Measuring the economic impact of a war is a daunting task. Common indicators like casualties, infrastructure damages, and gross domestic product effects provide useful benchmarks, but they fail to capture the complex welfare effects of wars. This paper proposes a new method to estimate the welfare impact of conflicts and remedy common data constraints in conflict-affected environments. The method first estimates how agents regard spatial welfare differentials by voting with their feet, using pre-conflict data. Then, it infers a lower-bound estimate for the conflict-driven welfare shock from partially observed post-conflict migration patterns. A case study of the conflict in Eastern Ukraine between 2014 and 2019 shows a large lower-bound welfare loss for Donetsk residents equivalent to between 7.3 and 24.8 percent of life-time income depending on agents' time preferences.