What Marginal Outcome Tests Can Tell Us About Racially Biased Decision-Making

Marginal outcome tests compare the expected effects of a decision on individuals who are of different races but at the same indifference point of the decision-maker. I present a simple formalization of how such tests can detect racial bias, defined as a deviation from accurate statistical discrimination. Namely, the tests can reject that the decision-maker ranks individuals according to some accurate prediction of a mandated outcome, given some unspecified race-inclusive information set. The frontier of marginal effects can furthermore rule out canonical taste-based discrimination. I relate this analysis to other interpretations of marginal outcome tests, other notions of racial discrimination, and recent identification strategies.