Exhausted? The Effect of Diesel School Bus Emissions on Academic Achievement

• In Event: Effects of Childhood Pollution Exposure on Development and Achievement

Abstract

Diesel emissions from school buses are harmful to children's health and development. Over 20 million children ride school buses each day in the United States, and over 90% of school buses in the US run on diesel. Diesel exhaust contains fine particulate matter (PM2.5), nitrogen oxides (NOx), and carbon monoxide, and is classified by the World Health Organization as a carcinogen. As a result, federal and state programs in the US have funded school bus retrofits, which installed emissions control devices to reduce pollution. While previous research found that retrofitting buses improved children’s health, little work has been done on other outcomes. This study examines over 9,000 school bus retrofits that occurred in North Carolina and Texas from 2004-2014 to evaluate the causal effect of this intervention on students' academic and behavioral outcomes.

I use educational records from the universe of public school students in North Carolina and Texas and take advantage of staggered timing in retrofit adoption and intensity across school districts using a generalized difference-in-differences design. Additionally, in the North Carolina sample, I compare students who live far away from school to their classmates who live nearby before and after retrofits were implemented using a triple differences strategy. This strategy more cleanly isolates exposure to school bus retrofits and permits flexible controls for district-level time trends. In both states, the rich individual-level data permits estimation of effects by race, gender, age of exposure, and length of exposure.

I find that for the average district that installed retrofits, students gained approximately 0.02 standard deviations in both reading and math standardized test scores. When scaled by the percentage of the bus fleet in each district that got retrofitted, I estimate that improving air quality by this margin for all bus riders would raise district-average test scores by 0.04-0.06 standard deviations. This magnitude is approximately equivalent to half a standard deviation gain in teacher quality. The estimates are consistent across the North Carolina and Texas samples. I find that treatment effects are stronger for students who live farther from school, and grow with years of exposure. Evidence on the effect of retrofits on absences and school discipline are less conclusive, which suggests that the mechanism for test score gains is direct improvements in cognitive ability.

This work contributes to literature about the effect of pollution on educational and labor market outcomes. Moreover, these results suggest that school bus retrofits were a cost-effective intervention that had substantive benefits beyond improvements in student health. When compared to other educational investments, I document that most other interventions either cost substantially more relative to the test score gains achieved, or otherwise deliver smaller gains in academic outcomes at a similar monetary cost as retrofits. Thus, recent efforts by federal and state governments in the US to replace diesel buses with newer, less polluting models or zero-emission electric buses may not just yield benefits for environmental and health outcomes, but may also be an effective investment in educational quality.

Author

• Maranna Yoder, Indiana University
  Presenting Author