Performance funding is an increasingly prevalent policy state officials use to allocate a portion of state funds to public colleges and universities. Researchers have begun to evaluate the effect of these policies, finding bleak evidence of their effectiveness in yielding intended outputs and suggesting the policies may even result in limited college access for underserved students. There may also be differences in policy effects depending on performance-funding policy designs, which vary considerably across states. Of particular interest to this study are premiums--financial bonuses to institutions--for promoting access and success for specified underserved student groups. Using difference-in-differences models and an original dataset on premiums in funding models, this study evaluates the impact of premiums for underserved students in performance-funding models on selectivity and the enrollment of minority and low-income students at 4-year universities from 1993 to 2014. We find that the share of both low-income and Hispanic students increases in institutions with performance-funding premiums for underserved students compared to institutions subject to performance funding without such premiums. Effects vary depending on premium type and longevity. The findings also reveal unexpected, negative effects of premiums on Black student enrollments. Our findings suggest that, by incorporating premiums, performance-funding model designers might prevent, minimize, or reverse the negative consequences of performance funding on vulnerable student groups. However, given variation in premium effects across student groups, performance- funding model designs should be tailored to local contexts.