96
Estimating the Long-run Relationship of State Cigarette Taxes and Life Expectancy
A large body of literature indicates that tobacco control legislation—including fiscal measures such as excise sales taxes—reduces tobacco smoking activity, and that tobacco use confers such high risk of mortality that it is plausibly related to large-scale variations in overall life expectancy. Yet there remains active debate about the regressivity over time of some tobacco control measures, suggesting that the net effect of taxes and other tobacco control measures on life expectancy (incorporating competing risks such as worsened poverty and attendant mortality risks if paying higher taxes and continuing to smoke) should be further quantified. This paper tests the hypothesis that increases in state cigarette excise taxes are positively associated with increases in life expectancy over the long-run. We additionally examine whether the relationship between cigarette taxes and life expectancy varies by sex or income and whether it is mediated by changes to prevalence of smoking and tobacco-related health conditions.
Methods
Our source of life expectancy data is from the Institute for Health Metrics and Evaluation, which provides estimated life expectancy at birth for all counties in the Unites States for each year 1980-2014 broken down by sex. Data on state cigarette excise tax rates by year are from the Centers for Disease Control State System, Tobacco Legislation Tax data. Our main specification is a multilevel dynamic panel regression model with fixed effects for county. We include time-varying county-level controls for yearly population composition, per capita income, and educational attainment. The dynamic model has the benefit of separately estimating short-run and long-run (equilibrium) effects, which is appropriate because life expectancy is not expected to respond immediately, but rather dynamically over time, to changes in cigarette taxes.
Results
The central finding is that for every 1 dollar increase in cigarette tax per pack, life expectancy increases over the long-run by approximately 6 months for men (0.47 years; 95% CI 0.27 to 0.68; p < 0.01) and 4 months for women (0.31 years; 95% CI 0.18 to 0.48; p < 0.01), on average. We find that the benefit of the policy is greater in lower income and rural counties. We provide evidence that the effect is mediated by long-run reductions in smoking (5.1% reduction among men; 95% CI -7.6 to -2.6; p < 0.01; and 2.2% reduction among women; 95% CI -2.7 to -1.7; p < 0.01).
Conclusions
The findings suggest that increasing cigarette excise tax rates may translate to improvements in overall life expectancy, especially among lower-income men in rural areas.