Winning the lottery is associated with a small but persistent reduction in ambient air pollution. This effect is similar for Black and White individuals, and driven entirely by those that move from their initial location.

These descriptive facts are compelling, but they don't tell us about the causal effect of a change in income or wealth on environmental quality. To get at that we exploit information on lottery winners to explore the extent to which a large windfall -> reductions in pollution

This pattern also holds within cities, and for other pollutants. The pollution-income gradients are slightly steeper, but still very inelastic. In 2016 a 1% increase in income for Black individuals is associated with a -0.0167% reduction in ambient air pollution.

We show that at every percentile of the national income distribution, Black, Hispanic, and Asian individuals are exposed to higher levels of ambient air pollution that White individuals. We show this pattern was true in 1984 and holds over time. This is how it looked in 2016.

We make progress using the universe of tax returns, high-res,remote sensing measurements of ambient air pollution, and linked individual-level demographics and residential address histories, which means we have individual, race-specific income at the same resolution as pollution

Answering this question has been hard. Measurement of air quality has been patchy, individual income data (by race) not typically available at the same spatial resolution as pollution varies, and most of the time we just end up with correlations rather than causal insights.