incentives for collaborative R&D improve SMEs’ ability to network with others

While policy interventions to support innovation have traditionally consisted in the provision of subsidies (including tax relief) to companies that invest in R&D, in recent years public incentives for collaborative innovation have become widespread. Through collaborative innovation schemes, companies receive public funds to engage in innovative projects together with other knowledge-intensive organisations such as universities, public and private research centres, and innovation intermediaries. These policies are usually evaluated for their effects on the participants’ economic and innovation performance. However, they might also enhance the participants’ ability to engage in further collaborative innovation – an effect whose importance should not be overlooked in a context where innovation processes are increasingly open and collaborative, but which has so far attracted little interest.

Two of our recent papers fill this gap in research by exploring empirically the effects of collaborative R&D policies on the participants’ networking activities.

In the first paper  (Caloffi, Mariani, Rossi and Russo, 2018), we analysed whether public subsidies supporting collaborative R&D projects were able to encourage persistent R&D investment and inter-organisational networking more than subsidies supporting individual R&D projects. We considered a matched sample of small and medium-sized enterprises (SMEs) that participated in two different policy interventions, one supporting collaborative R&D and one supporting individual R&D, both implemented in the same Italian region (Tuscany) in the same period.

We found that the two interventions had different effects. The subsidies for collaborative R&D increased the participants’ propensity to engage in further R&D activities, as well as their propensity to network with universities (but not with other companies), after the end of the policy. We also found that the networking effect was stronger for companies that did not have previous collaborative R&D experience, while the stimulus on R&D investment was stronger for companies that, prior to the policy, did not invest in R&D.

On the other hand, the subsidies to individual R&D projects led participants to increase their amount of R&D investment in the follow-up period, particularly for companies that, prior to the policy, were already R&D performers.

Our results suggest that different interventions are appropriate for different policy objectives. If the aim is to increase the number R&D performers, the most appropriate approach would be to target firms with modest R&D experience through an R&D collaboration policy. If the aim is to increase the total amount of R&D investment, the intervention should target SMEs that are already R&D performers, particularly with individual R&D subsidies. Finally, if the aim is to persistently increase networking, particularly with universities, the implementation of a R&D collaboration programme would likely bring some positive results.

In the second paper (Caloffi, Rossi and Russo, 2019), we untangled different possible network effects of collaborative innovation policies. We considered the following effects:

  1. persistence effect, which occurs when participants continue to collaborate with external organisations;
  2. breadth effect, which refers to an increase in the breadth of participants’ networks (companies create relationships with organisations with which they did not have any prior connections);
  3. composition effect, which occurs when companies change the type of organizations with which they collaborate; and,
  4. an intensification effect, which refers to a change in the intensity of collaborations.

The empirical analysis relied on a survey distributed to a sample of participants in a policy supporting collaborative R&D projects implemented in the Italian region of Tuscany, and a matched sample of companies that had applied to the scheme but were not funded.

We found that the policy generated network persistence particularly in terms of relationships with universities, and particularly for those treated companies that, prior to policy participation, were already used to collaborating with other organisations. In fact, for prior collaborators the probability to collaborate with universities increased by 27% (vs 14% for non collaborators), and the probability to collaborate with other companies increased by 22% (vs no effect for non collaborators).

The composition of the network changed too, as companies that, before the policy, did not have any collaborations with universities, began to create those links. On the other hand, we did not observe any significant effect on the increase in the number of external partners, nor on the intensification of innovation-related relationships.

Hence, this policy mainly supported the matching between SMEs and knowledge-intensive organizations such as universities or research centres.

Our results come from the analysis of a relatively small regional case study. Therefore, they should be corroborated by further empirical research conducted in other locations or regarding similar programmes of larger size before the last word is written on the topic. However, we believe that our contribution can stimulate further debate on whether, and for whom, subsidies to collaborative R&D are preferable to other, and maybe simpler, forms of public support to the innovative activity of SMEs.

The studies are presented in the following papers:

Annalisa Caloffi, Marco Mariani, Federica Rossi and Margherita Russo (2018) ‘A comparative evaluation of regional subsidies for collaborative and individual R&D in small and medium-sized enterprises’, Research Policy, 47(8), pp. 1437-1447.

Annalisa Caloffi, Federica Rossi and Margherita Russo (2019) ‘The Network Effects of Regional RD Collaboration Policy’, Scienze Regionali, 18(2), pp. 193-214.