. It is also interesting to compare whether domestic and foreign-owned firms behave in
different ways and to test the importance that belonging to an international network has in terms of competitiveness. We explore
these two aspects through the performance of four industries in the manufacturing sector: food, chemicals, electronics and
vehicles. In contrast to most available empirical evidence, this analysis focuses not only on high-tech industries but also on other
less technologically sophisticated ones. In order to test degrees of networking and competitiveness at a company level, micro data
were obtained from a survey on competitiveness conducted at company level in Spain.2
The paper is structured as follows: in the second section, we briefly review the main aspects of the relationships between
globalisation, networks, technology and competitiveness that frame our empirical analysis. In the third section, there is a
description of the competitiveness and technology-related variables from the Spanish survey results, differentiated according to
both industries and the ownership of companies. In the fourth, there is a characterisation of the relationships that companies
develop with other companies and also with other kinds of independent agents. In the fifth section, the empirical model is
developed in two steps: firstly, an explanation of the competitive profile of firms according to a set of indicators that were
specifically constructed following principal components' analysis; secondly, we explain the results obtained from the econometric
estimation based on the application of the Polytomous Universal Model (PLUM), which allow us to confirm the role played by
networking and innovation abilities in the level of competitiveness. Section six contains some concluding remarks.

2. Networks, technology and competitiveness

The establishment of networks has been an emerging issue over the last decades for understanding the improvement in
competitiveness of firms, generating increased interest in scholars of various disciplines as well as in practitioners. In accordance
with the literature, motivations for cooperation are grouped into two: the complexity of technological development and the
uncertain and costly nature of research, and market access and search for opportunity [10]. It is possible to look at them not on an
exclusive basis but taking its possible complementary character. The forces of the globalisation process imply an increasingly
international scope in competition and at the same time, there is a global dispersion of innovative activities caused by the
restructuring of the value chain in the most internationalised firms. Companies have evolved over recent decades toward the
development of new organisational forms of production and inter-organisation partnerships have become a core component of the
corporate strategy [9].
The arrival of new technology opportunities, the spread of innovation in the field of ITC and the shift of the external boundaries
of firms justify the contribution of networks to the innovative capabilities of firms [11]. Nevertheless, explanations concerning the
nature of these transformations are not completely new [12]. Companies' internal resources are in many cases insufficient to
achieve greater economies of scale, to reduce the levels of uncertainty involved in accessing new markets and to exploit new
business opportunities. For these reasons, it has been largely accepted that the reduction of transaction costs [13,14], such as those
related to negotiations and the establishment of contracts between firms, is behind the emergence of a new framework of
relationships. Networks can then be identified with a kind of hybrid form of organisation, with reciprocal patterns of
communications and exchange, that combines the incentive structure of the markets with some hierarchical elements [15].
On the other hand, the increasing internationalisation of companies explains why local productive systems are characterised by
contexts of competition in which both domestic and foreign-owned firms cohabit in the same geographical unit, in many cases
operating in the same sector and then competing in the same international market segment. In these circumstances, access to new
information and knowledge becomes a critical element and one of the most powerful motivations for the establishment of
different external ties. The complexity of technology and the uncertainty of innovation justify more than two agents being involved
in knowledge generation processes and external agents are observed more and more as a source of R&D. This would explain the
greater collaboration with universities and public research centres in the development of products and processes, and with
domestic and foreign companies as well as with customers [9]. Therefore, it is largely accepted that both competing and
cooperating relationships may work in favour of the enhancement of firms' competitiveness levels [16,17,18], the underlying
process being related to the growing importance that knowledge access has acquired in competitiveness, the role of technology in
the globalisation process and the need to reduce transaction costs related to technology transfer [10,19–22].
Pioneering contributions on this topic in the literature of industrial organisation and innovation focused on the importance of
geographical proximity in explaining the dynamic of company performance and innovation. While economic analysis has tried to
explain how locations affect a firm's decision to cooperate with other agents, there has been a growing interest in the conditions
under which knowledge flows take place and the issue of geographical proximity emerges as the main explanation for industrial
and research activity agglomerations [23]. This has been a primary aim of the literature on industrial districts, namely, looking at
firm clusters and their determinant factors. Available evidence confirms that networks may increase firms' competitiveness in

2
Although this analysis is based on firms in Spain, this paper was produced as part of the Project called Competitiveness, under the V Framework Programme of
the EU. The Spanish results come from a survey carried out in accordance with common methodological guidelines for all the European partners involved in the
project. The objective was to obtain comparable data for Spain, Poland, Hungary and the Czech Republic, in order to prepare further comparative papers.
Nevertheless, the authors are solely responsible for the views herein and they do not represent those of the EU, nor is the EU responsible for any use that might be
made of data appearing in this paper.
412 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421

general, even favouring the performance of small and medium enterprises [24,25]. Beyond the paradigmatic Italian case, other
experiences have been reported for other European countries as well as for North America and Japan [26–30]. Nevertheless,
proximity is an issue not necessarily bounded by space but one which has also been discussed from a cognitive point of view: the
importance of a common pool of shared knowledge is a driving factor in interaction between agents, organisations and the
institutional elements that contribute to define the benefits of the network dynamics [31,17]. It is commonly accepted that
technology – or more broadly, knowledge – becomes one of the key determinants of firm competitiveness.
Aspects such as trade performance, productivity and specialisation trends are some of the variables suitable for analysing
competitiveness. Even if the set of factors enhancing the competitive levels is broad, technological change is specially noted in the
neo-Schumpeterian tradition. Accordingly, international competitiveness would be related to two major factors: innovative
activities and inter-sectoral diffusion of advanced knowledge [32]. One branch of the spillover literature, on the other hand, is
devoted to understanding whether knowledge, not fully appropriable by the agent who generates it, can reach other nearby firms
or whether it spreads on an international level [33]. Two main forces seem to be at work simultaneously: the growing international
nature of research and development (R&D) activities [34] and the strength of local systems of innovation as the closest framework
for action [35]. A more complete view of the problem would require the joint consideration of them both in the analysis of how
cooperation boosts competitiveness.
The existence of market inefficiencies in the assignation of intangible assets, such as technology, has generated several
explanations about the internationalisation of companies based on market imperfections. While the eclectic theory turns to
ownership, location and internalisation (OLI) advantages to account for the strategies of multinational corporations (MNCs)
[36,37], the transaction costs theory promotes the advantages of vertical integration to avoid or reduce uncertainty [5,14,15]. On the
other hand, the evolutionary perspective takes innovation and technology development as core elements of market failures and
stresses the importance of company hierarchies and routines as the basis for understanding the process of internationalisation
[38–41]. It has also been demonstrated that the changing strategies of large MNCs have developed into the increasing
internationalisation of core activities such as R&D [42–44]. Although R&D activities are mainly concentrated in certain regions and
still rarely globalised [45], the way in which firms gradually become more internationalised in their innovation activities and
capabilities is still a source of current debate [46]. R&D internationalisation would give greater autonomy and independence levels
to affiliate companies established abroad and this aspect has inherent effects on host locations. The global strategies of large firms
and the increasing relationship between science and technology are identifying proximity as an issue of greater importance,
highlighting the relevance of local capabilities in the attraction of inward foreign capital. Public research centres, local
infrastructure, education systems and the local scientific base are elements particularly valuable for foreign-owned firms. These
elements, connected to the dynamics of intra-firm networks are conditioned by the independence level of the R&D activities
carried out by affiliates abroad and how they become competence-creating units [47,48].

3. Data description: Company performance and technology

The findings reported here are the results of an analysis conducted at a company level in four Spanish manufacturing industries:
food, chemicals, electronics and vehicles. The selection of industries ensures a better assessment of networking and
competitiveness in units of differentiated technological content. Data from a firm-level survey allow us to observe the information
provided directly by company managers and it refers to company results in two specific moments in time: 1998 and 2003. The aim
of the survey was to obtain data on competitiveness and to explore the impact of innovation and the network links of firms.
Selection of the companies followed the criteria of guaranteeing a representative sample and getting a balanced presence of units
from the different industries3: the smallest proportion of companies came from the vehicle industry, which made up 19%, and the
highest proportion was from the food industry, up to 30%. In terms of capital ownership, the survey included both domestic and
foreign-owned firms. In the company sample, 20% are foreign-owned and some differences exist across industries4: only 5% of the
food firms are foreign-owned, compared to over 37% in the chemical industry. In terms of company size, measured by the number
of employees, the results of the survey were mostly representative of medium-sized firms, with an average of 186 employees.
Again, differences across industries are noticeable, with the largest firms belonging to the motor vehicles industry and the smallest
being found in the food industry with an average size of 73 employees.
The homogeneous standard measure of competitiveness used in the survey is the company's market share of its main product.
The results obtained do not reveal a sole profile.5 A quarter of the firms face competition, with their sales capturing less than 10% of
the market. On the other hand, another quarter of the sample shows quasi monopolistic behaviour that dominates between 90 and
100% of the market. This is most notable in the food and vehicles industries where in both, more than 34% of domestic and foreign

3
The number of responses obtained was 134 from a sample of 554 firms, making the response rate 29% and thus representative. The error margin of the
sample is +/− 4.1% at 95.5% with p = q = 50%. The method for the sample selection is systematic random sampling that selects units at a fixed interval throughout
the sampling frame after a random start. This is done to obtain an unbiased variance estimator [49]. The sample includes firms with more than 50 employees with
production in Spain — and not only commercial. The database Fomento de la Producción 2002, which includes more than 30,000 firms in Spain, provided the
information for construction of the sample.
4
This coincides with figures on the presence of foreign firms in Spanish manufacturing industries: foreign firms represent 22% of manufacturing employment
[50].
5
Competitiveness has been measured according to categorisation of firms' market share in percentiles. Nevertheless, several other questions were introduced
into the survey to control for potential bias of this measure [51]. In addition, the information contained on the SABI database (Sistema de Análisis de Balances
Ibéricos, Bureau van Dijk, Belgium, 2000) enabled us to control outputs from the individual questionnaires.
 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421 413

Fig. 1. R&D intensity — percentage over sales, 1998 and 2003.

firms have reported a near complete domination of the market by their main product6 — see [51] for more details. Regarding shares
in the domestic market, the majority of companies in the four selected industries mainly operate under competitive conditions.
Nevertheless, just over one third of the firms in the vehicles and food industries show an important market power (38% and 36%,
respectively), an aspect that is less common in R&D-intensive industries. Comparing the results between 1998 and 2003, there has
been a trend towards the concentration of the market mainly in the chemical sector.7
Company managers were also asked to answer two questions related to their competitiveness profile in the international
market: their relative strength in both main products and in production technologies.8 With regard to products, the majority of
firms have declared a moderate level of competitiveness. However, almost 50% of firms in the chemical sector declare themselves
to be strongly competitive. Considering the differences between foreign and domestic firms, although most of them are placed in
the segment of moderate competitiveness, the level is notably higher among the former. On the other hand, competitiveness in
production technology is also considered moderate by most firms and differences across industries persist. In fact, for nearly half
the chemical companies their international competitiveness is strong; in contrast, nearly 28% of firms in the food sector declare
their competitiveness level as weak. Regarding ownership, a notable 43.5% of foreign-owned firms rate themselves strongly
competitive, whereas domestic firms have a moderate level of competitiveness. This result coincides with other available evidence
on Spanish manufacturing industries [52].
Assuming the relevance of the relationship between innovation and competitiveness, [53–55, 31], a diagnosis and exploration
of the companies' technology-related activities could provide a better understanding of their competitive profiles. Firstly, in terms
of the most formalised knowledge-generation activity, the institutionalisation of an R&D laboratory was carried out in 65% of firms
(out of 128 answers to this question). In the chemical industry, it is noteworthy that all the foreign firms and nearly 90% of the
domestic ones have an R&D or design unit, whereas in the food sector, R&D is less institutionalised, being more likely in the
foreign-owned firms of this industry. With regard to R&D intensity, measured as the proportion of sales revenues that firms devote
to R&D expenditures, 5% is the average value for the whole sample in 1998 and reaches 6% in 2003 (Fig. 1). The R&D intensity of
domestic vehicle and chemical firms is exceptionally high in 1998 — 8% and 7%, respectively; domestic firms in the chemical sector
retain this high intensity in 2003. Another R&D-intensive industry is electronics with 5.76% of sales in 1998 and 6% in 2003. The
lowest value of this indicator corresponds to foreign-owned firms operating in the food industry. Across periods, R&D carried out
by foreign firms rose in the other three industries, particularly in the vehicles industry it multiplied by more than 3.5. Although
domestic firms generally devote more resources to R&D than foreign-owned companies, the latter have more laboratories and their
formal innovation activities are more feasible.
The high costs and degree of uncertainty in knowledge generation is one powerful reason to explain why firms very often resort
to external sources of ideas [56,10]. In-house R&D is not always sufficient to accomplish complex projects of technological
development and in some cases other agents complement companies' in-house abilities. Consultation on this topic in Spain
provided us with a picture of the types of institutions favoured by cooperation and subcontracting. Relationships with domestic
universities, as well as with materials and machinery suppliers are the most habitual type of cooperative agreements, whereas

6
According to general data publicly available for all the manufacturing industries in Spain, the highest market concentration corresponds to vehicles and office
machinery industries, where the industry concentration (C4) index achieves values of 73% and 75% respectively [50].
7
These results are based on a more extensive analysis [51].
8
Although these answers are based on the subjective assessment made by company managers, a pilot analysis with a short selection of case studies was
elaborated prior to the survey to confirm the validity of the questions and to test potential bias in the executives' answers.
414 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421

Table 1
Cooperation and subcontracting R&D activities by type of institution, percentage of firms

Type of institution Cooperation Subcontracting

Private domestic research institutes 7.91 16.03
Public domestic research institutes 12.43 9.92
Domestic universities 20.90 16.03
Private foreign research organisations 2.26 6.11
Public foreign research organisations 1.69 3.05
Foreign universities 2.82 3.05
Raw material suppliers 20.34 11.45
Machinery and equipment suppliers 15.82 12.21
Independent researchers 5.08 15.27
Other firms with which yours has capital ties 8.47 5.34
Other 2.26 1.53

n = 89.

public and private research organisations and foreign universities are the least involved (Table 1). Regarding subcontracting,
Spanish universities and private research institutes are notably involved in cooperative R&D with firms, and individual researchers
are also intensively subcontracted by firms. In terms of the kind of activities, scientific research and design are the activities which
are most likely to be performed from outside the company. However, there is a predominance of in-house R&D in relation to
product and process innovation and quality control9 [51].
The number of patents granted to firms, which is generally used in the innovation literature as an indicator of technological
performance, shows that patenting propensity differs by industries but also because of other structural characteristics of firms,
such as their ownership [57–59]. Although the firms included in the sample do not declare a high number of patents in the last five
years, international patents have a higher representation in these industries than those granted in Spain.10 Comparing industries,
the chemical firms show a higher propensity to patent in both markets. The value of the two kinds of patents for domestic firms in
the vehicles industry is also especially important, being more notable for the international type. Finally, according to both the
related literature and the Oslo Manual guidelines, the introduction of new products to the market is one of the most outstanding
innovation indicators [60]. On average, nearly a quarter of company sales are due to new products, those generated in the last two
years (21% of sales in 1998 and 25% in 2003). Innovative capacity is even higher in the vehicles industry: more than 30% of its sales,
the gap with other branches having decreased between 1998 and 2003. Among the firms in this industry, domestic companies
seem to be the most innovative in products: 45% in 1998 and 40% in 2003.

4. The network relationships of manufacturing firms

Networks provide firms with the possibilities to enlarge the flows of resources, to access new ideas and to increase the benefits
from transfer of knowledge [61]. Companies' interactions are in some cases horizontal – with competitors – and also vertical —
mainly with customers and suppliers; the type of complementary assets could define the most preferable relationships and
condition the firm's choice of partners. In an increasingly internationalised world, with the predominance of a large dispersion of
innovative activities in the international networks of MNCs, an additional issue is whether cooperation is established with foreign
or with domestic partners [44,62,63].
The results obtained from the survey allow us to ascertain what type of activities are carried out in cooperation by firms as well as
the type of partners; that is, whether firms cooperate with competitors, customers and suppliers, these then further differentiated by
their nationality [51]. Subcontracting, technical assistance and acquisitions are the most cited types of cooperation, although the
differences between other types of cooperation are not large. With competitors, the pattern is similar for both domestic and foreign
organisations, and the preferred operations are those related to acquisitions and strategic alliances.11 These are also the most likely
forms of cooperation with domestic and foreign customers. Regarding the relative importance of the type of cooperation by the type of
agent and its nationality, the most relevant cooperative activities between foreign partners are company acquisitions, strategic
alliances and cooperation with competitors: more than 20% of firms cooperate through these three ways. On the other hand,
geographical proximity is an important factor for the establishment of networks since cooperation with domestic organisations is more
likely than with foreign agents. In fact, the highest values correspond to cooperation with domestic suppliers, whereas around a quarter
of the companies subcontracted and cooperated in technical assistance with them. Strategic alliances are the most relevant type of

9
These results differ partially from the findings of the same survey in the Czech Republic, Poland and Hungary where the most frequent areas for cooperation
are product quality and design, R&D and delivery terms [57].
10
According to the results of the survey, firms obtained 134 international patents on average, compared with only 75 in Spain [51]. Nevertheless, this is also a
shared general trend in the Spanish manufacturing industry.
11
The reason why we include acquisitions as a type of cooperation is related to the fact that on many occasions, these deals result from previous agreements
between firms. When firms decide to invest and expand into new markets, they may be forced to look for a partner in order to share risks and uncertainty
stemming from the development of new capabilities [62]. When a joint venture is formed, firms negotiate the option to carry out possible acquisitions in future.
Therefore, if the market for the new product is successfully proven, the option to acquire is likely to be exercised, especially if the valuation of the venture has
increased [64]. For an analysis of mergers and acquisitions in Spain, see [65].
 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421 415

Table 2
Objectives of the R&D activities (% of firms)

Food Chemicals Electronics Vehicles Total

Dom For Dom For Dom For Dom For

Development and improvement of products Yes 52.94 100 78.95 72.73 75.00 83.33 47.06 83.33 66.39
No 47.06 0 21.05 27.27 25.00 16.67 52.94 16.67 33.61
Development and improvement of process Yes 55.88 100 72.22 63.64 61.54 50.00 35.29 83.33 58.82
No 44.12 0 27.78 36.36 38.46 50.00 64.71 16.67 41.18
Scientific research Yes 8.82 100 52.63 36.36 23.08 0 5.88 16.67 21.67
No 91.18 0 47.37 63.64 76.92 100 94.12 83.33 78.33
Applied research Yes 11.76 0 68.42 45.45 37.04 33.33 11.76 16.67 30.58
No 88.24 100 31.58 54.55 62.96 66.67 88.24 83.33 69.42
Design Yes 23.53 100 15.79 54.55 62.96 66.67 35.29 66.67 40.50
No 76.47 0 84.21 45.45 37.04 33.33 64.71 33.33 59.50
Quality control Yes 57.58 100 52.63 54.55 52.17 66.67 29.41 66.67 52.59
No 42.42 0 47.37 45.45 47.83 33.33 70.59 33.33 47.41

n = 96.

cooperation when it comes to domestic competitors, whereas technical assistance is the type of cooperation with customers in which
the highest proportion of firms in the sample was involved. This last result is probably due to the relatively low level of both transaction
costs and the resources committed by the companies that this kind of agreement requires [66].
According to the survey results, the impact of networks for R&D activities is noteworthy, as nearly 20% of firms declared having
interactions with domestic universities and research institutes [51]. To a lesser extent, the importance of relationships with both
industrial and end-product customers, as well as with material suppliers was observed. These results highlight, at least partially,
the user–producer relationships as a basic external source for improving firms' internal innovation abilities, which is stressed in
the literature on systems of innovation [18,7]. By contrast, foreign partners, universities, research laboratories and consulting firms
have a slightly more limited relevance. Finally, Table 2 shows whether firms look for cooperation with a finalist or instrumental
end. Two main objectives of firm R&D activities are related to innovation, that is, the development and improvement of products
and processes, whereas scientific and applied research is less often prioritised. Domestic firms are more likely than foreign ones to
devote their R&D efforts to product innovation although this difference is less clear in the chemical industry in which more firms
perform research-oriented R&D: 14 chemical firms out of 26 declared scientific research as an objective, and 18 out of 37 declared
applied research.

5. Empirical model and estimation results

The aim of our empirical model is to discover the explanatory power that networking abilities have on the competitive profiles
of companies. For the econometric analysis we make use of the polytomous logistic regression. The main advantage of this kind of
estimation in relation to the traditional logit regression is that it permits the analysis of categorical dependent variables, allowing
us to analyse several ordered outcomes [67,68]. The dependent variable measures a company's competitiveness in the Euclidean
space — as explained below. The value calculated for each firm indicates its rank in relation to others. In our case, the PLUM analysis
will show the variables that would affect the relative probability of achieving a better competitive position.

5.1. Variables and indicators

A set of indicators is built with the micro data obtained from the survey of four Spanish manufacturing industries and serves to
obtain the competitiveness profile of firms. We first carry out principal components analysis for each group of variables related to
products, market characteristics, company performance and financial items, in order to obtain competitiveness indicators at a
company level.12 The indicator CED (Competitiveness Euclidean Distance) is used to denote the distance of every firm to the
frontier – this is defined as the maximum value of each factor – according to the following formula:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2
CEDj = ∑ 1−xij ð1aÞ
i

where

yij
xij =
max yik
k

and where yij is a value of the ith competitiveness variable in the jth firm.

12
Due to problems of space in this paper, we do not give all details on the set of variables included in the analysis but detailed results are available from the
authors on request.
416 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421

Table 3
CED competitiveness indicator by industry and ownership a

Maximum Mean Median Minimum Std. Dev.

Food Domestic 2.93 2.32 2.49 0.00 0.63
Foreign 2.69 2.11 2.11 1.52 0.83
Total 2.93 2.16 2.48 0.00 0.83
Chemical Domestic 2.78 2.28 2.53 0.00 0.64
Foreign 2.80 2.35 2.44 1.20 0.42
Total 2.80 2.13 2.46 0.00 0.81
Electronic Domestic 2.76 2.37 2.48 1.22 0.37
Foreign 2.81 2.17 2.46 1.03 0.62
Total 2.81 2.34 2.46 1.03 0.42
Vehicles Domestic 2.87 2.25 2.34 1.39 0.41
Foreign 2.47 2.26 2.26 2.08 0.16
Total 2.87 2.12 2.28 0.00 0.53
Total Domestic 2.93 2.32 2.48 0.00 0.53
Foreign 2.81 2.26 2.38 1.03 0.45
a
Full details about the definition of the variables included in Factorial analysis in Álvarez et al. (2004) [51].

Therefore, indicator CEDj measures the firm's competitiveness in the Euclidean space and the lower the value of the indicator,
the more competitive the firm. We also calculated several indicators corresponding to competitiveness determinants, making use
of the same method. These include not only networking variables but also human capital and innovation features of the
companies; the indicator CDED integrates human capital and innovation aspects.13 In addition, another six indicators refer to the
networking variables: R&D cooperation (NED1), cooperation with sister companies and subsidiaries (NED2), cooperation with
suppliers (NED3), cooperation with customers and competitors (NED4), benefits from cooperation for customers and suppliers
(NED5), and areas of benefits from cooperation (NED6).14
The bulk of empirical analysis highlights the results from the regression of the CED indicator against those potential
determinants. The generated indicators provide a ranking of firm competitiveness and for this reason, the PLUM seems to be the
most suitable estimation technique. As will be shown below, the CED indicator obtained for every firm has a continuous nature,
adopting values from 0 to 3, although in this estimation method it is necessary to consider it as a categorical variable. According to
the values obtained, we adopt the following criterion:
8
< 0 if 0VCEDV1
CED = 1 if 1bCEDV2
:
2 if 2bCED

where 0 = the best competitive position, 1 = an intermediate profile and 2 = a less competitive company.
It is reasonable to suppose that a company's competitiveness profile could differ according to its ownership and consequently,
we opted to include the nationality of companies as a control variable in the regression.
Table 3 shows the values of the CED Indicator and there are no significant differences between industries. The vehicle sector shows a
slightly higher competitive performance compared to the others, revealed by the lowest mean and median values. The food industry
presents the highest CED values, denoting the worst relative competitive position; however, this industry shows the highest
heterogeneity. In terms of ownership in each sector, domestic firms perform only slightly better in the vehicle and chemical industries
but in general terms, foreign firms show a higher performance, being more competitive than domestic firms and less heterogeneous.
Nevertheless, there are no significant differences between foreign and domestic firms when we look across industries.
Regarding other determinants, a brief description allows us to observe some common patterns.15 Firstly, taking human capital
and innovation as a competitiveness determinant (CDED), the best results of this indicator correspond to chemical and vehicle
industries, denoting the importance of both aspects in these sectors. The higher heterogeneity corresponds to the food industry,
and in terms of company ownership there are no great differences between the two groups. As expected, foreign firms generally
concede a greater importance to aspects such as workforce training levels and technological innovation than domestic-owned
firms, and the former presents the lowest dispersion.16

13
We carry out an analysis for each group of variables related to both competitiveness and networking. For the first group, two principal components were run
in order to obtain the CED indicator: one regarding product and market characteristics and the other one related to performance and financial items. In the
former, five components were obtained and the total variance explained was 70.9%. In the latter, the output was six components and the total variance explained
was 74.1%. A third analysis was carried out in order to obtain the competitiveness determinants related to human capital and innovation (CDED); in this case, six
components were obtained and the total variance explained was 67.1% [51].
14
With regard to the networking variables, six different analyses were run, one for each kind of cooperation aspect. First, for the cooperation in R&D activities
(NED1), eleven components were extracted with a total variance explained of 69.5%. Second, for cooperation with sister companies (NED2), we obtained twelve
components and the total variance explained was 81.6%. Third, cooperation with suppliers (NED3) gave us an output of nine components with a total variance
explained of 78.3%. The indicator related to cooperation with customers and competitors (NED4) extracted three components and the total variance explained
was 85.2%. Finally, the benefits of cooperation (NED5 and NED6) resulted in ten and nine components respectively, the total variance explained being 82.5% and
82.0%. In each of the nine analyses, eigenvalues were higher than one.
15
A more detailed description can be found in [51].
16
These affirmations are based on both average and median values.
 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421 417

Table 4
Differences between foreign and domestic firms according to competitiveness and networking indicators a

Total sample Food Chemicals Electronics Vehicles

Mean difference t-value Mean difference t-value Mean difference t-value Mean difference t-value Mean difference t-value
CED 0.052 0.46 0.219 0.48 −0.070 −0.34 0.206 1.15 −0.011 − 0.06
CDED 0.002 0.04 −0.054 −0.18 0.089 0.61 −0.033 −0.40 −0.166 ⁎ − 1.83
NED1 −0.033 −0.98 0.087 0.71 0.003 0.05 −0.100 −1.50 −0.048 − 0.69
NED2 −0.020 −0.98 0.029 0.50 −0.060 −1.32 0.024 0.67 −0.022 − 0.55
NED3 0.082 1.67 0.121 0.58 0.169 ⁎⁎ 1.97 −0.093 −0.75 0.060 0.54
NED4 0.030 0.60 0.419 ⁎⁎⁎ 2.77 0.180 ⁎⁎ 1.78 −0.110 ⁎⁎ −2.25 −0.067 − 0.77
NED5 −0.002 −0.10 0.002 0.02 −0.007 −0.36 0.028 0.67 0.010 0.24
NED6 0.012 0.56 0.060 1.16 0.019 0.37 −0.002 −0.05 0.000 0.01

According to Levene's test the equality of variance can be assumed.

a
t-test of the difference between mean values.
⁎ Significant at 10%.
⁎⁎ Significant at 5%.
⁎⁎⁎ Significant at 1%.

Regarding the main results of networking indicators (from NED1 to NED6), the following remarks summarise the comparative
behaviour of domestic and foreign-owned firms across industries and some of them are confirmed in the t-test, reported in Table 4:
– The values of NED1, the indicator of R&D cooperation, show that this is an aspect which is less important for foreign firms.
Comparing the results across industries (Table 4), there are no significant statistical differences between them. The same
happens in terms of cooperation with sister companies and subsidiaries; the NED2 indicator reveals no relevant differences
between industries. Although domestic and foreign firms may differ in these two issues, no statistical differences have been
confirmed.
– According to the description, industries with the lowest technological content give the highest importance to upstream
relationships, outsourcing and cost reduction being some of their main motivations. In terms of the cooperative behaviour of
the NED3 indicator of cooperation with suppliers, there are no significant differences for the total sample. Having said that,
cooperation with suppliers is one aspect in which the behaviour of foreign firms in the chemical industry statistically differs
from domestic companies (Table 4).
– Regarding downstream cooperation and horizontal type, NED4, that is, cooperation with customers and competitors, the
results reveal that the highest importance is given to it by the chemical sector. Taking into account the capital ownership
variable, the NED4 does not show significant dissimilarity between industries when considering all the firms in the sample.
Nevertheless, there are notable and significant differences in the behaviour of domestic and foreign-owned firms in food,
electronic and chemical industries (Table 4).
– Finally, the NED5 indicator considers the benefits obtained by customers and suppliers from cooperation activities and it does
not reveal any particular differences across industries. The same occurs in the network indicator concerning the areas of
benefits from cooperation, in which there is a great similarity among industries as well as among companies. The indicator
NED6 seems to be independent of the ownership structure of the firms, without showing any clear differences with regard to
this issue.

5.2. Estimation results

The CED Competitiveness indicator was regressed against the different independent variables – CDED and NED indicators –
previously described, in order to analyse the impact of both innovation and networking on the competitiveness of firms. Regarding
the expected signs and the meaning of the estimated coefficients and according to the interpretation of these kinds of models,
negative coefficients would indicate an inverse relationship between dependent and independent variables, whereas positive
signs show a positive relationship between the predictor and the outcome. It is important to notice that higher values of the
indicators would reveal a worse competitive profile. Therefore, lower CED values mean a higher degree of competitiveness
whereas lower NED and CDED values signify better abilities to network and stronger non-network determinants of
competitiveness, respectively. Consequently, positive signs of the coefficients would mean higher networking and innovation
activities (NED and CDED indicators) and they probably would result in a better competitiveness profile for a company.
The firms' ownership is introduced as a control variable in the analysis and its interpretation is different; negative signs would
improve the competitiveness profile of firms. The reason is that this variable accounts for the nationality of the firm whereas the
dependent variable (CED) refers to its competitiveness. Thus, the lower the value of the CED indicator, the higher the
competitiveness will be. Therefore, a negative sign in a firm's nationality would indicate that foreign firms are likely to be in the
lower CED categories and in consequence, they would perform better.
Estimated results show that two networking indicators tend to increase the probability of firms' improvement in
competitiveness: the one related to cooperation with sister companies and subsidiaries (NED2), and the indicator of downstream
and horizontal cooperation (NED4) (Table 5). The indicator corresponding to human capital and innovation (CDED) is also
418 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421

Table 5
Results of the regression of Competitiveness in products and markets Indicator

Estimate Std. Error Wald df Sig.

Threshold [CED = 0] 4.52 2.60 3.03 1 0.082
[CED = 1] 7.55 2.74 7.62 1 0.006
[CED = 2] 28.81 6.06 22.64 1 0.000
Location NED1 0.77 0.72 1.14 1 0.287
NED2 4.90 1.53 10.32 1 0.001
NED3 0.78 1.33 0.34 1 0.559
NED4 2.67 1.08 6.06 1 0.014
NED5 0.66 0.61 1.20 1 0.274
NED6 −0.66 0.78 0.71 1 0.401
CDED 2.64 1.37 3.68 1 0.055
Foreign −1.23 0.57 4.68 1 0.031
Domestic 0a 0

Link Function: Logit.

a
This parameter is set to zero because it is redundant.

significant. Their positive coefficients show that a better position in these indicators implies a greater probability of obtaining
higher competitiveness levels.
These findings confirm the importance that internal networks inside large business groups may acquire due to the choice of
firms with which to share complementary skills. The character of the sample, which is composed mainly of medium-sized firms in
some manufacturing industries may condition the coefficient achieved by NED2. Therefore, these results might be different if the
analysis were repeated on a sample made up of small firms or even integrating firms from the service sectors. A plausible
explanation for the results obtained in this study could be the strong presence that foreign MNCs have in Spain in the four sectors
analysed. It would also capture the influence of the headquarters of Spanish companies, the relationships with their subsidiaries
abroad as well as the strength of the international networks to which the foreign subsidiaries belong [44].
The significance achieved by the NED4 coefficient reveals the importance of inter-firm cooperation in improving a company's
position in the market. The relationships that firms establish with their customers seem to be positive for the improvement of
competitiveness, which confirms that information about particular interests and market demand has in improving the
performance of firms. This would contribute, for instance, to the reduction of risks associated with the introduction of a new
product into the market. The formal relationships that some Spanish companies have with their competitors also affect the
performance. The high complexity that is intrinsic to both products and processes, and the necessity to combine efforts are behind
these kinds of company strategies; this positive effect of cooperation with competitors on productivity levels has been also
confirmed, for instance, in the case of Dutch companies [16].
The indicator of human capital and innovation (CDED) is also significant and its sign is positive; the high values of this indicator
would imply a greater likelihood of competitiveness improvements. These findings denote the importance of qualitative aspects
beyond traditional cost determinants; definitively, gaining more market share requires greater efforts by firms in hiring qualified
personnel and a high degree of innovation. Moreover, this goal coincides with the idea that cooperating firms are generally
engaged in higher levels of innovation activities [25,16,32] and this collaboration takes place when there are gains from knowledge
integration [69]. This behaviour is especially important in those industries where innovation is complex and technological changes
are frequent, such as electronics and chemicals [70], but it is also in those characterised by high economies of scale, such as the
vehicles sector [39].
Finally, in terms of the ownership of capital assets, foreign-owned firms present a better competitiveness profile than domestic
ones. This is reflected by the negative sign of the coefficient, being more likely that foreign firms present lower CED values. This is
fully consistent with other analyses carried out in manufacturing industries in Spain. Foreign firms, in general, present better
productivity levels than Spanish firms [71,72,52,50,73]. This would confirm the assumption generally made about the superior
performance of foreign firms in host economies and justifies further research into the possibilities of positive spillovers and
derived impacts in the competitiveness of local companies.

6. Concluding remarks

Much of the empirical evidence built up on the behaviour of firms to undertake R&D cooperation addresses the question of
firms' motivations, whether they cooperate with external agents because of technological complexity, the costs and uncertainty of
research, market access and the search for opportunities. Most of the analyses confirm the idea that cooperation is a more common
aspect among large firms and those with high R&D intensity in continental Europe — i.e. results for Belgium, French, German and
Spanish firms [74–77]. External relationships for innovation are confirmed to be complementary rather than replacing internal
R&D, with some exceptions in German, Irish and British cases [78]. Moreover, the level of innovativeness – radical versus
incremental – seems to be significant in explaining inter-firm cooperation in the UK [79]. A resource-based approach is
predominant and the main motivation seems to be technology-seeking rather than market access.
The findings of this paper confirm the role that networking has in the competitiveness of manufacturing firms. The increasing
internationalisation of markets contributes to redefine the importance of cooperative relationships for firms, and we demonstrate
 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421 419

that their effects and the impacts on competitiveness differ according to the type of networks. Moreover, the presence and
influence of foreign capital in local productive systems varies among sectors in the manufacturing industry, contributing to the
definition of the trade-off with their technological opportunities. In a similar vein to the analysis presented here, the co-evolution
of domestic and international knowledge linkages in Italian firms seems to be related with a shift in the technological paradigm
[80], confirming the dynamic approach based on the response of network structure to different knowledge regimes [69].
A first aspect to highlight in this empirical paper is the relevance of the analysis of competitiveness and networking at a micro
level, which is more suitable for a better understanding of individual behaviour. This has obviously been possible because of the
decision to collect data at a company level. A second novel aspect in this research is the inclusion of some industries with different
technological content; most of the evidence about firm cooperation is based mainly on the technological alliances and agreements
of high-tech content industries. A third issue to note is the introduction of company ownership as a control variable, revealing that
foreign firms achieve better competitive results; their position differs by industry in terms of their ability to collaborate.
Cooperation would enhance the greater efforts of domestic firms to maintain and even to improve their position in the
international markets. Fourthly, this methodology would confirm the theoretical foundations of the innovation literature because
the results underline the importance of the amount of resources devoted to R&D activities, the efforts to hire and train a qualified
workforce, as well as to establish cooperation with other organisations in order to speed up innovation.
From our findings, it is possible to pinpoint some arguments related to the underlying process of improving competitiveness
through the networking abilities of firms: in particular, the importance of user–producer relationships for the four sectors
analysed, an aspect underlined in the literature on systems of innovation; the role of qualitative determinants of competitiveness,
such as human capital and innovation; and the relevance of intra-company networks, as shown by the significant role of the
cooperative relationships that firms establish with both sister and subsidiary companies to improve their competitive position.
Further theoretical and practical developments in this area may develop from newer evidence about the competitive patterns
of SMEs. New directions for research could be to carry out deeper comparative analyses of competitiveness and the influential
differences and conditioning factors across industries and countries. Other outstanding questions relate to the different levels of
competitiveness between domestic companies and the subsidiaries of foreign companies, since the presence of the latter is greater
in some economies and industries. Therefore, this dual presence would condition the definition of policy instruments more in
some locations than in others; foreign-owned firms in some industries with a high technological complexity still have aspects that
are worth making more explicit in the definition of actions aimed at improving local competitiveness.
Other ideas that can derive into policy implications from our findings are related to the variety of relevant elements for
innovation and competitiveness that could affect several fields of policies, such as instruments promoting high-quality basic and
applied research capabilities and those aimed at the establishment of greater levels of interaction between firms and other agents
in the local systems, i.e. through the reinforcement of interface mechanisms. Moreover, the significant role of user–producer
relationships leads us to underline the reinforcement of collaborative industrial associations as well as the provision of services
designed to facilitate contacts between companies. Thus, improvement of local capabilities for the generation and diffusion of
knowledge is still an issue for defining public policies in industrial and technological fields in many countries. One final aspect to
underline here is the necessity for a greater interrelation between different policy fields. In general terms, successful actions for
promoting competitiveness must include a broad range of aspects, from workforce qualifications to advertising campaigns or the
promotion of local products abroad.

Acknowledgements

The authors acknowledge the extremely helpful suggestions made by the anonymous referees of TFSC to an earlier version of
this paper. The authors also thank Susan Lees for English editing. The remaining weaknesses are the sole responsibility of the
authors.

References
[1] P. Krugman, Competitiveness: a dangerous obsession, Foreign Aff. 73 (1994) 28–44.
[2] J. Fagerberg, Technology and competitiveness, Oxf. Rev. Econ. Policy 12 (1996) 39–51.
[3] J.D. Roessner, A.L. Porter, N. Newman, D. Cauffiel, Anticipating the future high-tech competitiveness of nations: indicators for twenty-eight countries, Technol.
Forecast. Soc. Change 51 (1996) 133–149.
[4] J. Cantwell, Innovation and competitiveness, in: J. Fagerberg, D. Mowery, R. Nelson (Eds.), The Oxford Handbook of Innovation, Oxford University Press, Oxford,
2005.
[5] O. Williamson, Markets and Hierarchies, Free Press, New York, 1975.
[6] G.G. Hamilton, R. Feenstra, Varieties of hierarchies and markets: an introduction, in: G. Dosi, D.J. Teece, J. Chytry (Eds.), Technology, Organisation, and
Competitiveness, Oxford University Press, Oxford, 1998.
[7] C. Edquist (Ed.), Systems of innovation. Technologies, institutions and organisations, Pinter Publishers, London, 1997.
[8] M. Granovetter, The strength of weak ties, Am. J. Sociol. 78 (1973) 1360–1380.
[9] W.W. Powell, S. Grodal, Networks of innovators, in: J. Fagerberg, D. Mowery, R. Nelson (Eds.), The Oxford Handbook of Innovation, Oxford University Press,
Oxford, 2005.
[10] J. Hagerdoorn, Strategic technology alliances and modes of cooperation in high-technology industries, in: G. Gernot (Ed.), The Embedded Firm: on the
Socioeconomics of Industrial Networks, Routledge, New York, 1993.
[11] J.C. Fernandez-Arroyabe, N. Arranz, Principles for the design of management control systems in knowledge networks. Experiences involving the European
technological networks, Technol. Forecast. Soc. Change 69 (2002) 703–719.
[12] G. Hamel, C.K. Prahalad, Strategic intent, Harvard Bus. Rev. 67 (3) (1989) 63–78.
[13] R.H. Coase, The nature of the firm, Economica 4 (16) (1937) 386–405.
[14] O. Williamson, The modern corporation: origins, evolution, attributes, J. Econ. Lit. 19 (1981) 1537–1568.
420 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421

[15] O.E. Williamson, Transaction cost economics and organisational theory, in: G. Dosi, D.J. Teece, J. Chytry (Eds.), Technology, Organisation, and Competitiveness,
Oxford University Press, Oxford, 1998.
[16] R. Belderbos, M. Carree, B. Lokshin, Cooperative R&D and firm performance, Res. Policy 33 (2004) 1477–1492.
[17] M.S. Freel, Sectoral patterns of small firm innovation, networking and proximity, Res. Policy 32 (2003) 751–770.
[18] B.-A. Lundvall (Ed.), National Systems of Innovation, Pinter Publishers, London, 1992.
[19] M.A. Porter, Competitive Advantage, Free Press, New York, 1985.
[20] C.K. Prahalad, G. Hamel, The core competence of the corporation, Harvard Bus. Rev. 68 (1990) 79–91.
[21] J.H. Dunning, The Globalization of Business, Routledge, London, 1993.
[22] J. James, Technology, Globalisation and Poverty, Edward Elgar, Cheltenham, 2002.
[23] P. Krugman, Increasing returns and economic geography, J. Polit. Econ. 99 (3) (1991) 483–492.
[24] J. Humphrey, H. Schimitz, The triple C approach to local industrial policy, World Dev. 24 (12) (1996) 1859–1877.
[25] S.A. Rosenfeld, Does cooperation enhance competitiveness? Assessing the impacts of inter-firm collaboration, Res. Policy 25 (1996) 247–263.
[26] F. Phillips, Social Culture and High-Tech Economic Development: The Technolpolis Columns, Palgrave Macmillan, New York, 2006.
[27] E.J. Malecki, Technology and Economic Development, Longman, Essex, 1991.
[28] A. Saxenian, Regional Advantage —Culture and Competition in Silicon Valley and Route, vol. 128, Harvard University Press, Cambridge MA, 1994.
[29] A.J. Scott, The role of large producers in industrial districts: a case study of high technology systems houses in Southern California, Reg. Stud. 26 (3) (1992)
265–275.
[30] M. Storper, The transition of flexible specialisation in the US film industry: external economies, the division of labour and the crossing of industrial divide,
Camb. J. Econ. 13 (2) (1989) 273–305.
[31] P. Saviotti, Innovation systems and evolutionary theories, in: C. Edquist (Ed.), Systems of innovation, Technologies, Institutions and Organisations, Pinter
Publishers, London, 1997.
[32] F. Castellacci, Innovation and the Competitiveness of Industries: Comparing the Mainstream and the Evolutionary Approaches, Technological Forecasting and
Social Change forthcoming (2008).
[33] A.B. Jaffe, M. Trajtenberg, R. Henderson, Geographic localisation of knowledge spillovers as evidenced by patent citations, Q. J. Econ. 108 (3) (1993) 577–598.
[34] D. Archibugi, J. Michie, The globalisation of technology: a new taxonomy, Camb. J. Econ. 19 (1995) 121–140.
[35] J. Howells, Regional systems of innovation? in: D. Archibugi, J. Howells, J. Michie (Eds.), Innovation Policy in a Global Economy, Cambridge University Press,
1999.
[36] J.H. Dunning, The determinants of international production, Oxf. Econ. Pap. 25 (3) (1973) 289–336.
[37] J.H. Dunning, Toward an eclectic theory of international production: some empirical test, J. Int. Bus. Stud. 11 (1) (1980) 9–31.
[38] D.J. Teece, Technology transfer by multinational firms: the resource cost of transferring technological know-how, Econ. J. 87 (1977) 242–262.
[39] K. Pavitt, Sectoral patterns of technical change: towards a taxonomy and a theory, Res. Policy 13 (1984) 343–373.
[40] K. Pavitt, Public policies to support basic research: what can the rest of the world learn from US theory and practice? (And what they should not learn), Ind.
Corp. Change 10 (2001) 761–779.
[41] G. Dosi, D.J. Teece, J. Chytry, Technology, Organisation and Competitiveness, Oxford University Press, Oxford, 1998.
[42] R.D. Pearce, Decentralised R&D and strategic competitiveness: globalised approaches to generation and use of technology in multinational enterprises, Res.
Policy 28 (1999) 157–178.
[43] D. Archibugi, S. Ianmarino, Multinational corporations, and the location of technological innovation in the UK regions, Reg. Stud. 34 (2000) 317–332.
[44] J. Cantwell, J. Molero, Multinational Enterprises, Innovative Strategies and Systems of Innovation, Edward Elgar, Cheltenham, 2003.
[45] P. Patel, K. Pavitt, National system of innovation under strain: the internationalisation of corporate R&D, in: R. Barrell, G. Mason, O'Mahoney (Eds.),
Productivity, Innovation and Economic Performance, Cambridge University Press, 2000.
[46] G. Duysters, J. Hagerdoon, Internationalisation of corporate technology through strategic partnering: an empirical investigation, Res. Policy 25 (1996) 1–12.
[47] J. Cantwell, R. Mudambi, MNE competence-creating subsidiary mandates, Strateg. Manage. J. 26 (12) (2005) 1109–1128.
[48] J. Cantwell, L. Piscitello, The location of technological activities of MNCs in European regions. The role of spillovers and local competencies, J. Internat. Manag. 8 (1)
(2002) 69–96.
[49] M.N. Murthy, Sampling Theory and Methods, Statistical Publishing Society, Calcutta, 1967.
[50] I. Alvarez, Empresas extranjeras y efectos de derrame tecnológico o spillovers: una aplicación a las manufacturas españolas, Tesis Doctoral, Universidad
Autónoma de Madrid, (2003).
[51] I. Alvarez, A. Fonfria, R. Marin, Results from a Competitiveness Survey to the Manufacturing Spanish Firms, COMPETITIVENESS European Project WP 6 Del. 4,
2004, (available at www.case.pl).
[52] O. Bajo, C. López, La inversión extranjera directa en la industria manufacturera española: 1986–1993, Pap. Econ. Esp. 66 (1996) 176–190.
[53] G. Dosi, K. Pavitt, L. Soete, The Economics of Technical Change and International Trade, Harvester Wheatsheaf, London, 1990.
[54] L. Soete, B. Verspagen, Competing for growth: the dynamics of technology gaps, in: L.L. Pasinetti, R.M. Solow (Eds.), Economic Growth and the Structure of
Long-Term Development, St. Martin's Press, New York, 1994.
[55] B. Verspagen, K. Wakelin, Technology, employment and trade: perspectives on European integration, in: J. Fagerberg, P. Hansson, L. Lundberg, A. Melchior
(Eds.), Technology and International Trade, Edward Elgar, Aldershot, 1997.
[56] E. Von Hippel, The sources of innovation, Oxford University Press, Oxford, 1988.
[57] R. Woodward, P. Wójcik, Networking and competitiveness, in: I. Hoshi, P.J. Welfens, A. Wziatek-Kubiak (Eds.), Industrial Competitiveness and Restructuring in
Enlarged Europe, Palgrave Macmillan, New York, 2007.
[58] J. Howells, The internationalization of R&D and the development of global research networks, Reg. Stud. 24 (1990) 495–512.
[59] A. Fonfria, Patrones de innovación en la empresa española: un dato para la política tecnológica, Pap. Econ. Esp. 81 (1999) 182–195.
[60] OECD, The Oslo Manual, OECD, Paris, 2005.
[61] A. Kleinknecht, J. Reijnen, Why do firms cooperate on R&D? An empirical study, Res. Policy 21 (1992) 347–360.
[62] R. Narula, A. Zanfei, Globalisation of innovation, in: J. Fagerberg, D. Mowery, R. Nelson (Eds.), The Oxford Handbook of Innovation, Oxford University Press,
Oxford, 2005.
[63] J. Hagedoorn, G. Duysters, External sources of innovative capabilities: the preference for strategic alliances or mergers and acquisitions, J. Manag. Stud. 39 (2)
(2002) 167–188.
[64] B. Kogut, Joint ventures and the option to expand and acquire, Manage. Sci. 37 (1) (1991) 19–33.
[65] I. Álvarez, R. Marin, El impacto de las adquisiciones empresariales en la actividad tecnológica. Una aplicación al sector manufacturero español, Principios
Estudios de Economía Política 5 (May 2006) 57–76.
[66] J.A. Alonso, International process and forms of market penetration: a dynamic proposal, in: J. Molero (Ed.), Technological Innovation, Multinational
Corporations and New International Competitiveness, the Case of Intermediate Countries, Harwood Academic Publishers, Reading, 1995.
[67] P. McCullagh, J.A. Nelder, Generalized Linear Models, Chapman & Hall, London, 1989.
[68] N.J.D. Nagelkerke, A note on the general definition of the coefficient of determination, Biometrika 78 (3) (1991) 691–692.
[69] M. Ozman, Knowledge integration and network formation, Technol. Forecast. Soc. Change 73 (2006) 1121–1143.
[70] R.W. Rycroft, Does cooperation absorb complexity? Innovation networks and the speed of complex technological innovation, Technol. Forecast. Soc. Change 74
(2007) 565–578.
[71] F. Merino, V. Salas, Empresa extranjera y manufactura española: efectos directos e indirectos, Rev. Econ. Apl. 9 (3) (1995) 105–131.
[72] F. Merino, V. Salas, Diferencias de eficiencia entre empresas nacionales y extranjeras en el sector manufacturero español, Pap. Econ. Esp. 66 (1996) 191–208.
[73] I. Alvarez, J. Molero, Technology and the generation of international knowledge spillovers: an application to Spanish manufacturing firms, Res. Policy 34
(2005) 1440–1452.
 I. Álvarez et al. / Technological Forecasting & Social Change 76 (2009) 410–421 421

[74] R. Veugelers, B. Cassiman, R&D cooperation between firms and universities. Some empirical evidence from Belgian manufacturing, Int. J. Ind. Organ. 23 (2005)
355–379.
[75] L. Miotti, F. Sachwald, Co-operative R&D: why and with whom? An integrated framework of analysis, Res. Policy 32 (2003) 1481–1499.
[76] M. Fritsch, R. Lukas, Who cooperates on R&D? Res. Policy 30 (2001) 297–312.
[77] C. Bayona, T. García-Marco, E. Huerta, Firms' motivations for cooperative R&D: an empirical analysis of Spanish firms, Res. Policy 30 (2001) 1289–1307.
[78] J.H. Love, S. Roper, Location and network effects on innovation success: evidence for UK, German and Irish manufacturing plants, Res. Policy 30 (2001)
643–661.
[79] B.S. Tether, Who co-operates for innovation, and why. An empirical analysis, Res. Policy 31 (2002) 947–967.
[80] P. Guerrieri, C. Pietrobelli, Industrial districts' evolution and technological regimes: Italy and Taiwan, Technovation 24 (2004) 899–914.

Isabel Álvarez, Doctor in Economics, is currently Associate Professor at the Department of Economics of Universidad Complutense de Madrid. Her research
interests include transnational corporations, innovation studies and the relationship between technology and development.

Raquel Marin is Lecturer at the Department of Business Economics of Universidad Europea de Madrid. She is at present lecturer at the European University of
Madrid and her research interests are modes of foreign market entry and the internationalisation of technology in developed and developing countries.

Antonio Fonfría, Doctor in Economics from the Universidad Complutense de Madrid, is Associate Professor of Applied Economics at the same university. His main
research interests are the economics of innovation, the evaluation of policies for SMEs and the economics of defence.

All of the above are researchers in the Research Group on Innovation Economy and Policy (GRINEI) at the Universidad Complutense de Madrid.