The Elusive Environmental Justice Area: Three Waves of Policy in the U.S. Environmental Protection Agency

Introduction

In 1994, President Clinton's Executive Order 12898 directed all federal agencies in the U.S. to “[identify] and [address], as appropriate, disproportionately high and adverse human health or environmental effects of their programs, policies, and activities on minority populations and low-income populations.” ¹ Almost twenty years later, the U.S. Environmental Protection Agency (EPA) still lacks a precise, consistent definition of a disproportionately impacted population. Consequently, it remains challenging for the agency to monitor its own progress in achieving environmental justice (EJ). One of the goals articulated in the EPA's current Plan EJ 2014, unveiled by Administrator Lisa Jackson in July 2010, is to “develop a nationally consistent environmental justice screening tool”—one that would finally enable the EPA to identify disproportionately impacted populations—by the year 2014. ² However, a robust definition of the “environmental justice area,” or “EJ community,” remains elusive.

One obstacle to defining environmental justice areas consistently is the technical challenge of operationalizing “disproportionate impact.” Determining whether a community or population is disproportionately affected by pollution is far from straightforward. It depends, for instance, on the scale and resolution at which the community is defined geographically, on the criteria used to determine whether it counts as disproportionately low-income or minority, and on how adverse effects are measured as disproportionate relative to other communities. ³ But a second obstacle has been the range of political agendas and administrative strategies that have shaped U.S. environmental justice policy at the federal level. I argue that the EPA's efforts to define EJ areas have been complicated by three overlapping “waves” of environmental justice policy. The first wave expanded environmental justice policy from its origins in concerns about environmental racism to encompass both people of color and low-income populations, and it established a decentralized regional approach to identifying and addressing disproportionately impacted communities. Although the second wave retained these regional protocols, it developed national guidance based on redefining environmental justice as applicable to all communities and populations. The current third wave restores the earlier focus on minority and low-income populations, but aims to replace regionally specific protocols for identifying EJ areas with an approach applicable throughout the country. I suggest that lingering problems will make the latter task extremely challenging to accomplish by 2014.

The First Wave: The Executive Order and Regional Interim Protocols

Clinton's Executive Order represents the culmination of several years of effort that initiated the first wave of environmental justice policy in the EPA. Protests against a PCB landfill in Warren County, North Carolina in 1982 helped galvanize the broader movement against environmental racism and inspired classic studies, such as the 1987 study by the United Church of Christ Commission on Racial Justice, finding that people of color were disproportionately exposed to toxic pollution. ⁴ Another pivotal event was the 1990 Conference on Race and the Incidence of Environmental Hazards at the University of Michigan, which not only laid the groundwork for the First National People of Color Environmental Leadership Summit in 1991, but also generated the “Michigan Coalition” of academics and advocates that convinced the EPA and other federal agencies to attend to environmental equity during the Bush, Sr. administration. ⁵

Despite the grassroots movement's focus on environmental racism, the first wave of federal policy on environmental equity and justice—exemplified both by Clinton's Executive Order and by earlier documents of the Environmental Equity Workgroup during the Bush, Sr. administration—encompassed both people of color and low-income populations. ⁶ Initially, including low-income communities within the scope of environmental justice policy generated controversy. For example, in 1995 some members of the newly established National Environmental Justice Advisory Council (NEJAC) found it “degrading” that the EPA's first Environmental Justice Strategic Plan “includ[ed] low income groups as comprising a portion of the environmental justice community.” ⁷ Nonetheless, this was consistent with the Executive Order, and the EPA's early efforts to identify and address disproportionately impacted communities targeted these two constituencies.

During this first wave, the EPA delegated the task of developing interim geographic information systems (GIS) protocols for identifying “potential environmental justice areas” to its ten regional offices. ⁸ By 1998, EPA Regions 2 and 5 had each developed such protocols, and Region 4 followed in 1999; other regions issued protocols within the next few years. However, in the absence of national guidance on “disproportionate impact,” the regions based their definitions of “potential EJ areas” solely on minority or low-income demographics.

The Second Wave: Redefining the Constituency of Environmental Justice

The inauguration of President George W. Bush in 2001 ushered in a second wave of environmental justice policy. Its most striking characteristic was a broad redefinition of environmental justice, which carried significant implications for defining EJ areas. When Bush's first EPA Administrator Christine Todd Whitman reaffirmed the agency's “commitment to environmental justice” in 2001, she characterized it as a “goal to be achieved for all communities and persons across this Nation.” ⁹ In the following year, an internal memo from the EPA's Office of Environmental Justice explicitly disconnected environmental justice from minority and low-income populations: “Senior management should recognize that the environmental justice program is not an affirmative action program or a set-aside program designed specifically to address the concerns of minority communities and/or low-income communities.” ¹⁰

Although regional interim protocols for identifying EJ communities remained in place, the national Office of Environmental Justice developed a technical guidance document reflecting the reorientation of environmental justice as an objective for “all communities and persons”: the Toolkit for Assessing Potential Allegations of Environmental Injustice. ¹¹ This document, designed as a guidebook for regional EJ coordinators, presented a framework designed to “promote national consistency in how environmental justice concepts are understood and addressed.” ¹² It also went further than the regional interim policies by proposing specific indicators of adverse effects, divided into four broad categories: environmental, health, social, and economic. It thereby offered the potential to enable identification of EJ areas based on actual disproportionate impact, rather than demographics alone.

However, tensions between first wave and second wave interpretations of environmental justice pervade the Toolkit. Although the document acknowledges Clinton's Executive Order as a “fundamental basis for EPA's environmental justice program,” it subsequently rejects one of its central emphases: “However, while the Executive Order focused on minority and low-income populations, EPA's mission statement demonstrates that the environmental justice concepts should be applied to all communities regardless of race, ethnicity, or income status.” ¹³ It also interprets the “first principle” articulated in Whitman's 2001 memo as “that environmental justice is not limited to low-income and/or minority populations only.” ¹⁴ This tension between the Executive Order and the Whitman memo generated an ambiguous protocol for the identification of potential EJ areas. For example, the Toolkit divides “disproportionately high and adverse effects or impacts” into two categories: the first “is predominantly borne by any segment of the population, including, for example, a minority population and/or a low-income population”; the second “will be suffered by a minority population and/or low-income population and is appreciably more severe or greater in magnitude than the adverse effect or impact that will be suffered by a non-minority population and/or non-low-income population.” ¹⁵ But even within the second category—closer to the Executive Order definition—the Toolkit prescribes using “the [neutral] term ‘geographic areas’ instead of the term ‘environmental justice communities’…when conducting environmental justice assessments.” ¹⁶

Both the first-wave delegation of EJ policy to the regions and the second-wave redefinition of environmental justice as applicable to everyone came under attack in a 2004 evaluation report by the EPA's Office of the Inspector General (OIG). ¹⁷ This evaluation focused on two objectives—the EPA's implementation of Executive Order 12898, and its efforts to define environmental justice areas—and concluded that the agency had failed at both. With respect to the latter, the report emphasized that the Office of Environmental Justice had defined neither the meaning of disproportionate impacts nor the attributes of a minority or low-income community. Instead, each region used not only different terms for EJ areas, but also different definitions and thresholds to identify low-income or minority populations. ¹⁸ As a result, whether one lived in a “potential EJ area” depended heavily on the EPA administrative region in which one resided.

To demonstrate the point, the report compared results of three regional GIS-based protocols for identifying potential EJ areas in Worcester, Massachusetts. While Region 5's protocol would include only 34.5% of the city's population within potential EJ areas, and Region 1's only 41.9%, Region 6's protocol would include 59.5%. ¹⁹ The OIG report contended that this regional variation obstructed the EPA's capacity to measure its progress and demonstrate its compliance with the Executive Order: “The lack of a generic environmental justice definition for minority and low-income is … impacting the Agency from being able to quantify its accomplishments.” ²⁰ Second, it created the possibility of regional inequities: “Due to regional variations, populations in some States do not receive the same level of environmental justice action as in other States.” ²¹ Consequently, the report characterized standardizing the EJ community at the national level as an “urgent need.” ²²

The report also argued that the second wave reinterpretation of environmental justice “moved the Agency's environmental justice focus away from minority and low-income populations” and implied a return to “pre-Executive Order status.” ²³ The EPA's Office of Environmental Justice raised strong objections to the report, including its “central premise” that Executive Order 12898 “requires the Agency to identify and address specific communities and to define disproportionate impact.” ²⁴ It argued that the Executive Order required no national standard, threshold, or “brightline” for identifying EJ communities, calling these “dead-end questions.” ²⁵ However, the shift toward “environmental justice for everyone” had already generated strident objections from major environmental justice organizations and their legislative allies. ²⁶ By 2005, although the EPA formally retained its new definition of environmental justice, it began a third wave by restoring the emphasis on low-income and minority populations.

The Third Wave: EJ Seat and Beyond?

Two additional characteristics distinguish the third wave of policy for identifying EJ areas: the emphasis on replacing regional protocols with a consistent national approach, and the effort to move toward definitions based on actual disproportionate adverse effects. Although both characteristics emerged in the second-wave Toolkit, they reached a higher level of prominence and sophistication in the Environmental Justice Strategic Enforcement Assessment Tool (EJ SEAT). ²⁷ Designed by and for the EPA's Office of Enforcement and Compliance Assurance, EJ SEAT aims to assign scores to every census tract in the country, using an index compiled from eighteen indicators in four categories: environmental, human health, compliance, and social/demographic. Data for EJ SEAT must come from federally recognized sources, and the EPA can only use data available nationwide at the census tract level. According to one proposed method, the top-scoring 20% of census tracts in each state would become potential EJ areas. In contrast to the regional protocols, EJ SEAT uses uniform definitions of “minority” and “low-income” throughout the United States.

It became clear at the September 2007 NEJAC meeting, where EJ SEAT was first introduced for public scrutiny, that the tool's implementation would be difficult. First, incorporating data from the four categories presented numerous technical problems. The most challenging category was human health, which the developers of EJ SEAT had incorporated on the recommendation of a NEJAC subcommittee. One problem was that few health indicators were “available at the census tract level across the nation in a way that is accessible due to privacy issues.” ²⁸ The only indicators the workgroup included were low birth weight and infant mortality, and even these required the tool to incorporate the incorrect assumption that values at the county scale also applied to each census tract within the county. A NEJAC subcommittee reviewed EJ SEAT and recommended in a 2010 report that the EPA either eliminate the health indicators or reduce their weight. ²⁹

NEJAC's review found that other categories also presented serious problems. For example, the environmental data focus almost exclusively on airborne toxics and include controversial self-reported data from the EPA's Toxics Release Inventory program. Meanwhile, some compliance indicators—such as numbers of inspections or violations—are ambiguous; do they imply strict enforcement or neglect? Finally, the primary basis for demographic data is the U.S. Census, in which several key environmental justice constituencies, such as tribal communities or migrant workers, might not participate. In addition to these and other problems with the indicators and data sources, NEJAC members cited the potential for misusing EJ SEAT, whether as a means to exclude self-identified EJ communities with low scores, or conversely, as a source of stigmatization in communities seeking to attract development. ³⁰

Conclusion

Although the EPA's goal to develop a nationally standardized approach to identifying EJ areas remains active, EJ SEAT remains in draft form, and NEJAC's critical evaluation of the tool generated doubts about whether it can ultimately meet the needs of both the agency and the broader environmental justice community. Efforts to develop EJ SEAT have been plagued by many of the same technical challenges that have complicated regional protocols, from limited data to the difficulty of defining “disproportionate impact.” They also remain politically vulnerable. Just as the Bush, Jr. administration's EPA sought to reorient environmental justice policy away from its original focus on minority and low-income populations, so too might a new presidential administration reinterpret or even rescind Executive Order 12898, introducing a potential “fourth wave” of environmental justice policy.

The history of the EPA's ongoing efforts to identify EJ areas raises an important question for future research and policy deliberation: should this identification remain based on spatial analysis? First, defining environmental justice areas as census tracts—or other geographic units—with a particular range of scores on a standardized index may overlook the place-specific circumstances and experiences that lead a community to understand itself as the victim of an environmental injustice. For example, one community might regard an industrial cluster as an unwanted and unjust environmental hazard, while another with the same EJ SEAT score might regard a similar cluster as a tolerable economic opportunity. Second, distinctive communities—such as tribal communities—not only face unique risks, but also hold particular places as sacred and meaningful in ways that conventional GIS-based protocols may fail to recognize. ³¹ The EPA's Plan EJ 2014 itself emphasizes the need to “ground truth” the scores and to use EJ SEAT only as an initial screening tool, but perhaps it is time to reexamine the assumption that disproportionate impact can be measured similarly everywhere. Although defining environmental justice areas in a way that recognizes the specificity of different places would undoubtedly be challenging, it may represent the most productive way forward in the EPA's quest to replace its regional interim policies with a national EJ protocol.