Abstract
This study focuses on integration processes in European R&D by analyzing the spatio-temporal dimension of two different R&D collaboration networks across Europe. These networks cover different types of knowledge creation, namely co-patent networks and project based R&D networks within the EU Framework Programmes (FPs). Integration in European R&D – one of the main pillars of the EU Science Technology and Innovation (STI) policy – refers to the harmonisation of fragmented national research systems across Europe and to the free movement of knowledge and researchers. The objective is to describe and compare spatio-temporal patterns at a regional level, and to estimate the evolution of separation effects over the time period 1999–2006 that influence the probability of cross-region collaborations in the distinct networks under consideration. The study adopts a spatial interaction modeling perspective, econometrically specifying a panel generalized linear model relationship, taking into account spatial autocorrelation among flows by using Eigenfunction spatial filtering methods. The results show that geographical factors are a lower hurdle for R&D collaborations in FP networks than in co-patent networks. Further it is shown that the geographical dynamics of progress towards more integration is higher in the FP network.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The literature on R&D networks underlines the crucial importance of cooperative agreements between universities, companies and governmental institutes, for developing and integrating new knowledge in the innovation process (see Powell and Grodal 2005). This is explained by considerations that innovation nowadays takes place in an environment characterised by uncertainty, increasing complexity and rapidly changing demand patterns in a globalised economy. Organisations must collaborate more actively and more purposefully with each other in order to cope with increasing market pressures in a globalizing world, new technologies and changing patterns of demand. In particular, firms have expanded their knowledge bases into a wider range of technologies (Granstand 1998), which increases the need for more different types of knowledge, so firms must learn how to integrate new knowledge into existing products or production processes. It may be difficult to develop this knowledge alone or acquire it via the market. Thus, firms form different kinds of co-operative arrangements with other firms, universities or research organisations that already have this knowledge to get faster access to it.
- 2.
The theory of endogenous growth, and the geography-growth synthesis both consider that economic growth and spatial concentration of economic activities emanate from localised knowledge diffusion processes, in particular transferred via network arrangements between different actors of the innovation system.
- 3.
Hoekman et al. (2010) and Scherngell and Lata (2013) investigate the ongoing process of European integration by determining the impact of geographical distance and territorial borders on the probability of research collaborations between European regions. By analysing co-publication and FP network patterns and trends, the authors show that geographical distance has a negative effect on co-publication activities and FP cooperation, while for the FP networks this effect decreases over time. The study of Maggioni and Uberti (2009) focuses on the structure of knowledge flows by analysing four distinct collaboration networks, including co-patenting. Hoekman et al. (2013) focus on the effect of participation in FP networks on subsequent international publications, showing that the FPs indeed positively influence international co-publications, and, by this, seem to enhance integration across Europeans research systems.
- 4.
Since their introduction in 1984, different thematic aspects and issues of the European scientific landscape have been addressed by the FPs. Although the FPs have undergone different changes in their orientation during the past years, their fundamental rational remained unchanged (Roediger-Schluga and Barber 2006).
- 5.
Although substantial size differences and interregional disparities of some regions exist, these units are widely recognized to be an appropriate level for modelling and analysis purposes (see, for example, LeSage et al. 2007).
- 6.
Note that we do not distinguish between the FP network and the co-patent in the formal description of data as well as the modelling approach in the section that follows.
- 7.
We use a full counting procedure for the construction of our collaboration matrices (see, for example, Katz 1994). For a project with, for example, three different participating organizations a, b and c, which are located in three different regions, we count three links (from a to b, from b to c and from a to c).
- 8.
From a theoretical perspective the spatial autocorrelation of R&D collaboration flows may be explained by the assumption that the collaboration behaviour of one region influences the collaboration behaviour of neighbouring regions because – as described in various empirical studies – contiguity of regions may induce knowledge flows between them, to them, and from them, and, thus, evoke the transfer of information on potential collaboration partners that are located further away (Scherngell and Lata 2013). To give an example, if region A has many collaborations with region B (that is no neighbour of region A), region A may influence a neighbouring region C also to collaborate with region B due to information flows between region A and region C, in particular flows of ‘know who’ type information (see Cohen and Levinthal 1990).
- 9.
Spatial interaction models are widely used for modelling origin-destination flows data and were used to explain different kinds of flows, such as migration, transport or communication flows, between discrete units in geographical space (see, for instance, Fischer and LeSage 2010 among many others).
- 10.
One way to capture spatial autocorrelation of flows is the use of spatial autoregressive techniques (LeSage and Pace 2008). An alternative approach is the use of spatial filtering methods. The key advantage of the spatial filtering approach is that it can be applied to any functional form and thus, does not depend on normality assumptions (Patuelli et al. 2011). Consequently, we prefer the spatial filtering approach over spatial autoregressive model as we are dealing with a Poisson spatial interaction framework.
- 11.
The extracted eigenvectors have several characteristics. First, as shown by Griffith (2003), each extracted eigenvector relates to a distinct map pattern that has a certain degree of spatial autocorrelation. Second, the selected eigenvectors are centered at zero due to the pre and post multiplication of W by the standard projection Matrix \( \left(\boldsymbol{I}-\boldsymbol{1}\;{\boldsymbol{1}}^T{\scriptscriptstyle \frac{1}{n}}\right) \). Third, the modification of W ensures that the eigenvectors provide mutually orthogonal and uncorrelated map patterns ranging from the highest possible degree of positive spatial correlation to highest possible degree of negative spatial correlation as given by the Moran’s I (MI). (Griffith 2003). Hence, the first extracted eigenvector is the one showing the highest degree of positive spatial autocorrelation that that can be achieved by any spatial recombination; the second eigenvector has the largest achievable degree of spatial autocorrelation by any set that is uncorrelated with until the last extracted eigenvector will maximize negative spatial autocorrelation (Griffith 2003).
- 12.
We use an time invariant specification of the spatial filter as we assume an time invariant underlying spatial process.
- 13.
In order to determinate changes of our separation variables we include interaction terms (see, for an overview, Wooldridge 2008). In this procedure, variables of interest, for example R&D (see, Griliches 1984), interact with time dummy variables and illustrate if effects changed over a certain time period or not. In our case (time) interaction terms represent the interaction between our separation variables and the time dummies and determinate how separation effects have changed over time. These interaction terms pick up the inter-temporal variation of our separation effect and remain only cross-sectional variation.
- 14.
Note further that according to Bröcker (1989), we calculate the intraregional distance as s (1) ii = (2/3) (A i /π)0.5, where A i denotes the area of region i, i.e. the intraregional distance is two third the radius of an presumed circular area.
- 15.
Language areas are defined by the region’s dominant language. However, in most cases the language areas are combined countries, as for instance Austria, Germany and Switzerland (one exception is Belgium, where the French speaking regions are separated from the Flemish speaking regions).
- 16.
The dispersion parameter is statistically significant in both model versions, indicating that the Negative Binomial specification is essential to account for overdispersion in the data. A likelihood ratio test which compares the panel estimator with the pooled estimator confirms the appropriateness of the random effects specification.
References
Baltagi B (2008) Econometric analysis of panel data, 4th edn. Wiley, Chichester
Barber MJ, Scherngell T (2011) Is the European R&D network homogeneous? Distinguishing relevant network communities using graph theoretic and spatial interaction modelling approaches. Reg Stud 46:1–16
Berglund S, Karlström A (1999) Identifying local spatial association in flow data. J Geogr Syst 1(3):219–236
Boyer R (2009) From the Lisbon agenda to the Lisbon treaty: national research systems in the context of European integration and globalization. In: Delanghe H, Muldur U, Soete L (eds) European science and technology policy. Towards integration or fragmentation? Edward Elgar, Cheltenham, pp 101–126
Breschi S, Cusmano L (2004) Unveiling the texture of a European research area: emergence of oligarchic networks under EU framework programmes. Int J Technol Manag 27(8):747–772, Special Issue on Technology Alliances
Bröcker J (1989) Partial equilibrium theory of interregional trade and the gravity model. Pap Reg Sci 66:7–18
Brunet R (2002) Lignes de force de l’espace Européen. Mappemondo 66:14–19
Cameron AC, Trivedi PK (1998) Regression analysis of count data. Cambridge University Press, Cambridge/New York
Castells M (1996) The rise of the network society. Blackwell, Malden
CEC (Commission of the European Communities) (2007) Green paper the European research area: new perspectives. {SEC(2007) 412}, COM(2007)161 final, Brussels, 4 April 2007
CEU (Commission of the European Union) (2008) Council conclusions on the definition of a “2020 vision for the European research area”. 16012/08 RECH 379 COMPET 502, Brussels, 9 Dec 2008
Chun Y (2008) Modeling network autocorrelation within migration flows by eigenvector spatial filtering. J Geogr Syst 10:317–344
Cohen WM, Levinthal DA (1990) Absorptive capacity: a new perspective on learning and innovation. Adm Sci Q 35(1):128–152
Delanghe H, Muldur U, Soete L (eds) (2009) European science and technology policy. Towards integration or fragmentation? Edward Elgar, Cheltenham
Ejermo O, Karlsson C (2006) Interregional inventor networks as studied by patent coinventorships. Res Policy 35(3):412–430
European Council (2000) Presidency conclusions, Lisbon European Council, 23–24 March 2000
European Council (2010) Conclusions on innovation union for Europe. Brussels, 26 Nov
European Commission (2011) Progress report on the Europe 2020 strategy. Brussels, European Commission, Com (2011) 815 final
Fischer MM, Griffith D (2008) Modeling spatial autocorrelation in spatial interaction data: an application to patent citation data in the European Union. J Reg Sci 48:969–989
Fischer MM, LeSage J (2010) Spatial econometric methods for modeling origin destination flows. In: Fischer Manfred M, Getis A (eds) Handbook of applied spatial analysis. Springer, Berlin/Heidelberg, pp 409–432
Fischer MM, Scherngell T, Jansenberger E (2006) The geography of knowledge spillovers between high-technology firms in Europe. Evidence from a spatial interaction modelling perspective. Geogr Anal 38:288–309
Granstand O (1998) Towards a theory of the technology-based firm. Res Policy 27:465–489
Griffith D (2003) Spatial autocorrelation and spatial filtering: gaining understanding through theory and scientific visualization. Springer, Berlin Heidelberg
Griliches Z (1984) R&D, Patents, and Productivity, NBER Books, National Bureau of Economic Research. Chicago
Guzzetti L (2009) The European research area idea in the history of community policy-making. In: Delanghe H, Muldur U, Soete L (eds) European science and technology policy. Towards integration or fragmentation? Edward Elgar, Cheltenham, pp 64–77
Hoekman J, Frenken K, Tijssen RJW (2010) Research collaboration at a distance: changing spatial patterns of scientific collaboration in Europe. Res Policy 39:662–673
Hoekman J, Scherngell T, Frenken K, Tijssen R (2013) Acquisition of European research funds and its effect on international scientific collaboration. J Econ Geogr 13:23–52
Jaffe AB, Trajtenberg M, Henderson R (1993) Geographic localization of knowledge spillovers as evidenced by patent citations. Q J Econ 108(3):577–598
Katz JS (1994) Geographical proximity and scientific collaboration. Scientometrics 31:31–43
LeSage J, Pace K (2008) Spatial econometric modeling of origin-destination flows. J Reg Sci 48:941–967
LeSage J, Fischer MM, Scherngell T (2007) Knowledge spillovers across Europe. Evidence from a Poisson spatial interaction model with spatial effects. Pap Reg Sci 86:393–421
Maggioni MA, Uberti TE (2009) Knowledge networks across Europe: which distance matters? Ann Reg Sci 43:691–720
Moreno R, Paci R, Usai S (2005) Spatial spillovers and innovation activity in European regions. Environ Plann A 37:1793–1812
Patuelli R, Griffith D, Tiefelsdorf M, Nijkamp P (2011) Spatial filtering and eigenvector stability: space-time models for German unemployment data. Int Reg Sci Rev 34(2):253–280
Polt W, Vonortas N, Fisher R (2008) Innovation impact, Final report to the European commission, Brussels, DG Research
Powell WW, Grodal S (2005) Networks of innovators. In: Fagerberg J, Mowery DC, Nelson RR (eds) The Oxford handbook of innovation. Oxford University Press, Oxford, pp 56–85
Roediger-Schluga T, Barber M (2006) The structure of R&D collaboration networks in the European Framework Programmes. Unu-MERIT Working Paper Series 2006-36, Economic and Social Research and Training Centre on Innovation and Technology, United Nations University, Maastricht
Romer PM (1990) Endogenous technological change. J Polit Econ 98:71–102
Scherngell T, Barber MJ (2009) Spatial interaction modelling of cross-region R&D collaborations. Empirical evidence from the 5th EU framework programme. Pap Reg Sci 88:531–645
Scherngell T, Barber MJ (2011) Distinct spatial characteristics of industrial and public research collaborations: evidence from the 5th EU framework programme. Ann Reg Sci 46:247–266
Scherngell T, Lata R (2013) Towards an integrated European research area? Findings from eigenvector spatially filtered spatial interaction models using European framework programme data. Paper Reg Sci 92:555–577
Svanfeldt C (2009) A European research area built by the member states? In: Delanghe H, Muldur U, Soete L (eds) European science and technology policy: towards integration or fragmentation? Edward Elgar, Cheltenham/Northampton, pp 44–63
Wooldridge JM (2008) Econometric analysis of cross section and panel data. MIT Press, Cambridge
Acknowledgments
This work has been funded by the FWF Austrian Science Fund (Project No. P21450). We are grateful to Manfred M. Fischer (Vienna University of Economics) and Michael Barber (AIT) for valuable comments on an earlier version of the manuscript.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Lata, R., Scherngell, T., Brenner, T. (2013). Observing Integration Processes in European R&D Networks: A Comparative Spatial Interaction Approach Using Project Based R&D Networks and Co-patent Networks. In: Scherngell, T. (eds) The Geography of Networks and R&D Collaborations. Advances in Spatial Science. Springer, Cham. https://doi.org/10.1007/978-3-319-02699-2_8
Download citation
DOI: https://doi.org/10.1007/978-3-319-02699-2_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-02698-5
Online ISBN: 978-3-319-02699-2
eBook Packages: Business and EconomicsEconomics and Finance (R0)