On Balance: Social Welfare Effects of Smoking Bans in Bars and Restaurants

Anne BurtonMarch 6, 2019

By: Anne Burton

Since the 1964 Surgeon General’s report linking smoking cigarettes to adverse health outcomes, numerous federal, state, and local governments have passed regulations designed to reduce the prevalence of smoking and related externalities. Examples of such regulations include cigarette taxes, public health campaigns, minimum purchasing ages for tobacco, and – the focus of this post – smoking bans in bars and restaurants. While the links between smoking and health are clear, the effects of these bans on social welfare, which includes other types of risky behavior as well as smoking, are less well understood. This post describes work-in-progress to address some of the gaps. Preliminary results will be presented at the Annual Conference and Meeting of the Society for Benefit-Cost Analysis in March 2019.

The earliest smoking ban in the United States appeared in 1990. Since then, 35 states and hundreds of municipalities have implemented bans on smoking in bars and restaurants, with the goal of minimizing both exposure to secondhand smoke and the prevalence of smoking.

The findings on the effects of smoking bans on health have been mixed. While the evidence suggests these bans have reduced smoking by pregnant women (Bharadwaj, Johnsen, and Løken 2014), which implies positive health effects for both the pregnant women and their fetuses, bans also induce smokers to smoke more in private places. The increase in private smoking increases the secondhand smoke exposure of other individuals, particularly the children of smokers (Adda and Cornaglia 2010).

Moreover, smoking is a risky health behavior that can be associated with other risky behaviors, such as alcohol consumption. Some researchers have studied the effects of smoking bans on alcohol consumption by different types of smokers (Picone, Sloan, and Trogdon 2004; Koksal and Wohlgenant 2016). Picone and coauthors find that smoking bans reduce alcohol consumption for women. On the other hand, Koksal and Wohlgenant find that smoking bans increase restaurant alcohol consumption while decreasing at-home alcohol consumption. However, these papers only studied the effects of state-level smoking bans, whereas many individuals are affected by county or city-level bans.

Little attention, however, has been paid to the interaction between various risky health behaviors and their externalities. A notable exception is Adams and Cotti (2008), who examine the effect of smoking bans in bars on drunk driving and find that the bans lead to an increase in drunk driving fatalities.

The current work-in-progress extends the literature in two important ways. First, the study includes an empirical analysis of the effects of smoking bans in bars and restaurants on the amount and location of alcohol consumption and on alcohol-related externalities such as drunk driving, domestic violence, and bar fights. Second, the study conducts a cost-benefit analysis of the smoking bans.

The empirical analysis allows for the smoking bans to have heterogeneous effects on the cigarette and alcohol consumption of different smoking and drinking “types.” For example, smoking bans can be expected to make bars and restaurants more attractive to nonsmokers and less attractive to smokers. As a result, nonsmokers may spend more time drinking at bars and smokers may spend less time drinking at bars. The effects of the bans on individuals’ total alcohol consumption will depend on whether drinking at a bar and drinking at home are complements (e.g., through habit formation or addiction) or substitutes.

To uncover these effects, the analysis uses a difference-in-differences method. Essentially, this method compares the amount and location of alcohol consumption before and after the smoking bans are put into effect, between counties that implement smoking bans and counties that do not implement smoking bans. The analysis covers all counties in the United States for which data are available. Data on alcohol consumption by smoking type come from the Behavioral Risk Factor Surveillance System

(BRFSS). The BRFSS is an annual telephone survey conducted across the United States; it is representative at the state level and collects information about the health status and health behaviors of individuals. The Nielsen Consumer Panel provides data on alcohol purchases for home consumption by smoking type. Combining these two datasets makes it possible to estimate the effects on alcohol consumption at bars and restaurants, by different smoking types.

The study also assembles data on alcohol-related externalities from U.S. government sources: the Fatality Analysis Reporting System, the Uniform Crime Reports, and the Supplemental Homicide Reports. The Fatality Analysis Reporting System is a census of fatal motor vehicle incidents on public roadways in the U.S. and includes measures of drunk driving. The Uniform Crime Reports compile crime statistics from local jurisdictions and include measures of homicide, sexual assault, and aggravated assault. The Supplementary Homicide Reports provide additional information about victim and offender characteristics, which can be used to generate measures of domestic violence. Using data for all available U.S. counties, this analysis compares the aforementioned outcomes for counties before and after the smoking bans are implemented, between counties that do and do not implement smoking bans.

The final component of this study is a cost-benefit analysis of policies that ban smoking. Specifically, the objective is to determine how effective these policies are and whether they achieve their desired policy goals without generating too many unintended consequences.

It is well established that smoking generates negative externalities and therefore governments may want to regulate it. In regulating to address the market failure, however, it is important to fully understand the effects of smoking regulations on both smoking and other risky health behaviors, such as alcohol consumption and related externalities. In the best-case scenario, the smoking bans do have their intended effect of reducing smoking without leading to an increase in the alcohol consumption or alcohol-related externalities of any smoking type. In a less-than-ideal scenario, smoking bans may lead to an increase in the alcohol consumption of nonsmokers, particularly at bars, with a corresponding increase in associated alcohol-related externalities. This study aims to figure out which scenario bears out in reality.

Anne Burton is a PhD candidate in Economics at Cornell University. Her research focuses on the social welfare implications of risky health behaviors and crime.

Disclaimer: Researcher(s) own analyses calculated (or derived) based in part on data from The Nielsen Company (US), LLC and marketing databases provided through the Nielsen Datasets at the Kilts Center for Marketing Data Center at The University of Chicago Booth School of Business. The conclusions drawn from the Nielsen data are those of the researcher(s) and do not reflect the views of Nielsen. Nielsen is not responsible for, had no role in, and was not involved in analyzing and preparing the results reported herein.

References

  • Adams S, Cotti C. 2008. “Drunk Driving After the Passage of Smoking Bans in Bars.” J. of Public Economics 92:1288-1305.
  • Adda J, Cornaglia F. 2010. “The Effect of Bans and Taxes on Passive Smoking.” American Economic J.: Applied Economics 2(1):1-32.
  • Bharadwaj P, Johnsen JV, Løken, KV. 2014. “Smoking Bans, Maternal Smoking, and Birth Outcomes.” J. of Public Economics 115:72-93.
  • Koksal A, Wohlgenant MK. 2016. “How Do Smoking Bans in Restaurant Affect Restaurant and At-Home Alcohol Consumption?” Empirical Economics 50:1193-1213.
  • Picone GA, Sloan F, Trogdon JG. 2004. “The Effect of the Tobacco Settlement and Smoking Bans on Alcohol Consumption.” Health Economics 13:1063-1080.