The Swedish Outdoor Smoking Ban of 2019: An Application of Synthetic Control and Matrix Completion Methods in Policy Evaluation
Sammanfattning: Smoking bans constitute an important part of the tobacco control policies used globally to improve public health. In July 2019, Sweden implemented new regulation which banned smoking in several outdoor areas nationwide, including outdoor serving areas of e.g. restaurants and bars, whose scope is unique in an international context. This paper aims to estimate the short-term impact of this ban on cigarette consumption using retail scanner data in a cross-country context. Employing Synthetic Control methods as well as the recently proposed Matrix Completion with Nuclear Norm Minimization estimator we find no evidence for the existence of an effect of the ban, nor do we find any visual signs of trend breaks in related outcomes such as sales of a major substitute tobacco product or proxies for quit attempts. Our findings contribute to the literature on tobacco control measures by providing estimates based on sales data instead of self-reports, and to the policy evaluation literature in general by providing an example of how recent developments in methodological tools can be used in application. Finally, we provide decision makers deliberating on whether to follow Sweden's example in legislation with scientific evidence on the topic.
HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)