Henar Alcalde Heras, Dr, Researcher, Orkestra-Basque Insttute of Compettveness and Deusto Business School C/ Mundaiz, 50 20012, Donosta-San Sebastán Spain, tel: +34 943 297 327, fax: +34 943 279 323, henar.alcalde@orkestra. deusto.es.

Abstract

More and more, the ability to innovate can be considered as an explanatory factor in determining the long-term potental of frms to be compettve. Therefore, it is of increasing importance to understand the critcal success factors behind notably radical product innovatons. The present paper explores the yields and results in terms of a series of compettveness indicators that domestc and foreign frms in the Basque Country obtain from technological collaboraton practces. In partcular, the study seeks to assess differences in the way these two groups of frms organize their technological partnerships (in terms of the geographical spread of partners with whom they cooperate and the purposes for which they deploy collaboraton: for commercial or science/knowledge generaton), and the comparatve differences that stem from their respectve practces. The study uses frm level data from the Euskadi Innovaton Survey 2011, for frms located in the Basque Country. The paper fnds that (a) technological collaboratons comprising different types of partners have the greatest positve impact on innovaton novelty, and (b) when looking at the frm’s natonality, collaboraton strategies developed by foreign frms have a higher impact on achieving novel innovaton. We posit that the higher degree of product innovaton we observe among foreign frms – as opposed to domestc frms in the Basque Country – relies on their ability to beneft from both inter-regional partnerships and commercial-based networks for the sake of innovaton purposes.

INTRODUCTION

This research explores the yields and compettveness that domestcowned and foreign-owned frms located in the Basque Country obtain from collaboraton practces. Specifcally, this research tries to determine whether a frm’s natonality (foreign versus local knowledge base) moderates the impact of technological collaboraton agreements on that frm’s ability to bring novel products.

Literature has largely shown that collaboraton is a good method of improving frms’ innovaton capabilites. When relevant resources are not available in the organizaton itself, technological collaboraton offers a good soluton for product innovatons. As long as innovaton complexity increases, frms’ ability to identfy and absorb relevant knowledge will support the development of new products. (Becker and Dietz, 2004; Belderbos, Carree, Diederen, Lokshin and Veugelers, 2004; Nieto and Santamaria, 2007).

The ratonale behind the research follows on from an analysis of innovaton and compettveness indicators with regard to domestc-owned companies from the Basque Country (CAPV). These companies can be considered highly compettve according to their levels of innovatve performance in the natonal market. However, their positon is far below average when comparisons are made with foreign-owned companies located inside the CAPV. For example, in 2010 the percentage of domestc-owned frms that introduced a novel product in the market was 8.8%, while the average was 20.89% for foreign-owned frms located inside the CAPV. However, when we look at frms engaged in technological collaboraton we fnd that 45.71% of domestc-owned companies has developed technological collaboraton; this fgure is very similar to the 42.3% of foreign-owned frms (located inside the CAPV) engaged in technological partnerships.

Therefore, previous fgures suggest that domestc-owned frms in the Basque Country do not perform efcient innovaton practces, and their underperformance can be a consequence of unsuitable collaboraton practces. In order to build a compettve innovaton strategy, this research will explore what would be suitable collaboraton paterns for frms located in the CAPV to achieve novel product innovatons. Specifcally, as explained in Figure 1, we assess whether the nature of the knowledge base (defned by a frm’s natonality) moderates the impact of technological collaboraton agreements on the degree of product newness. We focus on product characteristcs to determine whether an innovaton can be considered radical or incremental. Radical innovatons describe innovatons with a higher degree of novelty, new or signifcantly improved products that involve a novelty for the market in which the frm operates. On the other hand, incremental innovatons describe incremental product innovatons without modifcatons or with slight modifcatons.

The paper proceeds as follows. The next secton discusses different collaboraton paterns, and sets out the hypotheses. Secton 3 discusses the methodology and data; Secton 4 presents the results; and the fnal secton presents the conclusions.

Figure 1. Model specifcaton for hypotheses development

THEORIZING COLLABORATION PATTERNS

Traditonal literature has been largely concerned about external knowledge and innovaton, devotng partcular atenton to performance issues. In today’s knowledge-based economy, the ability to innovate is more important than cost efciency in determining frms’ sustainable performance. The ability of the frm to beneft from innovatons in such environments quickly erodes and frms must obtain relevant knowledge in order to develop efcient innovatons. Consequently, companies must balance their capacity to develop know-how with their ability to introduce novel products.

Innovatons occur as a result of interactons between various actors rather than as a result of a solitary genius (Von Hippel, 1988); therefore, we must explore how collaboraton mediates on it.

Traditonal literature initally examined how markets for technology would enact innovaton performance (Caves, Crookell and Killing, 1983; Hennart, 1988; Williamson, 1989; Mitchell and Singh, 1992; Hagedoorn, 1993). However, more recent studies have been focused on the effect of different innovatve collaboraton strategies in technological evoluton (Rosenkopf and Nerkar, 2001), and the implicatons of an innovaton search strategy on the fnal product introducton (Ahuja, 2000; Chesbrough, 2003; Laursen and Salter, 2006; Alcalde 2014). Another stream of research focuses on the motvaton behind the type of collaboraton and its impact on innovaton performance (Miot and Sachwald, 2003; Becker and Dietz, 2004); Belderbos et al., 2004; Nieto and Santamaría, 2007; Tsai and Wang, 2009).

However, the willingness to engage in collaboratve agreements is conditoned by the internal organizatonal ability to explore the external environment. Thus, the knowledge base of the company defned by a frm’s natonality will be crucial to understanding the frm’s strategic behavior and innovatve performance in internatonal markets. Previous literature is equivocal about the innovatveness efciency of domestc versus foreign companies.

Researchers conclude that the fnal impact of external knowledge acquisiton on performance depends on many elements: theoretcal assumptons; context; specifcs of knowledge and its sources; type of innovaton; and type of performance variable analyzed (Frenz and IetoGillies, 2009).

On one hand, one literature stream argues that foreign frms face home country knowledge restrictons; therefore, these companies suffer a pervasive dependency on home country sources as a main driver of innovaton. Research posits that foreign frms provide important vehicles for transferring research and knowledge within MNE units while external collaboraton can lead to a depreciaton of internal capabilites and coordinaton costs (Weigelt and Sarkar, 2009).

Another literature stream considers external knowledge as a means to foster innovaton by getng access to specialized resources and learning opportunites from the host country. More specifcally, external knowledge flows have a positve impact on frm performance both through their direct effect on innovaton by the subsidiary and, indirectly, through stmulatng knowledge flows between the subsidiaries and other units within the MNE (Yasmin and Oto, 2004).

In the same way, research shows evidence that foreign frms use host knowledge to a greater extent than similar domestc frms (Almeida, 1996) and are most likely to result in innovaton (Almeida and Phene, 2004, 2008).

These debates on the development of efcient collaboratve agreements, a frm’s natonality and innovatve performance supports this paper to explore how foreign-domestc frms differ in product innovaton novelty through technological collaboraton practces in the Basque Country. To accomplish this research queston, we addressed the following issues: the differentaton of technological collaboraton strategies according to both the knowledgebased dimension and geographical locaton, and the empirical assessment of such collaboraton strategies on different degrees of product innovaton novelty (incremental and radical innovaton).

Hypotheses

As pointed out before, companies must create new knowledge and renew their technological competences in order to be innovatve. A frm’s innovaton strategy involves a complex decision: whether to rely on internal R&D sources, external R&D sources, or apply both of them.

On the one hand, reliance on internal R&D has tme and cost advantages, but over-reliance can lead to organizatonal rigidites and competency traps (Leonard-Barton, 1992).

On the other hand, external knowledge sourcing allows frms to explore new windows for opportunity, enlarge the innovaton scope and more flexibility to face market uncertainty. Companies that achieve this flexibility are able to cope beter with the speed, cost and complexity of technological development (Vanhaverbeke, Duysters, and Noordhoven, 2002; Tsai and Wang, 2009) and improve the organizatonal performance (Grant, 1996; Zahra and Nielson, 2002).

In the atempt to build a compettve innovaton patern, frms evaluate different partnership combinatons, according to innovatve purposes and frms’ internal ability to proft from external linkages.

The goal of this study is to understand the ratonale behind a compettve innovaton patern for companies located in the CAPV. In order to fulfll this objectve from a reliable perspectve, the study accounts for different technological collaboraton dimensions: the geographical spread of partners with whom they cooperate (geographical locaton), and the purposes for which they deploy technological collaboraton (knowledge-based dimension).

As long as the geographical locaton of partners shapes the content, the scope of the cooperatve agreement and external knowledge should be analyzed according to the geography of collaboraton in order to understand its effectveness in innovaton output. Specifcally, this study distnguishes between different geographical levels, and regional and inter-regional technological collaboratons. On the other side, external sources involve linkages with different organizatons such as: corporate groups, customers, suppliers, compettors, consultants, technological centers, laboratories, universites and government agencies. Thus, this organizatonal heterogeneity must be analyzed carefully to understand its implicatons in the innovaton output, making it essental to analyzing the purpose for which they deploy technological collaboraton. This study follows (Yamin and Oto 2004) and distnguishes between business and non-business linkages. Specifcally, this study differentates between science-based and commercial-based technological collaboratons. While commercial-based partners are located across the value chain and expected to be more directly related to problem-solving, science-based linkages may be more exploratory in nature. Thus, the impact of science system collaboraton may help frms to redirect their efforts towards innovaton sources. However, commercial-based agents would help frms to exploit their current knowledge paterns and search for new product solutons.

Foreign-domestc frms collaboraton and product innovaton

In order to determine how domestc frms could increase their innovaton compettveness in the market, this study explores if a frm’s natonality moderates its ability to exploit collaboraton agreements and its impact on product innovaton.

As pointed out in the literature, technological collaboraton practces are an important source of compettve advantage. Access to external relevant sources allows organizatons to obtain relevant knowledge which cannot be produced inside the organizaton. Collaboraton networks would support the innovaton actvites of collaboraton partners, increasing the ability to introduce new products (Nieto and Santamaria, 2007; Becker and Dietz, 2004). Firms will consider external partners as sources of valuable knowledge, which cannot be produced internally, maximizing frm value through partners’ resources combinaton and complementarites exploitaton (Kogut, 1988).

However, the success of collaboraton is conditoned by internal organizatonal characteristcs and the willingness to engage in different collaboratve agreements. In this sense, the latent nature of frms’ knowledge base defned by natonality will be crucial to understanding their collaboraton path and product innovaton.

The startng point of this research analyzes if a frm’s natonality moderates the impact of technological collaboraton agreements on product innovaton novelty.

The Basque Country is an example of sustainable regional development (Orkestra, 2008; OECD, 2011) and is the result of an effectve cluster policy. The knowledge base in cluster emerges on the basis of technological complementarites (Porter 2003), claiming that specializaton of related industries is benefcial for regional development (Boschma, Minondo and Navarro, 2010). This knowledge is supported by the dynamic flow of industryrelated informaton, and the involvement of local culture with specifc norms, values and insttutons (Malmberg and Maskell, 2002). This partcular specializaton from the knowledge base affects what is done within and among the frms and therefore what is learnt, how things are done and consequently how learning takes place (Lundvall and Maskell, 2000). According to previous assumptons, cluster industries would support an intensive collaboraton among frms to beneft from specialized local synergies (Maskell, 2001); this restrictve collaboraton will constrain the benefts from a broader collaboraton strategy.

On the other hand, foreign frms own an internatonal knowledge base from internatonal market and heterogeneous partners. The subsidiaries of multnatonal corporatons (MNC) have the potental to develop their knowledge base from two distnct knowledge contexts. As Almeida and Phene (2004) posited: “First, subsidiaries are, a part of an MNC that has the capacity to share knowledge across its various units (Bartlet and Ghoshal, 1989). Second, subsidiaries are located in host country regions that ofen embody social, professional, and technological relatonships among frms permitng inter-frm knowledge flows”. The ability to develop technological agreements through these two different contexts would have a positve effect in innovaton achievement (Andersson and Forsgren, 2000).

From literature sources, it is well known that innovaton opportunites exist because of informaton asymmetry, and that frms that have access to a larger variety of sources of informaton are in a beter positon to identfy and develop innovaton opportunites and introduce products with a higher degree of novelty (Venkataraman, 1997; Amara and Landry, 2005). Due to foreign-owned frms beneftng from subsidized and more diverse networks, they would overcome over-specializaton strategies supported by Basque region clusters. It seems reasonable to hypothesize that technological collaboraton strategies developed by foreign frms have a higher impact on achieving novel innovaton over the collaboraton of domestc frms.

H1: “Among frms competng in the CAPV, the collaboraton of foreign frms has a higher positve impact on novel product innovatons than the collaboraton of domestc frms”.

Foreign-domestc frms’ geographical collaboraton and product innovaton

In this secton, we explore how a frm’s natonality moderates the impact of different types of geographical technological collaboraton on product innovaton novelty. Specifcally we want to answer:

  1. Which type of geographical partnerships have a positve impact on novel product innovaton
  2. In which type of geographical partnerships do foreign-owned frms achieve superior innovaton performance than domestc-owned companies.

In this sense, it would be interestng to analyze if the geographical dimension of the network would dictate frms’ innovaton compettveness. In this sense we distnguish between three different types of collaboratons paterns: regional (within CAPV), inter-regional (outside CAPV) and diverse geographic networks (regional and inter-regional partnerships).

As long as the geographical locaton of the partner just focuses on the geographical locaton (and does not distnguish between the knowledge content of the partnership), regarding the inter-regional we could posit the following arguments according to the spatal collaboraton typology and product innovaton novelty.

First, frms located in the CAPV are involved in a localized cluster, where their knowledge base is based on specializaton from related regional industries. If frms decided to support regional networks involving short distances across collaborators they would beneft from knowledge externalites: bringing people together, favoring informaton contacts and facilitatng the exchange of tacit knowledge and innovaton performance (Jaffe, Trajtenberg, Henderson, 1993; Audretsch and Feldman, 1996; Boschma, 2005). However, over-reliance on closer partners would lead to a spatal lock-in. When regions become too inward looking, the learning ability of local actors may be weakened to such an extent that they lose their innovaton capacity and cannot respond to new environments (Boschma, 2005).

On the other hand, inter-regional networks would imply the access to collaborators outside the CAPV and enrich their knowledge base. This collaboraton would allow frms to avoid a spatal lock-in risk and encourage interactve learning. Nevertheless, an exclusive reliance on inter-regional networks will make frms lose cluster advantages from closer interactons and tacit knowledge. Firms’ distance from current knowledge sources would have a negatve impact on innovaton performance (Jaffe et al. 1993). Therefore, inter-regional collaboraton would have a positve impact on product innovaton if frms had the capabilites to integrate external knowledge and share the same values and expectatons with inter-regional partners.

Finally, frms located in the CAPV which develop both regional and interregional networks will have a greater positve impact on the degree of product innovatons than collaboraton with only one type. An exclusive regional compettve patern leads to an emphasis on the potental advantages of close and familiar cooperatve exchange and encourages an excessive knowledge specializaton. Thus, supportng excessive regional external sources will allow for the emergence of “lock-in” risk that deters frms’ ability to develop product innovaton. However, a combinaton between regional and inter-regional networks will allow frms to overcome experiental learning disadvantages from unfamiliar markets and cultural barriers. Looking at a frm’s natonality we can posit that diversity in geographical collaboraton would beneft all frms located in the CAPV. Foreign frms would reinforce their innovaton positon (through both host country networks and inter-regional opportunites), and it would allow domestc frms to support new diversifcaton paterns, proftng from inter-regional partners.

H2a: “Among frms competng in the CAPV, diversity in geographical collaboraton will have a more signifcant impact on product innovaton novelty than collaboraton with only one type of geographical partner”.

According to the second queston related to foreign frms’ advantage in geographical collaboraton, we posit the following reasoning.

If we look at the different geographical partnership paterns, it seems reasonable that exclusive dependence on regional networks by foreign frms will be harmful for the MNC knowledge base. More specifcally, MNC’s advantages due to experiental learning advantage and knowledge specifcity would be lost due to exclusive reliance on regional partnerships that would involve subsidiary autonomy, which does not match with headquarter dependency. In the same way, we cannot assume the superiority of foreign frms in exploitng diverse geographical collaboraton. Thus, while this diversity would beneft foreign frms in the reinforcement of their innovaton positon (through both host country networks and inter-regional opportunites), domestc frms would overcome lock-in risk, proftng from inter-regional and local partners.

So, the advantage for foreign frms in product innovaton according to geographical collaboraton would emerge from exclusive inter-regional technological collaboratons. Foreign frms have easier access to these partnerships than domestc ones due to the nature of their knowledge base. More specifcally, they have greater abilites to proft from inter-regional collaboratons because they already beneft from current internatonal knowledge sources, and openness culture (Ebersberger and Herstad, 2011). This knowledge orientaton provides internatonal platorms for collaboraton in the form of subsidiaries abroad (Phene and Almeida, 2008) and learning advantages by identfying place-specifc opportunites and partners with whom tght interacton is required (Lowe and Wrigley, 2010). On the other hand, frms in localized clusters stress informal networks and face-to-face contact to facilitate the exchange of specialized knowledge.

It can be concluded that foreign frms would beneft from easier access to internatonal networks and experiental learning advantages to exploit inter-regional knowledge. However, domestc frms rely on relatonships based on closer contact, and fnd it difcult to overcome experiental learning disadvantages in inter-regional contexts.

H2b: “Among frms competng in the CAPV, the inter-regional collaboraton of foreign frms has a higher positve impact on novel product innovatons than the inter-regional collaboraton of domestc frms”.

Foreign-domestc frms’ knowledge-based collaboraton and product innovaton

In this secton, we explore the role of different types of knowledge-based technological collaboraton and the effect on innovaton output. Specifcally, we want to answer:

  1. Which type of knowledge-based partnerships has a positve impact on novel product innovaton
  2. In which type of knowledge-based partnerships do foreign frms achieve superior innovaton performance than domestc companies.

In this hypothesis there is an analysis of whether the knowledge orientaton of the collaboraton differs in product novelty. We distnguish between three different types of collaboratons: science-based, commercial-based and diverse knowledge-based partnerships (scientfc and commercial).

Science-based partners involve research organizatons that encourage collaboraton based on the frst stage of generic knowledge (Belderbos, Carree, Diederen, Lokshin, Veugelers, 2004), providing new scientfc and technological knowledge rather than exploitng innovaton investments in the short-term and fnal product innovaton. (Lundvall,1992; Drejer and Jørgensen, 2005). An exclusive reliance on scientfc collaboraton would be contrary to the assumptons involved in cluster industries, which support specialized industrial knowledge bases (Malmberg and Maskell, 2002). More specifcally, an over-reliance on scientfc collaboratons would prevent these frms beneftng from interactve learning, which requires people sharing the same knowledge base and expertse to learn from each other (Boschma, 2005). In other words, a unique reliance on scientfc partnerships involves too much distance across collaborators’ knowledge bases, leading to difcultes in knowledge absorpton and exploitaton (Cohen and Levinthal, 1990). Therefore, we can conclude that a lack of complementarites in knowledge sources would be detrimental to the development of a novel product innovaton.

Commercial partnerships involve collaboraton among different collaborators in the industrial value chain. These technological collaboratons will look for a fast return on their investment in order to obtain product commercializaton benefts and involve knowledge that ofen is tacit and context-specifc. Thus, an exclusive technological collaboraton with commercial partners would be detrimental to learning and innovaton due to a lock-in risk. Innovaton requires complementary but dissimilar bodies of knowledge; thus homogeneous collaboratons of specifc knowledge would lead to competency traps, and a lack of novel sources (Boschma, 2005). Therefore, frms will suffer from myopia by restrictng their innovaton output to current technological combinatons and deterring further innovaton.

According to the previous argument, the unique reliance on commercial collaboraton would make frms focus on incremental rather than novel innovaton.

Nevertheless, Maskell (2001) posits that commercial collaboratons can fulfll requirements for effectve product innovaton in frms in geographical clusters if it involves dissimilar and complementary knowledge leading to sustainable specializaton, where the knowledge bases of the frms diverge to such an extent that interactve learning is stmulated.

However, a commercial partnership focused on specifc and similar knowledge would be detrimental to innovaton novelty. From this premise we cannot hypothesize either a positve or negatve impact on novel product introducton; the fnal effect would depend on a frm’s ability to encourage collaboratons based on either dissimilar or traditonal similar knowledge.

On the other side, diverse collaboraton involves a great beneft for frms located in the CAPV, supportng new innovaton paterns and encouraging novel product innovatons. Diversity in technological collaboratons would allow frms to beneft from scientfc generic knowledge that is useful for developing new diversifcaton paterns and avoiding lock-in risk. In the same way, commercial collaboraton would allow frms to exploit their current knowledge base with different actors in the commercial value chain and the advantages of a cluster industrial network. Therefore, we can conclude that the variety of partners will have a greater positve impact on the degree of product innovaton novelty than collaboraton with only one type of partner (Nieto and Santamaría, 2007).

H3a: “Among frms competng in the CAPV, diversity in knowledge-based collaboraton will have a more signifcant impact on product innovaton novelty than collaboraton with only one type of knowledge-based partner”.

According to the second queston related to foreign frms’ advantage in these types of collaboraton, we posit the following reasoning.

As long as both domestc and foreign frms located in the CAPV rely on specifc and tacit knowledge, exclusive reliance on scientfc collaboraton would have a negatve impact on novel product innovaton. In the same manner, diversity in knowledge-based collaboraton will beneft both domestc and foreign frms, allowing them to beneft from generic knowledge exploraton (scientfc collaboraton) and specifc knowledge exploitaton (commercial collaboraton).

However, we can expect a different impact of commercial collaboraton on innovaton output according to frms’ natonality. As posited above, foreign frms beneft from performing in two different contexts. Thus, foreign frms can beneft from commercial collaboratons within the corporaton but also from external partners. First, headquarters encourage the transference of knowledge across afliates through an asset exploitng strategy. Second, foreign afliates can support an asset-seeking strategy, augmentng existng assets by absorbing and acquiring technological spillovers from specifc collaborators in the host-country. Therefore, pressure from headquarters to exploit R&D investments and frms’ ability to access complementary and dissimilar knowledge would encourage frms to develop novel product innovatons.

H3b: “Among frms competng in the CAPV, the commercial collaboraton of foreign frms has a higher positve impact on novel product innovatons than commercial collaboraton of domestc frms”.

DATA AND METHODOLOGY

Eurostat innovaton survey

The analysis uses frm level data from the Euskadi Innovaton Survey, collected by Eustat (the Basque Insttute of Statstcs) in 2010 and sampled to be representatve at the regional level (Eurostat, 2006). The data is generated by a self administered survey questonnaire based on the homogenized and thoroughly tested European Community Innovaton Survey (CIS). CIS data is used for generatng ofcial innovaton statstcs on the EU and its member countries and has been used extensively for analysis in economics (Cassiman and Veugelers, 2002; Cefs and Marsili, 2006; Czarnitzki, 2005), in management studies (Laursen and Salter, 2004, 2006; Frenz and Ieto-Gillies, 2009; Schmiedeberg, 2008; Grimpe and Kaiser, 2010), and in economic geography (Simmie, 2003, 2004; Ebersberger and Herstad, 2011).

The total data set contains 4,220 frms, but the empirical analysis is restricted to 1,290 frms that report informaton about technological collaboraton agreements during the three-year reference period 2008-2010. We include innovatng and non-innovatng to avoid biased results (Tether, 2002; Cassiman and Veugelers, 2002).

Dependent variables

  1. Radical Innovatons: describes innovatons with a higher degree of novelty. It takes the value 1 when the frm declares that some percentage of total turnover in 2010 comes from new or signifcantly improved products that were new to the market; otherwise its value is 0.
  2. Incremental Innovatons: describes innovatons with lower degree of novelty. It takes the value 1 when the frm declares that some percentage of total turnover in 2010 comes from new or signifcantly improved products that were only new to the frm; otherwise its value is 0.

Independent variables

  1. Collaboraton (COLLABORATION). Firms indicated whether or not they had engaged in technological collaboraton agreements during the period 2008-2010. COLLABORATION
  2. Natonality. We create a dummy variable (FOREIGN) indicatng if the frm has a foreign company as owner (≥50% of the total stock). It takes the value of 1 if the company is owned by foreign frm; otherwise its value is 0.
  3. Type of technological collaboratons: geographical collaboratons and knowledge-based collaboratons signed during the period 2008-2010. We create six dichotomous variables to measure the effect of different types of technological partnerships. According to the locaton of the collaboraton we distnguish between:
  1. Regional collaboraton exclusively (REG). The frm only collaborates with local partners during the period 2008-2010.
  2. Inter-regional collaboraton exclusively (INTERREG). The frm only collaborates with partners outside the region during the period 2008- 2010.
  3. Diverse-geographical collaboraton (DIVERSEREG). The frm collaborates with both regional and inter-regional partners during the period 2008-2010. According to the knowledge-based character of the collaboraton we distnguish between:
  4. Science collaboraton exclusively (SCIENCE). The frm only collaborates with science-based partners during the period 2008-2010. Science collaboraton includes: universites, technological centers, R&D laboratories, and government organizatons (Yamin and Oto, 2004).
  5. Commercial collaboraton exclusively (COMMERCIAL). The frm only collaborates with commercial-based partners during the period 2008- 2010 (Commercial collaboraton includes: afliates and subsidiaries, customers, suppliers, compettors, and consultancy frms.
  6. Mult-knowledge collaboraton (DIVERSEKNOW). The frm collaborates with both science and commercial partners during the period 2008- 2010.

These mutually exclusive variables avoid potental problems of multcollinearity and capture the impact of each partner more clearly by separatng it from the effects atributable to other partner types in heterogeneous networks (Nieto and Santamaría, 2007).

Control variables

We realize that the levels of innovaton in frms will depend on internal and external environmental factors regardless of collaboratve agreements; therefore, we include different control variables for frm specifc and industry characteristcs.

According to frm specifc characteristcs we control for the size of the company, export intensity, group afliaton, R&D internal intensity, knowledge protecton. Firstly, several studies fnd that the differences between domestc and foreign frms are mainly due to the different frms’ group size (Falk, 2008); therefore, we control for the logarithm of company turnover (NETSALES). EXPSHARE measures the internatonal orientaton of the frm by its export share. Company group, irrespectve of the locaton of company subsidiaries, is positvely associated with innovaton performance, we use a binary variable (GROUP) coded 1 if an enterprise is part of a company, and coded 0 if it is a single-unit frm. We include an indicator for internal R&D intensity, measured as the rato of internal R&D expenses to the frm’s total turnover (R&DEXP.). An indicator for formal IPR protecton is introduced because protecton enables the frm to protect proprietary knowledge during collaboratve work and R&D contractng (PROPAT) (Ebersberger and Herstad, 2011).

We control for industry characteristcs by the introducton of 43 industry dummy variables, albeit their coefcients are omited from our tables (INDUSTRYDUMMIES).

The model

As both dependent variables (Radical and Incremental) are dichotomous, estmaton models such as logit or probit (Greene, 2000) would normally be appropriate. However, as the error terms of the two models are likely to be correlated, an extension of probit known as bivariate probit (Greene, 2000) is usually a more appropriate estmator. The bivariate probit model has the following specifcaton:

Zi1 = β'1 Xi1 + εi1 ; if γi1 = 1 if Zi1 > 0, γi1 = 0 if Zi1 ≤ 0,
Zi2 = β'2 Xi2 + εi2 ; if γi1 = 1 if Zi1 > 0, γi2 = 0 if Zi2 ≤ 0,
(ε, ε) ≈ N (0, 0, 1, 1, ρ)    

The bivariate probit model was estmated using the Stata 11 routne, based on the method of simulated maximum likelihood. The difference between the specifcatons of each model lies in the explanatory variables (collaboraton, and type of network).

RESULTS

Table 1 below gives descriptve statstcs, collinearity statstcs, and correlatons of all variables.

Table 1. Descriptve statstcs
VariableMeanStd. dev.RADICALINCREMENTALSCIENCECOMERCIALDIVERSEKNOREGINTERREGDIVERSEREGFOREIGNGROUPR&DEXPPROPAT
RADICAL 0.095 0.294 1                      
INCREMENTAL 0.962 0.192 -0.303 1                    
SCIENCE 0.050 0.219 0.074 -0.034 1                  
COMMERCIAL 0.052 0.222 0.038 -0.031 -0.054 1                
DIVERSEKNOW 0.203 0.402 0.264 -0.072 0.456 -0.118 1              
REG 0.103 0.304 0.011 -0.029 0.202 0.277 0.190 1            
INTER-REG 0.023 0.151 0.126 -0.012 -0.012 0.358 0.037 -0.052 1          
DIVERSEREG 0.180 0.384 0.214 -0.064 0.261 0.063 0.747 -0.159 -0.072 1        
FOREIGN 0.055 0.228 0.093 -0.002 0.089 0.045 0.072 -0.014 0.259 -0.003 1      
GROUP 0.712 0.453 -0.142 0.056 -0.058 -0.145 -0.198 -0.083 -0.132 -0.189 -0.347 1    
R&DEXP 0.183 3.795 0.010 -0.029 0.010 -0.016 0.029 -0.007 -0.012 0.039 -0.011 0.020 1  
PROPAT 0.062 0.241 0.406 -0.155 0.074 -0.045 0.211 -0.037 0.017 0.226 0.028 -0.084 0.026 1
EXPSHARE 0.064 0.180 0.162 -0.091 0.112 0.042 0.072 0.063 0.030 0.062 0.166 -0.190 -0.008 0.213
NETSALES 14.110 2.242 0.197 -0.069 0.081 0.135 0.246 0.070 0.134 0.258 0.218 -0.432 -0.057 0.162

Table2 presents the regression results testng our frst hypotheses. The ρ parameter is highly signifcant, this supports the choice of biprobit instead of the probit model. The Wald test also points out high signifcance of both variables for the models.

Table 2. Collaboraton impact on product innovaton
 Model 1aModel 1b
RADICALINCREMENTALRADICALINCREMENTAL
COLLABORATION 0.613*** -0.225* 0.534*** -0.247*
FOREIGN 0.516** 0.008 0.186 -0.134
COLLABORATION*FOREIGN     0.802** 0.321
GROUP -0.147 -0.019 -0.164 -0.024
R&DEXP -0.056 0.009 -0.056 0.009
PROPAT 0.755*** -0.194 0.762*** -0.199
EXPSHARE 0.379* -0.613* 0.418* -0.599*
NETSALES 0.013 0.074* 0.013 0.073**
INDUSTRY DUMMIES Signifcant
LR rho=0 39.974 40.448
Wald test of full Model 303.98 305.35
Log pseudolikelihood -913.985 -910.21
Number of observatons 1290 1290
One-tailed t-test applied. ***p > .01 **p > .05 *p > .10

In Model 1a we test the impact of collaboraton on the development of novel products. As expected, the collaboraton effect is positve and signifcant in frms’ ability to develop radical product innovatons; however, collaboraton has a negatve and signifcant impact on incremental innovatons. The effect of a foreign natonality is highly signifcant in radical innovatons; however, this foreign natonality takes a negatve but insignifcant effect on the development of incremental innovatons.

In Model 1b we test Hypotheses 1 to analyze if the effect of collaboraton on product innovaton would be higher in foreign rather than domestc companies. We fnd support for our Hypothesis 1 provided that the coefcient of the interacton term between the variables COLLABORATION*FOREIGN is positve and statstcally signifcant in radical innovatons. The impact of collaboraton remains positve and signifcant, but the effect of the foreign natonality of the company becomes insignifcant in the development of novel products. This makes us believe that foreign frms’ superiority in radical innovaton relies mainly on its superior ability to beneft from collaboraton, rather than its internatonal nature. According to incremental innovatons, collaboraton is negatve and signifcant, while foreign natonality does not have a signifcant effect. Consequently, the collaboraton of foreign frms does not have a higher positve impact on incremental product innovatons than the collaboraton of domestc frms; therefore, COLLABORATION*FOREIGN is not signifcant in achieving less novel innovatons. The effect of the control variable PROPAT on the likelihood of achieving innovatons is positve and signifcant in the case of radical innovatons. The variable for EXPSHARE has a positve and signifcant effect on the development of radical innovatons, while it takes a negatve and signifcant value in incremental innovatons. The size of the frm controlled by NETSALES is a positve and signifcant factor for the achievement of less novel innovatons.

Table 3 presents the estmated results for the impact of different types of geographical collaboratons on the degree of novel product innovaton; and the signifcance of natonality in determining geographical collaboraton impact on product innovaton. The ρ parameter is highly signifcant, this supports the choice of biprobit instead of the probit model. The Wald test also points out the high signifcance of both variables for the models.

Model 2a estmates the impact of different types of partnerships according to geographical locaton in frms located in the CAPV. Results suggest that the effect of collaboraton depends on the type of partner.

Table 3. Geographical collaboraton impact on product innovaton
 Model 2aModel 2b
RADICALINCREMENTALRADICALINCREMENTAL
REG 0.369** -0.280 0.324** -0.226
INTER-REG 0.968*** -0.320 0.507 -0.578
DIVERSEREG 0.709*** -0.181 0.672*** -0.227
FOREIGN 0.174** 0.009 0.181 -0.152
REG*FOREIGN     0.455 -0.569
INTER-REG*FOREIGN     1.401** 1.029
DIVERSEREG*FOREIGN     0.548 5.385
GROUP -0.150 -0.024 -0.165 -0.023
R&DEXP -0.054 0.009 -0.055 0.009
PROPAT 0.730*** -0.205 0.732*** -0.231
EXPSHARE 0.422* 0.620** 0.475** -0.508**
NETSALES 0.005 0.072** 0.006 0.074**
INDUSTRY DUMMIES Signifcant
LR rho=0 39.810 41.561
Wald test of full Model 309.690 308.600
Log pseudolikelihood -910.318 -902.871
Number of observatons 1290 1290
One-tailed t-test applied. ***p > .01 **p > .05 *p > .10

While exclusive regional collaboraton (REG) has a positve and signifcant effect on the development of radical product innovaton, exclusive interregional collaboraton (INTERREG) does not have a signifcant impact on the achievement of novel products. As expected, diversity in geographical collaboraton (DIVERSEREG) has the largest positve effect and signifcance in the achievement of radical innovatons. Therefore, we can confrm our Hypothesis 2a related to the highest signifcant impact of heterogeneous geographical networks in the development of new products.

Model 2b estmates the impact of different geographical technological networks in product innovaton according to foreign natonality. In order to analyze this issue, we create three interactons capturing the effect of different geographical networks in foreign frms: REG*FOREIGN, INTERREG*FOREIGN, DIVERSEREG*FOREIGN. The introducton of these interactons allows us to isolate the effect of collaboraton performed by foreign frms in product innovaton. Through the applicaton of these interactons we can confrm our Hypotheses 2b. Foreign frms’ advantage on geographical collaboraton relies on inter-regional collaboraton, INTERREG*FOREIGN is positve and signifcant in the achievement of novel products. However, foreign frms do not take a higher advantage from exclusive regional or diverse geographical networks than domestc frms, as long as REG*FOREIGN, DIVERSEREG*FOREIGN does not have a signifcant effect.

According to the development of incremental innovatons, neither the type of geographical collaboraton nor foreign-collaboraton interactons have a signifcant impact.

The effect of control variables is similar to the previous model.

Table 4. Knowledge-based collaboraton impact on product innovaton
 Model 3aModel 3b
RADICALINCREMENTALRADICALINCREMENTAL
SCIENCE -0.408** 0.106 -0.523** 0.046
COMMERCIAL 0.389** 0.441** 0.220 -0.478**
DIVERSEKNOW 0.949*** -0.136 0.923*** -0.189
FOREIGN 0.526** 0.006 0.261 -0.245
SCIENCE*FOREIGN     0.650 4.575
COMMERCIAL*FOREIGN     1.540** 0.384
DIVERSEKNOW*FOREIGN     0.252 0.785
GROUP -0.131 0.020 -0.143 0.032
R&DEXP -0.044 0.009 0.046 0.046
PROPAT 0.696*** -0.230 0.695*** -0.250*
EXPSHARE 0.472** -0.648** 0.524** -0.605**
NETSALES 0.004 0.076** 0.002 0.074**
INDUSTRY DUMMIES Signifcant
LR rho=0 42.350 44.354
Wald test of full Model 321.850 324.440
Log pseudolikelihood -901.416 -894.226
Number of observatons 1290 1290
One-tailed t-test applied. ***p > .01 **p > .05 *p > .10

Table 4 presents the estmated results for the impact of different types of knowledge-based networks on the degree of novel product innovaton, and the signifcance of natonality in determining the knowledge-based collaboraton impact on product innovaton. The ρ parameter is highly signifcant, supportng the choice of biprobit instead of the probit model. The Wald test also points out a high signifcance of both variables for the models.

Model 3a assesses the impact of different types of knowledge-based technological partnerships in frms located in the CPAV. Results suggest that the effect of collaboraton on novel innovaton depends on the type of partner.

As posited in the theory, exclusive science collaboraton (SCIENCE) has a negatve and signifcant effect on the development of radical product innovaton; while commercial collaboraton (COMMERCIAL) has a positve but insignifcant impact on the achievement of novel products. The largest positve effect of knowledge-based networks is found in diverse partnerships (DIVERSEKNOW), which are positve and signifcant in the achievement of radical innovatons. Therefore, we can confrm our Hypothesis 3a related to the highest signifcant impact of diversity on knowledge-based networks in the development of new products.

Model 3b estmates the impact of different knowledge-based networks on product innovaton according to foreign natonality. In order to analyze this issue, we create three interactons capturing the effect of different knowledge-based networks in foreign-owned frms: SCIENCE*FOREIGN, COMMERCIAL*FOREIGN, DIVERSEKNOW*FOREIGN. The introducton of these interactons allows us to isolate the effect of collaboraton performed by foreign frms on product innovaton. Through the applicaton of these interactons we can confrm our Hypothesis 3b. Foreign frms obtain greater performance than domestc frms on commercial collaboraton, COMMERCIAL*FOREIGN is both positve and signifcant in the achievement of novel products. This result confrms previous literature about frms’ commercial collaboraton in spatal clusters. Malmberg (2003) pointed out the disappointng results of commercial collaboratons within clusters, and the increased atenton to more informal unintended interactons.

According to the development of incremental innovatons, only exclusive commercial collaboraton is signifcant, which is a negatve sign. Exclusive science collaboraton has a positve but insignifcant effect, and diverse knowledge-based networks have a negatve but signifcant effect. Looking at foreign frms’ advantage from knowledge-based networks, any foreigncollaboraton interacton has a signifcant impact on the achievement of incremental innovatons. Once again, the effect of control variables is similar to the previous model.

DISCUSSION AND CONCLUSIONS

Economic development is connected with the generaton of innovaton. Collaboraton has been largely considered as an important enactor of frms’ innovaton ability. However, relevant literature is equivocal about the impact of different type of networks on innovaton output.

In this sense, frms’ partcular context will dictate the best collaboraton patern to achieve a higher rate of innovaton novelty. This research focuses on the development of an effectve collaboraton patern for frms located in the Basque Country. Traditonally the Basque Country has been considered an example of a geographical cluster. The aim of this paper is to understand why domestc frms achieve lower innovaton performance than foreign frms located in the CAPV. Specifcally, we assess whether the nature of the knowledge base (defned by a frm’s natonality) moderates the impact of technological collaboraton agreements on the degree of product newness. Therefore, we assess differences in the way these two groups of frms organize their technological collaboraton practces and the comparatve differences that stem from their respectve practces.

This study confrms that collaboraton strategies developed by foreign frms have a higher impact on achieving novel innovaton (Frenz and IetoGillies, 2004). In order to explore the source of this foreign advantage, we analyze how frms beneft from different networks (according to geographical and knowledge-based dimensions). This distncton between technological networks leads us to explore traditonal cluster advantages focused on geographical proximity and specialized industry collaboratons.

Firstly, when we focus on technological geographical collaboraton we fnd that diversity in the make-up of geographical networks favors innovaton novelty more than collaboraton with a single type of geographical partner does. This is in line with recent cluster literature focused on spatal collaboraton diversity (Boschma, 2005). By considering the individual impact of each type of geographical partner, results confrm that regional partners are the single partners who have the biggest impact on the achievement of novel product innovaton. This is consistent with the argument that short distances across collaborators beneft from knowledge externalites (Jaffe et al. 1993; Audretsch and Feldman, 1996; Boschma, 2005).

Looking at foreign frms’ superiority in innovaton performance, we fnd that inter-regional collaboratons seem to be the key source of advantage. Being part of a MNC allows afliates to be involved in a multnatonal network and develop an open-vision of spatal collaboraton. Thus, foreign frms beneft from inter-regional collaboratons due to their internatonal learning ability (Phene and Almeida, 2008; Lowe and Wrigley, 2010; Ebersberger and Herstad, 2011).

Second, knowledge-based technological collaboratons may be a makeor-break decision for the success of novel innovatons across frms located in the CAPV. Our results show that diversity in knowledge networks has the greatest positve effect on radical innovatons. Being integrated in a diverse knowledge-based network allows frms to beneft from generic knowledge from scientfc partners in order to explore new knowledge, but also from commercial collaborators exploitng current specialized knowledge.

Looking at foreign frms’ superiority in innovaton performance, we can confrm commercial collaborators as a source of innovaton advantage. Foreign frms have a greater chance of combining dissimilar and complementary knowledge in commercial-based collaboratons. Therefore, these commercial partnerships developed by foreign frms enable them to proft from a variety of specifc knowledge (across MNC and the host country) with a positve impact on novel innovatons creaton.

We contribute to literature exploring new global dynamics of collaboraton in frms located in geographical clusters. In this sense we challenge traditonal cluster assumptons by estmatng the impact of difference on innovaton novelty. Traditonal cluster literature has stressed the rigid “local” focus; however, we take a contemporary vision by integratng ownership natonality in geographical cluster research. The introducton of foreign companies in regional systems research has been neglected due to the partcular atenton paid to locally owned medium-sized frms, while MNC are seen as alien in the idea of a dynamic spatal cluster (Malmberg, 2003). We provide a global patern of effectve collaboratons and determine the foreign frm’s collaboraton advantage. In this sense, frms located in the Basque Country should encourage diversity in geographical and knowledge-based networks.

This study is a novel empirical study in the cluster theory, traditonally focused on the development of case-studies. Malmberg (2003) already suggested the importance of assessing the implicatons of global and local circuits in an empirical way.

Our fndings provide useful managerial implicatons. Managers should be aware of the importance of parent choice and the diversity of collaboraton strategy in order to achieve sustainable innovaton. Furthermore, managers should look at gaining advantage from inter-regional and commercial partnerships in order to gain compettveness from foreign frms. As pointed out above, radical innovaton is essental for organizatonal compettveness.

From the point of view of policy-makers, its mission is essental in promotng efcient practces. As long as the Basque Country is the result of an efcient regional policy, regional policy-makers should encourage a sustainable innovaton strategy. In this sense, they should encourage science organizatons to develop generic knowledge closer to domestc frms’ background to motvate innovaton development. In the same way, regional government should explore the ratonale behind informal commercial transactons in order to promote a commercial collaboraton efcient patern. According to geographical collaboraton, inter-regional partnerships should be supported, allowing frms to develop an open geographical perspectve.

Finally, this work is not free from limitatons. It would be interestng to enlarge empirical analysis through a larger sample from different periods. Due to data availability we have focused the analysis on product innovaton, but it would be interestng to replicate this study regarding other types of innovaton (process, organizatonal, commercial) and extrapolate global conclusions. The results warrant further study of geographical and knowledge-based networks at different levels and analyze its implicatons in innovaton performance. Related to the empirical validaton of the model, our fndings could be supported by the development of an in-depth case study. Finally, although radical innovaton is the most powerful source of innovaton performance, we cannot underestmate the role of incremental innovaton, as long as this is the dominant form of innovaton. Moreover, the diffusion of radical innovatons nearly always depends on incremental improvements, refnements and modifcatons, the development of complementary technologies, and organizatonal change and social learning. In the same way, the contributons of incremental innovatons to address socioeconomic challenges are substantal and may be even more important in a development context. Therefore, it would be interestng to go further on the implicatons of different collaboraton agreements on different degrees of incremental innovaton.

References

  • Ahuja, G. (2000). Collaboraton networks, structural holes and innovaton: a longitudinal study. Administratve Science Quarterly, 45, 425-455.
  • Alcalde, M. D. H. (2014). Building product diversifcaton through contractual R&D agreements. R&D Management, 44(4), 384-397.
  • Almeida, P., Phene, A. (2004). Subsidiaries and knowledge creaton: the influence of the MNC and host country on innovaton. Strategic Management Journal, 25, 847-864.
  • Almeida, P. (1996). Knowledge sourcing by foreign multnatonals: Patent citaton analysis in the US semiconductor industry. Strategic Management Journal, 17 (Winter Special Issue): 155-165.
  • Amara, N., Landry, R. (2005). Sources of informaton as determinants of novelty of innovaton in manufacturing frms: evidence from the 1999 statstcs Canada innovaton survey. Technovaton, 25, 245-259.
  • Andersson U, Forsgren M. (2000). In search of centers of excellence: network embeddedness and subsidiary roles in multnatonal corporatons. Management Internatonal Review, 40(4), 329-338.
  • Audretsch, D. B., Feldman, M. P. (1996). Innovatve clusters and the industry life cycle. Review of Industrial Organizaton, 11(2), 253-273.
  • Bartlet C, Ghoshal S. (1989). Managing across borders: the transnatonal soluton. Harvard Business School Press: Boston, MA.
  • Becker, W., Dietz, J. (2004). R&D co-operaton and innovaton actvites of frms-evidence for the German manufacturing industry. Research Policy, 33, 209-223.
  • Belderbos, R., Carree, M., Diederen, B., Lokshin, B., Veugelers, R. (2004). Heterogeneity in R&D co-operaton strategies. Internatonal Journal of Industrial Organizaton, 22, 1237-1263.
  • Boschma, R.A. (2005). Proximity and innovaton. A critcal assessment. Regional Studies, 39(1), 61-74.
  • Boschma, R.A., Minondo, A., Navarro, M. (2010). Related variety and regional growth in Spain. Papers in Evolutonary Economic Geography 10.12, Utrecht University.
  • Cassiman, B., Veugelers, R. (2002). R&D co-operaton and spillovers: some empirical evidence from Belgium. The American Economic Review, 92(4), 1169-1185.
  • Caves, R. E., Crookell, H., Killing, J. P. (1983). The imperfect market for technology licenses. Oxford Bulletn of Economics and Statstcs, 45, 249–267
  • Cefs, E., Marsili, O. (2006). Survivor: the role of innovaton in frm’s survival. Research Policy, 35, 626-41.
  • Chesbrough H. (2003). Open Innovaton. Harvard University Press: Cambridge, MA.
  • Cohen, W.M., Levinthal, D.A. (1990). Absorptve capacity: a new perspectve of learning and innovaton. Administratve Science Quarterly, 35(1), 128–152.
  • Czarnitzki, D. (2005). The extent and evoluton of productvity defciency in Eastern Germany. Journal of Productvity Analysis, 24, 211-231.
  • Drejer, I., Jørgensen, B.H. (2005). The dynamic creaton of knowledge: analysing public–private collaboratons. Technovaton, 25, 83-94.
  • Ebersberger, B., Herstad, S. (2011). Go abroad or have strangers visit? On organizatonal search spaces and local linkages. Journal of Economic Geography, 1-23.
  • Falk, M. (2008). Effects of foreign ownership on innovaton actvites: empirical evidence for twelve European countries. Natonal Insttute Economic Review, 204(1), 85-97.
  • Frenz, M., Ieto-Gillies, G.(2009). The impact on innovaton performance of different sources of knowledge: Evidence from the UK Community Innovaton Survey. Research Policy, 38, 1125-1135.
  • Grant, R.M. (1996). Towards a knowledge-based theory of the frm. Strategic Management Journal, 17 (Winter Special Issue), 109-122.
  • Greene WH. (2000). Econometric Analysis (4th edn).Prentce-Hall: Upper Saddle River, NJ.
  • Grimpe, C., Kaiser, U. (2010). Balancing internal and external knowledge acquisiton: The gains and pains from R&D outsourcing. Journal of Management Studies, 47, 1483-1509.
  • Hagedoorn, J. (1993). Understanding the ratonale of strategic technology partnering: interorganisatonal modes of co-operaton and sectoral differences. Strategic Management Journal, 14, 371-395.
  • Hennart, J.F. (1988). a transacton cost theory of equity joint ventures. Strategic Management Journal, 9(4), 361-374.
  • Jaffe, A.B.,Trajtenberg M., Henderson R. (1993). Geographic localizaton of knowledge spillovers as evidenced by patent citatons. Quarterly Journal of Economics, 108, 577-598.
  • Kogut, B. (1988). Joint Ventures: Theoretcal and empirical perspectves. Strategic Management Journal, 9, 312-332.
  • Laursen, K., Salter, A. (2006). Open for innovaton: the role of openness in explaining innovaton performance among UK manufacturing frms. Strategic Management Journal, 27(2), 131-150.
  • Leonard-Barton, D. (1992). Core capabilites and core rigidites: a paradox in managing new product development. Strategic Management Journal (Summer Special Issue), 13, 111-125.
  • Lowe, M., Wrigley, N. (2010). The ‘‘Contnuously Morphing’’ retail TNC during market entry: interpretng Tesco’s expansion into the United States. Economic Geography, 86, 381-408.
  • Lundvall, B.D., Maskell, P. (2000). Naton states and economic development from natonal systems of producton to natonal systems of knowledge creaton and learning. The Oxford Handbook of Economic Geography Eds G.L., Clark, M. P., Feldmann, M.S., Gertler. Oxford: Oxford University Press, 353-372.
  • Malmberg, A. (2003). Beyond the cluster: local milieus and global connectons. In: Peck, J. and Yeung, H. W.-c. (Eds.): Remaking the global economy. London: Sage.
  • Malmberg, A., Maskell, P. (2002). The elusive concept of localizaton economies: towards a knowledge-based theory of spatal clustering. Environment and Planning, 34, 429-449.
  • Maskell, P. (2001). The frm in economic geography. Economic Geography, 77(4), 329-344.
  • Miotti, L., Sachwald, F. (2003). Co-operatve R&D: Why and with whom? An integrated framework of analysis. Research Policy, 32, 1481-1499.
  • Mitchell W, Singh K. (1992). Incumbents’ use of pre-entry alliances before expansion into new technical subfelds of an industry. Journal of Economic Behavior and Organizaton, 18(3), 347-372.
  • Nieto, M.J., Santamaría, L. (2007). The importance of diverse collaboratve networks for the novelty of product innovaton. Technovaton, 27, 367- 377.
  • OECD (2011). OECD Reviews of Regional Innovaton: Basque Country, Spain. Paris: OECD.
  • Orkestra (2008). 2008. Informe de compettvidad del País Vasco. Hacia una propuesta única de valor. Bilbao: Publicaciones de la Universidad de Deusto.
  • Phene, A., Almeida, P. (2008). Innovaton in multnatonal subsidiaries: The role of knowledge assimilaton and subsidiary capabilites. Journal of Internatonal Business Studies, 39(5),901-919.
  • Porter, M. E. (2003). The economic performance of regions. Regional Studies, 37(6/7), 549-578.
  • Rosenkopf L, Nerkar A. (2001). Beyond local research: boundary-spanning, exploraton, and impact on the optcal disk industry. Strategic Management Journal, 22(4): 287-306.
  • Schmiedeberg, C. (2008). Complementarites of innovaton actvites: An empirical analysis of the German manufacturing sector. Research Policy, 37: 1492-1503.
  • Simmie, J. (2003). Innovaton and urban regions as natonal and internatonal nodes for the transfer and sharing of knowledge. Regional Studies, 37, 607-620.
  • Simmie, J. (2004). Innovaton and clustering in the globalised internatonal economy. Urban Studies, 41, 1095-1112.
  • Tether, B. (2002). Who co-operates for innovaton, and why. An empirical analysis. Research Policy, 31, 947-967.
  • Tsai, K.H., Wang, J.C. (2009). External technology sourcing and innovaton performance in LMT sectors: An analysis based on the Taiwanese Technological Innovaton Survey. Research Policy, 38, 518-526.
  • Vanhaverbeke W, Duysters G, Noorderhaven N. (2002). External technology sourcing through alliances or acquisitons: an analysis of the applicatonspecifc integrated circuits industry. Organizaton Science, 13(6), 714- 733
  • Venkataraman, S. (1997). The distnctve domain of entrepreneurship research. In: Katz, J., Brokhaus, R. (Eds.), Advances in Entrepreneurship. Firm Emergence, and Growth, JAI Press, San Francisco, CA.
  • Von Hippel, E. (1988). Sources of Innovaton. Oxford University Press, Oxford.
  • Weigelt, C., Sarkar, M. (2009). Learning from supply-side agents: The impact of technology soluton providers’ experiental diversity on clients’ innovaton adopton. Academy of Management Journal, 52(1), 37-60.
  • Williamson, O.E. (1989). Transacton cost economics. In: Schmalensee, R.Willig, R.D. (Eds.). Handbook of Industrial Organizaton. North- Holland, Amsterdam, 135-182.
  • Yamin, M., Oto J. (2004). Paterns of knowledge flows and MNE innovatve performance. Journal of Internatonal Management, 10, 239-258.
  • Zahra, S. A., Nielsen, A. P. (2002). Sources of capabilites, integraton and technology commercializaton. Strategic Management Journal, 23, 377- 398.

Biographical note

Henar Alcalde Heras holds Doctorate in Business Administraton and Quanttatve Methods (European Menton) from the University of Carlos III, Madrid. Part of doctoral studies she conducted at the Copenhagen Business School in the Department of Innovaton and Organizatonal Economics (INO). Since 2011 she acts as researcher at Orkestra-Basque Insttute of Compettveness, and teaches at Deusto University Strategy Deparment. Previously she has been teaching in the Business Economics Department at the University of Carlos III, Madrid. Her principal areas of interest are subjects relatng to Open Innovaton Strategy. In partcular, the analysis of collaboraton between different agents, knowledge management, technology transfer and the final impact on the results for innovaton achieved by the company.