On February 4^th 1994 President Clinton issued Executive Order 12898 requiring Federal agencies to “identify… and address…, as appropriate, disproportionately high and adverse human health or environmental effects” of agency programs, policies, and actions on minority populations and low-income populations. Prior to EO 12898 the EPA created an Office of Environmental Justice (OEJ) in 1992 to ensure “the fair treatment and meaningful involvement” of individuals in the development, implementation, and enforcement of its regulations and policies. Recently OEJ has been working with the EPA’s Office of Policy to incorporate more thorough analyses of environmental justice (EJ) into its regulatory impact analyses for rulemakings.  According to the Office of Environmental Justice at EPA, environmental justice exists when “no group of people, including racial, ethnic, or socioeconomic group, … bear[s] a disproportionate share of the negative environmental consequences resulting from industrial, municipal, and commercial operations.” This paper summarize the history of EJ analysis at the EPA and then uses a case study approach to describe the analytic approaches taken thus far and the challenges that the EPA has encountered.  Techniques for identifying potential effects on the basis of race and income range from basic summary statistics to proximity-based analysis.

We also use the academic literature as a guide to key issues and analytic methods. The early EJ literature was a direct response to media and activist attention brought to the siting decisions of landfills.  Despite the fact that these papers used small data sets and relatively simple empirical techniques, they are significant because they were the first to point out a potential causal relationship between the location of economic activity and poor and minority populations (e.g. Bullard 1983, GAO 1983, United Church of Christ 1987). The academic literature, using larger data sets and more sophisticated empirical techniques, has explored the extent to which EJ issues have affected the  location decisions of polluting plants in general and the level of their emissions (e.g., Hamilton 1995, Been 1994, Gray and Shadbegian 2004, Wolverton 2009), finding mixed evidence for the role that race and income play. Key issues that arise in the academic literature are sensitivity to how comparison groups are defined, the geographic scale of neighborhoods, and the way in which exposure to emissions is proxied. However, these studies are different from an EJ analysis to inform rulemaking in several key respects.  First, they are often sub-national in scope and, by their very nature, ex-post assessments. For the EPA, the Executive Order implies analyses should be conducted at the national level, due to the nature of the regulation, and should be an ex-ante assessment of the incremental effects of the rule.