Environmental Racism Do governments, firms, and individuals discriminate against the poor and racial minorities? Reminder: Midterm distributed on Tuesday.

Environmental Racism

Do governments, firms, and individuals discriminate against the poor and racial minorities?

Reminder: Midterm distributed on Tuesday No groups– work must be alone.

1. Introduction

“Environmental Racism refers to those institutional rules, regulations, and policies of government or corporate decisions that deliberately target certain communities for least desirable land uses…it is the unequal protection against toxic and hazardous waste exposure…” (Bryant, 1995)

Some definitions: LULUs

LULUs: toxic waste sites, bus depots, solid waste transfer stations, airports, sewage treatment facilities, recycling centers, etc.

Early movements:

1967 University of Houston riots over the death of an African American girl at a garbage dump that was located in the middle of a predominantly African American neighborhoodEarly 1980s protests over the landfilling of PCBs and oil-toxics in Warren County, North Carolina (mostly African American population)

More recent action:

The 1987 United Church of Christ Commission for Racial Justice’s Toxic Wastes and Race in the United States “shows that communities of color are disparately impacted by the nation’s environmental, industrial, and land use policies” All of these events point to discrepancies in the local, state, and federal permitting process that allow facilities to locate and pollute (disproportionately,) in minority and low-income neighborhoods

2. Motivation

The “Chicken vs. Egg” controversy – the traditional environmental racism model only looks at a point in time, developing correlations between pollution levels and racial composition

This doesn’t answer the question, because firms may only be locating based on cost…

a) Research question: Is there discrimination, or are we merely experiencing an edogeneity problem where firms (and people) are locating in the least expensive places?

b) Policy questions: Have zoning regulations targeted (or allowed firms to target) minority populations, or those that lack a “voice” in the political arena?

Previous LiteratureAlesina, et al, 1999 QJE article “Public Goods and Ethnic Divisions”: Results show that the shares of spending on public goods in U.S. cities are inversely related to the city's ethnic fragmentation

Cutler and Glaeser 1997 QJE Article “Are Ghettos Good or Bad”: They find that African Americans in more segregated areas have significantly worse outcomes than African Americans in less segregated areas.

3. Models (read: how can we improve them)

Dependent Variable: Pollution

Independent Variables: Indices of Segregation, Socioeconomic variables, etc.

 Pollution estimates: Total releases (volume)

Toxicity-weighted releases (acute, chronic)

Additional characteristics (odor, color)

Indices of Segregation (from Denton and Massey; Alesina)

Evenness (dissimilarity)Concentration (people per physical space)Centralization  Exposure (potential for interaction, or isolation)Clustering (spatial proximity)

Socioeconomic Variables:

Income/Capita (by race, by age, by gender)City Size (log of population)Educational AttainmentIncome InequalityPoverty Rate% Foreign BornHome Ownership and Age of StockTransportation (modes, access, and time needed)

Level of Aggregation: City, County, MSA/PMSA

Spatial ImprovementsUsing U.S. EPA-Office of Solid Waste guidelines, Meteorological Data, and Digital Terrain Data, ISC-AERMOD software computes the risk receptor grid for multiple pollutants—“Industrial Source Complex AMS/EPA Regulatory Model”

In short, the various pollution sources are modeled along with the surface contours, air dispersion. The model takes into account the dose-response assessment for the different pollutants, creates an exposure assessment, the risk characterization from that exposure.

Theoretical ImprovementsTraditional Env. Justice argument vs. Median Voter argument vs. Principle-Agent argument…How can we incorporate political representation, interest group practices, campaign contributions, etc. into the model?How do we model perceived risk vs. actual risk? Do we need to?Include non-homogenous (ubiquitous) mobile source pollution?

Multiple dimensions to segregation (spatial) as opposed to traditional measures which just look at % of total populationDoes our model differ from what the median voter model might predict? Perhaps the spatial distribution of races (and incomes) is a better predictor or pollution exposure and the provision of public environmental goodsWith the goals of:

a)Determining if environmental racism existsb)Figuring out how it came about in the first placec)Attempting to predict where it may occur in the futured)Creating ways to prevent it from happening again

4. Conclusions