Version classiqueVersion mobile

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

 | 
Mirja Meyborg

12. Final Conclusion, Reflection and Policy Recommendations

Texte intégral

1In the following, the final conclusion firstly presents all relevant results of the empirical analysis, illustrating in how far the four main hypotheses could have been confirmed or had to be rejected. Afterwards, a comprehensive reflection of the empirical results is shown accompanied by several policy recommendations. In this context, it is aimed to highlight potentials of the German academic landscape as well as possible constraints in order to be prepared for new challenges in a world where competition has become increasingly knowledge-based. In doing so, it is aimed to ensure that the German knowledge-based economy can further develop, providing sustainable economic growth rates and remaining competitive in our ever more globalised world.

12.1. Final Conclusion

2In order to present a comprehensive and well structured final conclusion of this PhD thesis, all relevant results of the four main hypotheses are illustrated separately. Hence, the following subsections refer to all important findings with regard to the knowledge generation, innovation and collaboration activities of the German universities.

Knowledge Generation, Innovation and Collaboration

3To be able to show a first broad overview of the empirical results of this first main part, the following table shows the first hypothesis and its outcome:

Table 38: Summary of the empirical Results (own illustration).

4Starting with the publication activity of the German universities, it is to be stated that they have shown high rates of published papers in renowned journals over the past ten years. In 2000, they have already been involved in around 44,000 publications, but could increase this number by 70 % in 2009. Besides, while they have still published more often on their own in 2000, this development has absolutely strongly changed during the last ten years. In 2009, they have been overall engaged in almost 80,000 publications at all, but only about 28,000 have come from single-authorship. Thus, publication and collaboration activity continuously rose over the time period from 2000 until 2009, while single-authorship has increased by only 18% compared to 125% regarding co-authorship. Thus, hypothesis 1 is confirmed in terms of publications.

5Besides, this PhD thesis could also confirm that the German universities have seen a huge increase regarding their number of filed patents. While they have filed around 281 patents from 1999 until 2000, they could already triple their number of patent applications within the time period 2007 until 2008. In all, it can be concluded that the behaviour of the German universities has strongly changed over the past ten years, as they operate much more often as applicants themselves, hence, being less dependent on other institutional actors anymore. Overall, it can be stated that not only the patenting activity of the German universities in terms of filed patents has highly increased over the past ten years, but the number of co-applicants, too. Over the past ten years, the German universities have seen an increase of around 150% regarding the number of co-applicants. Hence, hypothesis 1 is also confirmed in terms of patents.

From merely Knowledge Producers towards Knowledge Mediators

6The following part refers to the increasing meaning of the German universities as central nodes for knowledge and innovation transfer. Table 39 summarizes the empirical results of this part:

Table 39: Summary of the empirical Results (own illustration).

7Coming to hypothesis 2, it has to be mentioned that referring to the institutional distribution of the German university co-authors, network collaborations with universities are still most important, but university-industry linkages have had a more dynamic development. Further, the value of normalized degree centrality has, amongst other, shown that the German universities could have increased their distinct cooperation partners over the past ten years, as the value has risen by around 20%. Thereby, the top three German universities in 2009 are Heidelberg (0.091701), the Technical University of Munich (0.0865566) and the LMU Munich (0.796075). But, regarding the mean value of betweenness centrality, it has been eye-catching that it has not increased over the past ten years, but decreased from around 0.026 to 0.024. Thus, the German university networks have indeed become larger at the expense of a decreasing mean value of betweenness centrality which probably means that the overall network itself has become less dense. Overall, hypothesis 2 is largely confirmed in terms of publications.

8Regarding the institutional distribution of the cooperation partners of the German universities in terms of joint patent applications, it is obvious that the German universities have had only 209 co-applicants within the first time period but could raise this number up to 468 within the time period from 2007 until 2008. Further, they have most often cooperated with enterprises, but the highest increase of co-applicants can be observed within the group of research institutes (plus 540%), even though the absolute number of those partnerships has always been much lower compared to the one of the enterprises. In all, a strongly growing importance of the German universities as co-applicants in innovation networks can be observed, especially with regard to close network collaborations with enterprises. Further, as it applies for the publication data, the German universities have also highly increased their distinct co-applicants over the past ten years, as the value has increased by around 120%. Of course, the dynamics regarding the development of the number of distinct co-applicants is much higher compared to the one of the co-authors as in terms of patents, they have possessed a rather low value within the first time period but could greatly advance themselves. Further, the top three German universities within the last time period are Erlangen-Nurnberg (0.0608496), Wurzburg (0.0608696) and the KIT (0.0521739). Other than the mean value of betweenness centrality based on publications, the German universities could also highly increase their mean value of betweenness centrality regarding their innovation networks. Thus, hypothesis 2 is fully confirmed in terms of patents.

9In all, it is obvious that the German universities have shown different behavioural patterns regarding their publication and patenting activities, as, for example, the top three German universities regarding the highest value of normalized degree centrality differ in this regard. Thus, up to now, it has already been discovered that there are first differences regarding the behavioural patterns of the German universities.

10Last, it has also been shown that third-party funds have had a highly significant effect on the collaboration activity of the German universities with enterprises regarding joint papers. Reasons for the significance of third party funds can be constituted in the fact that most of those funds have come from the private sector which, in turn, has led to more collaborations and co-authored publications with enterprises. Besides, third-party funds have had also a significant effect on the patenting activity of the German universities. This finding is not surprising either, as linkages to enterprises have made up the highest share of collaborations. In all, the OLS regression model has underpinned the results that university-industry interactions are not only prevalent but also gaining much more importance so that industry partners are highly valuable in the system of knowledge generation and innovation for the German universities.

Proximity Patterns in Times of Globalization

11The concluding remarks regarding proximity patterns in times of globalisation also starts with an overview of the empirical result in tabular form:

Table 40: Summary of the empirical Results (own illustration).

12Factors as the growth of multinationals, global markets and the fast diffusion of the ICT may have led to the assumption that proximity patterns matter less nowadays. There are many new opportunities for enterprises and institutions which make it possible to nearly completely neglect spatial proximity. However, the results of this PhD thesis have given further insight into the question of proximity patterns in an ever more globalised world.

13Considering all co-authors within the 1,000 km radius, it could have been discovered that especially those co-authors who have been located within a radius of 100 km have been most important for the German universities as potential cooperation partners. Further, scaling down the size of the distance corridor to 25 km, it could have been even emphasized that especially locality still plays a major role for the German universities. While in 2009, around 2,500 co-authors have been located within the first (0 km - >25 km) distance corridor, the second one (25 km - >50 km) has only possessed around 500 co-authors. Overall, the values of Spearman's rank correlation coefficient have shown a quite strong and negative effect. In 2000, the value has been -0.80 and in 2009 even -0.85 regarding the spatial distribution of the German university co-authors between 0 km and 1,000 km. This means that the number of co-authors declines with the growth of distance. Thus, up to now, above hypothesis needs to be rejected.

14But, it seems likely that the value of Spearman's rank correlation coefficient has been just as high because of the huge number of co­authors within the first distance corridor (0 km - <25 km). Moreover, it has been shown that those co-authors who have been located anywhere between 25 km and <475 km have been more or less equally distributed within their particular distance corridors, while spatial proximity did again matter from about 475 km onwards. Thus, in case that the co-authors of the first distance corridor have not been considered anymore as well as those that have been anywhere located >474 km, it could have been shown that the German universities did not further consider distance for the choice of their cooperation partners. Hence, above hypothesis is to be partially confirmed.

  • 73 See chapter six for the determination of the distance classes.

15To briefly sum up above findings, it has to be said that while locality still seems to highly matter regarding the choice of the cooperation partners of the German universities, regional distance between 25 km and <475 km does not seem to play a role, but from 475 km onwards the friction of distance have started weight again. Hence, above hypothesis is finally to be partially confirmed.73

  • 74 See chapter six for the determination of the distance classes.

16Last, considering all co-authors of the German universities, either being regional (0 km - 468 km) or supra-regional (>468 km) located, it has been discovered that the spatial distribution of the German university co-authors is more or less equally distributed.74 Further, regarding the spatial distribution of the different institutional cooperation partners of the German universities, it has been shown that especially research and industry partners have become more regionally distributed, while the share of universities being regionally or supra-regionally distributed has remained quite similar from 2000 until 2009, even though the German universities have slightly increased their cooperations with supra-regionally located universities. However, in all it can be said that regional cooperations have slightly gained in importance, while supra-regional co-authors have lost a bit in importance over the past ten years.

Analysis of the three different groups of German universities

17After having concluded how the German universities as the whole group did behave regarding their knowledge, innovation and collaboration activities with special consideration of the spatial dimension, this final part refers to the behavioural patterns of the three different types of German universities. Hence, the following table firstly illustrates the outcome of all relevant hypotheses in this regard:

Table 41: Summary of the empirical Results (own illustration).

18Up till now, there is no doubt that the overall publication, patenting and collaboration activity of all German universities has highly increased over the past ten years. However, the empirical results of the last main hypothesis have discovered significant differences between the three different groups of German universities as above table already indicates.

19Before referring to hypothesis 4a in terms of publications, a non parametric median test has previously been conducted which has discovered that the medical as well as the elite universities have generally published more often compared to their counterparts. This does not apply for the technical universities as they have a similar median as the non-technical universities. Coming to their cooperation activity, above table shows that in terms of publications the non­technical universities have more often cooperated than the technical universities, even though they have not offered any differences regarding their publication activity. Further, in 2000, the non-medical universities have been more likely engaged in co-authored publications, while in 2009, it is the group of medical universities that has cooperated more often with other partners. Hence, the behavioural patterns in this regard have changed in favor of the medical universities within the last time period. This changing development could have been largely explained by the value of normalized degree centrality, as it has highly increased for the medical universities over the past ten years, while it has even declined for the non-medical universities. Last, no significant results have been found for the group of the elite and non-elite universities in this regard, even though the elite universities have indeed published more often compared to the non-elite universities. Regarding the institutional distribution of the German university co-authors, it has been demonstrated by hypothesis 4b that the technical universities have cooperated more often with enterprises than the non-technical universities. Further, the non-medical universities have had more linkages to enterprises over the past ten years compared to the medical universities. Finally, concerning the group of elite and non-elite universities, no significant differences could have been found in this regard again.

  • 75 The empirical results have been only significant for the first time period in case of the medical (...)

20Coming to the innovation and collaboration activity of the three different groups of German universities in terms of patents, a non parametric median test regarding the value of betweenness centrality has been conducted in order to explore hypothesis 4a. Thereby, it has been discovered that the medical and elite universities have been more likely engaged in close network cooperations compared to their counterparts. For the group of technical and non-technical universities the median test has not brought up a significant result. Further, coming to the institutional distribution of their cooperation partners, it has firstly been shown that the absolute number of enterprises as well as its increase is very strong in each group. In this context, it has been especially striking that the medical universities have experienced the largest increase of private enterprises as institutional partners which amounts to 1,000%. In all, it is the enterprises which seem to require more academic knowledge as the rising number of partnerships expresses. However, the empirical results have anyway confirmed hypothesis 4b, as the elite and technical universities have more likely cooperated with enterprises than their counterparts, and the medical universities have cooperated more often with research institutes and universities compared to the non-medical universities.75

21Further, hypothesis 4c assumes that there are significant differences regarding proximity patterns when the three different groups of German universities are examined and compared with each other in terms of their publication activity. Above table well documents that hypothesis 4c could be entirely confirmed. Thus, it is the elite and the technical universities that have more likely cooperated with supra-regional co-authors compared to their counterparts. Besides, the medical universities have been more likely to cooperate with regionally located partners compared to the non-medical universities. Hence, it is still prevalent that due to the important role of trust in medical knowledge collaborations, proximity patterns still highly matter.

22Due to the previous results, it has been further assumed that the technical and elite universities have been rather engaged with supra-regionally located enterprises compared to their counterparts, while the medical universities have more often cooperated with regionally located research institutes and universities compared to the non-medical universities. In this context, it is further surprising that it has been the non-technical universities that have rather cooperated with supra-regionally located enterprises and not the technical ones. Thus, above made assumption has to be rejected as the technical universities have been indeed more engaged with enterprises and with supra-regionally located cooperation partners in general, but not with supra-regionally located enterprises but with supra-regionally research institutes and universities. Besides, it has been the medical universities that have more likely cooperated with regionally located research institutes and universities compared to their counterparts and the elite universities have been more engaged with supra-regionally located enterprises compared to the non-elite universities as assumed above.

23Last, it has been looked at the spatial distribution of the German university co-authors based on their country codes. In this context, it has been basically proved whether the medical, technical and elite universities have been more likely to cooperate with partners coming from North America and from EU14 than from emerging countries as BRIC. Thereby, it has been discovered that the medical universities have most likely cooperated with partners coming from North America, while partners from EU14 have occupied second place. It has been further conspicuous that the medical universities have more likely either cooperated with North America or with EU14 than with the emerging countries as BRIC compared to their counterparts. Thus, the medical universities have not changed their cooperation manner in favour of the emerging countries, but still highly trust in the well-known ′Western knowledge′. Also in case of JANZ, the medical universities have more likely cooperated with North American partners. In contrast, the technical universities have indeed more often cooperated with partners coming from EU14 than from North America compared to the non­technical universities. But in case of potential collaborations with partners either coming from BRIC or from North America and EU14, the technical universities have decided in both cases in favor of BRIC. Besides, they have continuously been more likely engaged in close network collaborations with JANZ than with North America and EU14. Thus, the technical universities highly trust in knowledge-intensive interactions with the emerging countries as well as with JANZ which are furthest away from Germany. In case of the elite and non-elite universities, there are only a few significant results. In all, they have also been more engaged in close network collaborations with North America and the EU14 than with the emerging countries as BRIC compared to the non-elite universities.

12.2. Reflection and Policy Recommendations

24To come back to the statement of Francis Bacon from the introduction of this PhD thesis, it is beyond question that the German universities highly contribute to knowledge generation, innovation and are highly involved in close network collaborations so that the old statement gains even more in importance. In an ever more globalised world, it is not an easy task to remain competitive and to retain sustainable economic growth rates. Thus, it is of high interest to invest in human capital as human capital remains economic power.

25The empirical results of the first part regarding the knowledge generation, innovation and collaboration potential of the German universities have especially shown that in case of joint publications, university-industry interactions have observed a more dynamic development compared to the still important university-university linkages. But in absolute numbers, it is of course the universities themselves who are more likely to cooperate with each other. Thus, policy measures should further aim to stimulate cooperations especially between universities and enterprises. Thereby, it is of high interest that both sides, industry and university, benefit from the development of sustainable structures from policy makers. Hence, policy programs as the Hightech Strategy initiated by the BMBF which, amongst other, support knowledge-intensive cooperations between industry and the research community, have already been the right policy choice but should be intensively pursued and further extended.

  • 76 A German ′Gesellschaft mit beschränkter Haftung (GmbH)′ is synonymous with the English Limited (Lt (...)

26Besides, it is also beyond question that an increasing integration of university research and non-university research is also of high interest and remains trend-setting. At the moment, collaboration activities between the German universities and research institutes have not grown as fast as interactions among universities. While collaborations with research institutes have been more frequently occurred in 2000 compared to the famous university-university interactions, today, this development has changed in favor of the universities. Thus, when the German universities aim to get on a level with, for example, their US competitors, public policy measures should concentrate on stimulating those interactions, too. The KIT which was founded by a merger of Forschungszentrum Karlsruhe and Universität Karlsruhe or the foundation of the RWTH Campus GmbH76 which, amongst other, incorporates university and non-univeristy research as well as industry research are good examples how to shape the German knowledge-based future in order to be able to compete with famous institutions such as the MIT.

27Moreover, there is no doubt that the patenting activity of the German universities has increased over the past ten years, but it is still far away from industry patenting as well as from industry cooperation activity in terms of joint patents. Thus, regarding the innovative potential of the German universities regarding the number of filed patents and co-applicants, policy measures need to be created that explicitly affect these concerns. Against this background, it could be beneficial to foster and to improve transfer processes by establishing transfer models in order to increase the level of innovation among universities, enterprises and research institutes. In this context, it is further evident that third-party funds have increased the overall patenting activity of the German universities. Bearing this in mind, the logical conclusion is to provide policy measures that further stimulate collaborations between the broader research community and industry, in order to ensure valuable joint innovations. Here, it could be possible that missing frame conditions regarding property rights constitute an important issue as they are probably debilitating for joint patents. It is surely difficult for universities, research institutes and enterprises to agree on manifold questions concerning intellectual property rights. Therefore, a proper framework should be developed in this regard in order to avoid that inventions remain secretly, as an economy is only able to benefit from new knowledge if it is aware of it. Here, the aforementioned transfer models could be productive, too, amongst other.

  • 77 A traditional university can of course also be very active regarding joint patenting as an entrepr (...)

28In all, the SNA has already discovered that the German universities behave differently regarding their publication and patenting activity. As already discussed in the previous subchapter, it is the KIT, the University of Erlangen and the University of Wurzburg that have had most distinct co-applicants, while the Technical University of Munich, the LMU Munich and the University of Heidelberg have been mostly engaged in close network collaborations in terms of joint publications. Thus, it seems reasonable that some universities are more traditionally oriented than others who have rather turned to more entrepreneurial units.77 However, it is natural to explore why a given university is more engaged in either joint patenting or joint publications to be further able to provide the right policy choice. In this context, the separate analysis of the KIT and Heidelberg, as a technical compared to a medical university, has shown that the KIT has been more engaged in joint patents, while Heidelberg has been more likely to publish in cooperation. Hence, further studies should explore this phenomenon in order to provide each university with the right policy choice. Besides, it has been discovered that both in terms of publications and patents, the technical universities have been more likely to cooperate with industrial partners, while the medical universities have cooperated more often with knowledge-intensive institutions compared to their particular counterparts. Thus, it might be further worth to examine why they behave accordingly, aiming to simultaneously develop specific public policy measures that support newer forms of cooperations as public-private-partnerships (PPP) to join industry partners as well as the research community likewise. It is beyond question that diversification is especially worth regarding the emergence of new knowledge, technology and finally innovations.

29This PhD thesis has further demonstrated that proximity patterns still matter in an ever more globalised world as especially locality still plays a crucial role regarding the choice of the cooperation partners of the German universities. Against this background, public policy measures should of course further aim to support local cooperations, as the integration of various organizations within a particular region can bring about economic strength and can also attract the setting up of other foreign organizations. Thus, policy measures that strengthen a particular region and help to build up trust in this region may simultaneously stimulate national and international partnerships. Of course, there is no doubt that it is compulsive to foster knowledge-intensive cooperations within Germany or within a particular region of Germany; nevertheless, knowledge and experience from different countries or from different cultural regions may further stimulate the emergence of high-quality innovations that can be valuable for everyone. As especially linkages to supra-regionally-located enterprises have lost in importance over the past ten years, policy measures should of course not only foster domestic partnerships, but also consider global players and foreign markets in order to benefit from national as well as from international knowledge.

30It has been further discovered that the elite and the technical universities have more likely cooperated with supra-regional co-authors compared to their counterparts, and the medical universities have been more likely to cooperate with regionally located partners compared to the non-medical universities. These findings could have been also confirmed by the separate analysis of the KIT and Heidelberg. Hence, it is still prevalent that due to the important role of trust in medical knowledge collaborations, proximity patterns still highly matter in this regard. Further, in case the medical universities have cooperated with partners coming from abroad, they mainly trust in the well-known Western knowledge as they have been most likely to cooperate with partners coming from North America than from BRIC. Also in case of JANZ, the medical universities have more likely cooperated with North American partners. It might be worth offering policy measures that aim to deepen trust, even though the cooperation partner is far away or comes from an emerging country. There are certainly hidden potentials that can be further used to strengthen competitiveness and to bring up new knowledge and technologies. In this context, exploring whether the medical universities have been more likely to cooperate with partners coming from North America or from EU14, it has been shown that they have decided in favour of North America compared to their counterparts. Hence, it should be worked on better framework conditions in terms of inventive programmes at the European level, too, in order to strengthen Europe. Regarding the technical and non­technical universities, it is somewhat the other way round, as they have cooperated more often with BRIC than with EU14 or North America. Thus, it is generally beneficial to develop and provide comprehensive public policy measures that also affect at the European level and can be further followed within the European research frame programme Horizon 202078.

31In conclusion, it has been shown by this PhD thesis that the German universities are not only powerful with regard to their knowledge generation and innovation function but they are also highly involved in knowledge-intensive network interactions. Thus, as knowledge remains economic power, the German universities strongly contribute to the local, regional and national development of the German knowledge-based economy.

Notes

73 See chapter six for the determination of the distance classes.

74 See chapter six for the determination of the distance classes.

75 The empirical results have been only significant for the first time period in case of the medical and non-medical and elite and non-elite universities. The technical and non-technical universities have had still slight significant differences within the last time period.

76 A German ′Gesellschaft mit beschränkter Haftung (GmbH)′ is synonymous with the English Limited (Ltd). For further information on this topic, see http://www.rwth-http://aachen.de/cms/root/Wirtschaft/~ekt/Campusprojekt// 15.04.2013.

77 A traditional university can of course also be very active regarding joint patenting as an entrepreneurial university is also highly engaged in joint publications.

78 For further information on this issue see http://ec.europa.eu/research/horizon2020/index_en.cfm, 27.04.2013.

Table des illustrations

Légende Table 38: Summary of the empirical Results (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/242/img-1.jpg
Fichier image/jpeg, 89k
Légende Table 39: Summary of the empirical Results (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/242/img-2.jpg
Fichier image/jpeg, 109k
Légende Table 40: Summary of the empirical Results (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/242/img-3.jpg
Fichier image/jpeg, 91k
Légende Table 41: Summary of the empirical Results (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/242/img-4.jpg
Fichier image/jpeg, 283k

CC-BY-SA-4.0

Le texte seul est utilisable sous licence CC BY-SA 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Lire

Open access

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search