A model of the impact of smoking bans on smoking with evidence from bans in England and Scotland

This paper evaluates the impact of smoking bans on smoking using a policy change introduced by the UK government. We present a theoretical model of smoking that defines an individual’s life-cycle addiction and cigarette consumption in the presence and in the absence of a public smoking ban. The model shows that the imposition of a public smoking ban should affect individuals differently according to their age, gender and previous level of cigarette consumption. We test the predictions of the model using data drawn from the British Household Panel Survey (BHPS). We assess the short-term impact of the UK smoking ban using differences-in-differences (DD) and changes-in-changes (CC) methods. We make use of the longitudinal nature of the data and implement a set of more flexible DD fixed effects panel data models. We find that the introduction of the smoking bans in Scotland and England had no impact on overall smoking prevalence. However, we do find some evidence on the effects of the ban on the levelof cigarette consumption, especially among male heavy smokers, female moderate and heavy smokers and young people.