International R&D cooperation within the EU Framework Programme: empirical evidence for Spanish firms

Abstract

This work aims to explain which factors determine the participation of Spanish companies in R&D consortia within the EU's Framework Programme (FP), considering that the probability that a firm takes part must be regarded as the result of a two-phase process. First, the firm decides whether to apply for an FP cooperation project or not and secondly, the project is approved or rejected. The empirical results confirm that the probability of application depends on some firm-specific variables such as its commercial presence in foreign markets, its absorptive capacity regarding incoming spillovers and its prior experience in FP proposals. With regard to the proposal award, variables related to coordination capability (leader capacity, geographical distance among partners and prior experience in R&D cooperation) are valuable determinants.

Keywords:

• international R&D cooperation
• Framework Programme
• coordination capability
• Spanish firms

JEL Classification :

• O32
• O38

Acknowledgements

The authors are particularly grateful to the CDTI for providing useful comments and access to databases. Comments by the participants in the Concord 2007 Conference, especially by two anonymous referees and Franco Malerba, have particularly improved the final draft of this paper. This research has been partially funded by the CICYT project SEJ2004-02525/ECON. Elena Huergo also acknowledges financial support from the CDTI.

Notes

http://cordis.europa.eu/.

However, cost reduction and access to public support are considered to be less important.

In the case of Spain, the CDTI offers financial support for elaborating a proposal.

Notice that more than one Spanish firm can participate in the same proposal. The relevant variable in this case is the proposal.

To guarantee the homogeneity of the sample, only specific targeted research projects (STREPs) and integrated projects are considered. The present work does not consider the specific instruments designed for SMEs (CRAFT). Nevertheless, it should be noted that the STREPs and the integrated projects accounted for 74% of the total aid received by Spanish SMEs in FP6.

This period covers part of FP4 (1994–1998), all of FP5 (1999–2002) and part of FP6 (2002–2006).

The Survey on Business Strategies is an example of statistical research that uses this criterion. See http://www.funep.es/esee/ing/i_esee1.asp.

Appendix 1 gives the correspondence of activities and the resulting number of observations in the sample for each two-digit NACE class.

There is no information in the database about R&D intensity or other measures of the company's technological performance, such as innovations or patents.

The exact definitions of the variables are given in Appendix 2.

Initially, the whole set of industry dummies corresponding to the NACE two-digit classification (Appendix 1) was incorporated in the specification, with the exception of agriculture, hunting and related service activities. However, the estimations in column (2) correspond to the model in which, apart from high- and medium-high manufacturing and services sectors, only the dummies for industries with statistically significant coefficients are included.

Alternatively, we have estimated the specification as a random-effects probit model, which takes into account that the same Spanish firm can participate in different proposals during the period. The assumption behind this second estimation is that the composition of the consortia in which the same Spanish firm participates is also likely to be more constant. Therefore, we would be controlling for the presence of consortium effects. However, in both estimates, the results are nearly identical, suggesting that this correction is not very important in our sample.

We have also tentatively introduced the total budget of the project but it was ultimately dropped because of the high correlation between the two variables.