The determinants of R&D cooperation

Ample empirical research and examples exist covering the incentives of engaging in R&D cooperation. Using Microelectronics and Computer Technology Corporation as a case, Gibson and Rogers (1994) summarize the motivations to form R&D consortia, including the following: efficiencies of shared cost and risk, exploration of new concepts, pooling scarce talent, sharing research or manufacturing facilities, desire for research synergy, diversification of a technology portfolio, developing frameworks into which other technology modules or tools can fit, setting standards, marketing products, pre-competitive sharing of research results, industrial organization and accelerated technology development, big science and large projects, infrastructure development, and facilitating technology transfer or partnering, whether domestic or foreign. Using a database of European research joint ventures (RJVs),

Hernan, Marin, and Siotis (2003) find that R&D intensity, industry concentration, firm size, technological spillovers, and post-RJV participation all positively influence the probability of forming RJVs. Belderbos, Carree, Diederen, Lokshin, and Veugelers (2004) explore the determinants of innovating firms’ decisions to engage in R&D cooperation. They observe that the determinants of R&D cooperation differ significantly across cooperation types. The positive impacts of firm size, R&D intensity, and incoming spillovers are weaker for competitor cooperation. Based on German manufacturing enterprises, Fritsch and Lukas (2001) analyze the propensity to maintain different forms of R&D cooperation with customers, suppliers, competitors and public research institutions. According to their results, enterprises that maintain R&D cooperation relationships tend to be relatively large and have a high share of R&D.

R&D cooperation can be considered to restore private incentives, because of internalizing the knowledge spillovers between cooperating firms (e.g. D’Aspremont and Jacquemin 1988; Kamien et al. 1992; Ishii 2004). Peters and Beck (1997-98) analyze the role of knowledge spillovers between automakers and suppliers in vertical corporate networks, both theoretically and empirically. In the empirical results they find evidence for the importance and effects of the transfer of technological information between manufacturers and their suppliers in the R&D process to develop and produce a custom-tailored good. Cassiman and Veugelers (2002) empirically explore the effects of knowledge flows on R&D cooperation. They discover that there is a significant relation between external information flows and the decision to cooperate in R&D. Firms that generally rate available external information sources as more important inputs to their innovation process are more likely to be actively engaged in cooperative R&D agreements. At the same time, firms that are more effective in appropriating the results from their innovation process are also more likely to cooperate in R&D. Kaiser (2002a) uses innovation survey data from the German service sector to explore research expenditures and research cooperation. The main results show that RJVs are more widespread among vertically-related firms than among horizontally-related firms. An increase in horizontal spillovers tends to increase incentives to collaborate in R&D. In addition, R&D efforts are larger under RJV than under R&D competition with a sufficiently large spillover. Sakakibara and Dodgson (2003) evaluate the role that strategic research partnerships (SRPs) play in

Asia and conclude that SRPs are formed to facilitate technological diffusion in Taiwan.

Milliou (2004) analyzes the impact of R&D information flow on the incentives of innovation and social welfare under vertical integration. His results show that R&D information flow has a positive impact on innovation, outputs, and profits for R&D-integrated firms, but a negative impact for the R&D non-integrated firm.

Absorptive capacity and uncertainty¹⁰ are also deemed as crucial factors that influence the decision of R&D cooperation. Bayona, Garcia-Marco, and Huerta (2001) test firms’ motivations for cooperative R&D using Spanish firms that carried out R&D activities. The results obtained therein suggest that firms’ motivations for cooperative R&D include technology’s complexity and the fact that innovation is costly and uncertain. To undertake cooperative R&D, it is necessary to have certain internal capacities in this area. Becker and Dietz (2004) investigate the role of R&D cooperation in the innovation process. The results suggest that joint R&D is used to complement internal resources in the innovation process, enhancing the innovation input and output. The intensity of in-house R&D (absorptive capacity) also significantly stimulates the probability and the number of joint R&D activities with other firms and institutions. Caloghirou et al. (2003) investigate partnership performance and find that partnership success depends on the closeness of the cooperative research to the in-house R&D efforts of the firm, as well as on the firm’s effort to learn from the partnership and its partners. Sakakibara (2002) investigates economic and strategic incentives of R&D cooperation. She finds that a firm’s R&D capabilities, network formation through past consortia, encounters with other firms in the product market, age, and past participation in large scale consortia also positively affect its tendency toward consortia formation. Corey (1997) indicates that the risk-reduction opportunities provide an incentive for collaboration on large-scale projects with a relatively high degree of uncertainty. Another form of risk reduction that a collaborative venture provides is the opportunity to monitor technological advances in competitors’ R&D programs. Caloghirou et al. (2003) also find that firms use partnerships as vehicles of risk and uncertain reduction by collaborating with competitors as well as with suppliers and buyers.

Mathews and Cho (2000) indicate the importance of collaborative research

relationships for the development of the industry in Taiwan. However, very little research has focused on R&D cooperation activity because of data availability. In this study I provide a comprehensive analysis of R&D cooperation within Taiwan’s high-technology industries. Furthermore, a common feature in the prior R&D cooperation literature is the absence of uncertainty. Therefore, in addition to the factors used in Sakakibara’s research, I discuss the relationship between uncertainty and R&D cooperation.

See Table 5 for a literature summary of the determinants of R&D cooperation.

Table 5: Literature summary of the determinants of R&D cooperation

Author/Year Research topic Research method Research conclusion and research implication Atallah (2002) Author studies vertical

R&D spillovers between upstream and downstream firms.

1. Analytical research.

2. The model includes two vertically related industries, with horizontal spillovers within each industry and vertical spillovers between the two industries.

Research conclusion:

Author finds that vertical spillovers affect R&D investments directly and indirectly through their influence on the impact of horizontal spillovers and R&D cooperation. In addition, no matter what type the cooperation is, vertical spillovers always increase R&D efforts and welfare.

Research implication:

Based on the two vertical industry model, Atallah (2002) includes four R&D scenarios: R&D competition, vertical R&D cooperation, horizontal R&D cooperation, and

generalized R&D cooperation. In this study I adopt his R&D cooperation scenarios in my theoretical model.

Cassiman and Veugelers (2002)

Authors empirically explore the effects of knowledge flows

1. Empirical research.

2. The data are drawn from the Community

Research conclusion:

They discover that there is a significant relation between external information flows and the decision to cooperate in

Author/Year Research topic Research method Research conclusion and research implication (knowledge spillovers) on

R&D cooperation.

Innovation Survey (CIS) conducted in Belgian

manufacturing firms in 1993.

R&D. Firms that generally rate available external information sources as more important inputs to their innovation process are more likely to be actively engaged in cooperative R&D agreements. At the same time, firms that are more effective in appropriating the results from their innovation process are also more likely to cooperate in R&D.

Research implication:

Authors use survey data to explore the effects of knowledge flows on R&D cooperation and suggest that incoming

spillovers and appropriability have important effects on R&D cooperation. In this study I will use archival data to test the determinants of R&D cooperation.

Kaiser (2002a) Author uses innovation survey data from the German service sector to explore research

expenditures and research

1. Empirical research.

2. The empirical analysis is based on the survey data of the Mannheim

Research conclusion:

The main results show that RJVs are more widespread among vertically-related than horizontally-related firms. An increase in horizontal spillovers tends to increase incentives to

collaborate in R&D. In addition, R&D efforts are larger under

Author/Year Research topic Research method Research conclusion and research implication

cooperation. Innovation Panel in

the Service Sector (MIP-S).

RJV than under R&D competition with a sufficiently large spillover.

Research implication:

The types of R&D cooperation of my study are the same with those of Kaiser’s research. This makes my research results are more comparable with Kaiser (2002a)’s.

Sakakibara (2002)

Author investigates economic and strategic incentives of R&D cooperation and focus on factors that affect a firm’s rate of participation in R&D consortia.

1. Empirical research.

2. Research sample includes 312 Japanese firms in 74 industries between 1969 and 1992.

Research conclusion:

She finds that a firm with weak competition and higher

spillover has a higher rate of R&D cooperation. A firm’s R&D capabilities, network formation through past consortia,

encounters with other firms in the product market, age, and past participation in large scale consortia also positively affect its tendency at consortia formation.

Research implication:

A common feature in the prior R&D cooperation literature is the absence of uncertainty. Therefore, in addition to the

Author/Year Research topic Research method Research conclusion and research implication factors used in Sakakibara’s research, I discuss the relationship between uncertainty and R&D cooperation.

Sakakibara and Dodgson (2003)

Authors evaluate the role that strategic research partnerships (SRPs) play in Asia.

1. Descriptive research.

2. Asian countries include Korea, Japan, and Taiwan.

Research conclusion:

Authors indicate that the networks created among small Taiwanese firms through their research links with research organizations and conclude that SRPs are formed to facilitate technological diffusion in Taiwan.

Research implication:

Mathews and Cho (2000) indicate the importance of

collaborative research relationships for the development of the industry in Taiwan. However, very little research has focus on R&D cooperation activity because of data availability. In this study I provide a comprehensive analysis of R&D cooperation on Taiwan’s high-technology industries.

2.3 The relationship between R&D cooperation, R&D investments, R&D