This paper follows an instrumental variable strategy using wind currents and meteorological data to evaluate the impact of air pollution on several Brazilian labor market outcomes. I find that a change in one standard deviation of small particulate matter (PM 2.5) leads to a statistically significant fall in the overall labor market size of 7.6%, a reduction in hiring of 3%, a fall in quitting of 0.05 percentage points, and a rise in wages of 1.4%. There is no statistically significant change in firing rates. The wage and firing rate results are less reliable since they vary based on specification. Older workers, women, and low-skill employees are more heavily impacted than others. I confirm this effect by exploiting a 2009 pollution control policy. With a synthetic difference in difference strategy, I find that the policy resulted in statistically significant improvements in air pollution and employment outcomes roughly 3 times the size of my IV estimation.