Faculty Details

Photo of Edson R Severnini

Edson R Severnini

Assistant Professor of Economics and Public Policy

Voice: (412) 268-2329
Email: edsons@andrew.cmu.edu
Personal Website


Edson Severnini joined Carnegie Mellon University in the Fall 2013, as an Assistant Professor of Economics and Public Policy. His interests lie at the intersection of energy, economics, and the environment.


Research Interests

Primary Fields: Urban, Energy, and Environmental Economics, and Economic History Secondary Fields: Applied Econometrics, Labor Economics


B.A., Economics (summa cum laude), University of São Paulo (USP) (2004).
M.Sc., Economics, Pontifical Catholic University of Rio de Janeiro (PUC-Rio) (2007).
Ph.D., Economics, University of California at Berkeley (UC Berkeley) (2013).
Dissertation: Essays in Applied Microeconomics
Committee: David Card, Enrico Moretti, Patrick Kline, Steven Raphael

Working Papers

The Power of Hydroelectric Dams: Agglomeration Spillovers

 How much of the geographic clustering of economic activity is attributable to agglomeration spillovers as opposed to natural advantages? I present evidence on this question using data on the long-run effects of large scale hydroelectric dams built in the U.S. over the 20th century, obtained through a unique comparison between counties with or without dams but with similar hydropower potential. Until mid-century, the availability of cheap local power from hydroelectric dams conveyed an important advantage that attracted industry and population. By the 1950s, however, these advantages were attenuated by improvements in the efficiency of thermal power generation and the advent of high tension transmission lines. Using a novel combination of synthetic control methods and event-study techniques, I show that, on average, dams built before 1950 had substantial short run effects on local population and employment growth, whereas those built after 1950 had no such effects. Moreover, the impact of pre-1950 dams persisted and continued to grow after the advantages of cheap local hydroelectricity were attenuated, suggesting the presence of important agglomeration spillovers. Over a 50 year horizon, I estimate that at least one half of the long run effect of pre-1950 dams is due to spillovers. The estimated short and long run effects are highly robust to alternative procedures for selecting synthetic controls, to controls for confounding factors such as proximity to transportation networks, and to alternative sample restrictions, such as dropping dams built by the Tennessee Valley Authority or removing control counties with environmental regulations. I also find small local agglomeration effects from smaller dam projects, and small spillovers to nearby locations from large dams. Lastly, I find relatively small costs of environmental regulations associated with hydroelectric licensing rules.

Who to Marry and Where to Live: Estimating a Collective Marriage Market Model

 I study the joint choice of spouse and location made by individuals at the start of their adult lives in the U.S. I assume that potential spouses meet in a marriage market and decide who to marry and where they will live, taking account of varying economic opportunities in different locations and inherent preferences for living near the families of both spouses. I develop a theoretical framework that incorporates a collective model of household allocation, conditional on the choice of spouse and location, with a forward-looking model of the marriage market that allows for the potential inability of spouses to commit to a particular intra-household sharing rule. Estimation results for young dual-career households in the 2000 Census point to three main findings. First, I find excess sensitivity of the sharing rule that governs the allocation of resources among couples to the conditions in the location they actually choose, implying that spouses cannot fully commit to a sharing rule. Second, I show that the lack of commitment has a relatively larger effect on the share of family resources received by women. Third, I find that the failure of full commitment can explain nearly all of the gap in the intra-state migration rates of single and married people in the U.S.

Research In Progress

Air Pollution, Power Grid, and Infant Health: Evidence from the Shutdown of Nuclear Power Plants in the Tennessee Valley Authority in the 1980s

 When environmental regulations focus on a subset of power plants, the ultimate goal of human health protection may not be reached. Because power plants are interconnected through the electrical grid, excessive scrutiny of a group of facilities may generate more pollution out of another group, with potential deleterious effects to public health. I study the impact of the shutdown of nuclear power plants in the Tennessee Valley Authority (TVA), in the 1980s, on infant health outcomes. After the Three Mile Island accident in 1979, the Nuclear Regulatory Commission (NRC) intensified inspections in nuclear facilities leading to shutdown of many of them, including Browns Ferry and Sequoyah in the TVA area. I first show that, in response to the shutdown, electricity generation shifted mostly to coal-fired power plants within TVA, increasing pollution in counties where they were located. Second, I find that babies born after the shutdown had lower birth weight in those counties with coal-fired power plants. Third, I provide suggestive evidence of heterogeneity in those effects depending on how much more electricity those coal-fired facilities were generating in response to the shutdown. These findings may shed light on some potential consequences of the retirement of the San Onofre Nuclear Plant in California, and the ongoing nuclear power phase out in Germany, greatly intensified after the Japanese Fukushima disaster. 

Short and Long Run Effects of Disease and Pollution: Evidence from the 1918 Influenza Pandemic Interacted with Air Pollution from Fossil-Fuel Power Plants (joint with Karen Clay (CMU) and Joshua Lewis (University of Montreal))

 Urban outdoor air pollution is responsible for 1.3 million premature deaths annually (WHO, 2009). In addition to these direct effects, poor air quality may also be harmful to the immune system, lowering underlying health, and leaving individuals more susceptible to the consequences of a negative health shock. In this study, we examine whether exposure to air pollution exacerbates the effects of a negative health shock. We investigate the interaction effect between the 1918 influenza pandemic and pre-determined levels of air pollution arising from fossil fuel power plants, based on a newly collected data on those plants in the 1910s. We study the short run impact on mortality using county level vital statistics data, the medium run effects on labor market outcomes and educational attainment using county level data computed from the 1940 U.S. Census, and the impact on the long run outcomes studied by Almond (2006) using state level data from the 1960-80 decennial U.S. Census. To the best of our knowledge, this is the first attempt to understand how pollution interacts with health conditions in determining short and long-term outcomes.

Benefits and Costs of Electricity in pre-Clean Air Act United States (joint with Karen Clay (CMU) and Joshua Lewis (University of Montreal))

China and India are opening thermal power plants at an extremely rapid pace. Many have little or no emission controls. This is similar to the United States in the age of electrification, particularly the period from the 1930s-1960s, when transmission lines and economies of scale allowed utilities to build large generation plants outside of cities, and generation plants did nothing to limit their emissions beyond building moderately tall smokestacks. Using county-level data for 1930-1960, this paper investigates the economic gains from local electrification and the pollution costs associated with fossil fuel generation. The empirical analysis combines newly digitized data on power generation and transmission with county-level data on mortality and land values. In the baseline analysis, we relate annual changes in county-level all age and infant mortality rates to the construction of fossil-fuel power plants in a county-fixed effects framework. Local availability dictated the types of fuels used, and fuels varied in their emissions from low emission natural gas to high emission eastern bituminous coal. We employ a difference-in-differences approach to compare counties with power facilities using fossil fuels with different emission levels per megawatt hour. We also compare the net benefits of local thermal power generation to local hydropower generation and expansion of the power grid. This analysis sheds light on historical consequences of electrification in the United States, and is relevant for ongoing electrification policy in developing countries.

The Effects of the Clean Air Act on Productivity in the Electricity Industry: Evidence from the U.S. 1938-1993 (joint with Karen Clay (CMU) and Joshua Lewis (University of Montreal))

 The costs of environmental regulations have been widely debated in the U.S. since the passage of the Clean Air and Water Acts beginning in the 1960s.  Using data from 1972-1993, a recent paper by Greenstone, List and Syverson (2012) showed large productivity losses in manufacturing from air quality regulations. The authors’ paper also highlighted the need for data spanning longer time periods, since it was impossible to follow plants during a significant portion of the period of regulatory change.  This study provides new evidence on the effects of regulation by digitizing and analyzing data on steam power electricity plants covering the period 1938-1993. Federal Power Commission data begins in 1938. Data in the Federal Power Commission Reports is very detailed and so will allow analysis of the effects of regulation on steam power plants that burn coal, natural gas, and oil.  By following specific plants over time, one can observe fuel switching by existing plants, fuel choices by new plants, and changes in output resulting from regulation-induced technological modifications.  These are relevant for understanding productivity losses in electricity generation in developing countries that are beginning to implement or strengthen enforcement of air quality regulations. 

Rural Electrification, Urban Growth, and Economic Development: Evidence from the Rollout of the U.S. Power Grid (joint with Joshua Lewis (University of Montreal))

 Between 1930 and 1970, the Southern United States experienced a striking transformation, moving from an economy characterized by low-wage rural employment to a modern industrialized economy. This paper examines the role that rural electrification played in this transition. We study whether increases in agricultural productivity associated with rural electrification lowered the demand for farm labor, allowing for a shift from the “Old South” to the “New South” (Wright, 1986). The empirical analysis relies on a newly digitized dataset of all major transmission lines in 1923 and 1962, along with detailed information on the timing of power plant openings. To begin, we study the impact of electrification on farm productivity – as measured by land values and average farm size – in a county-level fixed effects framework. Next, we use county-level information on rural and urban populations to examine the link between rural electrification and urban growth. This analysis is supplemented by a study of individual-level migration patterns, based on retrospective data from the census.  Our study contributes to our understanding of the Southern economic transition throughout this period, and the role of electricity as an engine of urban growth.

Energy Mix and Political Affiliation: Evidence from U.S. States Since 1960 (joint with Karam Kang (CMU))

 The two major political parties in the U.S. appear to take opposite views on energy and environmental policies, such as whether CO2emissions should be regulated under the Clean Air Act. We intend to investigate the historical relationship between the energy mix and the political environment related to party affiliation of state government in continental U.S. If we find that a specific political environment affects the role of fossil fuel in the energy mix, measured in terms of share of fossil-fuel installed capacity and electricity generation, we question how this impacts the distribution of emissions across the nation. Further, we study how political environment affects the decisions of firms regarding where to open a new plant or expand existing ones.

A Note on the Trade-off between Ecosystem Preservation and Air Quality: Evidence from Hydroelectric Licensing Rules

 Do environmental regulations aimed at preserving natural ecosystems really protect the environment? The answer seems to be not really. I present suggestive evidence that, while hydroelectric licensing rules conserve the wilderness and wildlife by restricting the development of hydro projects, they lead to more greenhouse gas emissions. Basically, land conservation regulations give rise to a replacement of hydropower, which is a renewable, non-emitting source of energy, with conventional fossil-fuel power, which is highly polluting. I find that, on average, each megawatt of hydropower that is not developed because of these regulations induces the same amount of carbon dioxide that a U.S. coal-fired plant would emit in producing a megawatt of electricity. Environmental regulations focusing only on the preservation of ecosystems appear to encourage electric utilities to substitute dirtier fuels for electricity generation.