Initiative on Equity in Energy and Environmental Economics: 2023-2024 Graduate Student Project Descriptions
Long-run Impacts of Historical USDA Discrimination: Examining the Agricultural Adjustment Act of 1933
Sheah Deilami
This project estimates the impacts of historical discrimination by the United States Department of Agriculture using the 1933 Agricultural Adjustment Act. Implementation of the AAA allowed white landowners to retain benefits that should have been passed on the Black tenants. I aim to exploit county differences in AAA spending in agricultural areas to measure the impacts on Black farmer exit, and explore the possibility of USDA discrimination as a force contributing to the Great Migration. Additionally, I plan to track county committee membership using archival data to link members to AAA appeal decisions and benefit distribution. This research relates to a broader interest in the contribution of historically discriminatory institutions to Black farmer land loss in the United States.
How does minority representation impact local environmental outcomes?
Kendra Marcoux
Racial minority groups and low-income households tend to be disproportionately exposed to various measures of environmental harm. This gap is often referred to as environmental inequity and is thought to be attributable to four separate mechanisms: residential sorting, firm sorting, coordination between residential and firm sorting, and discriminatory policy and enforcement. Few studies have explored the role that discriminatory politics plays, despite elected officials playing a large role in setting and enforcing local environmental policy. At the same time, racial minorities are historically underrepresented in U.S politics, and minority representation can lead to a higher provision of public goods in majority-minority neighborhoods. I would like to understand the intersection of these trends and ask the question: What is the effect of minority representation on environmental inequity?
Forecasting far-future emissions using Bayesian tools
Emily Martell
My project aims to forecast global emissions using a Bayesian model of GDP per capita, population, and carbon intensity (carbon per unit of output). Previous forecasts by Raftery et al. (2017) use strong assumptions about the evolution of these variables. I aim to improve the process for the evolution of carbon intensity. First, I will understand what trends exist in the data using country-level data from the Global Carbon Project, Penn World Tables, and Intergovernmental Panel on Climate Change. Then, I will make modeling decisions to forecast carbon intensity.
Natural Disasters and Housing Wealth Accumulation: A Long-run View
Elena Ojeda
The frequency and severity of natural disasters are growing. Understanding how marginalized populations respond to different types of disasters is crucial for effective policy response. This project explores the long-run consequences of exposure to varying types of natural disasters using linked Census records from 1900 to 1940, real estate, insurance, and federal disaster assistance data. The project studies how individuals respond to a natural disaster in the short-run through decisions on homeownership and location preferences, and if these behaviors are passed on to their children. I start by asking, what is a homeowner’s short-run response to a local natural disaster? Then, using linked Census data I follow their children to understand their location and housing decisions. Do the children of parents exposed to a natural disaster also become homeowners? I plan to use a spatial regression discontinuity design across the boundary of the impacted areas to identify the effect of a natural disaster on those who lost their home.
Food Waste and Food Deserts: Evidence from Walmart's “Imperfect but Good" Produce Initiative in Mexico
Daniela Paz
Approximately 15.3% of all food produced globally goes to waste at the farm stage, resulting in significant environmental consequences such as the emission of greenhouse gases and the depletion of natural resources. At the same time, low- and middle-income households face challenges in accessing nutritious food, with projections indicating that obesity rates could exceed 51% of the population by 2030, leading to a range of adverse health impacts such as diabetes, heart disease, and stroke. Food insecurity is particularly acute in developing countries with high poverty rates and limited access to healthy food. Recently, supermarkets in different countries started to sell “misshape” food - fruits and vegetables that do not meet traditional color or shape standards- at a reduced price. In this project, we aim to study if this initiative can impact these coexisting problems, reduce nutritional inequality in consumption in developing countries, and decrease food waste at the farm level.
Local Spillovers and Managed Retreat: Evidence from Home Buyout Programs in Harris County, Texas
Suvy Qin
This project aims to estimate the local housing market impacts of Harris County’s home buyout program. By enabling government agencies to purchase and demolish flood-prone properties, buyout programs can permanently reduce flood risks, but little is known about their effects on neighboring properties. This project studies the effectiveness and equity implications of home buyout programs by combining parcel-level data on residential properties, buyout locations, historical flood maps, flood insurance claims, and demographic information. I first construct counterfactual buyout areas using Harris County’s buyout eligibility criteria to identify properties that could become buyouts in the future. I then estimate the spillover effects of buyouts on the value and sales likelihood of neighboring (non-buyout) homes. These estimates allow me to then consider the welfare consequences of buyouts, with a focus on equity.
Optimal Transportation Investment and Environmental Externalities
Alice Schmitz
This project seeks to build a spatial equilibrium model to estimate the effects of highway systems on deforestation in the Brazilian Amazon. The model considers not just how state road construction influences land use, but also how state roads give rise to the construction of an informal road network. Incorporating the social costs of deforestation into my analysis to determine optimal highway routing might identify routes which circumvent natural areas, or suggest the partial construction of a route which will likely be completed by informal road construction. More broadly, I hope to contribute to the economic geography's literature on optimal transportation investment by considering how environmental externalities should be included in welfare analyses.
Does the National Flood Insurance Program Crowd Out Investments in Mitigation?
Max Snyder
Governments are increasingly interested in policies which protect households from flooding. One recent federal initiative, Risk Rating 2.0, uses new measurements of flood risk to adjust the prices that households pay for flood insurance. The initiative aims to create more equitable flood insurance premiums by increasing prices for high-risk homes and decreasing rates for low-risk homes. Policymakers have also expressed hope that the new pricing scheme will reduce climate risk. This project will evaluate whether this initiative can 1) promote social equity by lowering the proportion of insurance claims distributed to high-value homes, and 2) promote climate resilience by encouraging private and public investments in flood mitigation. In doing so, I aim to contribute to broader conversations about the uses and consequences of scientific advances in measuring climate risk.
Optimal Siting of Air Quality Monitors in Ambient Pollution Regulation
Aaron Watt
This project starts with a simple mathematical model of air pollution regulation and builds computational tools around it to estimate the optimal locations for future air pollution monitors. A key part of the regulation modeling is understanding how pollution measurements from the monitors are used to regulate areas with high pollution. One reason this type of analysis is relatively rare is the high computational complexity, so we will be exploring three tool sets that can help us: (1) high performance computing resources (Savio); (2) programming packages that make some of the computations much quicker; and (3) a type of machine learning that can be trained to output optimal locations given input data. We will be adding mathematical complexity to our model and eventually take it to US data from the EPA and NASA. Julia is the main programming language for this project -- a relatively young, high performance, scientific programming language.