Research

Publications 

1.  "Air Pollution Exposure and Donation to Its Victims: Evidence From Online Charitable Giving" (with Peng Shen, Xincheng Wang, Yinxiao Wang, Yucheng Wang and Shuhuai Zhang) Journal of Environmental Economics and Management (2025): 103245.

2.  "Modeling Behavioral Response to Infectious Diseases Under Information Delay” (with Haosen He and Frederick Chen) Review of Economic Design (2025): 1-22.

3. "Observational Studies Generate Misleading Results About the Health Effects of Air Pollution: Evidence From Chronic Air Pollution and COVID-19 Outcomes." (with Marc N. Conte, Matthew Gordon,Nicole A. Swartwood,and Rachel Wilwerding) PloS One 19.1 (2024): e0296154.

4.  "An Agent-based Model of Elephant Crop Consumption Walks Using Combinatorial Optimization" (with Haosen He, Erin Buchholtz, Frederick Chen, and Susanne Vogel) Ecological Modelling  (2022): 464, 109852.  

5. "COVID-19's U.S. Temperature Response Profile"  (with Richard Carson, Samuel Carson, Thayne Dye, Samuel Mayfield, and Daniel Moyer) Environmental and Resource Economics (2021): 1-30. 

Working papers

"Mismeasured and Misunderstood: Unmasking the Temperature Proxy Problem in Climate Impact Estimates" with Richard T. Carson. Accepted at Journal of the Association of Environmental and Resource Economists.

Thousands of studies estimate prospective climate change impacts by examining how temperature affects a wide range of outcomes of interest. Potential measurement error in temperature variables is routinely ignored, and when acknowledged, typically assumed to be small in magnitude with a classical (i.i.d. normal) distribution, leading estimates to be attenuated toward zero. We formally document the sources of errors introduced in the construction of temperature proxy variables and show this widely held assumption is wrong using a best-case approach that treats readings at 1,120 U.S. weather stations as truth. Overall, the average downward bias in estimated impacts is substantial. However, due to the non-classical nature of the measurement errors, overestimates are not rare. Estimation problems are concentrated where attention focuses—the tails of the temperature distribution.

"Co-Pollutant Bias in Air Pollution IV Designs" with Zhenxuan Wang.  Under review.

This paper studies air pollution IV estimates when instruments move several pollutants simultaneously and provides diagnostic tools for interpreting them. We distinguish two interpretations: a stimulus effect of the named pollutant holding co-pollutants fixed, and a proxy effect of the pollution bundle indexed by that pollutant. An empirical evaluation shows that adding controls, swapping the focal pollutant, or jointly instrumenting multiple pollutants can still leave nontrivial bias. We derive conditions under which single- and multi-pollutant IV estimates support either interpretation and provide a framework for quantifying bias when they do not. A crime application illustrates the framework.

"Exposure, Vulnerability and Environmental Injustice: Air Pollution and Emergency Medical Care in NYC" with Marc N. Conte and Rachel Wilwerding. Under review.

Air pollution can create inequitable environmental burdens through differences in hazard exposure and vulnerability. Using Emergency Medical Service (EMS) dispatch records from New York City, we find that air pollution significantly increases emergency responses, with larger effects in disadvantaged communities (DACs). Integrating EPA monitors with a denser local network improves measurement of within-city pollution exposure and reveals stronger pollution impacts in DACs than estimates based on EPA data alone. Under everyday pollution levels, disadvantaged communities show limited vulnerability to certain respiratory conditions, consistent with the use of accessible defensive technologies that reduce reliance on emergency care. During the 2023 Canadian wildfire smoke episode—an extreme event with similar outdoor exposure across neighborhoods—structural constraints reduced the effectiveness of these defenses, revealing greater vulnerability in DACs. These findings highlight the importance of improved monitoring and policies that reduce structural barriers to protection.

"When IV Does Not Fix Measurement Error: Evidence from Satellite PM2.5" with Zhenxuan Wang.  Draft available upon request.

Instrumental variables are often used to address measurement error in air pollution exposure. This paper shows why that logic can fail when thermal inversion instruments are paired with satellite-based PM2.5. IV recovers the effect of true pollution only if proxy error is uncorrelated with the instrument. Using Chinese monitor, satellite, reanalysis, and meteorological data, we show that satellite PM2.5 error is strongly predicted by thermal inversions. In a crime application, satellite-based proxies materially change IV estimates despite strong first stages. The key diagnostic is whether proxy error is orthogonal to the identifying variation.

"When Flights Land Elsewhere: Airport Network Spillovers and Local Health" with Zhanhan Yu. Draft available upon request.

This paper studies how disruptions in an airport network reallocate local environmental and health burdens across neighborhoods. Flight diversions are a useful setting because they change not only when aircraft arrive, but where they arrive. Combining flight-level diversion records with ZCTA-day EMS dispatch data in New York City, we show that neighborhoods more exposed to received diverted arrivals experience increases in EMS demand, especially near airports and during late-night periods. We then use the 2016 relaxation of slot controls at EWR to show that policy changes at one airport can shift operational burdens through the network toward communities near other airports.

Work in progress

"Unintended Digital Precaution in Extreme Weather" with Tin Cheuk Leung [slides]

Resting papers

"Estimating the Impact of Climate Change: An Exploration of the Bin Regression Model" with Richard T. Carson and Dalia Ghanem

The “bin” regression model has been put forward as a flexible semi-parametric method for representing a climate variable and it has emerged as the workhorse approach for empirical work (e.g., Deschênes and Greenstone, 2011). Our paper is the first to formally explore econometric properties of the bin regression approach. We show that, although the bin regression approach often produces reasonable results, that the approach produces consistent parameter estimates only under very stringent and highly unlikely assumptions about the true data generating procedure. Problems with the bin regression approach are likely to be most severe in the tail bin categories, where most policy interest with respect to climate change impacts lies. We propose alternatives to bin model for the climate change impacts that produce consistent estimates and generally have better efficiency properties.