Spatial Resolution, Costs, and Equity in Air Toxics Regulation

Abstract

Concern about environmental injustice has been driving the recent effort to characterize risks from exposures to air toxics at very fine spatial resolutions. However, few studies seek to understand the potential policy implications of regulating risks at increasingly finer spatial resolutions and the impact of resulting policies on distribution of risks. To address this gap, the broad question for this research is how could the choice of spatial resolution for regulation of risks from toxic air pollutants affect emission controls and the consequences thereof? This research develops a formal model of a hypothetical decision maker choosing emission controls within a risk-based regulatory framework. The model suggests that optimal controls on air toxics emissions vary depending on the spatial resolution chosen to regulate risks; net social costs are non-decreasing as one regulates at finer and finer spatial resolutions.

An empirical application of the model using air toxic emission data for Escambia and Santa Rosa Counties in Florida demonstrates the sensitivity of optimal emissions to spatial resolution chosen for regulation. The research then investigates the equity implications of regulating at different spatial resolutions with regard to the spatial distribution of cancer risks. The empirical results indicate that regulation at finer spatial resolutions could involve a tradeoff between costs and equitable distribution of risks. For example, at a threshold cancer risk of 100 in a million, regulating at census block level resolution could be twice as costly as regulating at census tract resolution while reducing the maximum individual risk by almost half. Further, regulation at finer spatial resolutions might not address environmental injustice by itself unless such concerns are more explicitly incorporated into emission control decisions. Finally, this research shows that spatial resolution at which air toxics risks are regulated could matter in predictable ways even after taking into account the uncertainties that the decision maker faces.