The effect of smoking habit changes on body weight: Evidence from the UK

- We estimate the causal effect of quitting or reducing smoking on BMI.
- We use a DID strategy to account for time-invariant unobservables.
- We account for endogeneity by using different control groups.
- We provide estimates at different quantiles of the BMI distribution.

This paper evaluates the causal relationship between smoking and body weight through two waves (2004-2006) of the British Household Panel Survey. We model the effect of changes in smoking habits, such as quitting or reducing, and account for the heterogeneous responses of individuals located at different points of the body mass distribution by quantile regression. We test our results by means of a large set of control groups and investigate their robustness by using the changes-in-changes estimator and accounting for different thresholds to define smoking reductions. Our results reveal the positive effect of quitting smoking on weight changes, which is also found to increase in the highest quantiles, whereas the decision to reduce smoking does not affect body weight.