We use a vignette-based survey experiment to elicit respondents’ assessments of the fairness of race-based hiring decisions and compare these assessments with the predictions of four preregistered ethical frameworks. While conservative respondents are much more accepting of discriminatory actions than others, respondents of all political leanings rate the relative fairness of different actions in a very similar way. A two-group framework in which one group (mostly self-described conservatives) values employers’ decision rights, the other has utilitarian concerns, and both groups use the same race-blind rules to assign relative fairness levels to actions explains our data well.