Current work
Indian air and water pollution
Indian cities are among the most polluted in the world. A major driver of this pollution is agriculture. In a series of projects I am examining the role of anti-poverty efforts in increasing the use of burning to clear crop residue from fields, the consequences of crop burning for health outcomes, and the impact of changes in agricultural practices on water pollution.
Working papers:
Man or Machine ? Environmental Consequences of Wage Driven Mechanization in Indian Agriculture
This paper uses an exogenous shock to wages from the world’s largest anti-poverty program to show that higher wages can lead to increased air pollution, likely by inducing farmers to shift into a labor-saving and mechanized production process. Using a difference-in-differences approach on the staggered roll-out of India’s Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA), combined with data on nearly 1 million fires, the paper shows that the frequency of agricultural fires increases by 21 percent after the shock. The increase in fires is concentrated in districts that appear more likely to mechanize the harvest. MNREGA did not lead to changes in area planted or tonnage produced in fire intensive crops. The estimates show that nationally, the shock increased the rate of particulate emissions from biomass burning by 30 to 50 percent. The results suggest that absent policies to correct for environmental externalities of mechanization at all stages of development, labor market shocks may lead to inefficient levels of mechanization.
Air pollution reduces economic activity: Evidence from India (with Rishabh Choudhary and Dhruv Sharma)
Exposure to fine particulate pollution (PM2.5) increases mortality and morbidity and reduces human capital formation and worker productivity. As a consequence, high levels of particulate pollution may adversely affect economic activity. Using a novel dataset of annual GDP changes in Indian districts, we investigate the impact of changes in the level of ambient PM2.5 on district-level GDP. Using daily temperature inversions as an instrument for pollution exposure we find that higher levels of particulate pollution reduce GDP. The effect is non-trivial - the median annual increase in PM2.5 levels reduces year to year changes in GDP by 0.56 percentage points.
On-going projects:
The impact of NREGA on Indian groundwater availability (with Hemant Pullabhotla)
Depletion of groundwater is a major challenge in India. We examine how a major rural public works program (NREGA) that financed the construction of surface water infrastructure that may have plausibly increased aquifer recharge rates impacted groundwater levels. Using a difference-in-differences approach on the staggered roll-out of NREGA we show that groundwater levels increased after the implementation of NREGA. These increases were concentrated in states that constructed the largest number of NREGA financed surface water projects. The increases we observe in groundwater appear to have led to increases in the irrigated area of high value crops and greater overall irrigation during the dry season. We estimate positive increases in farmer profits due to these changes in groundwater but the change in profits is small relative to overall NREGA expenditures.
Indian court decisions and air pollution (with Shareen Joshi, Daniel Chen, Olexiy Kyrchenko, and Shashank Singh)
Using the full corpus of Indian court cases over 40 years we examine how the relationship between the Indian judiciary and pollution regulators impacts observed levels of ambient air pollution throughout India.
Credit access and firm pollution in India (with Piyush Gandhi, and Teevrat Garg)
Climate change, pollution, and crime in Texas
It is well known that high temperatures can lead to increases in crime. Using a new, comprehensive data-set on arrests and court actions in Texas co-authors and I are examining how this relationship is modified by neighborhood characteristics.
Working papers:
Adapting to Heat: Evidence from the Texas Criminal Justice System (with Valentin Bolotnyy) - Revise & Resubmit
Using administrative criminal records from Texas, we show that heat increases crime in a heterogeneous way across neighborhoods with different housing and economic characteristics. The heterogeneity allows us to predict how effective certain forms of adaptation will be at reducing the impacts of climate change on criminal activity. Our simulations show adaptation reducing, but not completely offsetting, these impacts. Differential rates of adaptation across neighborhoods will likely exacerbate the consequences of already unequal exposure to climate change across society.
Moving to Adaptation? Understanding the Migratory Response to Hurricanes in the United States (with Valentin Bolotnyy)
Using data on the paths of all hurricanes in the Atlantic Basin from 1992 to 2017, this paper studies whether migration has served as a form of adaptation to hurricane risk. The findings show that on average hurricanes have little to no impact on county out-migration, with population-weighted exposure to hurricanes increasing slightly over the sample period. Counties with high economic activity see net in-migration in the years after a hurricane. Further, return migration likely plays a role in offsetting any out-migration in the year of the storm. The intensity of pre-hurricane migration between county pairs is a strong predictor of excess migration after a hurricane, suggesting that existing economic and social ties dominate in post-hurricane migration decisions. Given existing policies and incentives, the economic and social benefits that people derive from living in high-risk areas currently outweigh the incentive to adapt to future storms by relocating across counties.
On-going projects:
Air pollution and criminal activity (with Valentin Bolotnyy)
Heat exposure, human capital, and workers
Heat reduces human performance across a range of tasks and settings. I have a series of on-going projects with Professor R. Jisung Park examining the consequences of heat exposure for workers and students across a variety of outcomes, including productivity and safety, and the role of policy in promoting adaptation to reduce the consequences of heat exposure.
Working papers:
High Temperature and Learning Outcomes : Evidence from Ethiopia (with Kibrom Tafere and Bhavya Srivastava)
This paper uses data from 2003–19 on 2.47 million test takers of a national high stakes university entrance exam in Ethiopia to study the impacts of temperature on learning outcomes. It finds that high temperatures during the school year leading up to the exam reduce test scores, controlling for temperatures when the exam is taken. The results suggest that the scores of female students are less impacted by higher temperatures compared to their male counterparts. Additionally, the analysis finds that the scores of students from schools located in hotter regions are less impacted by higher temperatures compared to their counterparts from cooler regions. The evidence suggests that the adverse effects of temperature are driven by impacts from within-classroom temperatures, rather than from indirect impacts on agriculture.
Temperature, Workplace Safety, and Labor Market Inequality (with Jisung Park and Nora Pankratz)
Using data covering the universe of injury claims from the nation’s largest worker’s compensation system (2001-2018), we explore the relationship between temperature and workplace safety and its implications for labor market inequality. Hotter temperature increases workplace injuries significantly, causing approximately 20,000 injuries per year. The effects persist in both outdoor and indoor settings (e.g. manufacturing, warehousing), and for injury types ostensibly unrelated to temperature (e.g. falling from heights), consistent with cognitive or cost-related channels. The risks are substantially larger for men versus women; for younger versus older workers; and for workers at the lower end of the income distribution, suggesting that accounting for workplace heat exposure may exacerbate total compensation inequality. We document a decline in the heat-sensitivity of injuries over the study period, suggesting significant scope for adaptation using existing technologies.
Will we adapt? Temperature, Labor, and Adaptation to Climate Change (with Jisung Park)
We explore the labor-related production impacts of temperature stress both for its own interest and to understand the role of adaptation in responding to climate change. Focusing on non-agricultural sectors in the United States, we find that hot temperatures exert a causal, negative impact on county-level payroll -- reducing payroll by several percentage points in a 2C hotter year -- with larger impacts in highly exposed industries such as construction and manufacturing. We assess differences in implied adaptation investments across regions with varying incentives for long-run adaptation, and find evidence consistent with hotter climates being better adapted to hot weather.
Press coverage:
New York Times (part 2); LA Times; Washington Post; The Guardian; Vox; Economic Times; Science News; NPR; Seattle Times; Scientific American; US Council on Economic Advisors; Voice of America; Marketplace
On-going projects:
Temperature, safety and compensating variation (with Jisung Park and Michael Dalton)
Air pollution and agriculture
Smoke exposure and crop yields (with Sherrie Wang) - Provisionally accepted
Wildfires throughout western North America produce smoke plumes that can stretch across the crop growing regions of the American Midwest. Climate change is likely to increase the number and size of these fires and subsequent smoke plumes. These smoke plumes change direct, diffuse, and total sunlight during the crop growing season and consequently influence yields of both corn and soy. We use a ten year panel of county level yields of both crops from all counties east of the 100th meridian combined with measures of exposure to smoke plumes of low and high density during the growing season to show that low density plumes enhance yields, likely due to increases in the fraction of diffuse light, while high density plumes have negative impacts on yields. Because there are more low density plumes today the net effect is a slight increase in yields on average. As climate change makes wildfires larger and more frequent the overall impact of smoke on yields is projected to be substantially more negative.