Does quitting smoking increase obesity? Evidence that accounts for misreporting

Abstract

Studying the relationship between smoking and obesity, the leading causes of preventable deaths in the U.S., helps assess the potential unintended consequences of policy efforts to reduce smoking. Because existing literature is mixed among studies using experimental and observational data, we investigate the role of misreporting in observational data. We use the Behavioral Risk Factor Surveillance System, cigarette taxes to instrument for smoking, and survey completion to instrument for misreporting. Starting with the seminal two‐stage least squares (2SLS) approach, our estimates similarly suggest quitting smoking substantially reduces body mass index (BMI). However, the relationship between cigarette taxes and BMI has shrunk over time, and the 2SLS estimates are sensitive to specification, functional form, and misreporting. Accounting for misreporting using the 2‐step estimator from Nguimkeu et al. (2019) yields estimates consistent with the experimental literature; quitting smoking modestly increases BMI. Our preferred estimate suggests reduced smoking accounts for 6% of the concurrent rise in obesity.