Working Papers
Abstract: I estimate the effect of Medicaid coverage of GLP-1 medications for obesity, such as Wegovy, on total and category-level household grocery spending and nutrient purchases. Using household level grocery spending data from Nielsen IQ linked to state level Medicaid prescription records, I exploit variation in the timing of state Medicaid adoption of obesity indicated GLP-1 medications to compare obese and non-obese Medicaid households before and after coverage adoption. I find that coverage adoption reduces per-person quarterly grocery spending among obese Medicaid households by approximately $49, or about 12% of the sample mean, relative to non-obese Medicaid households. This intent-to-treat estimate reflects the policy-level effect across all obese Medicaid households. These reductions in spending is concentrated in calorie dense categories, namely sweet snacks, ice cream, candy and salty snacks. I find little evidence of offsetting increases in fresh produce or other healthier alternatives. Nutrient level estimates show corresponding declines in calories, fat, sugar, and saturated fat purchased. Event study estimates show no differential pre-adoption trends and effects that grow over time. The results show that Medicaid coverage decisions affect household food purchasing behavior among low income populations, a behavioral margin relevant for evaluating the broader returns to public coverage of obesity treatment.
Abstract: China’s 2017 National Sword cut allowable contamination in imported recyclables to 0.5%. We study the U.S. recycling sector’s response in trade, processing, dumping, and prices. We build a two-country open circular-economy model where households generate waste, and waste managers either recycle or dump domestic and imported waste to supply a recycled input used with labor and capital. Calibrated to U.S. plastic-waste data, the model quantifies changes in recycling rates, trade volumes, and allocations under alternative policies.