Does cigarette smoking affect body weight? causal estimates from the clean indoor air law discontinuity

This paper examines the causal effects of smoking behavior on body weight in Italy. In 2005, the Italian government introduced a smoking ban in all indoor public places. We use a regression discontinuity design, which exploits this exogenous variation across cohorts to achieve identification in our model. Our estimates indicate that the smoking ban reduced cigarette consumption and the smoking participation rate. Most interestingly, we estimate a significative, although not very large, effect of nicotine reduction on weight increases. Heterogeneous effects are also estimated, with smaller impact on men and employees and, conditionally on BMI distribution, overweight and obese people