Using regression discontinuity and difference-in-differences designs to compare prices of California homes across a new wildfire risk zone boundary, I identify null effects in the immediate vicinity of the new risk zone boundary. In contrast, prices of homes newly assigned to a riskier zone drop by 5.7% further away from the new risk zone boundary. These findings suggest that households do not respond to the risk salience triggered by the new risk zone assignment. Instead, they update their risk beliefs in response to rezoning in a way that is consistent with the actual wildfire risk.