2.1 Concept of Proximity and Proximity Paradox

Developing successful technological innovation requires more in-depth understanding of what facilitates the interaction of actors with diverse and complementary pieces of knowledge, skills, and experiences. The term proximity refers here to the degree to which “one actor relates to the another”. The following section discusses the above mentioned subject related literature in more details with special emphasis on the role of proximity in networks formation. The analysis starts from the first works on proximity dynamics originating from the French school and the works of Rallet (1993) and Rallet and Torre (1999). In their research the authors conclude that other forms of proximity besides geographical may facilitate interactive learning and the formation of innovation networks. Similarly, Dutch regional economist Ron Boschma has claimed that geographical proximity is neither necessary nor sufficient for inter-firm learning and innovation (2005). In the network based system, innovation is the outcome of an interactive and transdisciplinary process involving at least three types of stakeholders – academia, industry and government – as active players involved in co-creation, experimentation and valorization of the new knowledge (converting knowledge into commercial, feasible products/processes, services and/or societal value) (Leloux et al., 2009; Drooge et al., 2011; Arits and Duijvesteijn, 2012). The three helices commit themselves to cooperation based on their own overlapping interest or “boundary objects” through common trust, needs and stakes. Consequently, their common boundaries have greater chance to stimulate innovation within the particular industries. In general, the notion “proximity” is related to the interaction of actors with common boundaries within the networks. In this sense, geographical proximity or the physical closeness of two or more actors, provides opportunities for frequent, interpersonal contacts, allowing the reduction of information asymmetries, whereas organizational and cultural proximities influence social interactions and social capital formation (Fitjar et al., 2016; Lorenzen, 2007). While a high degree of proximity might be considered a prerequisite to make agents connected, some authors claim that proximity between agents does not necessarily increase their innovative performance and may even harm it. Boschma (2005), Broekel and Meder (2008) and Boschma and Frenken, (2009) refers to this as the “proximity paradox”. They claim that higher level of innovative performance depends on the existence of the “optimal” level of proximity between agents. Another topic dealing with the formation and performance of innovation networks is the path dependency in the network evolution. Ter Wal (2009) claims that the networks of innovation become less geographically proximate over time during the course of an industry lifecycle. Whereas several other authors, including Boschma and Frenken (2009) and Gluckler (2007) argue that the different proximities induce path dependency in network evolution, and may cause retention or lock-in in the local network. Therefore, Boschma and Frenken (2009) and Ter Wal and Boschma (2008) emphasize the importance of applying the dynamic network approach in the studies of networks formation, accounting for the evolution of network structures and its relationship to the different types of proximity. The section below discusses the importance of various types of proximities for the interactions and exchange of knowledge within innovation networks.

2.2 Types of Proximity and Innovation Networks Formation

Proximity does not create innovation, but serves as an enabling factor for it to happen. As stated by Boschma (2005), there are several types of proximities facilitating interactions as well as an exchange of knowledge and information between innovation ecosystem stakeholders, such as a geographical, cognitive, organizational, institutional, social and cultural one.

Geographical proximity appears as a distinctive element that leads to a clustering effect, especially useful for the transfer of tacit knowledge (Audretsch and Feldman 2004; Koschatzky et al. 2009). It is defined as the physical distance between actors. Porter (1998) showed that industry clusters provided some benefits in terms of entrepreneurship and innovation potential. The increasing role of geographical proximity in shaping economic and social interactions has triggered the “new” economic geography literature, and in particularly, the concept of clusters in facilitating knowledge exchange and innovation activities. Koschatzky et al. (2009) pointed out that in case of advanced technology sectors both territorial and sectoral determinants influence the development of innovative activities.

Cognitive proximity and technological relatedness or “technological proximity”, are less investigated by the researchers. Some notable contribution comes from the works of Petruzzelli (2011), Nooteboom (2000), Nahapiet and Ghoshal (1998), Brockhoff and Teichert (1995). In their findings, cognitive proximity is manifested by the homogeneity of competencies, capabilities and skills as well as the homogeneity of knowledge bases (Nooteboom 2000: 3–11). The first level of homogeneity refers to cognitive similarity between individuals: communication codes, written specific technical language, common professional or scientific backgrounds. Whereas the second level of homogeneity refers to the cognitive similarity between independent organizations (in their knowledge bases, capabilities, competences, and experiences). Having an overlapping knowledge base and a shared technical vocabulary enhances the actors’ ability to communicate and exchange information (Nahapiet and Ghoshal 1998). In relation to partners’ technological relatedness, Petruzzelli (2011) suggests that in order to increase innovative performance, a certain threshold of similar technological competencies between partners is required. However, too much similarity may in turn have a detrimental effect on the actors’ innovative performance since the development of valuable innovations may require dissimilar but also complementary sources of knowledge.

Institutional proximity refers to the interaction among actors from various institutions within innovation networks. Much of the literature focuses on the three institutional, Triple Helix (TH) spheres – university, industry and government. However, as Jensen and Tragardh (2004) put it, cooperation within the TH model is complex, dynamic and ambiguous, thus the institutional architecture of particular university-industry-government innovation network models may differ by sector, for example in the case of aerospace the government would occupy a larger role than in the life sciences. Moreover, as evidenced in the research studies by Su and Wu (2015) and Silveira Luz and Monteiro Salles-Filho (2011), institutions can be greater enablers of innovation networks in the case of science-based industries, such as biotechnology, then in the case of the science-applied industries, like aviation.

Organizational proximity is related to the organizational dimension of the proximity. It identifies two levels of analysis: the inter-organizational and the intra-organizational (Antonelli, 2000). The latter division results from the fact that knowledge spills over from one to another organization, but also among different units within the same organization. People are simultaneously proximate to everyone else in their organization, as they move about the organization. The latter facilitates interaction both intentional and accidental. The inter-organizational proximity can be further distinguished from the low (loosely coupled) networks and weak ties between autonomous organization, to the highly networked, such as ownership and wholly-owned subsidiaries. In terms of the intra-organizational level, strong ties among different units define high organizational proximity, whereas weak ties correspond to a low proximity.

Social proximity and its concept originates from the literature on embeddedness. It claims that economic behaviour is heavily embedded in social relations, in which behavioural factors such as trust, openness, professionalism, complementarity and transparency are of key importance. The trustful relations among actors, driven by friendships or common experiences, encourage further development of new networks and the exchange of tacit-knowledge between related actors (Maskell and Mallberg, 1999; Ziemiański, 2018; Etzkowitz and Leydesdorff, 2000).

Finally, cultural proximity refers to unwritten social and cultural rules for how people should behave in various contexts. Research show that shared norms and beliefs in networks play an important role (Adler and Kwon 2000; Runiewicz-Wardyn, 2020b). Nahapiet and Ghoshal (1998) state that norms represent a degree of consensus in a social system and so called the “norms of cooperation”. These norms have influence on people’s attitudes and motivations towards interactions and information and knowledge exchange. Culture affects how people perceive and interpret their environment. The latter implies that individuals sharing a common language and culture are more likely to perceive the social interactions and exchanges in similar ways.

2.3 Analytical Framework for Empirical Study

The literature review on the innovation networks, presented in the Introduction section, and the concept of proximity, discussed in the earlier section, show that both spatial and non-spatial dimensions of proximity are crucial for the concepts of Entrepreneurial University and “innovation ecosystem”. The following types of proximity – geographical, institutional, cognitive/technological, organizational, social and cultural – make important non-spatial characteristics of innovative ecosystems and influence the formation of innovation networks between their actors (Figure 1).

View Full Size

Innovative ecosystem and innovation networks concepts, and related with them types of proximity

Citation: Triple Helix 9, 2 (2022) ; 10.1163/21971927-bja10025

Source: Own elaboration based on the reviewed litera ture sources

• Download Figure
• Download figure as PowerPoint slide

Furthermore, it is hardly possible for one of the dimensions to generate innovation process alone, nor the simultaneous presence of all six can guarantee successful innovation networks formation. The actors must have the active will to do so. Thus, the focus of the research in the innovation networks must be scaled to the individual level. After all, in innovation ecosystems the network of formal relationships among organizations merges with the personal networks that every individual has in the ecosystem. The boundaries of every innovation ecosystem depend on its participants, which identify with the same ecosystem community. The innovation ecosystem is composed of the multitude of actors. Each have different roles to play (Figure 1). Scientists (SCs) conduct knowledge intensive research (basic and applied) in technology research areas. Managers (MNs) have technical and practical knowledge on technology markets. Technology transfer officers (TTOs) manage, consult and commercialize the IP generated by the research centre. Intermediary organizations and brokers (NP and BPs) bridge cognitive and technological distances through better understanding and recombining cross-disciplinary ideas, problems and solutions. Notwithstanding, it is social proximity that builds the trustful relations between the above actors, and encourage knowledge spillovers, and the exchange of tacit-knowledge between actors. The latter one is essential in solving real problems related to the commercialisation, adoption and diffusion of new technologies. The geographical proximity alone is not sufficient to explain how innovations happen within ecosystems. Rather, it is an enabling factor for the social interactions to happen. The above presented relations derived from the literature on innovation ecosystems, Entrepreneurial University, innovation networks and proximity constitute the analytical basis for the further empirical analysis in the article.

4.1 Biotechnology vs Aviation Industry

The modern biotechnology is a relatively young branch of bioscience, developed by the biopharmaceutical industry in the late 2000s. According to the literature, biotechnology industry started to form its shape in the early 1980s, when it improved the regulatory and patenting and licensing systems and launched government-lead research initiatives, especially in the United States (Zhang and Patel, 2005). Based on applications, there are five main branches of biotechnology: (1) Animal biotechnology, (2) Medical biotechnology, (3) Industrial biotechnology, (4) Environmental biotechnology and (5) Plant biotechnology (www.biotechhealth.com).The innovation processes within the biotechnology industry is not just one life cycle curve, but a succession of curves from organic chemistry/pharmacology to biochemistry, molecular biology, etc. The waves of molecular biology overlap the waves of biochemistry and are about to leap upward. Scientists and researchers are currently attempting to exploit basic molecular research to identify new drugs, the production of which will be based on recent advances in genomics technology. However, since biotechnology companies often focus on just small number of compounds within a specific area, many companies lack both the resources and experience required to bring novel drugs and therapies to the market. Therefore, partnerships with other key stakeholders in the specific fields are very important, and can take several forms, the agreements covering joint research, development, production, marketing and promotion, second- sourcing arrangements and in- , out- and cross- licensing (Gulati and Singh, 1998: 782). Moreover, biotechnology companies often prefer a light internal infrastructure, creating the need for externalised clinical development with partners. One example can include partnerships with contract research organizations (CROs), that provide research services primarily to pharmaceutical, biopharmaceutical, and biotechnology companies. Services include bioanalytical studies, clinical-study and clinical-trial support for drugs and/or medical devices (e.g. Syneos Health or IQVIA enterprises). Moreover, in the biotechnology, as in the whole life sciences sector, the past decade has witnessed the emergence of numerous interdisciplinary areas – bioinformatics, nanobiology, synthetic biology, biomaterials, tissue engineering, computational biology, etc. These new research fields have one common scientific path of development resulting from convergence and collaboration between different disciplines. In fact, this major share of this convergence happened in the very innovative, entrepreneurial and rich in social capital environment of the US Silicon Valley and San Francisco Bay Area (Runiewicz-Wardyn, 2020a). This provided a unique opportunity for life sciences firms to merge with the IT sector. In the late 1980s, UC Berkeley’s postdoctoral fellow and research scientist, Stephen Fodor, came up with the idea that semiconductor manufacturing techniques could be used to build vast amounts of biological data on a glass chip, which would facilitate the analysis of complex genetic information (Zhang and Patel, 2005). Moreover, rapid advances in technological convergence and its applications in life sciences also induce changes in the market conditions, forcing the transformation of current business models, research networks models, and public innovation and R&D support policies. This trend, along with the increasing global biopharmaceutical competition, drives specialization and increases the role of business alliances and partnerships in research and innovation. Close collaboration is also important in the development of genomics technologies that requires massive amounts of information to be collected and analysed. In turn, the characterization of genes requires a means to manage, store and process enormous databases of biological information (bioinformatics).

Aviation is a science-based and science applied industry. It is related to several different engineering fields – mechanical, aeronautical, materials, electrical and electronic. From its beginnings in the mid1930s and increasingly after its dominant design appeared around the, aircraft was a complex product, composed of several major modules or sub-systems, including fuselages, wings, tails, engines, nacelles, avionics, landing gear, electrical system and interior finishing (Zhegu and Niosi 2008). Therefore, industry is often presented as a pyramid, with primary contractors at the top (conducting assembling and sales) to final clients. Thus, the aviation industry deals with all-things aircraft-related within the earth’s atmosphere. These dealings include the design, manufacture and operation of many types of aircraft within this airspace. There are a few different sectors of aviation – Aerial Firefighting, Agricultural Aviation, Business and Personal Travel, Law Enforcement, Medical Transport, Overnight Delivery, Search and Rescue, Sightseeing – with three being the main pillars that uphold the aviation industry as a whole: commercial, general, and military aviation. One typical feature of this industry is the high and rising development cost of aircrafts, and the high risk involved (Niosi, 2008; Mowery and Rosenberg, 1998). Also, development takes several years, and companies cannot forecast what the market conditions will be at the time of the launching of the product. This is particularly true because of the strong cyclical demand for aircraft (Niosi, 2008). Thus, it is not surprising the importance of private-public partnerships in this industry. many national agencies are involved in the support for aeronautical R&D. For instance, the Commission for the Future of the United States Aerospace Industry points to thirty-four different agencies and departments in the US federal government that share some responsibility and support for R&D in the aircraft industry. They include the Departments of Defence (DOD), Energy (DoE), Transportation (DoT), Education (DoE), and Commerce (DoC). A similar pattern can be found in the European Union (European Commission, HORIZON 2020).

The modularization of the aviation and more particularly aircraft industry has allowed international outsourcing of entire subsystems. The outsourcing practice allowed to reduce development costs in the industry by means of cost and risk-sharing agreements with foreign partners (on average, an airplane becomes a profitable product after 500 copies are sold, Niosi 2008). In the late 1970s, American companies started outsourcing entire subsystems to Japan, and in the late 1980s to China (Zhegu and Niosi 2008). These long distance partnerships allowed both to reduce production and R&D training costs, because foreign governments subsidized their local companies in order to increase their learning and inward technology transfer capabilities. The growth of the aerospace engineering services outsourcing (ESO) industry is also facilitated by intensifying partnerships between original equipment manufacturers (OEMs) and service providers for leveraging cost-effective access to domain-specific technical expertise. For example, in February 2019, Diehl Aviation signed a multi-year agreement with Altair Engineering, Inc., with the latter company becoming Diehl Aviation’s strategic computer-aided engineering supplier (Market Analysis Report, 2019). As in words of Frank Wengler, Aviation market segment leader: „(...) diverse expertise and innovative ideas are important elements bringing aviation projects to life, therefore (...) partnering with small and disadvantaged businesses firms opens opportunities to bring in a broader range of capabilities (Wengler, 2019). The smaller firms often hold specialty expertise and local knowledge, skill sets that allow the larger project team to be responsive to changing project needs and demands. The future air transport growth is another factor stimulating the close collaboration and technological convergence in the sector, in particularly with information and communication technology (ICT), display and electronics industries.

In sum, the above discussion confirms that both sectors – biotechnology and aviation – have both similar as well as very distinct patterns of technological innovation process. The table 1 summarizes the selected patterns of technological innovation process in these two sectors. Hence, aviation industry, compared to the biotechnology one, is more mature, characterised by longer technological cycles, multidisciplinary modes of research, has fewer people dedicated to R&D, higher capital intensity and a larger involvement of government funding in the purchase of specialized machinery and equipment.

View Full Size

Selected patterns of technological innovation: comparison of biotechnology and aviation industries

Citation: Triple Helix 9, 2 (2022) ; 10.1163/21971927-bja10025

Source: own elaboration based on Runiewicz-Wardyn (2020a), Zhang (2005), Zhegu and Niosi (2008) and Wengler (2019)

• Download Figure
• Download figure as PowerPoint slide

4.2 Innovation Life Cycle and Innovation Partnerships in Biotechnology and Aviation Industries

As was mentioned in previous section biotechnology industry is about very unique and complex products which have not one life cycle, but rather three different life periods: (1) an early development period, (2) a highly competitive mid-life period and (3) a late post-patent period (Bernard, 2013). At the early phase of preclinical research, various actors and teams of stakeholders may conduct closed-shop or collaborative research. Generally, the R&D is conducted in the academia and organizations research labs. The R&D function in an innovation-driven biopharmaceutical company has two important and interdependent roles. Firstly, to invent, evaluate, and later develop the molecules that eventually lead to products on the market. Secondly, to provide scientific expertise capable of identifying and evaluating external opportunities. These can be specific molecules at various stages in the value chain or technologies and ideas that can be directly developed into new molecules or contribute to other projects (Lipton and Nordstedt, 2016).

Given its length and the risks involved, a single biotechnology company is unlikely to undertake the whole R&D process alone. Presently, the global biotech companies pressure to partner with other biopharma companies and universities to embargo some research, with or without strong or immediate discovery commitment. Academia, on the other hand, finds it convenient to have financial support from industry, especially in the early-stage discovery and professional expertise on the possibilities of transferring their innovation into industry. Moreover, different stages of product development motivate different types of partnership (Rothaermel and Deeds, 2004). They also vary in duration. In terms of the time, upstream activities (such as basic and applied research development, verification and validation, prototype development, etc.) may take years. Similarly, downstream processes such as manufacturing and entering the market require substantial time. Moreover, the form of the value chain in the biopharma field is affected by the available resources and funding (Powell et al., 1996). Upstream activities are dependent on local resources and funding, whereas downstream activities (such as marketing) rely on global resources and developed through partnerships (Hine and Kapeleris, 2006). Partnerships may also be classified according to the objective: exploration or exploitation. Rothaermel and Deeds (2004) link such partnerships to the different stages of biotechnology product development process: discovery and commercialization. In either form of partnerships, on average, only one to two of every 10,000 substances synthesized in laboratories will successfully pass all stages of development required to become a marketable medicine. Moreover, by the time a medicinal product reaches the market, an average of 12–13 years will have elapsed since the first synthesis of the new active substance. This is why intensifying social ties with people on both sides, especially in the early stage of clinical trials greatly increases the success of the project and its money value. In the early development phase of the innovation life cycle, social networks allow researchers and entrepreneurs to engage in discussions, share information and connect with others in order to expand their professional network and raise funds. Similarly, the success of commercialization at the post-launch phase will, to a large extent, be determined by the social and interpersonal networks between different groups of stakeholders (Figure 2a).

Aviation industry sector (both the aeroplane and spacecraft industries) is the industry that has extremely long product-life-cycles and development periods (Figure 2(b)). It is characterised by its renowned heavy upfront investments and exceptionally long programme lives. Governments subsidize the industry through many different policies, i.e. tax credits for R&D, direct subsidies, export credits, R&D loans, sponsoring expensive equipment, training and educating highly skilled labour force.

View Full Size

Product innovation life cycle in the biopharmaceutical (a) and aviation sectors (b)

Citation: Triple Helix 9, 2 (2022) ; 10.1163/21971927-bja10025

Source: Runiewicz-Wardyn 2020a; Bernard 2013

• Download Figure
• Download figure as PowerPoint slide

Therefore, the industry tends to be very regulated and also requires huge stocks of the existing knowledge (even though a slow deregulation of the industry in last decades lowered the entry barriers into its market). The process of technological substitution and adoption of new products and processes is very long. In technical terms, the technology diffusion for the aircraft manufacturing industry has obtained a high rate of maturity as now anyone who can afford the products in that market is able to obtain them without encountering difficulties (Runiewicz-Wardyn, 2013). However because of big lack of expertise of suppliers and users of specific technological processes, the diffusion process may be an “incomplete success or at times a failure on the part of the user to completely absorb the technologies provided by the suppliers” (Rasheed and Manarvi, 2008). Because of the high investments into R&D and high costs related activities it seems the industry is fairly concentrated. Universities did not play as important a role in the innovation partnerships as they had done in biotechnology or nanotechnology. Instead, knowledge is accumulated in companies. Nevertheless engineering schools help in the development of mechanical, electrical and material components and provide with high-skilled labour. Moreover, as the study by Runiewicz-Wardyn (2013) showed that industrial diversity and co-location externalities turns out to be a particularly important input for new knowledge creation in the aviation industry. Therefore, some smaller firms with no access to adequate human resources, organisational and R&D capacities may be disadvantaged. In addition, the industry dynamics converge with the expansion of other industries. Access to IT and computer technology is becoming increasingly important and in general has become a necessity for many parts of operations and marketing in aviation industry (Subramanian et al., 1994; Anbil et al., 1991). Additionally, the industry has been facing a gradual deregulation of competition and recently noted consolidation in the form of mergers between companies or takeovers of failing carriers and a part of the birth-and-death process in the aviation industry. Therefore strong, coherent research networks and innovation partnerships may drive further aviation industry as it approaches its early maturity stage. Last but not least, the industry’s cyclical nature is determined by its complex and sensitive nature relative to the external factors such as the emergence of low-cost carriers, the safety, pandemics or potential terrorist attacks, and the increasing volatility in fuel prices. The implies that it is crucial for the airline and aviation-related companies to be able to adapt dynamically to changing market conditions on a constant basis.

5.1 Geographical Proximity

In all the three innovation ecosystems geographical proximity defined by the interviewees as close physical (walking) distance to both academic institutions and mature companies, was mentioned as a very important driver of potential innovation collaboration and knowledge spillovers via the job mobility or participation in different academic and scientific events. Moreover, the proximity to the lead users in their ecosystems – patients and personnel at the city hospitals – was another factor that allowed to meet the complex and diverse health care needs of the new biotechnology markets. Both, formal and informal physical interactions were equally important for all the ecosystem actors. In addition, the respondents from the Stanford and Cambridge University ecosystems (academic faculty, TTOs, business representatives) emphasized the importance of “social infrastructure – sport centers, clubs, bars and coffees” that created opportunities for informal interactions. Only in the opinion of the Medicon Valley cluster representatives, geographical proximity was somewhat less significant for the informal interactions.

5.2 Institutional Proximity

The study revealed that over 2/3 of biotechnology ecosystem respondents in the selected case studies kept collaboration with the representatives of diverse TH stakeholders. They confirmed that being institutionally proximate facilitated the knowledge transfer and research collaboration with universities. Moreover, the US respondents emphasized that intermediaries played essential role in narrowing both social and institutional distances between the interviewed stakeholders. They also noticed that once the established formal relationships between TH organizations i.e. industry-university or university-hospital, merged with the informal social networks, the institutional proximity became less important.

5.3 Organizational Proximity

The study findings reveal that the primary motives for the partners in the selected biotechnology ecosystems to start talking, about research or academic collaborations, were found in their cognitive and organizational proximities. For example, respondents from the Bay Area ecosystem mentioned social networking events in the field of life-sciences, organized by their own company or intermediaries, strengthened their intra- and inter-organizational networks. They also admitted that informal social networking events stimulated inter-organizational knowledge sharing and improved their competences, capabilities and resources. Furthermore, organizational proximity strengthened the social networks between the Bay Area universities and colleges representatives. Job mobility was mentioned as an important process of bridging and networking between the various organizations. In fact, the researchers in the Bay Area mentioned that staying with one organization for around 4–5 years was an optimal period for successful career and professional network development. This contrasted with the respondents from the Cambridge and Medicon Valley ecosystems, who demonstrated poor job mobility and preferred long-term or undefined work contracts.

5.4 Cognitive Proximity

The study findings provided with the evidence that distance in terms of a knowledge base and created an obstacle for the knowledge and innovative networks formation between the analysed biotechnology ecosystems agents and other non-local biotechnology colleagues. In fact, limited competence and poor absorptive capacities of the non-local actors made the successful research interaction harder. For example, several respondents from Cambridge life-sciences ecosystem found technological and cognitive proximities (the second level of homogeneity) to be a major challenge when expanding innovation-driven networks with partners from Poland or Hungary.

5.5 Social Proximity

In terms of social proximity, the respondents located in the San Francisco Bay Area (university affiliated agents) pointed out that social networking is especially important in the process of search of ideas (early preclinical stage) as well as upgrading of R&D ideas (early maturity stage). Yet, since the early or so called “idea search” stage is very often filled with certain precautions, some respondents (start-up and business agents) demonstrated their preoccupation with a possible copying or imitating of their ideas and concepts. Yet, for another group researchers and faculty members from Cambridge life-sciences innovation ecosystem, increasing intensity of social networking events presented some challenges. In general, the faculty members emphasized the importance of social relations and networking skills, however, they also pointed out to the potential challenges of putting too big pressure on scientists to invest into social “bridging” and “bonding” efforts.

5.6 Cultural Proximity

The findings from the three case studies revealed that at least for half of the respondents shared cultural, religious and linguistic backgrounds were very appreciated, however did not precondition successful social ties and interactions. Almost everybody, in the analysed biotechnology ecosystems, mentioned openness in sharing ideas and meeting people as essential for strong and long-lasting social networks. Furthermore, the respondents mentioned that cultural proximity, resulting in the same cultural norms (Swedish-Danish or Polish), habits and values, enabled one to build trust and thus a willingness to exchange information. For example, respondents with Polish, Chinese or Hindi cultural roots were involved also in their national-based scientific community networks.

In sum, social interactions constituted one of the central aspects that explained the capacity of the three world leading life-sciences ecosystems to foster innovation networks. Such interactions were fostered and moderated by several dimensions of proximity. Geographical proximity was an important precondition for the social/personal interactions within the ecosystems. It also moderated the cognitive and technological distances. The gap created by cognitive and technological distances (between the industry and university actors) was bridged by the geographical proximity. The local latter was done via the intermediaries i.e. non-profit professional organizations in San Francisco Bay Area and the entrepreneurial-scientists in the Cambridge University. On top of that, innovation networks were orchestrated and engineered by these intermediary actors. Social proximity was important in facilitating the interactions and mutual interactive learning. It reinforced the institutional proximity and moderated the cognitive distances occurred along the technology life-cycle. Consequently, strong and long-lasting social networks between the ecosystem actors were built on cognitive and technological proximities, whereas social proximity helped to coordinate and maintain the established innovation networks. Moreover, the weaker inter- and intra- organisational proximities were strengthened by the geographical and social proximities, via broker and intermediary organizations. Cultural proximity was appreciated by all the interviewed actors. Yet, in certain cases, trust was culturally and linguistically embedded.