New Evidence on Whether Cigarette Taxes Reduce Youth Smoking

Andrew Sfekas, Northwestern University

We present new evidence on the extent to which cigarette prices and taxes affect a youth’s decision to start smoking. We use longitudinal data from the Tobacco-Use Supplements to the CPS, PSID, and NLSY97 to show that prices and taxes matter. We also resolve a puzzle in the empirical literature. Most studies that use longitudinal data find that the probability of initiation is uncorrelated with changes in taxes. This result contradicts standard economic theory that demand falls when prices increase and it stands in contrast with cross-sectional evidence showing lower smoking prevalence among youth when taxes are higher. We resolve this puzzle by showing that taxes reduce smoking uptake, affects casual smoking much less than regular smoking, and that some of the empirical contradictions stem from basic specification errors flowing from the particular longitudinal data used. The findings have important implications for domestic and international tobacco control and public health policy.