Evaluating the Role of Income on the Impact of Fat Taxes

To tackle rising obesity rates, countries such as Mexico (2014), USA (2014), France (2012), Denmark (2011), and Finland (2011) have introduced a tax on snacks with high sugar or fat content. 2 These are known as “fat taxes”. As other countries, such as UK, Ireland, Romania and Italy, debate the introduction of “fat taxes”, there have been calls for additional empirical research to guide policy. Policy makers are concerned by the impact of “fat taxes” on consumers from the low-income group as they are more prone to obesity (Ford and Dzewaltowski 2008). Previous studies which have examined the impact of “fat taxes” have certain limitations. Sharma et al. (2014) use descriptive models which suffer from the Lucas critique. Jensen and Smed (2013) use data before and after a “fat tax” was implemented. But the authors caution that their short observation window after the tax makes the long term impact of the tax difficult to judge. Khan, Misra and Singh (2015) use a static structural model where consumers do not stockpile. However, studies show consumers stockpile frequently purchased goods, and find that static models overestimate price sensitivity when consumers stockpile (Hendel and Nevo 2006). Clinical studies suggest that a demand model to study the impact of a tax on consumption of sugar-based snacks needs to incorporate a reinforcement effect where past consumption can increase current consumption (Avena and Gold 2011).3 Reinforcement effect can be modelled using the rational addiction theory (Becker and Murphy 1988). Hence, to evaluate the role of income on the impact of “fat taxes”, we build a dynamic structural model of rational addiction with endogenous consumption and stockpiling. We extend Gordon and Sun’s (2015) dynamic