Cutting through the smoke: separating the effect of price on smoking initiation, relapse and cessation

I employ a large national representative dataset (Current Population Survey-Tobacco Use Supplements) to investigate how cigarette prices affect smoking decisions. A standard econometric approach is to estimate the relationship between cigarette prices and smoking participation at a point in time. I extend this approach to model past-year decisions to start, resume or quit smoking. Considering reverse causality, I apply an instrumental variable (excise taxes) for cigarette prices. I include an index of state-level anti-smoking sentiment to control for omitted variable bias. After estimating separate models for smoking initiation, relapse and cessation and for different age groups, I find no evidence that increasing taxes on cigarettes can prevent the onset of youth smoking. Neither does it effectively induce young smokers to quit. However, cigarette prices do play an important role to prevent relapse and encourage quitting at older ages.