Smoking bans in bars and restaurants are one example of the many ways in which governments intervene to correct market failures. These bans also represent a change in a non-price determinant of demand for alcohol consumed at bars, which could affect total alcohol consumption. This paper studies the effects of smoking bans on the amount and location of alcohol consumption, smoking, and alcohol-related externalities.
Abstract
I use a difference-in-differences method that exploits variation in the effective dates of smoking bans in bars and restaurants across cities, counties, and states. For individuals who drink, smoking bans result in an average increase in alcohol consumption of 1 drink per month, with heterogeneous effects by smoking status. These increases are entirely driven by changes in bar and restaurant alcohol consumption.
In addition, smoking bans may lead some former smokers to become occasional smokers. I find imprecisely estimated null effects of smoking bans on drunk driving fatalities and homicides. Taken together, these results imply that while smoking bans lead to unintended increases in alcohol consumption, there is no evidence to suggest that they lead to corresponding increases in the most socially costly alcohol-related externalities.
- More information
More information on this seminar can be found on VERBseminar.org. Registration is required and can also be done here.
Organisers
- Tinna Laufey Ásgeirsdóttir (University of Iceland)
- Ana Inés Balsa (Universidad de Montevideo)
- John Cawley (Cornell University)
- Hans van Kippersluis (Erasmus University Rotterdam)