Orphan (rare) diseases occur infrequently in the population, thus providing limited incentives for drug development. However, rare diseases affect a non-negligible part of the population, raising policy relevant issues. Since the early 80s, orphan specific legislations have addressed the lack of incentives to invest in innovation for rare diseases. We provide new evidence on the heterogeneous effect of these provisions on innovation targeting rare diseases belonging to different classes of prevalence. A theoretical model is developed that distinguishes market-related and input-related incentives to show how the different provisions are crucial in defining the relative convenience to invest across different classes of prevalence. Data on US orphan designations, as a proxy for R&D effort, are used to empirically test the model and to estimate the effect of orphan legislations. We show that, while the number of designations has increased over time for all orphan diseases, those in the class with the lowest prevalence have done so at a much slower pace. Our findings support the idea that the type of incentives in place may be responsible for this increase in inequality within orphan diseases.