While the costs of environmental policies are generally thought to be regressive, the distribution of beneﬁts is less understood. This paper explores the incidence of an unexpected decrease in air pollution in metropolitan Los Angeles by estimating the resulting change in housing costs and neighborhood demographics. The decrease in air pollution was caused jointly by the California Electricity Crisis of 2000 and the RECLAIM cap-and-trade program for NOx emissions and impacted neighborhoods differentially based on their location relative to major polluters and local wind patterns. I measure local exposure to this pollution shock using a dispersion model developed by atmospheric scientists which calculates the effect of individual ﬁrms’ emissions on the air quality of nearby locations. The estimates show that (a) housing rents increase signiﬁcantly and as much as house prices; (b) 9% of low-income households leave the sample area due to improved air quality; and (c) low-income households are rarely home owners who would beneﬁt from increased housing wealth. I show that a standard residential sorting model predicts that when low-income residents respond to improved amenities by leaving, the distribution of beneﬁts from the improvement is likely regressive. Together, these results suggest that the distribution of beneﬁts from improved air quality likely favors higher-income households.