Association for Public Policy Analysis & Management

Since the issuance of Executive Order 12898 in 1994, EPA has yet to clearly define potential environmental justice areas (PEJA) in terms of critical demographic and socioeconomic standards (e.g., minority, income and poverty status). Likewise, only a few states have defined and screened PEJAs and incorporated this issue into agency decision making. A lack of action to some extent reflects the contestation about a balance between two important administrative values: social equity and program effectiveness. Without identifying the communities of concern, policy stakeholders are unable to develop strategic planning, establish goals and performance measurement, allocate compensatory resources, or undertake targeted regulatory actions to ameliorate environmental inequities. However, a strict and inflexible definition limits agencies' ability of assisting the communities that are overburdened by environmental hazards in reality and hinders agencies in carrying out responsibility for protecting public health and the environment in an effective way.

In 2003, New York State adopted Commissioner Policy 29: Environmental Justice and Permitting, formally defining and screening PEJAs and mandating targeted permitting and enforcement activities of the Department of Environmental Conservation. This policy intervention provides an invaluable lens to evaluate one of the most critical but contested issues in the research on environmental justice: is there a tradeoff between program effectiveness and social equity? If so, to what extent does this circumstance exist and what are the potential implications for the relationships between public management, environmental governance, and citizenship?

Using census block group as the unit of analysis, this study empirically examines the implementation variations across the communities that reflect different types of issue salience. The core data on community demographic and socioeconomic characteristics are drawn from the U.S. Census 2000 and the 2005-2009 American Community Survey. The dependent variables are the agency's compliance monitoring and assurance activities (measured by the aggregated number of inspections and administrative enforcement actions) under the Clean Air Act. The environmental information of the regulated facilities is compiled from EPA’s Integrated Data for Enforcement Analysis - Air Facility System and Facility Registry System. As for the independent variables, the equity-oriented community is measured by the state-designated PEJAs (coded as 1 and otherwise as 0). The data is collected from the New York State GIS Clearinghouse database. A dichotomous variable is employed to represent the policy intervention (coded as 1 in the post-policy period, and as 0 in the pre-policy period). The needs for program effectiveness are measured by the status of high priority violator and the historical noncompliance record of a given block group. To evaluate the efficacy of the CP-29 and the possible tradeoff of two administrative values, interaction variables are created between the policy intervention and the measures of equity and equity concerns, respectively. This study also considers a variety of control variables, including minority, median household income, residents living below poverty level, English proficiency, linguistic isolation, higher education attainment, criteria pollutant standard nonattainment area, and a lagged dependent variable. Negative binomial regression models are employed to estimate the effects of explanatory variables on the count outcome variables.