by Cherice Bock
Editor, Whole Terrain
We’ve known for 30 years that race is a stronger predictor than economic status of proximity to toxic waste and other industrial pollutants. The work of Robert Bullard (“Solid Waste Sites and the Black Houston Community,” 1983), and the United Church of Christ’s Toxic Wastes and Race in the United States report (1987) illuminated this correlation. Unfortunately, issues of environmental justice have not been resolved in the intervening years (see Toxic Wastes and Race at Twenty, 2007, with Bullard as one of the principle authors).
One problem still difficult to unravel is why the populations around toxic waste facilities and other areas with concentrated pollutants are more likely to be communities of color. Do toxic waste and polluting facilities choose to locate near already-established communities of color? If so, is this intentionally racist, or due to economic factors? Do racial minorities move into such neighborhoods after the waste and polluting facilities are already there? What is the role of city zoning laws? Do they help or hinder? Are people being intentionally racist in their choices of where to build polluting facilities, or by their choices of neighbors? Although it has long been shown that race is a more predictive factor for proximity to toxic waste and polluting firms than is economic status, is wealth still an important factor in the equation? What other factors might be important? What might resolve the problem of environmental injustice?
In Rethinking Environmental Justice in Sustainable Cities: Insights from agent-based modeling, Heather E. Campbell, Yushim Kim, and Adam Eckerd tackle these questions in an unusual and ingenious way: by modeling a city using computer simulation software (they used NetLogo, and one could also use R, MASON, etc.) and watching what emerges as “agents” (individuals and firms) populate the city. This is called an environmental justice agent-based model (EJ ABM). Although one can learn about environmental injustice through data analysis of real cities, observing what happens in a city does not tell us why it happens. Therefore, agent-based modeling is a helpful tool because researchers can run various scenarios and see how they match up with reality, thus suggesting why we see the results that we do, and what we might do to remedy environmental injustice.
The authors meticulously explain their methods and results, making it possible for others to replicate the study or to build on the study using other variables. They built virtual cities where individuals and firms could move, choosing their location based on economic factors alone, or incorporating the factor of similarity to others in the neighborhood (race preference), outright racism, and proximity to amenities or disamenities. In this way, the authors can study the emergent properties that aggregate after a number of agents’ individual choices accrue. In some of the scenarios, the authors focused on firm behavior, while in others they focused on the choices of the individual people. They ran multiple iterations of each trial, and did separate trials with each factor holding a different weight. The book also includes a chapter by Hal T. Nelson, Nicholas L. Cain, and Zining Yang where these authors utilized the EJ ABM overlaid with GIS and census data, showing the utility of the EJ ABM to provide extra dimensions of information when used in concert with real-world data.
Campbell, et al. (2015) found that, while intentional racism did cause slightly heightened levels of environmental injustice compared to other scenarios, it was not the only factor involved in creating situations of environmental injustice, and it was not required in order for environmental injustice to occur. Economics is an important factor, in that the trials that randomly assigned incomes to all individuals did not see as pronounced of a difference between races in proximity to pollution. In many of the trials, individual agents were assigned incomes matching average incomes of those of their race in the 2010 US census. Since, in the US, people of color have a lower average income than whites, these trials more closely matched current reality. When the authors also combined this with strong, medium, or weak preference for living in a community with individuals who are mostly the same race as oneself, the results showed that communities of color ended up living nearer to disamenities, especially in the stronger-preference trials. Thus, they conclude:
Our most consistent finding in this research is that most of the common explanations for environmental injustice, like discrimination and mobility, have marginal effects that interact with one key facet of the story that had heretofore not been considered—our own preferences to live near others like ourselves. (Campbell, et al., 2015, 192)
The trials where individual agents held the highest similarity preferences and where economic ranges reflected census data showed the greatest disparity in environmental quality gaps between races.
This is not to say, stated the authors, that economic factors do not come into play. For one thing, they found that poorer white communities also ended up closer to disamenities. Second, the chapter combining EJ ABM with GIS showed that polluting firms may end up siting in communities without the collective means to combat the incoming firm due to lack of education. More educated communities, and those with greater wealth (and, presumably, more time on their hands), are more likely to protest polluting firms and other perceived disamenities—even environmentally-friendly ones. If individuals’ preferences for living near people of their own race reduced, education and economic factors could still create disparities in environmental quality. The authors state that Millennials seem to have less issues with living near people of different races, and in fact value diversity, so it may be that the problem of environmental justice as related to race may dissipate in coming decades if we can all overcome the preference to live nearby only those who look like us. Interestingly, the authors pointed out that in trials where the minority group was wealthier than the majority, the minority group ended up in more advantageous locations. They likened this to the situation in South Africa under apartheid.
Campbell, et al. (2015) also helpfully ran trials aimed at discovering which pollution mitigation and remediation policies prove most effective. They found that zoning is helpful in increasing environmental quality overall, but some zoning laws were more equitable than others. Also, one caveat is that by decreasing the gap in environmental quality, sometimes some communities’ environmental quality reduced while others increased. Remediation of polluted sites, however, increased environmental quality for everyone, as well as, in some cases, closing the gap of environmental quality between different populations. According to this study, placing and enforcing zoning laws from the beginning of a firm’s residency on a site and requiring cleanup of already-polluted sites were the best strategies for reducing the gap between communities’ environmental quality, and increasing overall environmental quality for everyone.
While the authors acknowledge the limited number of factors that can be combined in any modeling software, I found this approach to looking at environmental justice extremely useful, and I think it has potential to aid in understanding and addressing many areas of environmental injustice in the future. In some ways, the results may sound frustratingly like blaming the victim, showing that it is more the personal preferences of individuals that make the difference, and less the actions of evil corporations (who are easy to demonize). I think what these results show is an internalized, systemic form of racism which, although not intended by many of us, continues to pervade the US society: we choose to live near people who are like us, and due to our country’s history, this keeps people of color in communities with fewer educational and economic opportunities and nearer to higher levels of environmental degradation. How can we begin to address this? The answers are surprisingly simple, say Campbell, et al. (2015): clean up polluted areas, and create zoning laws that keep pollution to a minimum in any neighborhood. Be at least a little bit more willing to live near those who are racially different from ourselves.
The authors pragmatically assume that these polluting firms are going to be part of our future no matter what, but I would add that creating cleaner ways to produce the things we actually need, and not producing things that harm the environment, would also be a step in the right direction. If we don’t create unhealthy and environmentally degraded places, environmental injustice won’t be a problem. Therefore, making choices about what we buy and knowing how it is produced are important elements in our ability to reduce environmental injustice. These are personal choices that emphasize the individual, but they also impact the system as a whole, and give aggregates of individuals the power to influence the entire system. This is the premise on which the agent-based model is formed: our individual actions alone aren’t that significant, perhaps, but in aggregate, properties emerge that cannot be predicted by looking solely at individual beliefs.
This book will be helpful to academics interested in working on the wicked problems of systemic environmental injustice and breakdown of ecosystems. Graduate students and researchers alike will find the step-by-step explanations of how Campbell, et al. (2015) built this model very clear and helpful for replicating or expanding on this agent-based model. Those less interested in modeling and more interested in mitigating environmental injustice will also find this book’s conclusions illuminating, and the results can be used to advocate for better zoning laws and for pollution remediation projects. I agree with the authors’ assessment that environmental justice literature seems to be at an impasse, since it can describe the injustice that is happening, but can’t explain why. This study goes a long way toward helping us understand why, and providing some suggestions for ways to improve the situation.