Grantee Research Project Results
Final Report: A National Evaluation of Equity in Hazardous Waste Sites
EPA Grant Number: R823185Title: A National Evaluation of Equity in Hazardous Waste Sites
Investigators: Anderton, Douglas L.
Institution: University of Massachusetts - Amherst
EPA Project Officer: Chung, Serena
Project Period: July 1, 1995 through July 1, 1997
Project Amount: $263,470
RFA: Socio-Economics (1995) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
The purpose of this project is to assess plausible inequities or biases in the disproportionate siting of hazardous waste facilities and hazardous material handlers in neighborhoods characterized by a high percentage of minority or socio-economically disadvantaged residents.
This project pursued several cumulative analyses to address these objectives:
- The investigator's prior research on commercial TSDFs did not find substantial cross sectional evidence of siting biases or inequities. This research was extended and enhanced through the addition of environmental and industrial data from TIGER, TRI, Dunn and Bradstreet, etc.
- A longitudinal analysis of residential areas with and without commercial TSDFs was conducted to determine whether inequities existed at the time these facilities were sited and to assess whether inequities had emerged since siting (e.g., ?white flight').
- A similar analysis was conducted to determine the potential biases in the distribution of CERCLA/NPL sites across social groups and to examine longitudinal biases in the discovery and prioritization processes.
- A national survey of RCRA governed facilities was conducted to verify RCRIS data (especially facility location) and to obtain additional detailed facility information for similar analyses of inequities in the distribution of RCRA governed facilities.
- A more general analysis of industrial and residential segregation was pursued using Dunn and Bradstreet data on all manufacturing enterprises in the largest 25 metropolitan areas of the United States (excluding Washington, DC). This research addresses the important potential that concerns over environmental inequities may further reduce the access of minority and disadvantaged populations to employment opportunities.
- A variety of related methodological issues also were addressed including the proper methods of research design controlled comparisons for assessing environmental inequities, the proper measurement of industrial segregation, and methods for the analysis of ?cumulative' or multiple facility burdens.
The investigator's analyses have direct policy implications in determining the extent of inequities and providing guidance in the relative priorities of remediation. Specific policy implications regarding the prioritization process are suggested by findings on NPL inequities. The RCRA analysis similarly provides the first such policy relevant evaluation of possible inequities in the distribution of those facilities. This research is, however, directed at the issues of siting bias in immediate residential areas of facilities?it does not directly explore environmental health risks or exposures.
Summary/Accomplishments (Outputs/Outcomes):
The findings from each of these projects are summarized in the same order as above.
- All of the analyses of site specific biases utilized census tract demographics to characterize the population composition of areas near and isolated from facility sites. Prior research suggesting industrial features of the areas were most significant in accounting for the variation in facility siting and required additional area data. Accordingly, we extended commercial TSDFs to include the simultaneous use of TIGER file infrastructural features (e.g., highways, ports, etc.), Dun and Bradstreet data on the distribution of related manufacturing industries, detailed RCRA survey data differentiating types of facilities (e.g., on and offsite, public-private, incinerators, landfills), and TRI data on volume of activities at sites. These analyses support prior research findings. Commercial TSDFs, as defined by this original study, appear on average to be located in predominantly white working class neighborhoods, with significant industrial infrastructure and industrial workforce being the most significant predictors for their siting. Abutting neighborhoods near these industrial residential areas do tend to have greater proportions of minority residents.
- This project included a longitudinal evaluation of demographics within
communities with commercial TSDFs to determine potential historical siting
biases and/or site impacts. Published results from the project did not find
these hazardous waste industries to have any specific siting biases longitudinal
impacts beyond those found for industrial areas in general.
Published results "suggest that commercial TSDFs are, on average, sited in communities that are neither disproportionately poor nor minority communities. The few effects we found to be related to equity in cross-sectional descriptive comparisons and bivariate significance tests (e.g., lower median percentage Hispanic in TSDF areas) do not stand out as significant in longitudinal or multivariate analyses. We believe this research, in concert with our earlier findings, suggests that commercial TSDF census tract communities are best characterized as areas with largely white and disproportionately industrial working class residential areas, a characterization consistent with what one might historically expect near industrial facilities."
The longitudinal analysis also "suggests that these facilities have not increased the percentage of minority composition in surrounding communities and that longitudinal changes in TSDF communities are similar to those in other more industrial areas...since these areas are among those more likely impacted by recent general trends of deindustrialization, initial impressions may be misleading."
- This project also extended these findings to Superfund (NPL) sites and
CERCLA candidate sites. Again, in published findings, an analysis of CERCLIS/NPL
sites did not find substantial evidence of environmental inequities in the
population composition near such sites. The findings of this analysis did,
however, provide prima facie evidence of potential, although slight, bias
working against minority neighborhoods in the prioritization process that moves
sites from CERCLIS to NPL designation.
"Initial descriptive findings do not support the prevailing notion that minority or socioeconomically disadvantaged populations are much more likely to reside near CERCLIS or NPL sites. In addition, multivariate analyses predicting the numbers of such sites from neighborhood social characteristics indicate there are at best modest and mixed effects of these variables on the presence of either type of site. A greater number of CERCLIS sites is slightly more likely in census tracts with greater percentages of black and Hispanic residents, considerably more likely with higher percentages of Native Americans or families on public assistance, yet less likely with a higher percentage of poor families. In contrast, a greater number of NPL sites is less likely in tracts with higher percentages of black residents or poor families. All of these effects are modest at best and do not effectively predict the presence of sites in an area. Finally, in both of these analyses the single substantively largest effect on the distribution of both CERCLIS and NPL sites is simply the greater preponderance of such sites in metropolitan areas (where there are also often higher percentages of minority and socioeconomically disadvantaged residents)."
An analysis of the prioritization of NPL sites over time, however, "indicates some plausible biases in the likelihood and timing of NPL designation from among CERCLIS sites. In this analysis, a greater percentage of black residents or of poor families decreases the likelihood and pace of prioritization. A greater percentage of households receiving public assistance, however, increases the chance of prioritization over time. And, metropolitan differences which are pronounced in the distribution of sites are not significant with respect to the process of prioritization."
- This project also surveyed all RCRA governed sites to collect additional information and verify geocoding information that was used in an analysis of equity in the distribution of these facilities. In results currently under review and accepted for conference presentation, our national level analysis does not "indicate stark inequities in the distribution of RCRA facilities across neighborhoods of different socioeconomic composition. These results do suggest that facilities are primarily located in working class neighborhoods. However, these results also strongly suggest that these industrial areas abut nearby residential areas characterized by higher percentages of both minorities and economically disadvantaged populations. These results also strongly suggest that particular types of RCRA facilities (i.e., commercial private facilities) are more likely to be located in central urban-industrial areas that are characterized by higher percentages of minorities. And, our results suggest specific area issues (e.g., location of RCRA facilities in rural southern areas with higher percentages of black residents)."
- This research also has extended these concerns to the more general
distribution of manufacturing industries in the largest metropolitan areas using
Dun and Bradstreet data files on related manufacturing industries. In findings
both published and under peer review, this research finds that industrial
segregation underlies and reinforces residential segregation patterns in major
metropolitan areas. Results provide "very limited support for hypotheses of the
environmental equity model. The results indicate that hazardous industrial
manufacturing enterprises are not more heavily segregated into minority
neighborhoods. However, non-Hispanic whites are relatively more segregated from
such industries given their already low segregation from all industries.
Our findings appear to lend support to the opportunity exclusion model for blacks and Hispanics. It does appear that these disenfranchised minority groups are less likely to be located in the same area of an MSA as industrial manufacturing employment opportunities. Yet, the Hispanic population is more similar to the non-Hispanic white population in terms of access to high employment industry in many of the MSAs that we studied."
Regional findings are of considerable interest. Blacks are least segregated from all industries in the South and West, yet, relatively more segregated from high employment industries. In contrast, they are relatively less segregated from hazardous industries in the West, but remain relatively more segregated from such industries in the South. Hispanics are least segregated from all three types of industry in the South and West. They are also, not surprisingly, less relatively segregated from high employment industries in these two regions. Hispanics are substantially less relatively segregated from hazardous industries only in the East.
"Results also suggest very different patterns of association between residential and industrial segregation for black and Hispanic minorities. Black residential segregation is strongly related to more limited access to all industries, and especially high employment enterprises, as well as to a greater isolation from hazardous industries. Hispanic residential segregation is less strongly related to industrial segregation and isolation from hazardous industries. And, Hispanic residential segregation is not, in aggregate indices, significantly related to limited access to high employment industries."
- Finally, this research has explored a variety of methodological comparative designs in drawing environmental equity conclusions, in measuring industrial segregation, and in assessing cumulative (i.e., multiple facility) environmental burdens. These methods are integrated within published research findings and several have been discussed in two published methodological commentaries arising from the project. In two minor unpublished analyses: (1) matched case control comparisons were found infeasible for analysis of hazardous industry sites due to variability in basic census tract characteristics that did not afford sufficient numbers of matched cases for reliable analysis, and (2) analyses involving population density were found to be strongly affected by a multicollinearity with urban-rural biases, which should be controlled within research designs.
Journal Articles on this Report : 5 Displayed | Download in RIS Format
Other project views: | All 14 publications | 6 publications in selected types | All 5 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Anderton DL. Methodological issues in the spatiotemporal analysis of environmental equity. Social Science Quarterly 1996;77(3):508-515. |
R823185 (1997) R823185 (Final) |
not available |
|
Anderton DL, Oakes JM, Egan KL. Environmental equity in Superfund. Demographics of the discovery and prioritization of abandoned toxic sites. Evaluation Review 1997;21(1):3-26. |
R823185 (Final) |
|
|
Davidson P, Anderton DL. Demographics of dumping II: a national environmental equity survey and the distribution of hazardous materials handlers. Demography 2000;37(4):461-466. |
R823185 (Final) |
Exit |
|
Egan KL, Anderton DL, Weber E. Relative spatial concentration among minorities:addressing errors in measurement. Social Forces1998;76(3):1115-1121. |
R823185 (Final) |
Exit |
|
Oakes JM, Anderton DL, Anderson AB. A longitudinal analysis of environmental equity in communities with hazardous waste facilities. Social Science Research 1996;25(2):125-148. |
R823185 (Final) |
Exit |
Supplemental Keywords:
environmental justice, equity, locally unwanted land uses, hazardous waste, RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, environmental justice, Economics, decision-making, Ecology and Ecosystems, Economics & Decision Making, Social Science, census tract characteristics, hazardous waste siting, Superfund sites, hazardous environmental exposures, commercial TSDFs, environmental equity, RCRA sites, environmental values, bias, hazardous waste facilities, environmental policy, ethnicity, hazardous waste siting decisions, community based intervention, equity of distributionProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.