Power Plant Locations and Unemployment Rates

Introduction

The placement of energy grids and power plants across the United States impacts not only the physical landscape but also the economic, environmental, and social fabric of surrounding communities. These plants are often concentrated in economically disadvantaged or historically marginalized areas, raising critical concerns of environmental justice and energy equity. Although power plants can spur local employment and economic growth, they bring disproportionate environmental burdens such as air and water pollution to the very communities that do not have the resources to advocate for cleaner alternatives. The benefits of energy infrastructure, including access to reliable power and job opportunities, are not always equitably distributed, making it essential to examine who bears the costs and who reaps the rewards. In order to assess these disparities, this analysis maps the locations of power plants across the United States, overlays them with regional unemployment rates, and explores potential correlations. By doing so, we aim to highlight structural patterns and equity gaps in the ways that energy infrastructure intersects with socioeconomic vulnerability and environmental justice outcomes.

Motivation and Background

To understand these patterns more comprehensively, it is essential to examine the context in which energy infrastructure is developed and deployed. The distribution of power plants across the United States reflects and reinforces patterns of economic disparity, environmental burden, and policy neglect. As found by Johnson and Kim (2020), these facilities are often built in regions with higher unemployment levels and limited economic opportunity, where local governments may use infrastructure projects as a source of temporary job creation and tax revenue. This siting pattern, however, is not just a matter of economic pragmatism or feasibility, it is also shaped by longstanding practices of systemic inequality and environmental injustice. Scholars such as Robert Bullard (2000) have documented how communities of color and low-income populations have disproportionately shouldered the environmental burdens of industrial development, including energy production, while receiving fewer of the corresponding economic or infrastructural benefits (Bullard, 2000). 

Despite promises of job creation, the long-term economic benefits of power plant development are often overstated or unevenly distributed. For example, research conducted by Carley and Konisky (2020) concluded that while some fossil-fuel plants offer limited employment during the construction phase, the number of permanent jobs created is relatively small and increasingly threatened by automation and industry shifts toward renewables. Further, many of these jobs require specialized training not always accessible to residents in the surrounding communities. Even though there may be employment opportunities, they may not lead to lasting economic growth unless they are paired with broader investments in education, public health, and infrastructure. This is highlighted by a study on coal-dependent communities in Appalachia by Black, McKinnish, and Sanders (2005), who found that while mining booms did reduce unemployment temporarily, they also coincided with rising mortality rates due to pollution-related illnesses, occupational hazards, and declining educational outcomes, reflecting the trade-offs of these power plant development strategies.

Alongside these factors, these facilities also often impose substantial environmental and health costs. Power plants, particularly those that rely on coal and natural gas, are significant sources of air pollutants such as sulfur dioxide (SO₂), nitrogen oxides (NOₓ), and particulate matter (PM2.5), all of which are linked to respiratory diseases, cardiovascular conditions, and premature death (Graff Zivin and Neidell, 2012). For example, a study conducted in 2017 by Currie, Greenstone, and Meckel found that children living near coal-fired power plants experienced significantly higher rates of hospitalization for respiratory conditions, particularly asthma (Currie et al., 2017). These impacts are compounded by the fact that such harms are rarely distributed evenly. Communities located near energy production sites often already face other disadvantages including food insecurity, inadequate housing, and limited access to healthcare which makes them especially vulnerable to the cumulative effects of pollution. As Hernández (2016) explains, energy insecurity and housing insecurity often compound, creating “energy precarity” that disproportionately affects marginalized households and undermines long-term health and stability (Hernández, 2016).

These disparities are the result of policy choices that have historically excluded vulnerable populations from environmental decision-making. Siting processes frequently lack engagement from the surrounding local community, and environmental impact assessments often do not account for cumulative exposures. As a result, energy infrastructure becomes yet another mechanism through which environmental and economic harms are concentrated in politically powerless communities. According to the U.S. Environmental Protection Agency’s 2022 EJScreen tool, census tracts with the highest proximity to energy-related emissions also rank among the most socioeconomically disadvantaged (EPA, 2022). This supports the claim that infrastructure planning, particularly in the energy sector, must be evaluated for its impact on equity and justice to the surrounding community.

With these intersecting concerns, there is an emerging area of research calling for the integration of spatial and socioeconomic analysis in infrastructure policy. For instance, Carley and Konisky (2020) argue that energy justice must include both distributive and procedural components: it is not enough to ask where energy infrastructure is built, we must also ask how these decisions are being made and who gets a seat at the table (Carley and Konisky, 2020). Similarly, recent studies from the National Renewable Energy Laboratory emphasize the importance of mapping energy burdens alongside demographic indicators as a way to guide more equitable transitions to clean energy (Brown et al., 2022). Although much of this work has focused on the shift toward renewables, fewer studies have systematically examined the current geographic distribution of traditional power plants in relation to economic vulnerability.

This analysis aims to contribute to that gap. By mapping the locations of power plants across the United States and overlaying them with regional unemployment data, we seek to identify patterns that may indicate whether plant siting decisions are reinforcing existing economic and environmental disparities. This kind of spatial analysis is a crucial step for more equitable energy and environmental policy shifts. Understanding where energy infrastructure is built, who benefits, and who is harmed is essential to designing policies that promote not just energy efficiency but also environmental justice and economic equity.

Methodology

Data Sources

By Chloe Cowan, Updated April, 2025

Data Cleaning and Merging

The data cleaning began with standardizing text fields to ensure uniformity across the different datasets. All county names and state identifiers were converted to lowercase and stripped of trailing whitespace to prevent mismatches during joins. This was important given variations in naming conventions across sources (e.g. “St. Louis” and “Saint Louis”). Power plant geographic coordinates were validated and corrected using data from the U.S. Energy Information Administration (EIA) Form 860, which provided authoritative location and capacity data for active plants.

To merge the datasets, we used both exact and fuzzy matching techniques. County-level socioeconomic data including unemployment rates and population estimates were matched to the power plant dataset using ‘CountyFIPS’ as the common key when available. However, some entries lacked FIPS codes or had inconsistently formatted county names, so fuzzy matching was used as a fallback method. The FuzzyWuzzy Python library was applied to compute string similarity ratios, with a matching threshold to minimize false positives. We also used manual spot checks to check matches in borderline cases. In cases where geographic coordinates were missing or implausible (e.g., coordinates outside the expected state), external sources such as OpenStreetMap and the EIA’s Plant Master File were used to fill gaps or correct errors. Duplicate entries which arose from overlapping data sources were identified using a combination of county, plant name, and capacity, and were removed to avoid double-counting.

The final merged dataset contains standardized columns for county names, FIPS codes, state names, unemployment rates, population figures, power plant geocoordinates, generation capacity, primary fuel type, and operational status. These fields were selected for the spatial analysis of how energy infrastructure aligns with economic vulnerability. The analysis is a visual overlay of power plant locations with county-level unemployment heatmaps, allowing for exploratory pattern recognition across geographic regions. While no formal statistical tests (such as correlation or regression analysis) were conducted, the visual patterns provide insight into associations between infrastructure siting and economic conditions. These findings offer a valuable foundation for future quantitative work. Although the dataset is comprehensive, some other limitations remain, particularly with missing data in rural counties and the precision of fuzzy matches for counties with similar names in different states. These limitations were documented and flagged during processing. The cleaned and integrated dataset supports a reliable basis for identifying potential equity concerns and structural disparities in regional energy development.