Assessing the Impact of PFAS Water Regulation

• In Event: Cleaning the Water, Improving the Land: Policy Effects

Abstract

Per- and Polyfluoroalkyl Substances (PFAS), known as “Forever Chemicals”, are a class of man-made contaminants that are ubiquitous, long-lasting and difficult to remove. Recent evidence points to adverse health outcomes (Waterfield et al., Env. Health, 2020). One pathway through which humans are exposed is drinking water (DeLuca et al., Env. Int., 2022). In response, the Safe Drinking Water Act will bring PFAS under its purview with regulation starting in 2029. Because data collection is sparse, the impact these regulations have on drinking water quality and human health remain uncertain. The goal of this paper is to evaluate the impact of the reduction of Notification Levels for Perfluorooctanoic acid (PFOA) and Perfluorooctanesulfonic acid (PFOS) in California in 2019. Anecdotally, this low-cost policy led to improvements in water quality. A key challenge of this analysis is the wide prevalence of partial censoring in PFAS reporting. I first show that applying standard approaches, such as difference-in-differences or changes-in-changes, ignoring interval censoring, would lead to biased estimates of the causal impact of the regulation. To overcome this bias, I use a parametric approximation to recover the cumulative distribution function under interval censoring. I implement a changes-in-changes approach using the recovered distribution of PFAS concentration to estimate the impact of tightening standards for the regulated pollutants on water quality. Using PFAS concentration data at the sampling point - test level from the California Department of Drinking Water and the Environmental Protection Agency, paired with socioeconomic and demographic characteristics, weather data, and other variables, I quantify the effect of this policy on water quality. On the intensive margin, analyzing facilities that tested for these contaminants before and after the policy change, I find no significant effect on water quality throughout the distribution. However, public water systems may respond to the policy by adding new wells or treatment plants. Preliminary analysis shows the majority of this response comes from the addition of new and retrofitted treatment plants to address PFAS contamination. These results are important as federal environmental policy becomes uncertain, and we rely on state level policies to address contamination and reduce our exposure to harmful substances. Furthermore, the methods proposed can be used in various settings with interval censoring present.

Author

• Laura Alcocer Quinones, University of California, Davis
  Presenting Author