Fine Particulate Air Pollution and Accident Risk: Three Essays

Abstract

I present three essays on the effect of fine particulate air pollution (PM 2.5) on workplace and vehicular accident risk. The first essay examines how PM 2.5 exposure affects the probability of serious construction worksite accidents, using accident investigation data from the Occupational Health and Safety Administration (OSHA). The second essay uses this OSHA data to estimate the effect of PM 2.5 on manufacturing workplace risk. The third essay uses data from the National Highway Traffic Safety Administration to identify the effect of PM 2.5 on vehicular accident risk. In all cases, I find that increased PM 2.5 exposure significantly increases the probability of serious accidents, thereby imposing large costs on society.

Essay 1: PM 2.5 and Construction Worksite Safety This paper provides the first causal estimates of the impact of fine particulate matter air pollution(PM 2.5) on workplace accidents, which cost the U.S. economy an estimated$273.3 billion in 2019.I construct a novel dataset composed of the universe of serious workplace accidents investigated by OSHA from 2003 to 2015 and exploit plausibly exogenous variation in PM 2.5 caused by thermal inversions to implement an instrumental variables research design. Focusing on the U.S. construction sector, which employed 7.5 million workers (4.6% of the U.S. workforce) and accounted for 5.7% of nonfatal and 19.9% of fatal U.S. workplace injuries in 2019, I find that decreasing PM 2.5 exposure by 1 μg per cubic meter could lead to a 7% decrease in the risk of a serious construction worksite accident, representing an elasticity of 0.67.

Essay 2: Fine Particulate Air Pollution and Agricultural Worker Safety This paper provides the first causal estimates of the impact of fine particulate matter air pollution (PM 2.5) on manufacturing workplace accidents. I construct a novel dataset composed of the universe of serious workplace accidents investigated by OSHA from 2003 to 2015 and exploit plausibly exogenous variation in PM 2.5 caused by changes in the height of the atmospheric planetary boundary layer height and the shape of the temperature-altitude curve in the lower atmosphere to implement an instrumental variables research design. Focusing on the U.S. manufacturing sector, which employed 12.7 million workers (8.5% of the U.S. workforce) and accounted for 12.1% of nonfatal and 6.5% of fatal U.S. workplace injuries in 2018, I find that decreasing PM 2.5 exposure by 1μg per cubic meter could lead to a 2.8% decrease in the risk of a serious construction worksite accident, representing an elasticity of 0.25.

Essay 3: Fine Particulate Air Pollution and U.S. Vehicular Accident Risk This paper provides causal estimates of the impact of fine particulate matter air pollution (PM2.5) on vehicular traffic accidents in the United States, which resulted in 36,560 fatalities and 2.7 million injuries in 2018. I construct a panel dataset composed of the universe of fatal traffic accidents documented by NHTSA from 2003 to 2015 and exploit plausibly exogenous variation in PM 2.5 caused by changes in the height of the atmospheric planetary boundary height layer and the shape of the temperature-altitude curve in the lower atmosphere to implement an instrumental variables research design. I find that decreasing PM 2.5 exposure by 1μg per cubic meter could lead to a 2.4% decrease in the risk of fatal traffic accidents, representing an elasticity of 0.20.