Version classiqueVersion mobile

The role of German universities in a system of joint knowledge generation and innovation

Mirja Meyborg

2. The Concept of Knowledge and Innovation

Texte intégral

1The increasing importance of technical advance and innovation regarding economic performance and development is a well-discussed and highly-accepted issue in the older macroeconomic theory (Schumpeter 1911, Solow 1956, Romer 1986, Lucas 1988 and Lundvall 1992, Grupp 1998). Thereby, knowledge is a crucial factor for innovation activity, so that innovation can be seen as the result of an interactive process of knowledge generation, diffusion and application (Camagni 2001 and Katila and Ahuja 2002). Van Hippel (1988) already stated that the resource knowledge is an essential factor for innovation research and practice, thus playing an important role in generating successful innovations.

2As this PhD thesis deals with the increasing role of German universities regarding joint knowledge generation and innovation, this chapter now introduces the major concepts of knowledge and innovation. First of all, a short explanation of the term knowledge is given, as knowledge is a driving force for innovation activity. Further, in order to perceive a deeper understanding of the role of knowledge, the nature and concept of knowledge is illustrated. Afterwards, the concept of innovation research is pointed out. In this context, different innovation systems as well as several measures of innovation activity are illustrated. Further, the relevance of innovations for an economy is elaborated. Finally, the increasing importance of innovation networks is stressed followed by a short introduction of the importance of the spatial concentration of firms and institutions which lead to knowledge spillovers, thus, enhancing innovation activity, too.

2.1. The Importance of Knowledge

3It is more than four centuries ago, when Francis Bacon (1561-1626) an English philosopher, statesman, scientist, jurist and author, already referred to the power of knowledge. In his Meditationes Sacrae (1597), he wrote about his ambition to bring the human being on a higher level of existence through the power of knowledge. Later on, many scholars referred to his expression and further showed that the accumulation of human capital is the most strategic resource for regional economic development and competitiveness within our globalised world (Romer 1986, Lucas 1988, Grossman and Helpman 1991 and Lundvall 1992). Besides, Von Hippel (1988) also stated that knowledge is an essential factor for innovation research and practice. Hence, it is not surprising that many scholars illustrated that the expedient exploration of knowledge is essential for shaping successful innovations (Katila and Ahuja 2002). But how can knowledge be defined? There is indeed no single definition of the term knowledge as its meaning goes often in line with the context it is used for. Referring to the Oxford dictionary (2012), knowledge is defined as “facts, information, and skills acquired through experience or education; the theoretical or practical understanding of a subject”.

4Plotkin (1994), for instance, sees knowledge as a kind of dynamic system from which information can be stored, processed and understood, and Blackler (1995) generalises that knowledge is above all an active process that is mediated, situated, provisional, pragmatic and contested (Howells 2002). Of course, there have been many attempts to identify and classify different types of knowledge. One of the first authors who tried to present a first distinction was Michael Polanyi (1958). In his early works, he distinguishes between explicit and tacit knowledge. The difference lies in the degree of formalisation and the requirement of presence in knowledge formation. Thereby, explicit knowledge includes expertise that is transmittable in formal and systematic language, thus, it does not require direct experience of the knowledge that is being acquired. On the other side, tacit knowledge cannot be communicated in any direct way, so that it concerns direct experience which, in turn, means that disembodied expertise is acquired via learning by doing or any other procedures (Cowan et al. 2000). But how can knowledge be best dispersed? As knowledge comes from individuals and not from the institution itself, it is arguable how to transfer this knowledge best into the institution's knowledge base, where it then can be managed and shared. Nonaka and Takeuchi (1995) have developed a simple model what they call the knowledge-creating company. For them, the interaction of tacit and explicit knowledge is the driving force for knowledge creation in any institution and provokes the four phases of knowledge conversion. First, externalisation stands for the conversion from tacit to explicit knowledge. This is, for example, possible through articulating tacit knowledge, hence, bringing it into an explicit form, such as a report or forum discussion. Second, the adaptation from tacit to tacit knowledge through sharing experiences and practice is called socialisation. This kind of knowledge conversion occurs, for example, in apprenticeships and at seminars and conferences. Third, Internalisation implies the conversion from explicit knowledge back to tacit knowledge. In this context, it is mainly about learning by doing and leads to operational-procedural knowledge. Last, dissemination means the adaptation from explicit to explicit knowledge, and appears when the owner of certain knowledge shares it with one another. This kind of conversion is the most common one and takes place through technologies such as computer-based training and groupware (Fischer 2006).

5However, as it is common sense that knowledge is essential for innovation activity due to the fact that innovation is closely linked to the exchange and recombination of knowledge (Nonaka and Takeuchi 1995), the next subchapter now illustrates important details on innovation research.

2.2. Innovation Research

  • 3 Innovation is often classified as product and process innovation (see Schumpeter 1934 and Grupp 199 (...)

6In literature, there are various definitions of the term innovation which have been evolving over time. Joseph A. Schumpeter is thereby often thought of as the first economist who draws special attention to this topic. So, it is not surprising that he is seen as the father-figure of innovation research. Hence, his early work from 1911 “The Theory of Economic Development” demonstrates the cornerstone of the innovation research era. There, Schumpeter distinguishes between economic growth and economic development. He states that pure economic growth occurs, if either the population or the resources of an economy increase. Apart from this, economic development stands for a “[...] fundamental change in the sphere of production [...]” (Schumpeter 1911). For him, an innovation must be something new, i.e. either a new product or a new production process, which must also have an economic effect3. In the 1930s, Schumpeter further defined five types of innovation:

  • Introduction of a new product or a qualitative change in an existing product,

  • process innovation new to an industry,

  • the opening of a new market,

  • development of new sources of supply for raw materials or other inputs, and

  • changes in industrial organization (OECD 1997).

7These different kinds of innovation appear spontaneously as well as discontinuously within economic affairs. First, they occur next to and compete with the old constructions until they become superior and displace the old ones from the market. In all, for Schumpeter, an innovation implies the phenomenal fact in economic history of a capitalistic economy as well as the basic fundament for changes in economic processes. The innovation process itself is classified by Schumpeter into three steps, namely into invention, innovation and diffusion. For him, the entrepreneur does not have to be necessarily the inventor of a new product or process. The entrepreneur translates the new idea into a business or other useful application and turns the invention into an innovation. Thus, Schumpeter already distinguishes between the pure idea (invention) and its market launch (innovation). But, he already indicates that not any invention must necessarily turn into an innovation. Besides, the third step is called diffusion which means the distribution of the innovation on the market, i.e. the process of diffusion is essential for an innovation to become well-known between different members in different social systems and implies the driving force for economic development (Schumpeter 1939).

  • 4 Product refers to goods and services.

8Many authors now refer to Schumpeter's definition of innovation. Also the OSLO Manual which is produced by the OECD (1997) tries to set a kind of benchmark for innovation research. Today, the OECD identified four different types of innovation. The first one is called product innovation and can include either a new or an improved product4 whose characteristics are significantly different from previous ones (Rogers 1998), i.e. significant improvements in technical specifications, components and materials, incorporated software, user friendliness or other functional characteristics are conducted; a product innovation within the education sector can be a new or significantly improved curriculum or a new educational software (OECD 2005). The second type of innovation (process innovation) is the adoption of new or significantly improved production methods, including methods of product delivery (Rogers 1998). This encompasses significant changes in techniques, equipment and/or software; a process innovation within the education sector can for example be a new or significantly improved pedagogy. Besides these two different possibilities of innovation types, additional focus is laid on marketing and organizational innovation. The former includes a new marketing method involving significant changes in product design or packaging, product placement, product promotion or pricing. Again, the education sector can for example imply a new way of pricing the education service or a new admission strategy. Last, the organizational innovation can be pictured through the introduction of a new organisational method in the firm's business practices, workplace organisation or external relations. A new organization of work between teachers, or organizational changes in the administrative area can be part of an organisational innovation within the education sector (OECD 2005). However, the basic message from the work of the OSLO Manual is that there is no consistent definition for innovation at all (Rogers 1998).

9The following subchapters present different forms of innovation systems, as well as different measures of innovation activity, before illustrating the relationship between economic development and innovation activity.

2.2.1. Innovation Systems

10In the late 1980s, a new conceptual approach appeared in the science, technology, and innovation studies which is called National System of Innovation (NSI). The NSI approach generally stresses that the flows of technology and information among people, enterprises and institutions are key to the innovative process. Thus, innovation is the result of a complex set of relationships among actors in a system, which includes enterprises, universities, research institutes and the government. To understand the NSI can help to identify important points for enhancing innovative performance and overall competitiveness as it can permit to find disparities within the system. Policies which seek to improve networking among the actors and institutions in the system are most valuable in this context (OECD 1997). Most authors agree that the framework itself comes from researchers like Freeman (1987), Lundvall (1992), Nelson (1993), Patel and Pavitt (1997) and Edquist (1997) (Godin 2007 and Cooke et al. 1997). Freeman (1987), for example, puts the accent on the network of institutions in the public and private sectors whose activities and interactions accompany technical innovations. Lundvall (1992) defines NSI as “[...] the elements and relationships which interact in the production, diffusion and use of new and economically useful knowledge [...] and are either located within or rooted inside the borders of a nation state”. For him the domestic interaction is most important and basically explains the existence of NSI. Nelson (1993) emphasises that NSI are a function of government policy at the national level, and Patel and Pavitt (1997) see NSI as national institutions that determine the rate and direction of technological learning in a nation (Niosi 2002). Last, for Edquist (1997), NSI include all important economic, social, political, organizational, institutional factors that influence the development, diffusion and use of innovations. Thus, the concept of an innovation system basically emphasises that the flow of knowledge, technology and information among private persons, enterprises and institutions is essential to an innovative process. Thereby, the interaction between the actors who are needed in order to turn an idea into a process, product or service on the market is indispensable.

11The first author who used the expression NSI was Lundvall (1992) in his thought-provoking book “National Systems of Innovation. Towards a Theory of Innovation and Interactive Learning”. The Idea itself goes even back to the 19th century where List (1841) is firstly associated with ideas concerning the concept of innovation systems. In his early work “The National System of Political Economy”, he already records that every nation with its own language and literature, with its peculiar roots and history, with its special customs and traditions, laws and institutions is self dependent for its own development. Contrary to Adam Smith (1776) who advanced the view that the British capitalism was seminal for all other economies, List was convinced of the idea that proper economic development can be rather reached by considering the specific characteristics of economies than by imitating others (Freeman 1987).

12Further, as innovation systems can be applied and adapted to different research questions, three further systems of innovation have been discussed over the past years, namely regional, sectoral, and technological systems of innovation. As the NSI is already explained in detail above, the following short explanations refer to regional systems of innovation (RSI), technological systems of innovation (TSI) and sectoral systems of innovation (SSI). First, the RSI is comparable with the national one, but includes only regions within countries or includes parts of different countries, whereas technologies and sectors are still neglected within this special system. RSI have been introduced by Cooke et al. (1997), Asheim and Isaksen (1997) and Cooke (2001), amongst others. For Cooke et al. (1997), for example, regions are ‘[...] territories smaller than their state possessing significant supra-local governance capacity and cohesiveness differentiating them from their state and other regions [...]’. Second, other economists rather highlighted TSI by centering on technology fields, i.e. the TSI explores a specific technology and is mostly not restricted by geographical or sectoral boundaries (Edquist 2001). Third, Nelson and Rosenberg (1993) gave some good reasons that support the sectoral conceptualisation (Christ 2007). Following their ideas, “[...] the system of institutions supporting technical innovation in one field, say pharmaceuticals, may have very little overlap with the system of institutions supporting in another field, say aircraft” (Nelson and Rosenberg 1993). Besides, Christ (2007) visualized the variety of systems of innovation as follows:

Figure 2: Varieties ofthe Innovation System (Christ 2007).

13The above coordinate system now illustrates the interaction of the four different systems of innovation. The y-axis shows the sectoral scale of analysis, while the x-axis presents the spatial scale of analysis. Whereas the NSI and RSI refer to the geographical perspective, the innovation concept is also used for technological and sectoral perceptions as can be seen from the figure. Thus, academic literature illustrates some really good modifications and co-evolutions in reference to a more disaggregated level of analysis. Thereby, much literature is increasingly about a system of innovation that focuses on agglomeration and territorial innovation as Silicon Valley or Route 128 where sectoral specialization and causes of this local agglomeration overlap. Thus, it might be difficult to establish explicit distinctions between sectoral and local indications and perspectives. For this purpose, technological, organizational, institutional and economic changes within and between systems either on global, continental, national, or sub-national lines of argumentation have come to the fore. Additionally, technological systems can also be transnational and even global (Christ 2007). However, overall, this PhD thesis makes primarily use of the NSI, but also highly focussing on the regional perspective.

2.2.2. Measures of Innovation

14It has widely been accepted that it is of high importance to be able to measure innovation, due to the fact that innovation is one of the main drivers of sustained economic growth, if not the single most important driver (Romer 1986, Lucas 1988 and Grossman and Helpman 1991). As innovation is a complex concept, it is not easy to suggest proper innovation indicators. However, in order to measure something which is not measureable at the first glance, special indicators have to be used which, at least, indicate a good trend of the development path of the certain object of investigation. Even though, there is no generally accepted indicator for innovation, Grupp (1998) delivers an overview of several innovation indicators. The following figure now shows the classification of three groups of innovation indicators and their typology:

Figure 3: Overview of important Innovation Indicators and their Typology (Grupp 1998).

15As can be seen from the figure, he separates innovation indicators into input (also resource indicators) and output (also R&D results indicators) indicators. A third group implies progress indicators which affect innovation on a micro and macro level of an economy. The resource indicators flow directly into the innovation process, whereas the R&D results indicators implicate the real effects of the process. The progress indicators primarily refer to the real impact of the innovation itself (Grupp 1998). In general, as this issue is a complex one, validity of any innovation indicators is restricted and interpretations should be done with caution (Acs et al. 2002 and Kleinknecht et al. 2002). Nevertheless, the use of any innovation indicators strongly increases over the last two decades (Lepori et al. 2008) as the access to data-bases became simplified and as the interest of policy makers, enterprises and the society regarding the usage of such indicators highly increased over the past years (Freeman and Soete 2009). But what do the three groups of different innovation indicators exactly imply?

16The resource indicators stand at the beginning of the innovation process and encompass all measurable inflows into an innovation process. Due to this, they are also called input indicators. The overall input, thereby, often consists of several different inputs as R&D expenditure, R&D personnel and R&D-intensive equipment. Other inputs are governmental R&D subsidies, technical consulting, and all kind of expenditures and investments of knowledge transfer, such as fees, licenses, standards, as well as, material and components. In most cases, these indicators can be measured unambiguously in monetary units, even in different countries (Grupp 1998). Nevertheless, all resource indicators should be used with caution and demonstrate a rather restricted tool to measure innovation, as they only cover the inputs of the innovation process. Already Schumpeter (1939) stated that not all inventions must turn into an innovation, thus, also the results of the resource indicators must not necessarily lead to a market launch of a new product or process (Coombs et al. 1996 and Ejermo 2009).

17In contrast, according to Grupp (1998), R&D results indicators try to measure the output of an innovation process, hence, implying any returns that result from R&D efforts. As can be seen from the above figure, the most commonly used indicators in this sense are scientific publications, patent applications and norms. Thereby, scientific publications are the outcome of scientific work and fundamental research and patent applications imply the outcome of applied science and experimental development. Norms are defined as a commonly accepted document which covers rules, guidelines or general standards which, in turn, encompasses codified knowledge of the state of the art (Konrad and Zloczysti 2010). As this PhD thesis uses scientific publications and patent applications to measure knowledge generation and innovation, both indicators are explained in detail in chapter four.

18Last, progress indicators complete the set of innovation indicators. As already mentioned above, they do not refer to single R&D efforts but to the real impact of the innovation itself. They can further be divided into direct and indirect indicators. Direct progress indicators are, for example, innovation counts or the evaluation of product characteristics. But, it seems to be obvious that this kind of evaluation or the mere count of innovation implies problems, as it is indeed arduous and time consuming to carry out surveys which are engaged in such concerns. Long-run surveys are even more problematic as they are often not possible to conduct because of panel mortality and changes in the panel, such as insolvencies and mergers and acquisitions. Indirect progress indicators are production growth, factor productivity and a variety of trade and market share indicators. As trade numbers can be, for instance, extracted from federal ministries or annual reports of large multinational enterprises, they are easier to carry out compared to the direct progress indicators (Tran 2011). This PhD thesis makes use of the second group of innovation indicators, namely R&D results indicators, which are presented in detail in chapter four. Generally, it can be stated in this regard that knowledge creation, innovation and diffusion is very difficult to measure as there exist no conventional statistical variables.

2.2.3. Innovation and Economic Development

  • 5 Solow (1956) made use of the classical Cobb-Douglas production function.

19It is not a new phenomenon that economists appreciate the importance of technological progress in order to maintain sustainable economic growth rates. Adam Smith (1776) already referred in his book “An Inquiry into the Nature and Causes of the Wealth of Nations” to the advantages of technical advance through his famous description of productivity improvement in the making of pins. Later on, it was Schumpeter (1911) who pointed out that sustainable economic growth rates can be achieved through new technology and innovation. But, due to the limited mathematical capabilities applied in economics at this time, his work has not been displayed in a formalised way. Even though his work has not been considered for further developments of innovation theories for a long time, he has later on strongly influenced all further innovations theories. Elaborating upon the question of the relationship between innovation and economic growth, it is not surprising to go back to Solow's neoclassical growth model. Solow (1956) started to consider the influence of innovation by means of a classical production function5. First, one output (Y) and two inputs, namely labour (L) and capital (C), has been considered as demonstrated in the following equation (1):

Y = Lα *C1-α*TFP (1)

  • 6 Note that exogenous growth theories comprise a whole class of theories, among which the model of So (...)
  • 7 For a clear and detailed introduction of the model of endogenous technological change see Romer (19 (...)

20In doing so, he was able to approximate the relevance of technical advance for economic growth by illustrating that the two input factors (L) and (C) of the Cobb-Douglas production function do not fully explain economic growth in the United States (US). Hence, as there must be another factor for this economic growth, he considered a third input factor, the so-called Total Factor Productivity (TFP), which now fully explained economic growth in the US.6 Of course, also his traditional theory of economic growth does not yet clearly measure the role of human capital. Augmented theorizing on economic growth went beyond the limits of exogenous technological innovation and highlighted the importance of the accumulation of human capital as a determinant of economic growth (see Romer 1986, Lucas 1988 and Grossman and Helpman 1991), thereby, Lucas (1988) was one of the first researchers who considered human capital as an alternative to technological process to improve economic growth. Traditionally, education and learning by doing were considered as main sources of human capital, and, for that reason, many growth models have introduced these factors in their models. Following Romer (1996), countries need to understand the importance of technical advance, therefore, building up a knowledge-based economy to retain stable economic growth rates. Thus, it has widely been accepted that the ability to create, access and use knowledge and technology is becoming a fundamental determinant of long-term development and competitiveness. This PhD thesis exactly picks up this line of thought and explores the effect of the accumulation of human capital in the broad sense. Nevertheless, the entire Romer model7 will not be introduced in the following as it is out of scope of this work.

21However, as it is common sense that knowledge and the ability to absorb, create and exchange knowledge interdependently are essential for the generation of innovations, innovation and diffusion usually comes up as a result of collaboration activity within a net of individual and/or institutional connections. Thus, the following subsection now introduces the importance of networks within knowledge generation and innovation.

2.3. The Role of Networks within the Process of Knowledge Generation and Innovation

22Knowledge generation and innovation by a particular person or institution do not only depend on their own efforts but also on outside efforts. Hence, networking assists the individual or institution to get information and knowledge that they cannot generate internally. As a part of a network, the individual or institution can usually create greater opportunities for learning, hence, enabling to improve economic performance (Fischer 2006). The following subchapters now firstly present a short introduction of the analysis of networks and clusters. Then, the importance of innovation networks is stressed.

2.3.1. The Analysis of Networks

23The analysis of networks has a long history and goes even back to the beginning of the 20th century, where Marshall (1907) concentrated on industries in Sheffield, Lancashire and other British regions, firstly using the term 'industrial districts' in order to describe a geographical district.

24In this context, he considered the manufacturing industry where he observed groups of skilled labour and many small firms of similar characteristics. According to him, specific advantages could be associated with exactly these ‘industrial districts’ as, for example, specialization or the flow of knowledge, information and plenty of good ideas (Marshall 1907). However, the interest of network building highly increased not until the last two decades due to the fact that it has become a topical subject in industrial economic development and public policy (Sacchetti and Sugden 2003).

25As there is no single definition for the term knowledge (see subchapter 2.1), there is also no agreement on the appropriate definition of networks. In the broad sense, networks are groups of interactive actors or organizations which are legally and financially independent from each other. Through this kind of informal cooperation, all participants wish to gain advantages from one another, perceiving themselves as partners and not as rivals. Due to the fact that the success of the whole network depends on each individual actor, it is evident that all involved persons follow similar goals. Hence, they usually behave cooperative and coordinate and match their activities forbearing (Hunt and Morgan 2000). Besides, Aldrich and Zimmer (1986) define networks as investments into cooperative ties between actors, in order to exchange and share knowledge and resources (Borch and Arthur 1995). For Thorelli (1988), networks are an interaction of nodes and lines, where each individual actor wants to enhance its position within the network.

  • 8 Scholars have, however, pointed to negative consequences of forming networks, too, as for example f (...)

26According to Powell (1990), networks only exist, if all actors refrain from following their own individual interests to avoid that anyone else interests are jeopardized. Thus, all activities can be characterized as mutual and supportive for all actors likewise (Sacchetti and Sugden 2003).8

  • 9 See subchapter 2.4 for further information on geographical proximity.

27It is further important to distinguish between different types of networks. On the one hand, vertical networks take place along the production chain for particular products, whereas, on the other hand, horizontal networks occur between partners at the same level of the production process. Moreover, the shape of relationships may differ according to the relational structures between the independent actors involved, either internalized of externalized relationships. Referring to the externalized relationships, there are, on the one hand, formalized networks such as a joint venture which is formed by two or more partners which build up a separate company on shareholder basis. On the other hand, informal relationships, for example, between individual researchers who are located in different institutional settings, are highly trust-based as there is no contractual validation (Fischer 2006). This PhD thesis will focus on the latter, as it is about informal linkages between actors of different institutional settings, no matter if privately or publicly funded. Further, one important aspect within this PhD thesis implies the spatial dimension of network collaboration. Thus, the phenomenon of clusters should be also shortly mentioned in this context, as they are a special from of a network with a strong focus on the spatial aspect. Generally, clusters can be understood as networks of interconnected companies, specialized suppliers, service providers, firms in related industries and associated institutions, such as universities or research institutes, in a particular field that cooperate with but also compete against each other. Hence, according to Porter (2000), clusters do not only need local supplier chains to gain a competitive advantage but also close network collaborations. He further argues that competition rests on innovation and that close network collaboration with enterprises and other institutions is important, not only for efficiency, but also for the rate of improvement and innovation (Porter 2000). To sum up, network collaborations are able to create continuously innovations due to the fact that they concentrate knowledge and expertise (Simmie 2004), whereas clusters further focus on the importance of geographical proximity9.

2.3.2. Innovation Networks

28Innovation networks are a special form of a network, and consist of knowledge-intensive linkages between different actors. There are indeed many definitions of innovation networks. According to Koschatzky (2001), an innovation network can be understood as all possible organisational forms of different actors which contribute to the diffusion and exchange of information, knowledge and resources and simultaneously help to realise the innovation itself. Thus, innovation networks are social systems which concentrate on inventions, i.e. the development of an innovation and its market launch (Borchert et al. 2004). Another definition which is offered by Tijssen (1998) seeks to capture the most important features of the network mode. For him, an innovation network can be seen as “an evolving mutual dependency system based on resource relationships in which their systematic character is the outcome of interactions, processes, procedures and institutionalization. Activities within such a network involve the creation, combination, exchange, transformation, absorption, and exploration of resources within a wide range of formal and informal relationships”.

29Generally, much scientific work on economic development has focused on the increasing interest regarding innovation networks due to the fact that such networks facilitate sustained economic growth. Already Lucas (1988) and Grossman and Helpman (1991) have shown that knowledge diffusion between social actors within an institutional system is crucial for innovation and economic growth. Later on, Reid and Smith (2009) have stated that cross-institutional collaboration can be the catalyst for innovative economic development. Thus, innovation networks are seen as a kind of construction through which productive resources, social values and economic interests can freely circulate (Gluckler 2007). Also Breschi and Lissoni (2003) remarked that networks that include members from more than one country or institution allow for spreading knowledge freely among the various countries and institutions. For instance, many researchers have explored the relationship between networks and innovations by analyzing how these networks influence the frequency of patenting. They found out that there is a strongly positive relationship between network formation and innovation within diverse industries, and across industries. This positive relationship holds for network formation not only other enterprises but also universities. Further, within our knowledge-based economy, innovation networks facilitate the production of complex goods that are difficult to produce in their entirety within one single firm (Grodal 2004). Another stream of literature refers to advanced innovations which are said to draw on new scientific knowledge. This knowledge can be acquired through tight network collaborations with universities. Besides, incremental innovation are rather said to take place more in networks with partners from the business sectors (Todtling et al. 2009). However, as knowledge diffusion between different social actors is critical for innovation and economic growth, it is not surprising that the literature on collaborative research networks and their impact on knowledge diffusion and innovation have increased greatly over the last years.

30The following subchapter now finally introduces the phenomenon of geographical proximity in times of globalization, as this PhD thesis lays its special focus on the spatial dimension, too.

2.4. Geographical Proximity in Times of Globalisation

31In an ever more globalised economy, it is arguable whether proximity still matters or not. Due to the growth of multinationals and global markets, economic theories often proclaim that 'geography is dead'. Also the rapid diffusion of the information and communication technologies (ICT) offers new opportunities for enterprises and institutions as it makes it possible to nearly completely neglect geographical proximity (Morgan 2004).

  • 10 See Rallet and Torre (2000) for further information on this distinction.
  • 11 See Boschma (2005) for further information on this distinction.

32Before coming to the role of geographical proximity in times of globalisation, some concerns about the term proximity in the broad sense, and geographical proximity in the narrow sense should be made. A vast majority in the literature of various fields loosely refers to proximity as notions such as space, neighborhood, relationships and networks of professionals with recognized expertise and kinship (Torre and Rallet 2005). In the special field of economics, Rallet and Torre (2000) have presented a simple definition based on a distinction between two types of proximity, namely geographical and organized proximity10. Boschma (2005) goes even further and provides five dimensions of proximity, called cognitive, organizational, social, institutional and geographical proximity11. This PhD thesis refers to geographical proximity, which is, as its name already indicates, the kilometric distance that separates two units in geographical space. Whereas the term unit could imply individuals, institutions or cities. The purpose of examining the distance of two or more entities is simple as it is for determining whether one is 'far from' or 'close to' any other unit (Torre and Rallet 2005). However, as this PhD thesis is about knowledge-intensive collaborations and innovation networks, it is always about actors who are connected by sharing the same reference space or knowledge. Here, the reference space consists of joint scientific publications and patent applications. Thus, the extent to which relations are shared in an institutional arrangement, either within or between institutions, is always present.

  • 12 See Boschma (2005) for the classification of other possible forms of proximity that could matter in (...)
  • 13 As it is for collaboration networks in general, proximity may also have negative aspects on knowled (...)

33A key issue is now to determine whether the significance of geographical proximity on joint knowledge generation and innovation is still existent or decreased over time. A vast majority of the literature on economic geography suggests that the diffusion of knowledge clusters geographically, i.e. knowledge spillovers are rather found within a short distance. Jaffe et al. (1993) have shown this through patents that have been more likely cited by local patents than by patents that have been located farther away. According to Audretsch (1998), the diffusion of knowledge from the firm or university who create that knowledge to any other third-party is, on the one hand, essential to innovative activity, and tends, on the other hand, to be spatially restricted. Zucker et al. (2001), for example, have demonstrated that biotechnology firms are strongly influenced by the location of successful giants in academic research institutions. Lundvall (1992) has, further, argued that the importance of geography should differ by technology type. Thus, the importance of geographical proximity grows quickly for technologies that are undergoing incremental innovation and radical innovation. Other researchers show a geographic pattern in the context of European patent citations, as, for example, Maurseth and Verspagen (1999) have demonstrated in a study of over 100,000 patent citations between European regions that there is strong evidence of geographic clustering. Storper and Venables (2004) have stressed that the relevance of geographical proximity exists when there is uncertainty about the future and nature of a new product or process. Finally, Hamermesh and Oster (2002) have shown that the impact of information and communication technology on productivity or on knowledge diffusion across distance is not that great as expected. In contrast, the results of Arundel and Geuna (2004) have shown that firms interested in codified knowledge, e.g. in terms of patents or publications, are less likely to perceive geographic distance a barrier. Breschi and Lissoni (2003) expand the study of Jaffe et al. (1993) who has pointed to the importance of geographical proximity and found out that social proximity is more relevant for the degree of knowledge spillovers than geographical proximity (Cantner and Graf 2006). Besides, in times of globalization, Johnson and Lybecker (2012) have found that within the biotechnology sector, knowledge is more likely to diffuse over longer distances than it was true 20 years ago. Also Boschma (2005) provided a critical assessment on this topic. He has, generally, stressed that the importance of geographical proximity cannot be assessed in isolation12 so that he has come to the conclusion that the phenomenon of geographical proximity is neither a necessary nor a sufficient aspect for knowledge-intensive collaboration or innovation to take place.13

34This PhD thesis will now give a further insight into the phenomenon of geographical proximity in the context of knowledge generation and innovations and will show by means of scientific publications and patent applications if geographical proximity still matters as much as it did years ago.


3 Innovation is often classified as product and process innovation (see Schumpeter 1934 and Grupp 1997).

4 Product refers to goods and services.

5 Solow (1956) made use of the classical Cobb-Douglas production function.

6 Note that exogenous growth theories comprise a whole class of theories, among which the model of Solow is only one among many.

7 For a clear and detailed introduction of the model of endogenous technological change see Romer (1990).

8 Scholars have, however, pointed to negative consequences of forming networks, too, as for example firms could get too focused on each other that they are not aware of other developments within the industry (Portes and Sensenbrenner 1993).

9 See subchapter 2.4 for further information on geographical proximity.

10 See Rallet and Torre (2000) for further information on this distinction.

11 See Boschma (2005) for further information on this distinction.

12 See Boschma (2005) for the classification of other possible forms of proximity that could matter in terms of interactive learning and innovation.

13 As it is for collaboration networks in general, proximity may also have negative aspects on knowledge-intensive activity and innovation due to the problem of lock-in (Boschma 2005).

Table des illustrations

Légende Figure 2: Varieties ofthe Innovation System (Christ 2007).
Fichier image/jpeg, 72k
Légende Figure 3: Overview of important Innovation Indicators and their Typology (Grupp 1998).
Fichier image/jpeg, 103k


Open access

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search