Obtaining R&D Joint Venture Co-operation Under Prisoners’ Dilemma Incentives: Logic and Experiment RICHARD AREND, University of Nevada A subset of all R&D joint ventures is characterized by actions and payoffs most similar to a Prisoners’ Dilemma game. Firms may cooperate or defect in the venture. When it is unfeasible to obtain inde- pendent verification of the venture inputs and out- come, then the dominant solution to the Prisoners’ Dilemma – mutual defection – results. This paper proposes the use of a new revelation mechanism that truthfully determines whether defection has occurred in order to trigger penalties. Cooperation results. A simplified version of the solution is tested experimentally and shows a significant improvement in cooperation level results. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Joint venture, R&D, Prisoners’ dile- mma, Revelation mechanism Most research and development joint ventures (R&D JVs) fail, or simply perform poorly (e.g., Park and Russo, 1996; Parkhe, 1993; Ernst and Bleeke, 1993; Le- vine and Byrne, 1986). However, this is a similar fail- ure rate to alternative strategic actions aimed at creating new technology. New ventures, internal ventures, and acquisitions have high failure rates as well (e.g., Timmons, 1999; Block and MacMillan, 1993; Hunt, 1990; respectively). A substantial possi- ble upside, such as the creation of a valuable patent, is balanced by the uncertainties and economic haz- ards involved in each of these strategic actions. Firms continue to pursue all of these actions when the balance is tipped in favor of positive net expected rewards regardless of whether any individual pro- ject’s probable outcome is a loss. One aspect that differentiates the R&D JV from the alternative actions is the distinctive structure of control that the partners share. Each partner controls the quality of some in- puts to the project, splitting the benefits of any result- ing output. In some instances, these ventures play like a Prisoners’ Dilemma (PD), where each partner has no individual incentive to put forth its best qual- ity inputs. The output quality subsequently suffers, with failure of the venture being a possible if not probable outcome. This paper presents a new method for improving the efficiency of some R&D JVs, specifically those that have similarity to a Prisoners’ Dilemma in actions and payoffs. This paper proposes the use of a mech- anism that truthfully reveals whether defection – a failure to provide the highest quality inputs to the project – has occurred. Defection is revealed in order to issue appropriate penalties. The penalties are made credible through a third-party, one that im- poses penalties in the form of fines, some of which it keeps. When the firms play the venture strategies under such policing, they have an incentive to play cooperatively. When firms cooperate, the jointly opti- mal outcome of the R&D JV is achieved, and each partner’s performance improves. In this paper, a joint venture is a partnership formed among independent firms in order to access comple- mentary resources and capabilities with goals includ- ing the sharing of risks and costs, and the creation of 520 European Management Journal Vol. 23, No. 5, pp. 520–532, October 2005 doi:10.1016/j.emj.2005.09.007 European Management Journal Vol. 23, No. 5, pp. 520–532, 2005 Ó 2005 Elsevier Ltd. All rights reserved. Printed in Great Britain 0263-2373 $30.00