Three empirical studies on environmental hazards, public health, and policy responses

Abstract

The first essay examines the relationship between wildfire-driven ambient particulate matter (wildfire-PM) and health insurance enrollment in the United States. I leverage temporal and geographic variations in wildfire-PM to estimate their causaimpact on health insurance enrollment. The findings show that exposure to wildfire-PM significantly increases health insurance enrollment and this is largely driven by increases in Medicaid enrollment. I also show evidence that Medicaid enrollment can mitigate the mortality effects of wildfire-PM exposure. The results of this essay indicate that public health insurance plays a crucial role in buffering vulnerable individuals against health risks associated with environmental hazards like wildfire-PM. The second essay explores the relationship between artisanal small-scale mining (ASM) in Ghana and malaria transmission. In Ghana, ASM tends to be a highly transitory and illicit activity that overwhelms mining lands with uncovered excavation pits. Once these pits are filled with stagnant water, they turn into ideal breeding sites for mosquitoes which are the primary vectors for malaria transmission. Using a difference-in-differences approach, this study finds that proximity to newly established ASMs significantly increases malaria incidence among children under five years old. This relationship is robust across different methodological specifications and highlights the need for targeted interventions to mitigate disease transmission in mining-adjacent communities. The third essay shifts focus to a global scale, examining how countries responded to the the COVID-19 pandemic. Using nighttime lights (NTLs) satellite imagery, this study categorizes 170 countries into four adaptation pathways based on how NTLs change over time. The findings indicate that, as expected, countries implementing stringent lockdowns experienced steep declines in nighttime activity. Interestingly, those that rapidly adopted health policies such as contact tracing and testing exhibited faster recovery of their NTLs compared to those that relyied primarily on lockdowns. This research provides insights into how remotely sensed data can be used to supplement efforts to monitor impact and recovery at a fine temporal and spatial scale.