Is public housing bad for children? The net effect of moving into public housing on children's academic outcomes is theoretically ambiguous and likely to depend on changes to neighborhood and school characteristics. Drawing on detailed individual-level longitudinal data on New York City public school students, we exploit plausibly random variation in the precise timing of entry into public housing to estimate credibly causal effects of public housing residency on academic outcomes. Both difference-in-differences and event study analyses suggest positive effects of public housing on test scores, with larger effects after the initial year. Stalled academic performance at entry may reflect disruptive effects of residential and school mobility. Effects on test scores are larger among students who move from lower-income neighborhoods due, perhaps, to increases in neighborhood and school quality. For some subgroups, attendance improves and the incidence of obesity declines. Our results contradict the popular belief that public housing is bad for kids.