State and local governments throughout the United States attempted to mitigate the spread of Covid-19 using stay-at-home orders to limit social interactions and mobility. We study the economic impact of these orders and their optimal implementation in a fiscal union. Using an event study framework, we find that stay-at-home orders caused a 4 percentage point decrease in consumer spending and hours worked. These estimates suggest a $10 billion decrease in spending and $15 billion in lost earnings. We then develop an economic SIR model with multiple locations to study the optimal implementation of stay-at-home orders. From a national welfare perspective, the model suggests that it is optimal for locations with higher infection rates to set stricter mitigation policies. This occurs as a common, national policy is too restrictive for the economies of mildly infected areas and causes greater declines in consumption and hours worked than are optimal.