Association for Public Policy Analysis & Management

Smoking is linked with cancer. Recent estimates suggest that smoking accounts for 30% of cancer deaths in the U.S. and that 606,880 Americans will die of cancer in 2019, which implies that smoking will lead to 182,064 cancer deaths in that year. Indeed, medical research shows that tobacco products cause a variety of cancers, including lung, oral cavity, nasal cavity, larynx, oropharynx, hypopharynx, oesophagus, stomach, liver, pancreas, bladder, ureter, kidney, and cervix cancer, and myeloid leukaemia. In particular, the carcinogens contained in tobacco products are responsible for these cancers. Elevated cancer risk attributable to smoking accrues to both the smoker through direct inhalation of tobacco smoke and to non-smokers through secondhand tobacco smoke (SHS). Over the past several decades, all levels of government in the U.S. have engaged in an aggressive anti-smoking campaign that has included, among other activities, increased taxes on tobacco products and prohibition of smoking in public venues such as worksites, restaurants, and bars. The majority of the extant tobacco control research has explored the effect of these policies on smoking outcomes and exposure to SHS, but, surprisingly, no effort has been made to directly examine changes in cancer risk, which will be impacted by changes in where and how cigarettes are smoked (e.g. intensity of smoking) in addition more general decisions about whether to smoke or not and how many cigarettes to smoke. We aim to address this gap in the literature by studying the effects of two common state-level policies – cigarette taxes and bans on use in public venues – on biomarkers of carcinogen exposure. To this end, we use the continuous National Health and Nutrition Examination Survey (NHANES) restricted data from cycles 1999/2000 to 2015/2016 which includes both state of residence and biomarkers of exposure to carcinogens. These biomarkers of carcinogen exposure are measured in NHANES urine and blood samples by trained CDC professionals and are linked with cancer risk in numerous medical studies. We estimate differences-in-differences style models that control for a wide range of individual- and state-level variables, and leverage variation in state cigarette taxes and public use bans. As such, we provide the first causal evidence on the effects of state-level tobacco control policies on biomarkers of carcinogen exposure among smokers and non-smokers. Over our study period, numerous states increased, and in some cases decreased, cigarette taxes and the restrictiveness of public use bans. These policy changes offer us substantial variation with which to identify treatment effects. In extensions, we separately examine policy effects in sub-populations at elevated risk of smoking and cancer. Our results show that, following increases in both cigarette taxes and public use bans, biomarkers of carcinogen exposure decline among smokers and non-smokers. We find evidence of heterogeneity across policy lever and sub-group. These findings are robust to a range of sensitivity analyses.Our findings, for the first time, document using biomarker data that standard tobacco control policies, in addition to reducing smoking and SHS, reduce carcinogen exposure (and therefore possibly lower cancer risk), among smokers and non-smokers.