Long Run Effects of Pollution: Evidence from the Acid Rain Program

Abstract

Despite the emergence of high quality evidence on the short run effects of exposure to pollution, we know considerably less about its long run effects, largely because dynamic optimization leads to endogenous exposure in the long run. Obtaining causal estimates is essential because the most significant portion of the total benefits from improvements in environmental quality are attributed to long run changes in mortality. In this paper, we exploit quasi-experimental variation in pollution induced by the US Acid Rain Program, which led to immediate and persistent improvements in air quality through changes in production at coal power plants. The scope for behavioral responses is greatly limited because: 1) although the program centered on SO2, the biggest change in air quality came from a largely unknown pollutant at the time, fine particulate matter (PM); and 2) economic effects were quite small and localized, while pollution changes spanned hundreds of miles. Using an event study analysis to assess dynamic effects in a difference-in-differences framework, we find significant, persistent decreases in mortality rates in counties within 100 miles of a power plant, suggesting long-run improvements in mortality. Effects are largest for cardiovascular deaths, with no detectable effects for external causes, both supporting exposure to fine PM as the relevant mechanism. These findings, while informative for the cost-benefit calculation of the Acid Rain Program, are relevant for climate policies that aim to further limit emissions from power plants.