How do FDA policies impact demand for cigarettes and e-cigarettes in the US tobacco market? Predictions from discrete choice experiments
Dr John Buckell, Postdoctoral Associate, Health Policy and Management, Yale School of Public Health
Monday, 10 September 2018, 1.30pm to 2.30pm
1st Floor Main Meeting Room, Richard Doll Building, Old Road Campus, OX3 7LF
Hosted by HERC
Speaker’s Bio: John is a postdoctoral associate at Yale University under the supervision of Professor Jody Sindelar. His work involves using choice experiments to understand the behavior of smokers, with a particular focus on e-cigarettes. John’s previous experience includes working on efficiency and regulation in the English NHS. John has also held secondment and consultancy posts at NHS England and Cambridge Economic Policy Associates, respectively. John works chiefly in the field of health but has experience in regulated industries, namely utilities and transport. His research interests include health econometrics; choice models; econometric estimation of cost and production functions; economic regulation; and policy evaluation.
Abstract: Tobacco control regulations have been in force in the US for centuries, starting with excise taxes on tobacco. Despite these policies, nearly 40 million American adults still smoke and youths still continue to start to smoke. An estimated 480,000 Americans die each year from tobacco-related illness; as such, tobacco is the largest avoidable cause of death in the US. Recently the US tobacco market has changed considerably, particularly with the emergence of e-cigarettes. E-cigarettes have become the second most popular tobacco product behind cigarettes. Evidence suggests that, whilst not harmless, e-cigarettes are much less harmful than cigarettes. Given this, and their popularity, e-cigarettes offer substantial potential to benefit public health as a substitute for the cigarettes, but regulations have to be designed to maximize the net improvement.
In 2009, the Family Smoking Prevention and Tobacco Control Act (FSPTCA) gave the Food and Drug Administration (FDA) regulatory control over cigarettes and some other tobacco products mandating the goal of improving public health. In 2016, the FDA extended its authority to include e-cigarettes amongst other products. The FDA has the authority over the manufacture, distribution and marketing of combustible tobacco products. Specifically, they can ban flavors in tobacco products, reduce levels of nicotine in tobacco products and indirectly reduce health harms of tobacco products. To use these tools effectively, it is necessary to understand how smokers value tobacco products and their features; that is, to understand the demand for these products and their features. Further, understanding how smokers will respond to changes in the features of tobacco products is central to policymaking.
In this research, a series of discrete choice experiments (DCEs) are used to estimate smokers’ demand for cigarettes and e-cigarettes, and how demand varies as a function of key, policy-relevant attributes: flavor availability, levels of health harms and nicotine. In addition, other attributes to characterize the wider regulatory environment are examined: prices, use in public places, e-cigarettes’ effectiveness as a cessation aid. We collect large, representative samples of smokers and recent quitters. Preferences are estimated at the sample level and by key subpopulations. We then use the estimated preferences to predict the impact of a range of policies that the FDA could implement.We examine policies in isolation, but also policy bundles. Information on states’ currently policies is used to replicate real-world conditions when making predictions. Lastly, we consider and discuss future directions for providing tobacco-based policy evidence using DCEs.
We find that smokers’ demand for tobacco products is reduced by flavor bans, higher health harms, low nicotine, whether e-cigarettes are useful as a cessation aid, and higher prices. We estimate smokers’ relative preferences for these attributes, and assess the impact of varying these attributes on choices. Further, we find substantial heterogeneity in smokers’ tobacco product and attribute preferences. This is critical information for the FDA in assessing how policies will impact within the population. Predictions indicate that several of the tools available to the FDA are likely to have a significant impact on smokers’ choices. Further, impacts of the policies are sensitive to their specification. For example, flavor bans in e-cigarettes is likely to impact the market much differently than banning menthol in cigarettes. Finally, predictions indicate that the valence of FDA polices is mediated by otherregulators’ policies. Therefore, this should be considered when setting policy.
In sum, we provide evidence on smokers’ demand for tobacco products, and how possible FDA policies will impact
on the market. With this, the FDA is better-placed to deliver policies to protect the health of the public.