Innovation Through Optimal Licensing in Free Markets and Free Software

We consider openness in private and socially optimal licenses under conditions where network effects and multiperiod innovation are both possible. For private firms, we model a variety of possible business models from completely closed to fully open, and find that opening a platform can increase profits based on network effects exclusively, innovation exclusively, or both. A firm's ability to control downstream innovation gives it reaon to rationally behave more like a social planner and even tolerate limited levels of piracy, interpreted as free user access. Further, open contracts with modest royalties offered to all developers can dominate closed Nash bargaining subcontracts with lead developers. We also find conditions when firms choose proprietary licenses despite innovation and network effects. In social planning terms, we find that optimal protection for reusable information is not arbitrarily long. Overlong protection interferes the inputs to downstream innovation. Further, licenses must enforce shorter-than-privately-optimal disclosure terms. Otherwise, a prisoner's dilemma in private incentives limits free access to derivative work, essential for decentralized innovation. In modeling terms, we add to the recent literature on two-sided network effects by incorporating a production function on one side of the market. We also contribute a framing innovation that places several existing license types in a space suggesting that socially optimal but unexplored licenses might exist