A?rmative action policies in employment are proposed and used in both developed and developing countries with the goal of promoting economic progress for members of historically disadvantaged minority groups. Whether they actually help the targeted groups overall, and whether some subgroups bene?t more than others, are open questions. This paper evaluates the e?ects of one such policy–setting aside jobs for minorities–on minorities’ labor market outcomes. I take advantage of the fact that public sector jobs in India are set aside for minorities based on a strict policy rule to identify the causal e?ect of job reservations for minorities. The policy rule is stated in the Indian Constitution, and requires that the share of reserved jobs for scheduled castes and scheduled tribes–the two principal disadvantaged minority groups in India–be the same as their share of the total population in the most recent decennial census. The policy rule and the administrative lags in its implementation generate plausibly exogenous variation in share of jobs reserved for minorities. I implement this identi?cation strategy using individual-level data from multiple rounds of the National Sample Survey. The main ?nding is that increasing job reservations for the scheduled castes signi?cantly increases their presence in “good” jobs, but increasing job reservations for scheduled tribes has no signi?cant impact on scheduled tribes at conventional levels of signi?cance. The bene?ts for the scheduled castes appear to be more pronounced for members who reside in urban areas and who are less educated. That members of scheduled tribes do not bene?t may be due to their concentration in remote rural areas; there is a spatial mismatch between where most of them live and where most public sector jobs are. Thus, although scheduled tribes and scheduled castes both have much worse socioeconomic outcomes than non-minorities in India, the ?ndings suggest that distinct policies for each minority group may be needed to narrow the gaps.