Few large pharmaceutical companies have recently discovered out-licensing of terminated R&D results as a way to recoup some of the significant investments made in R&D and to improve R&D productivity. Our empirical investigation reveals that the licensing partners are preferably young, small and highly specialized companies. This reverses the traditional logic of out-licensing. While out-licensing is usually done because of downstream concerns, our analysis shows that the company which owns the necessary assets for further development (the large pharmaceutical company) sells the license to a firm (the small partner company) which has — at the time of deal closure — no track record to prove its ability to successfully develop the compound. As the lack of a track record does not allow the pharmaceutical company to distinguish between the partner firms based on their development capabilities, these out-licensing deals are characterized by an asymmetric distribution of information. The application of the theory of adverse selection allows deriving managerial recommendations along three dimensions of the out-licensing deal: product coverage, price setting and performance presumption. By making changes along these dimensions, R&D managers are able to reduce the information asymmetry and approximate an equilibrium in the out-licensing market.