Epsilon Archive for Student Projects

Abstract

While the tobacco epidemic is a worldwide problem continuing to kill millions of people each year, evidence on the behavioral impact of pictorial health warnings remains limited. This study uses the synthetic control method to evaluate the effect Australia’s 2006 implementation of pictorial health warnings had on smoking prevalence. A synthetic Australia is constructed, which is a weighted combination of 15 OECD countries that did not implement pictorial health warnings during the period of the study. To assess the impact of the policy, smoking rates in Australia are then compared with smoking rates in its synthetic control, over the time period 1980-2012. The results points at a five percent decrease in smoking prevalence over a seven year long post-intervention period, although placebo tests show limited robustness of this finding. Even though the independent policy implications of my findings are limited, they align with previous work evaluating the same policy. Hence, when considered alongside earlier research, this study provides suggestive evidence that the Australian policy did cause smoking prevalence to decrease.