This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

  1. Home
  2. ASSA Annual Meeting
  3. 2024
  4. Program
  5. Health, Human Capital, and Food
« Back to Results

Health, Human Capital, and Food

Paper Session

Saturday, Jan. 6, 2024 10:15 AM - 12:15 PM (CST)

Grand Hyatt, Bonham D
Hosted By: Association of Environmental and Resource Economists
  • Chair: Shana McDermott, Trinity University

Leveraging Unsupervised Machine Learning to Examine Women's Vulnerability to Climate Change

German Caruso
,
World Bank
Melanie Gross
,
World Bank
Marcos Puig Insua
,
World Bank
Valerie Mueller
,
Arizona State University and IFPRI
Alexis Villacis
,
Arizona State University

Abstract

A leading source of income risk in Malawi is the occurrence of droughts. Liquidity-constrained households engage in early age marital contracts as a form of household consumption-smoothing This can force young women to transition toward low-return domestic activities rather than high-return schooling activities and raise fertility rates affecting their ability to engage in paid employment. We estimate the effect of drought on early age marital decisions, schooling, and fertility rates in Malawi to inform expectations of future labor force losses attributable to climate change.

The impact of drought exposure is based on difference-in-difference estimates that rely on big data and machine learning techniques. We use administrative data from the 2008 and 2018 Malawi censuses and treatment status is determined by a k-means, machine learning algorithm. The algorithm classifies drought based on previous exposure to rainfall anomalies derived from the Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) data. We compare climate impact estimates based on the drought indicators established objectively from the k-means algorithm to more traditional measures implemented in the literature. Robustness checks and falsification exercises are used to support the interpretation of the drought impacts as causal. We find that young women exposed to a drought five years prior to their interview were 5 percentage points more likely to be married by 18 than those living in non-drought areas, Increases in early age marriage coincide with a rise in fertility (3 percentage points) and decline in the completion of primary (2 percentage points) and secondary school (1 percentage point) which can affect labor force participation rates. We use these figures to project that the increased odds of early age marriage by 2100 under the extreme RCP scenario will induce 3.3 million women to exit the labor force.

Cognitive Effects of Early Life Exposure to Clean Cooking Transitions

Emily L. Pakhtigian
,
Pennsylvania State University
Jennifer Orgill-Meyer
,
Franklin and Marshall College

Abstract

Between 2007-2012, Indonesia embarked on a wide-spread policy effort to transition household cooking fuels away from biomass and kerosene and towards liquid petroleum gas (LPG. Despite nearly a decade passing since the transition was completed, our understanding of the impacts of this program are incomplete, particularly the long-term impacts of this transition on children who experienced this clean cooking transition in early life. In this paper, we exploit variation in the rollout of the Kerosene-to-LPG conversion program across Indonesia to estimate its impacts on children's cognition. We use data on over 8000 children born within a decade of the conversion program and a difference-in-differences cohort study to identify the program's effects on children's cognition. Specifically, we compare the scores of a standard, 12-point cognitive test among children born within five years of the policy rollout (i.e., children exposed to clean cooking during their first five years of life) to children born between five and ten years before the policy rollout (i.e., children exposed to clean cooking after their first five years of life. Our preliminary results show that for each addition year of early life exposure to clean cooking, children's cognitive scores increase by 0.71 points; this effect represents an 8.3 percent increase in cognitive scores for children who were exposed to clean cooking for five early life years, relative to children exposed for zero early life years. We find that these impacts are larger for female children. Girls who were exposed to clean cooking for five early life years have cognitive scores that are 14.8 percent higher than girls who were not exposed to clean cooking during the early life period. This work contributes to a growing literature assessing the cognitive impacts of environmental health technology adoption and demonstrates the gendered effects of such investments.

Toxic Recycling: Used Lead-Acid Battery Processing's Negative Effect on Test Scores in Mexico

Erin Litzow
,
University of British Columbia
Tatiana Zárate
,
Texas A&M University
Bianca Cecato
,
Smart Prosperity Institute
Mauricio Romero
,
Mexico Autonomous Institute of Technology

Abstract

Lead is a very toxic pollutant; even trace amounts can affect the cardiovascular, reproductive, and neurological systems. It is especially harmful to children, whose brains are rapidly developing. Given the high cost of lead pollution, many countries, especially those in the rich world, have strictly regulated sources of lead exposure. We study the effects of a 2009 change in U.S. regulation which lowered the National Ambient Air Quality Standard (NAAQS) for lead and caused used lead-acid battery recycling (ULAB) to shift to Mexico. We quantify the negative externality that U.S. environmental regulation had on the Mexican population. Specifically, we examine the impact of increased exposure to lead pollution from ULAB recycling on academic performance among students in Mexico. We use data from Mexico’s ENLACE standardized test in Spanish and mathematics, conducted annually between 2006 and 2013. We implement a difference-in-difference design, comparing test scores before and after the 2009 NAAQS reduction between schools located close to and farther away from Mexican ULAB recycling facilities. Using this difference-in-difference design, we find substantial evidence that exposure to lead-acid battery recycling reduces academic achievement. Specifically, we find that schools located within four miles of a battery-recycling facility have lower math and Spanish scores relative to schools located slightly farther away after the 2009 U.S. policy change. This decline translates to an approximately 3% (2%) decrease relative to the mean and a roughly 15% (11%) reduction relative to the standard deviation in math (Spanish) scores. Rich countries are continuing to crack down on lead exposure, and recycling activities in places with weaker regulatory oversight are only expected to increase. Overall, our findings add to the literature estimating the costs of lead pollution exposure and shed light on how international regulation can adversely affect domestic well-being in contexts with relatively weaker environmental regulations.

Measuring the Cascading and Interstate Impact of a Drought Event on the U.S. Domestic Food Supply Chain

Hyungsun Yim
,
University of Illinois-Urbana-Champaign
Sandy Dall'erba
,
University of Illinois-Urbana-Champaign

Abstract

According to the current Intergovernmental Panel of Climate Change report, climate change will increase the probability of occurrence of droughts in some areas. Recent contributions at the international (e.g. Costinot et al., 2016) and US domestic level (e.g. Dall’erba et al., 2021) indicate that trade is expected to act as an efficient tool to mitigate the adverse effect of future climate conditions on agriculture. However, no contribution has focused on the capacity of trade to mitigate the impact of weather shocks throughout the entire domestic food supply chain. The U.S. is an obvious choice given that many climate impact studies focus on its agriculture and less than 10% of the U.S. agrifood production is exported (Bureau of Economic Analysis, 2023). Combining the recent FAF state-to-state trade flow dataset of the Bureau of Transportation Statistics with detailed drought records at a fine spatial and temporal resolution, this paper starts with a gravity model of trade to highlight that interstate exports of agricultural products (crops, fruits, vegetables, livestock) increase as the destination state experiences more drought and inversely in the origin state. Next, state-level estimates of the food manufacturing production function indicate the extent to which food processing in one state depends on agricultural inputs grown locally and in other states so that a drought event in any state perturbates the entire domestic food supply chain. Projections based on future weather data will allow us to measure the gap between the losses expected in the food manufacturing sector without trade with the lesser losses expected once interstate trade has been accounted for. The difference between the two forecasts will provide us with an accurate measurement of the capacity of trade to mitigate climate change impact. Future research shall consider other extreme weather events such as early frost and floods.

Discussant(s)
Ariel Ortiz-Bobea
,
Cornell University
Jayash Paudel
,
University of Oklahoma
Huan Li
,
North Carolina A&T State University
Shana McDermott
,
Trinity University
JEL Classifications
  • Q1 - Agriculture
  • Q5 - Environmental Economics