Marijuana Legalization and Opioid Mortality: A Causal Analysis Using DiD and IV Models
Econometric study of the effects of recreational cannabis legalization on opioid and cocaine mortality, using Difference-in-Differences, IV, and Event Studies.
This project examines the impact of recreational marijuana legalization on drug-related mortality, with a focus on opioids and cocaine.
The analysis applies causal inference methods including Difference-in-Differences (DiD), Instrumental Variables (IV, 2SLS & 2SRI), and Event Studies, using state-level panel data from the CDC.
Methods
- Panel dataset: Massachusetts (treatment) vs. New Hampshire (control)
- Data cleaning and imputation via Kalman filtering
- DiD and Event Study with monthly leads/lags
- IV estimation to address potential endogeneity
- Binary Logit/Probit robustness checks
Key results
- Significant decline in cocaine deaths after legalization (DiD & Event Study)
- No statistically significant effect on opioid mortality
- Placebo and robustness checks confirm validity of cocaine result
- Findings highlight implications for European countries facing emerging overdose risks
Causal econometric analysis of cannabis policy reform and drug-related mortality