This study investigates the role of job characteristics on an individual’s decisions to follow social distancing policies, work, and apply for unemployment insurance in the United States during the COVID-19 pandemic. We use data that track millions of mobile devices and their daily movements across physical locations to measure whether the devices’ owners leave their homes, or work part-time or fulltime on a given day, and we also collect data on weekly unemployment insurance claims. We find that the presence of jobs with a high work-from-home capacity in a region increases the ability of people to follow social distancing policies and decreases their unemployment risk, whereas the presence of jobs with high physical proximity decreases the incidences of following social distancing policies and unemployment and increases the incidence of work during the pandemic. These heterogeneous responses based on local job characteristics persist even conditional on a broad set of demographic and socioeconomic variables.