Venue: The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27708-0120
Presentation
Reduced Smoking and Rising Obesity: Does Workplace Smoking Ban Matter?
Medical research suggests that smokers tend to weigh less than non-smokers (Gorden et. al., 1975). However, it is unclear whether declined smoking rate has contributed to rising prevalence of obesity in the U.S. Smoking may be an endogenous decision since overweight people may initiate smoking to lose weight (Cawley et. al., 2004). Scholars find mixed results using different instrumental variables (IV) to control for endogeneity. Chou et al. (2004) found that lower smoking rates that were due to higher cigarette prices increased obesity; while Gruber and Frakes (2006) came to an opposite conclusion by using cigarette taxes as the instrument. Because of relatively little variations within-state overtime, cigarette prices and taxes are highly collinear with other state level variables. Such serious multi-collinearity may lead to unreliable results. To overcome this problem we employ individual variations in worksite smoking restrictions as instruments for the probability of smoking to estimate effects of smoking on obesity.
The data source is the Behavioral Risk Factor Surveillance System (BRFSS) for the years 1998-2005. We estimate the effects of smoking on Body Mass Index (BMI) and obesity using two-stage least square (2SLS) technique with state and year fixed effects. The IVs include whether smoking is allowed in working or public areas at workplace. We compare current smokers to never smokers and former smokers. We further conduct our analysis among sub-groups by gender, race, and education levels to test any differential effects.
Results show the BMI of current smokers is 1.2 and 3.1 lower than never and former smokers, respectively. Smokers are 0.06 and 0.14 percentage points less likely to be obese than never and former smokers, respectively. Effects are higher for females than for males.
Our study provides evidence that reduced smoking may lead to rising obesity. The central implication of this study is, while tobacco control policies such as worksite smoking bans seem effective at achieving the goals of reducing smoking, they may have significant unintended consequence of increasing obesity.