A growing body of literature suggests that economic downturns predict an increase in child maltreatment. However, to inform policies and practices to prevent and intervene in child maltreatment, it is necessary to identify how, when, and under what conditions community-level economic conditions affect child maltreatment. In this study, we use North Carolina administrative data from 2006 to 2011 on child maltreatment reports and job losses to distinguish effects on maltreatment frequency from effects on severity, identify the timing of these effects, and test whether community characteristics moderate these effects. To isolate effects of unanticipated job losses and to control for potential confounding factors, we use a fixed effects regression approach. We find that, though job losses did not affect the frequency of reports, job losses increased the share of reports that were relatively severe. This effect endured for 9 months following job losses and was only evident in economically disadvantaged communities.