Local Knowledge Resources and Knowledge Flows

BYLINE: Brenner, Thomas

In recent years, knowledge has become a major research topic in economics. In the literature it is even claimed that we have a knowledge-based economy and some researchers talk about the New Economy in this context. This might be an exaggerated view. Furthermore, in recent years technologies have been developed that have tremendously improved the ways of storing, transporting and transferring knowledge. Information technology makes it much easier nowadays to track down relevant available knowledge. This holds to a large degree for the scientific world where almost all knowledge is written down and published. Knowledge flows depend less on collaboration networks in the case of knowledge in scientific publications, which implies that knowledge flows differently in the scientific world than in industries.

FULL TEXT:

In recent years, knowledge has become a major research topic in economics. In the literature it is even claimed that we have a knowledge-based economy (Foray and Lundvall, 1996; Godin, 2006) and some researchers talk about the New Economy (Godin, 2004) in this context. This might be an exaggerated view. Knowledge always played an important role in the economy. However, the stock of knowledge on which economic activity is based today is definitely much larger than in previous eras. Foray and Lundvall (1996) argue that knowledge has become quantitatively and qualitatively more important than before. Furthermore, in recent years technologies have been developed that have tremendously improved the ways of storing, transporting and transferring knowledge.

This has caused a division of knowledge. It is long ago that one person was able to command the available knowledge to such an extent that he could work on the scientific frontiers in all disciplines. It seems as if it is today even impossible to command the knowledge that is necessary to develop one technology further. Often many different knowledge resources have to be employed in an innovation process. Hence, one major capability of a researcher or innovator is to be able to command knowledge flows and collect knowledge to such an extent that knowledge is at hand if needed.

Information technology makes it much easier nowadays to track down relevant available knowledge. This holds to a large degree for the scientific world where almost all knowledge is written down and published. It holds much less for research in private companies where technological knowledge often is a major factor of competitiveness (some proof for this difference is given by Sorenson and Singh in this issue). Researchers in private companies have usually no incentive to write down their knowledge. It holds also much less for knowledge that is related to research practices. While the methods applied have to be published in the scientific world many other aspects of conducting research are less well described in the literature. Private firms do not at all make public the way they organize and conduct research. Hence, we can state that information technology makes the transfer of knowledge easy, but that this does not mean that all existing knowledge is easily accessible.

Much of the knowledge that is required in an innovation process is not freely available. Other ways of getting hold of required information are therefore necessary. In the literature it is often argued that tacit knowledge plays an important role in this context because it cannot be (yet) codified and has to be transferred via personal (frequent) contact (Cooke et al., 1997; Audretsch, 1998; Edquist et al., 2001). However, it is unclear whether personal contact really allows for more knowledge exchange than distant contact with the use of the current information technology, such as video conferences (a more detailed discussion of the tacit argument is given by Brökel and Binder in this issue).

Therefore, let us state that in innovation processes sometimes-presumably quite often-knowledge is required that is not freely available, either because it is not codified, or because it cannot be codified, or because somebody is able to restrict access to it-in whatever form. We might now ask three questions:

* How is the knowledge transferred?

* What kind of knowledge is usually not freely available, necessary in innovation processes and transferred mainly locally?

* Why does the spatial dimension matter in this context?

Of course we could ask the same questions for codified, freely available knowledge. However, here it is argued that this is what the contemporary information and communication technologies have changed. Codified, freely available knowledge can be tracked down independently of its location. There is also no "important" additional cost involved, even if the knowledge is located in a far away place. Hence, the spatial dimension should not matter in the case of codified, freely available knowledge. One might argue that nevertheless knowledge in far away places might be difficult to detect or people might not know about the existence of such knowledge. However, this means that getting hold of the required knowledge involves two parts: knowing that and where it exists and obtaining the knowledge. Separating the two parts implies that the former part is again knowledge that might not be codified and freely available. For the latter part the above argument still holds.

Since we are concerned with the importance of knowledge flows within regions, we neglect codified, freely available knowledge in the following and focus on knowledge that is not freely available, for whatever reason. This should be the only kind of knowledge that has a spatial component, except for the fact that people might become aware of knowledge by chance with a higher probability within a region.

Such a spatial component has been empirically proven to exist. Many studies that examine the impact of local knowledge resources on the innovativeness of firms find a strong regional bias (see, e.g. Jaffe and Henderson, 1993; Anselin er al., 1997; Beise and Stahl, 1999). The usual way of conducting such empirical studies is to estimate knowledge production functions (see, e.g. Jaffe, 1989; Acs et al., 1994; Feldman and Florida, 1994; Anselin et al., 1997). As a result of these studies, we know that local knowledge resources are important for innovation processes and that the spatial dimension matters. However, these studies do not answer the above questions. They do not answer the question of what kind of knowledge is transferred, how it is transferred and why local proximity is important for this transfer. What these studies, however, do tell us is that knowledge transfers are important and that they have a spatial component.

A further shortcoming of the existing literature is that industrial differences are often ignored. It seems obvious that industries differ in the relevance of knowledge sources and in the importance of different ways of knowledge transfer. This is rarely discussed or studied in the literature (exceptions are Mansfield, 1995; Broke! and Brenner, 2005; Brenner and Greif, 2006). Some works study, at least, the impact of one factor for various industries separately (see, e.g. Jaffe, 1989; Anselin era/., 1997; Balconi et ai, 2004). However, we still know very little about which kinds of knowledge transfer matter in which industries. Definitely more research is needed on this issue.

In the following we will focus on what answers the literature provides to the three questions above, what the papers in this issue contribute to answering these questions and what open questions remain. This will be done step by step considering one question at a time.

Mechanisms of Knowledge Transfer

The ways in which knowledge is transferred locally is thoroughly discussed and examined in the literature. We distinguish three types of mechanisms here.2 First, if persons who hold knowledge are employed by a firm, they move from one place to another and transfer the knowledge they hold. second, most innovation projects are conducted in cooperation with other actors, who usually contribute complementary knowledge or capabilities. Third, a person involved in the innovation process might obtain necessary knowledge by personal, informal contact with other people. In this case knowledge is transferred from one person to another.

Let us start with the knowledge transfer that is caused by the movement of people. For innovation processes, four kinds of people that move to a firm, firm site or R&D department of a firm are of interest:

* Graduates from universities

* Researchers from public research

* People who have worked in other firms with similar technologies

* People who move within the firm, especially between different departments of the firm.

In the case of university graduates, there is evidence that the share of these in a region or firm is relevant for-or, at least, correlated with-innovation output (Audretsch, 1998; Rao et al., 2002; Romijn and Albaladejo, 2002). However, this does not tell us whether they are important because of the knowledge they bring with them or because of their skills and competences.

Some more details are provided by Audretsch and Stephan (1999) about researchers who start a firm and thus transfer technological knowledge into the private sector. A detailed discussion about the importance of researcher mobility for the transfer of knowledge between R&D departments of the same firm is also given by Criscuolo (2005). However, in total we do not know much about the importance of the people's mobility for the knowledge transfer that is involved in innovation processes.

The situation is even worse in the case of formal cooperations. There is a large literature that studies the relevance of cooperation for innovation. However, in the theoretical literature, the fact that knowledge might be transferred within such cooperation is rather considered to be an unwanted side effect. The empirical literature, meanwhile, focuses on measuring the extent of cooperation and its impact (see, e.g. Fritsch and Lukas 1999; Rao et al., 2002). Little is known about the impact of cooperation on knowledge flows. Fritsch and Franke (2004) find that formal R&D cooperation is only of minor importance for knowledge spillovers.

Informal contacts are often seen as important vehicles of local knowledge transfer as well. However, while there is much conceptual discussion about their importance (see, e.g. Camagni, 1995), they are very difficult to study empirically. There are some case studies that emphasize-on the basis of evidence gained from questionnaires-an importance of knowledge exchange via informal contacts in local clusters (see, e.g. Kozul-Wright, 1994; Brown and Hendry, 1998; Lawson and Lorenz, 1999).

Further evidence is provided by an OECD study (Arvanitis and Hollenstein, 2001). It finds that communication with customers/users of products is ranked as the most important knowledge source, while communication with suppliers of materials/components, with competitors, with suppliers of equipment and with subsidiaries/mother firms is ranked third, sixth, seventh and eighth, respectively. Employing people is ranked fifth, while consulting universities and government research institutions is only ranked ninth and eleventh. The other positions in the top 10 are taken by fairs/expositions (second), professional conferences/journals (fourth) and acquisition of capital goods (tenth). This means that the literature provides some evidence that informal contacts and communication between firms is an important way of transferring knowledge. According to the subjective evaluation in questionnaires, it seems to be the most relevant mechanism for knowledge flows.

In recent years, researchers have turned to study the role of networks in these knowledge transfers. Many different kinds of networks are studied, ranging from networks of co-patenting persons to networks of Internet links, cooperating in EU projects or students' mobility (Balconi et al., 2004; Breschi and Lissoni, 2004; Maggioni and Uberti, 2005). Besides describing the structure of the networks, these works also study whether there are differences between scientific and industry networks, whether they have a spatial dimension and how different networks are related. The issue of knowledge transfer enters the picture by the fact that the network links are defined as something, like co-patenting, that includes knowledge transfers. An alternative approach to measure the importance of networks is taken by Boschma and WaI in this issue (see description at the end of this introduction).

This literature sheds some light on the structure of research networks and their development, in the case of the paper by Boschma and WaI it also gives some answers to the question of how important these networks are for knowledge transfer. Nevertheless, we can conclude that there is only limited knowledge about the importance of different knowledge transfer mechanisms available in the literature. We have reported above that formal cooperation is found to be rather less important while communication seems to play a strong role.

Kind of Knowledge that Is Transferred Locally

The question of what kind of knowledge is mainly transferred locally is usually answered in the literature by pointing to tacit knowledge. However, as pointed out above and as discussed in detail in the paper by Broke! and Binder in this issue (see description at the end of this introduction), this is not as evident as it seems to be. The empirical literature does not try to answer the question of what kind of knowledge is transferred. This holds even more so for the question of what kind of knowledge is transferred in what way. This seems to most researchers not to be an interesting topic, although if we finally want to answer the question of why knowledge transfer has a spatial dimension, we have to answer this question.

Only if we have detailed knowledge about the combination of types of knowledge and ways of knowledge transfer, will we be able to explain what part of knowledge transfer takes place locally and why. Understanding the spatial component of knowledge transfer is, however, an important part of understanding why geography matters for economic activities. More research should be conducted on this issue.

An exception is the work by Zellner (2003). He provides information about what kind of knowledge is transferred by people moving, in his case from public research to private companies. Researchers themselves evaluate skills and methodological knowledge to be the most important part of what they transfer from their former work in public research to their work in private firms (see Zellner, 2003).

Some evidence, however, can be obtained by studying knowledge sources in more detail. Since we know something about what kinds of knowledge are provided by different knowledge sources, we can deduce from the importance of knowledge sources some insight into the importance of kinds of knowledge. The three papers by Andersson and Karlsson, Sorenson and Singh, and Fritsch and Slavtchev in this issue make contributions to this line of research (see description at the end of this introduction). The literature provides further studies of this kind, although not many. However, the issue of what categories of knowledge exist and how they are transferred is little discussed. Much more could and should be done in this direction.

Spatial Dimension of Knowledge Transfer

Considering the spatial dimension of knowledge transfer two questions are at stake: to what extent is there a spatial dimension in knowledge transfers and why does space matter for knowledge transfer? How well these two questions are answered in the literature differs tremendously. A quite detailed answer is provided for the former question, while only the very general answer based on tacit knowledge is given to the latter question.

Let us start with the former question. For a number of knowledge transfer mechanisms, the literature provides some evidence on the spatial range of these mechanisms. For example, Mohr (2002) has studied how far graduates move from where they studied. This study shows that there is a clear spatial component in the movement of university graduates and hence in the related knowledge transfer. In addition, Beise and Stahl (1999) study how far away research institutes are located that firms name as knowledge resources.

Furthermore, there are many studies that examine the impact of different factors, including knowledge resources such as other firms' R&D activities or public research, on the innovativeness of firms. Some of these include the spatial dimension in their analysis (see, e.g. Jaffe and Henderson, 1993; Anselin et al., 1997). They all find that the importance of knowledge sources depends on the distance between the source and the knowledge recipient.

Similar results are found for formal cooperation. There is a huge literature that examines the spatial distribution of cooperation (see, e.g. Fritsch and Lukas, 1999) and this literature finds that most of the cooperation between firms and between firms and public research occurs locally.

As a consequence, we know much about the fact that knowledge transfer takes place to a great extent locally. This holds in similar, but not identical ways, for the different ways of knowledge transfers and for the different knowledge sources. However, much less clear are the reasons for this regional bias.

In the case of labour mobility it is clear that the attitude of people matters. In some countries they are very unwilling to leave their region, while in other countries people are very flexible. Nevertheless, in all countries there is, at least, some regional bias in searching for jobs.

In the case of cooperation, we find different arguments such as that cooperation might be cheaper or easier within a region, that cultural similarities might make cooperation easier, that the necessary frequent interaction might only be possible locally, that potential cooperation partners might be better known locally, that the search for cooperation partners might have a local bias, that cooperation possibilities might be realized more often within a region, and that the choice of cooperation partners might be influenced by personal contacts, which are more frequent within regions. However, detailed studies that analyse the establishment of cooperation with respect to this issue are lacking.

In the case of informal contacts, the usual argument is that personal ties matter and that these are more frequent within a region. This means that we know a lot about the importance and impact of the different knowledge sources, such as universities and public research, on the innovativeness of firms. There is also a lot of empirical research about the spatial dimension of this impact. There are also several explanations provided for these findings. However, detailed descriptions of the processes and mechanisms are rare and empirical evidence for the relevance and functioning of the different processes is almost always not available. The paper by Brökel and Binder in this issue sheds some light on these issues (see description at the end of this introduction).

Papers in this Special Issue

This issue contains five papers that deal with local knowledge transfers from different perspectives.

Andersson and Karlsson study for Sweden whether various knowledge resources (university R&D, business R&D and number of patent applications) have a significant effect on the growth of value added for firms in the same municipality, in the same functional region or in the same country. They find evidence for positive effects of all three resources within municipalities and functional regions, but not beyond functional regions. All spatial autocorrelation is explained by such an approach that includes interaction within municipalities and functional regions. This can be interpreted as evidence for the fact that university and business research provides a kind of knowledge that is transferred regionally.

Brökel and Binder take up in more detail the question of why knowledge flows more easily within regions. They do this on a theoretical level and discuss on the basis of psychological knowledge why the search for knowledge has a local bias. Furthermore, they argue that other explanations such as tacitness of knowledge do not necessarily explain the spatial dimension of knowledge transfers. They, instead, argue that the cognitive restrictions of humans, which do not allow them to consider all options but to search on the basis of heuristics, cause the spatial dimension of knowledge transfer.

Boschma and WaI contribute to analysing the impact of networks on knowledge transfers. They analyse the importance of networks and the network position of a firm for its innovation activities. To this end, they examine the network in one specific industrial district, the Barietta footwear district in the south of Italy. Structured interviews are used to study the characteristics of the network in this district and examine the innovativeness of the firms therein. Besides explaining the network position by firm characteristics, Boschma and WaI examine how the innovative performance of firms depends on the network position. They find that local connections as well as outside connections increase the innovativeness of firms.

Another study of the importance of certain knowledge sources in the region is done by Fritsch and Slavtchev. They study the characteristics of universities that are important for the innovativeness of nearby firms. This means that they examine the question of how the quality of these resources can be measured and what role they play. To this end, they check the impact of universities on the patent activity in a region. Their study is based on German data. Two characteristics of universities are included in the empirical analysis: the regular funds of universities-which signify the size of the universities-and the external research funds-which they argue represent the quality of research in these universities. They find that only the external research funds matter. Furthermore, they analyse the geographic spread of this impact and find that a significant impact is found for the external research funds in surrounding districts within a range of 50 km.

The paper by Sorenson and Singh provides additional insights in the above directions of research. They include data about networks in an analysis of knowledge transfers visible in patent data. The data consists of 17,264 US utility patents and information about the patents and scientific papers cited in these patents. On a general level, they find that knowledge flows decrease with spatial and technological distance. They also decrease with the distance in collaboration networks. However, there is a clear difference between the diffusion of knowledge contained in scientific publications compared to knowledge contained in patents. Knowledge flows depend less on collaboration networks in the case of knowledge in scientific publications, which implies that knowledge flows differently in the scientific world than in industries.

2 We Ignore embodied knowledge transfer, In which knowledge Is transferred by the purchase of machinery, equipment and components, because this Is not the focus of the discussion here.

References

Acs, Z., Audretsch, D. B. and Feldman, M. P. (1994) R&D spillovers and recipient firm size, The Review at Economics ana Statistics, 78, pp. 336-340.

Anselin, L., Varga, A. and Acs, Z. (1997) Local geographic spillovers between university research and high technology innovations. Journal of Urban Economics. 42, pp. 422-448.

Arvanctis. S., Hollenstein, H. (2001) Innovative activity and firm characteristics: a duster analysis of Swiss manufacturing using firm-level data, In: OECD, Innovative Networks-Co-operation In National Innovation Systems, pp. 49-75 (Paris: OECO).

Audretsch, D. B. (1998) Agglomeration and the location of innovative activity, Oxford Review of Economic Policy. 14, pp. 18-28.

Audretsch, D. and Stephan, P. (1999) How and why does knowledge spill over in biotechnology, in: D. Audretsch & A. Thurik (Eds), Innovation, Industry Evolution, and Employment, pp. 216-229 (Cambridge: Cambridge University Press).

Bakxxii. M., Breschi. S. and Lissom, F. (2004) Networks of inventors and the role of académie: an exploration of Italian patent data, Research Policy. 33, pp. 127-145.

Boise. M. and Stahl, H. (1999) Public research and industrial innovations in Germany, Research Policy, 28, pp. 397-422.

Brenner, T. and Greif, S. (2006) The dependence of innovativeness on the local firm population-an empirical study of German patents, Industry and Innovation, 13(1), pp. 21-39.

Breschi, S. and LJssoni, F. (2004) Knowledge networks from patent data: methodological issues and research targets, in: H. F. Moed, W. Qlanzel & U. Schmoch (Eds), Handbook of Quantitative Science and Technology Research: The Use of Publication and Patent Statistics in Studies ol SST Systems, pp. 613-643 (Dordrecht: Kluwer Academic).

BrAkel. T. and Brenner, T. (2005) Local factors and innovatives-an empirical analysis of German patents for five industries. Papers on Economics & Evolution #0509, Max Planck Institute of Economics, Jena.

Brown, J. E. and Hendry, C. (1998) Industrial districts and supply chains as vehicles for managerial and organizational learning, International Studies of Management and Organization, 27, pp. 127-157.

Camagni, R. p. (1995) The concept of Innovative milieu and its relevance for public policies In European lagging regions, Papers in Regional Science, 74, pp. 317-340.

Cooke, P., Uranga, M. G. and Etxebarria. G. (1997) Regional innovation systems: institutional and organisational dimensions. Research Policy, 26(4/5), pp. 475-491.

Criscuolo, P. (2005) On me road again: researcher mobility inside the R&D network, Research Policy, 34, pp. 1350-1365.

Edquisl, C.. Hommen, L. and McKelvey, M. D. (2001) Innovation and Employment (Cheltenham: Edward Elgar).

Feldman, M. P. and Florida, R. (1994) The geographic sources of innovation: technological Infrastructure and product innovation in the United States, Annals of the Association ol American Geographers, 84, pp. 210-229.

Foray, D. and Lundvall, B.-A. (1996) The knowledge-based economy: from the economics of knowledge to the learning economy. In: OECO, Employment and Growth in the Knowledge-Based Economy, pp. 11 -32 (Paris: OECD).

Fritsch, M. and Franke, Q. (2004) Innovation, regional knowledge spillovers and R&D cooperation, Research Policy, 33, pp. 245-255.

Frltsch, M. and Lutes, R. (1999) Innovation, cooperation, and the region, In: D. B. Audretsch & R. Thurik (Eds), Innovation, Industry Evolution and Employment, pp. 157-181 (Cambridge: Cambridge University Press).

Godin, B. (2004) The new economy: what the concept owes to the OECD, Research Policy, 33, pp. 679-690.

Godin, B. (2006) The knowledge-based economy: conceptual framework or buzzword?, Journal at Technology Transfer. 31. pp. 17-30.

Jaffa, A. B. (1989) Real affects of academic research, The American Economic Review, 79, pp. 957-970.

Jaffa, A. and Henderson, R. (1993) Geographic localization of knowledge spillovers as evidenced by patent citations, Quarterly Journal of Economics, 108, pp. 577-598.

Kozul-Wrighl, T.. (1994) Technological diffusion In the Italian Marshallian industrial district of Brianza, in: United Nations, Technological Dynamism In Industrial Districts: An Alternative Approach to Industrialization In Developing Countries?, pp. 155-188 (New York: United Nations).

Lawson, C. and Lorenz. E. (1999) Collective learning, tacit knowledge and regional Innovative capacity, Regional Studies, 33, pp. 305-317.

Maggioni. M. A. and Uberti, T. E. (2005) International networks of knowledge flows: an econometric analysis. Papers on Economics & Evolution #0519, Max Planck Institute of Economics, Jena.

Mansfield, E. (1995) Academic research underlying industrial innovations: sources, characteristics, and financing, Review of Economics and Statistics, 77, pp. 55-65.

Mohr, H. (2002) Räumliche Mobilität von Hochschulabsolventen, in: L. Bellmann & J. Veiling (Eds). Arbeitsmarkte für Hochqualifizierte, pp. 249-281 (Nürnberg).

Rao, S., Tang, J. and Wang, W. (2002) The importance of skills tor innovation and productivity, International Productivity Manager, 4, pp. 15-26.

Romijn, H. and Albaladejo, M. (2002) Determinants of innovation capability in small electronics and software firms in Southeast England, Research Policy, 31, pp. 1053-1067.

Zellner, C. (2003) The economic effects of basic research: evidence for embodied knowledge transfer via scientists' migration, Research Policy. 32, pp. 1881-1895.

THOMAS BRENNER1

Max Planck Institute of Economics, Evolutionary Economics Group, Jena, Germany

Correspondence Address: Thomas Brenner, Max Planck Institute of Economics, Evolutionary Economics Group, Kahlalsche Straße 10, D-07745 Jena, Germany. Email: brennereecon.mpg.de

1 I want to thank Tom Brökel for helpful comments.