Scholarlink

<p>This study reassesses the efficacy of cigarette taxation in curtailing smoking by leveraging recent advancements in the difference-in-differences (DiD) literature to account for heterogeneous treatment effects. Using data from the Behavioral Risk Factor Surveillance System Selected Metropolitan/Micropolitan Area Risk Trend (BRFSS SMART) for the sample periods 2004-2010 and 2015-2020, the study reveals three key findings. Firstly, the TWFE estimate for the 2004-2010 sample is only 48% of the average treatment effect on the treated (ATT) estimate obtained through the DiD framework. Secondly, event-study-type estimates demonstrate a gradual increase in magnitude following the treatment year, highlighting dynamic treatment effects overlooked by the TWFE estimate. Third, the ATT estimate for the 2015-2020 sample is approximately 66% of the ATT estimate for the 2004-2010 sample. Overall, the study underscores the potential bias toward zero in elasticity estimates when relying solely on TWFE models.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Revisiting the effects of cigarette taxation on smoking outcomes

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2024年12月31日

查看全文
  • Vinish Shrestha

摘要

This study reassesses the efficacy of cigarette taxation in curtailing smoking by leveraging recent advancements in the difference-in-differences (DiD) literature to account for heterogeneous treatment effects. Using data from the Behavioral Risk Factor Surveillance System Selected Metropolitan/Micropolitan Area Risk Trend (BRFSS SMART) for the sample periods 2004-2010 and 2015-2020, the study reveals three key findings. Firstly, the TWFE estimate for the 2004-2010 sample is only 48% of the average treatment effect on the treated (ATT) estimate obtained through the DiD framework. Secondly, event-study-type estimates demonstrate a gradual increase in magnitude following the treatment year, highlighting dynamic treatment effects overlooked by the TWFE estimate. Third, the ATT estimate for the 2015-2020 sample is approximately 66% of the ATT estimate for the 2004-2010 sample. Overall, the study underscores the potential bias toward zero in elasticity estimates when relying solely on TWFE models.