The Effects of Socioeconomic and Racial Characteristics on Firm Pollution Decisions in the St. Louis MSA

Abstract

The purpose of this paper is to assess how the characteristics of class and race are associated with facility pollution levels. Other studies have examined the relationship between class and racial characteristics and firm emissions and have found mixed results. For example, Ringquist (1997) and Brooks and Sethi (1997) find a positive and significant correlation between income and emissions levels, while Centner (1996) fails to find a significant correlation between the two variables. Arora and Cason analyze neighborhood characteristics at the zip code level in an attempt to explain changes in toxic emissions releases between 1990 and 1993. My paper builds on the work of Arora and Cason (1996) but differs in several important ways. First, I focus on the St. Louis MSA. Second, I use more recent Toxic Release Inventory and Census Data. In my analysis, I address the following question: Are 2004 emissions (compared to 2000 emissions) disproportionately larger in minority and low-income populations in the St. Louis MSA or not? Using the Heckman two-step maximum likelihood model, I find that the level of pollution earlier in time is a positive predictor of pollution later in time. While environmental justice variables are not statistically significant in my results, further study over longer periods of time and over larger geographic areas may reveal ways in which these variables significantly affect communities.