Discussion Paper

No. 2009-28 | June 12, 2009
Endogenous Technology Sharing in R&D Intensive Industries


This paper analyses the endogenous formation of technology sharing coalitions with asymmetric firms. Coalition partners produce complementary technology advancements, although each firm determines its R&D investment level non-cooperatively and there is no co-operation in the product market. We show that the equilibrium coalition outcome is either one between the two most efficient firms, or a coalition with all three firms. The two-firm coalition is the preferred outcome of a welfare maximising authority if ex ante marginal cost is sufficiently high, and the three-firm coalition is preferred otherwise. Furthermore, we show that the equilibrium outcomes result in the lowest total R&D investment of all possible outcomes. Aircraft engine manufacturing provides a case study, and indicates the importance of anti-trust issues as an addition to the theory.

Paper submitted to the special issue
The Knowledge-Based Society: Transition, Geography, and Competition Policy

JEL Classification

L11 L13

Cite As

Derek J. Clark and Jan Yngve Sand (2009). Endogenous Technology Sharing in R&D Intensive Industries. Economics Discussion Papers, No 2009-28, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2009-28


Comments and Questions

Anonymous - Comment
June 15, 2009 - 23:55


Derek J. Clark and Jan Yngve Sand (2009). Endogenous Technology Sharing in R&D Intensive Industries. Economics Discussion Papers, No 2009-28.

As the authors suggest, this paper fills a gap in the existing literature: The issue of endogenous formation of technological innovation/sharing coalitions among ex ante ...[more]

... technologically asymmetric Cournot competitors in a homogeneous product market.
Generally speaking, it is easy but often uninteresting to criticize game-theoretic IO models on the basis of the restrictiveness of their assumptions. At the same time, adopting particularly special assumptions and unusual modeling strategies calls for a more careful motivation and justification in order for an IO paper to claim any generality and robustness of its results. For this to be done, as I will suggest below, knowledge of the relevant literature and discussion of specific peculiar assumptions in reference to a number of recent related articles is necessary and usually sufficient.
In this brief note, I would like to comment on the very particular way in which the issue of shared technological innovation is modeled and its implications for the resulting “cooperative solution” obtained. In the present manuscript, two R&D-cooperating firms benefit so much from each other’s R&D results, that their individual costs are reduced BY THE SUM of effects of their individual R&D investments. This is interpreted as perfect R&D complementarity, perfect spillovers, or R&D cooperation. Surprisingly, the authors make no effort to place this assumption within a growing recent literature on endogenous R&D-spillover absorptive capacity (Wiethaus, 2005), endogenous technology choice and thus endogenous spillovers (Gil-Molto, 2005) and most recently endogenous, non-cooperative (NON-) protection of R&D investments (Miliou, 2009). Especially with respect to this last strand of findings, it would be reasonable to ask whether the present manuscript’s positive findings on cooperation would hold robust if the firms were allowed to “open” their R&D secrets in a non cooperative way (in an non cooperative game involving the current assumption as one of its points) rather than as the result of an exogenously imposed type of cooperation contract imposing the extreme case of perfect sharing. Totally independently from this last comment, I am surprised that there is no reference (not even a simple comment) to a very plausible way of sharing technology information, which would be that both sharing partners start producing (after R&D investments reduce costs) their homogeneous product at the lowest cost resulting among all productive plants involved. In that sense, and following textbook knowledge on technology as a publicly available know-how for the most efficient production, summing the results of independent R&D efforts looks even more awkward.
Let us now move to the cooperative solution obtained under this particular R&D sharing assumption. It is usual in the literature that cooperative solutions with respect to a specific stage of a multistage game are obtained assuming maximization of the partners’ joint profit function with respect to a specific strategic variable. I do not say that the formation of R&D-related information sharing coalitions is not an interesting issue, but in all the R&D literature (I refrain from giving the large list of papers doing or suggesting so) Research Joint Ventures (RJV) have been seen as a way of coordinating R&D expenditures (apart from technology sharing environments), in order to avoid socially and privately undesirable overspending on R&D effort. A natural consequence is that partners in an RJV jointly decide on their common R&D expenditures (giving rise to an interesting issue of cooperatively deciding who should contribute what) as opposed to non cooperative investments. This certainly introduces TWO different LEVELS of cooperation which should be interesting to study jointly and separately from each other in order to assess the effects of RJVs under different scope definitions: FIRST, cooperatively setting R&D expenditures; SECOND, sharing information on R&D results. Apart from a limitation of the current analysis, this comment aims at pointing to a potentially very important extension of the model presented in this manuscript.


1. Chrysovalantou Milliou, 2009. "Endogenous protection of R&D investments," Canadian Journal of Economics, Canadian Economics Association, vol. 42(1), pages 184-205.

2. María José Gil Moltó & Nikolaos Georgantzís & Vicente Orts, 2005. "Cooperative R&D with Endogenous Technology Differentiation," Journal of Economics & Management Strategy, Blackwell Publishing, vol. 14(2), pages 461-476, 06.

3. Wiethaus, Lars, 2005. "Absorptive capacity and connectedness: Why competing firms also adopt identical R&D approaches," International Journal of Industrial Organization, Elsevier, vol. 23(5-6), pages 467-481.

Jan Sand - Reply to "Comments"
June 19, 2009 - 15:14

Please find a reply to "Comments" by anonymous author enclosed.

Anonymous - Referee Report
July 31, 2009 - 09:47

see attached file

Jan Sand - Response to the referee report
August 29, 2009 - 21:44

Attached is our response to the referee's comments.

Jan Sand - Revised version of the paper
August 29, 2009 - 21:45

Attached is the revised version of the paper.

Anonymous - Referee Report 2
October 26, 2009 - 10:06

see attached file

Jan Sand - Response to referee report 2
November 13, 2009 - 09:12

see attached file