Acemoglu, D. et all. (MIT). Experimentation, Patents, and Innovation.  2008.

...a simple patent system, where a copying fi…rm has to make a prespeci…fied payment to the innovator, can implement the optimal allocation. When the optimal allocation involves simultaneous experimentation, the patent system makes free-riding prohibitively costly and implements the optimal allocation as the unique equilibrium. When the optimal allocation involves staggered experimentation, the patent system plays a more subtle role. It permits ex post transmission of knowledge but still increases experimentation incentives to avoid delays.

The patent system can achieve this because it generates "conditional" transfers. An innovator receives a patent payment only when copied by other …firms. Consequently, patents encourage one fi…rm to experiment earlier than others, thus achieving rapid experimentation without sacri…ficing useful transfer of knowledge. Moreover, we show that patents can achieve this outcome in all equilibria. The fact that patents are particularly well designed to play this role is also highlighted by our result that while an appropriately-designed patent implements the optimal allocation in all equilibria, subsidies to experimentation, research, or innovation cannot achieve the same objective.

Is Open Source approach consistent with the proposed model? For example, when free-riding is not a problem, do we need to impose a "transfer" fee?