We model government takings of private property, embedding prominent theories of their economic consequences. Random assignment of U.S. judges and the fact that judges make takings precedent in a manner correlated with their race, political party, and prior government advocacy facilitate causal estimates of allowing government takings. Novel data on judicial biographies and takings decisions since 1975 indicate that decisions favoring physical takings increase growth by 0.2% points but reduce minority home ownership and employment by 0.5% and 0.3% points respectively, while those favoring regulatory takings initially depress property prices but spur economic growth in the medium-run by 0.7% points.