Can summer programs, as remedial supplements to regular schooling, extend learning opportunities and other benefits to disadvantaged students? To frame this question, we compare logics from "social reproduction" and "partial compensation" perspectives, and then apply them to a large mixed method study of four kinds of summer programs in Ontario. Drawing on quantitative data on over 10,000 students and qualitative data from interviews with over 200 teachers and parents, we examined patterns of student recruitment and participation, social valuations, and academic outcomes. We found that all summer programs successfully recruited disadvantaged students without stigmatizing them, and raised their average achievement without widening pre-existing gaps. We interpret these findings as being consistent with the "partial compensation" perspective, and discuss related policy implications that include COVID-19 learning recovery strategies.