Incentives to innovate is a central element of innovation theory. In the private-investment model, innovators privately fund innovation and then use intellectual property protection mechanisms to appropriate returns from these investments. In the collective-action model, public subsidy funds public goods innovations, characterized by non-rivalry and non-exclusivity. Recently, these models have been compounded in the privatecollective innovation model where innovators privately fund public goods innovations (von Hippel and von Krogh, 2003). Private-collective innovation can be illustrated in the case of open source software development. The current paper contributes to the work on private-collective innovation by investigating incentives that motivate innovators to share their knowledge in an initial situation devoid of community activity. We use game theory to predict knowledge sharing behavior, and test these predictions in a laboratory setting. The results show that knowledge sharing is a coordination game with multiple equilibria, reflecting the fragility of knowledge sharing between innovators with conflicting interests. The experimental results demonstrate important asymmetries in the fragility of knowledge sharing and, in some situations, much more knowledge sharing than theoretically predicted. A behavioral analysis suggests that knowledge sharing is not only affected by the material incentives, but also by social preferences. The results offer general insights into the relationship between incentives and knowledge sharing and contribute to a better understanding of the inception of privatecollective innovation.
innovation private-collective innovation model knowledge sharing incentive open source software experimental economics game theory