Association for Public Policy Analysis & Management

Over the past three decades, mounting evidence has suggested that people of color and low-income communities in the U.S. are more likely to live closely to the sources of a variety of environmental hazards, to be subject to higher levels of actual harms and poor environmental quality, and to have adverse health outcomes resulting from cumulative exposure to toxics. Equally important, a growing amount of studies show that these two segments of social groups are also less likely to receive government’s compliance inspections and punitive actions, particularly pointing to the lax regulatory enforcement by state governments. However, there is relatively scarce research exploring the policy implementation practices of the U.S. Environmental Protection Agency (EPA). Given the significant institutional authorities wielded by the EPA regional offices in environmental federalism and their heightened responsibilities in carrying out national environmental justice agenda, without evaluating the compliance monitoring and assurance activities of the EPA regional offices, we cannot grasp a comprehensive understanding of the federal government’s environmental justice performance. This study contributes to public administration and environmental policy literature by assessing how the construction of problem contexts shapes the EPA’s policy implementation. Specifically, we argue that political/social, health, risk, and convergence problem contexts drive the agency’s decision making. With census block group as the unit of analysis, this study examines EPA regions’ implementation variations in their jurisdictions from 2008 through 2012 under the Clean Air Act and Clean Water Act. The dependent variables are regional offices’ compliance monitoring and assurance activities (measured by the aggregate number of inspections and administrative enforcement actions). The environmental information of the regulated facilities is compiled from EPA’s Integrated Data for Enforcement Analysis and Facility Registry System. We measure the multi-dimensional problem contexts in the following ways: (1) political/social: the percentage of racial/ethnic minorities; (2) health: community health index; (3) risk: the number of alleged violators of regulatory compliance; and (4) convergence: environmental justice vulnerability index (convergence between political/social and health factors), overlapping policy goal index (convergence between political/social, health, and risk factors). The control variables include the EPA regional offices’ organizational capacity: the number of the EPA institutional establishments as well as the number of the EPA workforce at the state level (data collected from the EPA and the U.S. Office of Personnel Management), community demographic and socioeconomic characteristics (median household income, residents living below poverty level, household linguistic isolation, higher education attainment, unemployment rate, the universe of regulated facilities, manufacturing labor employment, criteria pollutant nonattainment area, population, and population density) (data drawn from EPA’s EJSCREEN information screening tool and the 2008-2012 American Community Survey). Count data model (the final model depends on the specific analysis) will be employed to estimate the effects of explanatory variables on the count outcome variables.