This paper studies the effects of Los Angeles Metro Rail on the spatial distribution of people and prices. Using a panel of bilateral commuting flows, I estimate a quantitative spatial general equilibrium model to quantify the welfare benefits of urban rail transit and distinguish the benefits of reduced commuting frictions from other channels. The subway causes a 7%-13% increase in commuting between pairs of connected tracts; I select plausible control pairs using proposed subway and historical streetcar lines to identify this effect. The structural parameters of the model are also estimated and are identified using a novel strategy that interacts tract-specific labor demand shocks with the spatial configuration of the city. These parameters indicate people are relatively unresponsive to changes in local prices and characteristics, implying that the commuting response corresponds to a large utility gain. The welfare benefits by 2000 are significant: LA Metro Rail increases aggregate welfare by $246 million annually. However, these benefits are only about one-third of annualized costs. While benefits did not outweigh costs by 2000, I employ more recent data to show that there are dynamic effects: Commuting continues to increase between connected locations.