How does the design of military institutions affect who bears the costs of war? We answer this question by studying the transformative shift from segregated to integrated US military units during the Korean War. Combining new micro-level data on combat fatalities with archival data on the deployment and racial composition of military battalions, we show that Black and white soldiers died at similar rates under segregation. Qualitative and quantitative evidence provides one potential explanation for this counterintuitive null finding: acute battlefield concerns necessitated deploying military units wherever soldiers were needed, regardless of their race. We next argue that the mid-war racial integration of units, which tied the fates of soldiers more closely together, should not alter the relative fatality rates. The evidence is consistent with this expectation. Finally, while aggregate fatality rates were equal across races, segregation enabled short-term casualty discrepancies. Under segregation there were high casualty periods for white units followed by high casualty periods for Black units. Integration eliminated this variability. This research note highlights how enshrining segregationist policies within militaries creates permissive conditions for either commanders' choices, or the dictates and variability of conflict, to shape who bears war's costs.